Chemistry of Zeolites and Related Porous Materials: Synthesis and Structure

Chemistry of Zeolites and Related Porous Materials: Synthesis and Structure RUREN XU Jilin University, China WENQIN PANG Jilin University, China JIHONG YU Jilin University, China QISHENG HUO Pacific Northwest National Laboratory, USA JIESHENG CHEN Jilin University, China

John Wiley & Sons (Asia) Pte Ltd

Chemistry of Zeolites and Related Porous Materials

Chemistry of Zeolites and Related Porous Materials: Synthesis and Structure RUREN XU Jilin University, China WENQIN PANG Jilin University, China JIHONG YU Jilin University, China QISHENG HUO Pacific Northwest National Laboratory, USA JIESHENG CHEN Jilin University, China

John Wiley & Sons (Asia) Pte Ltd

Copyright # 2007

John Wiley & Sons (Asia) Pte Ltd 2 Clementi Loop #02-01, Singapore 129809

Visit our Home Page on www.wiley.com All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, scanning, or otherwise, except as expressly permitted by law, without either the prior written permission of the Publisher, or authorization through payment of the appropriate photocopy fee to the Copyright Clearance Center. Requests for permission should be addressed to the Publisher, John Wiley & Sons (Asia) Pte Ltd, 2 Clementi Loop, #02-01, Singapore 129809, tel: 65-64632400, fax: 65-64646912, email: [email protected]. Designations used by companies to distinguish their products are often claimed as trademarks. All brand names and product names used in this book are trade names, service marks, trademarks or registered trademarks of their respective owners. The Publisher is not associated with any product or vendor mentioned in this book. This publication is designed to provide accurate and authoritative information in regard to the subject matter covered. It is sold on the understanding that the Publisher is not engaged in rendering professional services. If professional advice or other expert assistance is required, the services of a competent professional should be sought. Other Wiley Editorial Offices John Wiley & Sons, Ltd, The Atrium, Southern Gate, Chichester, West Sussex PO19 8SQ, England John Wiley & Sons Inc., 111 River Street, Hoboken, NJ 07030, USA Jossey-Bass, 989 Market Street, San Francisco, CA 94103-1741, USA Wiley-VCH Verlag GmbH, Boschstr. 12, D-69469 Weinheim, Germany John Wiley & Sons Australia Ltd, 42 McDougall Street, Milton, Queensland 4064, Australia John Wiley & Sons Canada Ltd, 6045 Freemont Blvd, Mississauga, ONT, L5R 4J3, Canada Wiley also publishes its books in a variety of electronic formats. Some content that appears in print may not be available in electronic books. Anniversary Logo Design: Richard J. Pacifico Library of Congress Cataloging-in-Publication Data Chemistry of zeolites and related porous materials synthesis and structure / Ruren Xu ... [et al]. p. cm. ISBN 978-0-470-82233-3 (cloth) 1. Zeolites. 2. Porosity–Congresses. I. Xu, Ruren. TP245.S5C52 2007 2007015329 6660 .86–dc22

ISBN 978-0-470-82233-3 (HB) Typeset in 10/12 pt. Times by Thomson Digital, India Printed and bound in Singapore by Markono Print Media Pte Ltd, Singapore. This book is printed on acid-free paper responsibly manufactured from sustainable forestry in which at least two trees are planted for each one used for paper production.

Contents Preface

xi

1. Introduction 1.1 The Evolution and Development of Porous Materials 1.1.1 From Natural Zeolites to Synthesized Zeolites 1.1.2 From Low-silica to High-silica Zeolites 1.1.3 From Zeolites to Aluminophosphate Molecular Sieves and Other Microporous Phosphates 1.1.4 From 12-Membered-ring Micropores to Extra-large Micropores 1.1.5 From Extra-large Micropores to Mesopores 1.1.6 Emergence of Macroporous Materials 1.1.7 From Inorganic Porous Frameworks to Porous Metal-organic Frameworks (MOFs) 1.2 Main Applications and Prospects 1.2.1 The Traditional Fields of Application and Prospects of Microporous Molecular Sieves 1.2.2 Prospects in the Application Fields of Novel, High-tech, and Advanced Materials 1.2.3 The Main Application Fields and Prospects for Mesoporous Materials 1.3 The Development of Chemistry for Molecular Sieves and Porous Materials 1.3.1 The Development from Synthesis Chemistry to Molecular Engineering of Porous Materials 1.3.2 Developments in the Catalysis Study of Porous Materials

1 2 2 3

2. Structural Chemistry of Microporous Materials 2.1 Introduction 2.2 Structural Building Units of Zeolites 2.2.1 Primary Building Units 2.2.2 Secondary Building Units (SBUs) 2.2.3 Characteristic Cage-building Units

4 5 6 7 8 9 9 10 11 13 13 14 19 19 23 23 24 25

vi

Contents

2.3

2.4

2.5

2.6

2.2.4 Characteristic Chain- and Layer-building Units 2.2.5 Periodic Building Units (PBUs) Composition of Zeolites 2.3.1 Framework Composition 2.3.2 Distribution and Position of Cations in the Structure 2.3.3 Organic Templates Framework Structures of Zeolites 2.4.1 Loop Configuration and Coordination Sequences 2.4.2 Ring Number of Pore Opening and Channel Dimension in Zeolites 2.4.3 Framework Densities (FDs) 2.4.4 Selected Zeolite Framework Structures Zeolitic Open-framework Structures 2.5.1 Anionic Framework Aluminophosphates with Al/P  1 2.5.2 Open-framework Gallophosphates with Extra-large Pores 2.5.3 Indium Phosphates with Extra-large Pores and Chiral Open Frameworks 2.5.4 Zinc Phosphates with Extra-large Pores and Chiral Open Frameworks 2.5.5 Iron and Nickel Phosphates with Extra-large Pores 2.5.6 Vanadium Phosphates with Extra-large Pores and Chiral Open Frameworks 2.5.7 Germanates with Extra-large Pores 2.5.8 Indium Sulfides with Extra-large-pore Open Frameworks Summary

3. Synthetic Chemistry of Microporous Compounds (I) – Fundamentals and Synthetic Routes 3.1 Introduction to Hydro(solvo)thermal Synthesis 3.1.1 Features of Hydro(solvo)thermal Synthetic Reactions 3.1.2 Basic Types of Hydro(solvo)thermal Reactions 3.1.3 Properties of Reaction Media 3.1.4 Hydro(solvo)thermal Synthesis Techniques 3.1.5 Survey of the Applications of Hydro(solvo)thermal Synthetic Routes in the Synthesis of Microporous Crystals and the Preparation of Porous Materials 3.2 Synthetic Approaches and Basic Synthetic Laws for Microporous Compounds 3.2.1 Hydrothermal Synthesis Approach to Zeolites 3.2.2 Solvothermal Synthesis Approach to Aluminophosphates 3.2.3 Crystallization of Zeolites under Microwave Irradiation 3.2.4 Hydrothermal Synthesis Approach in the Presence of Fluoride Source 3.2.5 Special Synthesis Approaches and Recent Progress 3.2.6 Application of Combinatorial Synthesis Approach and Technology in the Preparation of Microporous Compounds

29 32 33 33 34 39 41 41 43 47 47 72 72 88 92 93 95 97 100 101 104

117 117 117 119 120 122

123 123 124 144 157 161 164 168

Contents

3.3

Typical Synthetic Procedures for some Important Molecular Sieves 3.3.1 Linde Type A (LTA) 3.3.2 Faujasite (FAU) 3.3.3 Mordenite (MOR) 3.3.4 ZSM-5 (MFI) 3.3.5 Zeolite Beta (BEA) 3.3.6 Linde Type L (LTL) 3.3.7 AlPO4-5 (AFI) 3.3.8 AlPO4-11 (AEL) 3.3.9 SAPO-31 3.3.10 SAPO-34 (CHA) 3.3.11 TS-1 (Ti-ZSM-5)

4. Synthetic Chemistry of Microporous Compounds (II) – Special Compositions, Structures, and Morphologies 4.1 Synthetic Chemistry of Microporous Compounds with Special Compositions and Structures 4.1.1 M(III)X(V)O4-type Microporous Compounds 4.1.2 Microporous Transition Metal Phosphates 4.1.3 Microporous Aluminoborates 4.1.4 Microporous Sulfides, Chlorides, and Nitrides 4.1.5 Extra-large Microporous Compounds 4.1.6 Zeolite-like Molecular Sieves with Intersecting (or Interconnected) Channels 4.1.7 Pillared Layered Microporous Materials 4.1.8 Microporous Chiral Catalytic Materials 4.2 Synthetic Chemistry of Microporous Compounds with Special Morphologies 4.2.1 Single Crystals and Perfect Crystals 4.2.2 Nanocrystals and Ultrafine Particles 4.2.3 The Preparation of Zeolite Membranes and Coatings 4.2.4 Synthesis of Microporous Material with Special Aggregation Morphology in the Presence of Templates 4.2.5 Applications of Zeolite Membranes and Films 5. Crystallization of Microporous Compounds 5.1 Starting Materials of Zeolite Crystallization 5.1.1 Structures and Preparation Methods for Commonly Used Silicon Sources 5.1.2 Structure of Commonly Used Aluminum Sources 5.2 Crystallization Process and Formation Mechanism of Zeolites 5.2.1 Solid Hydrogel Transformation Mechanism 5.2.2 Solution-mediated Transport Mechanism 5.2.3 Important Issues Related to the Solution-mediated Transport Mechanism 5.2.4 Dual-phase Transition Mechanism

vii

172 172 173 175 176 177 178 178 179 180 181 181

191 192 192 194 197 199 201 212 215 218 226 226 235 241 248 251 267 268 268 284 285 287 289 294 305

viii

Contents

5.3

5.4

Structure-directing Effect (SDE) and Templating in the Crystallization Process of Microporous Compounds 5.3.1 Roles of Guest Molecules (Ions) in the Creation of Pores 5.3.2 Studies on the Interaction between Inorganic Host and Guest Molecules via Molecular Simulation 5.3.3 Conclusions and Prospects Crystallization Kinetics of Zeolites

6. Preparation, Secondary Synthesis, and Modification of Zeolites 6.1 Preparation of Zeolites – Detemplating of Microporous Compounds 6.1.1 High-temperature Calcination 6.1.2 Chemical Detemplating 6.1.3 Solvent-extraction Method 6.2 Outline of Secondary Synthesis 6.3 Cation-exchange and Modification of Zeolites 6.3.1 Ion-exchange Modification of Zeolite LTA 6.3.2 Modification of FAU Zeolite through Ion-exchange 6.4 Modification of Zeolites through Dealumination 6.4.1 Dealumination Routes and Methods for Zeolites 6.4.2 High-temperature Dealumination and Ultra-stabilization 6.4.3 Chemical Dealumination and Silicon Enrichment of Zeolites 6.5 Isomorphous Substitution of Heteroatoms in Zeolite Frameworks 6.5.1 Galliation of Zeolites – Liquid–Solid Isomorphous Substitution 6.5.2 Secondary Synthesis of Titanium-containing Zeolites – Gas–Solid Isomorphous Substitution Technique 6.5.3 Demetallation of Heteroatom Zeolites through High-temperature Vapor-phase Treatment 6.6 Channel and Surface Modification of Zeolites 6.6.1 Cation-exchange Method 6.6.2 Channel-modification Method 6.6.3 External Surface-modification Method 7. Towards Rational Design and Synthesis of Inorganic Microporous Materials 7.1 Introduction 7.2 Structure-prediction Methods for Inorganic Microporous Crystals 7.2.1 Determination of 4-Connected Framework Crystal Structures by Simulated Annealing Method 7.2.2 Generation of 3-D Frameworks by Assembly of 2-D Nets 7.2.3 Automated Assembly of Secondary Building Units (AASBU Method) 7.2.4 Prediction of Open-framework Aluminophosphate Structures by using the AASBU Method with Lowenstein’s Constraints 7.2.5 Design of Zeolite Frameworks with Defined Pore Geometry through Constrained Assembly of Atoms

307 307 324 325 326 345 345 345 347 348 350 351 351 357 361 361 362 364 373 374 377 378 379 380 381 383

397 397 398 399 401 406 412 415

Contents

7.3

7.4

7.2.6 Design of 2-D 3.4-Connected Layered Aluminophosphates with Al3P4O163 Stoichiometry 7.2.7 Hypothetical Zeolite Databases Towards Rational Synthesis of Inorganic Microporous Materials 7.3.1 Data Mining-aided Synthetic Approach 7.3.2 Template-directed Synthetic Approach 7.3.3 Rational Synthesis through Combinatorial Synthetic Route 7.3.4 Building-block Built-up Synthetic Route Prospects

8. Synthesis, Structure, and Characterization of Mesoporous Materials 8.1 Introduction 8.2 Synthesis Characteristics and Formation Mechanism of Ordered Mesoporous Materials 8.2.1 Mesostructure Assembly System: Interaction Mechanisms between Organics and Inorganics 8.2.2 Formation Mechanism of Mesostructure: Liquid-crystal Template and Cooperative Self-assembly 8.2.3 Surfactant Effective Packing Parameter: g and Physical Chemistry of Assembly and Interface Considerations 8.3 Mesoporous Silica: Structure and Synthesis 8.3.1 Structural Characteristics and Characterization Techniques for Mesoporous Silica 8.3.2 2-D Hexagonal Structure: MCM-41, SBA-15, and SBA-3 8.3.3 Cubic Channel Mesostructures: MCM-48, FDU-5, and Im3m Materials 8.3.4 Caged Mesostructures 8.3.5 Deformed Mesophases, Low-order Mesostructures, and Other Possible Mesophases 8.3.6 Phase Transformation and Control 8.4 Pore Control 8.4.1 Pore-size and Window-size Control 8.4.2 Macroporous Material Templating Synthesis 8.4.3 The Synthesis of Hierarchical Porous Silica Materials 8.5 Synthesis Strategies 8.5.1 Synthesis Methods 8.5.2 Surfactant, its Effect on Product Structure and Removal from Solid Product, and Nonsurfactants template 8.5.3 Stabilization of Silica Mesophases and Post-synthesis Hydrothermal Treatment 8.5.4 Zeolite Seed as Precursor and Nanocasting with Mesoporous Inorganic Solids 8.5.5 Synthesis Parameters and Extreme Synthesis Conditions 8.6 Composition Extension of Mesoporous Materials 8.6.1 Chemical Modification 8.6.2 Synthesis Challenges for Nonsilica Mesoporous Materials

ix

426 429 430 430 433 454 455 459 467 468 472 472 478 489 494 494 497 505 508 520 525 526 526 529 531 533 533 535 541 547 550 558 558 561

x

Contents

8.7

8.8

8.6.3 Metal-containing Mesoporous Silica-based Materials 8.6.4 Inorganic–Organic Hybrid Materials 8.6.5 Metal Oxides, Phosphates, Semiconductors, Carbons, and Metallic Mesoporous Materials Morphology and Macroscopic Form of Mesoporous Material 8.7.1 ‘Single Crystal’ and Morphologies of Mesoporous Silicas 8.7.2 Macroscopic Forms Possible Applications, Challenges, and Outlook 8.8.1 Possible Applications 8.8.2 Challenges and Outlook

562 563 565 572 573 575 583 583 584

9. Porous Host–Guest Advanced Materials 9.1 Metal Clusters in Zeolites 9.1.1 Definition of Metal Clusters 9.1.2 Preparation Approaches to Metal Clusters 9.1.3 Alkali Metal Clusters 9.1.4 Metal Clusters of Silver 9.1.5 Noble Metal (Platinum, Palladium, Rhodium, Ruthenium, Iridium, Osmium) Clusters 9.1.6 Other Metal Clusters 9.1.7 Clusters of Metal Oxides or Oxyhydroxide 9.2 Dyes in Zeolites 9.3 Polymers and Carbon Materials in Zeolites 9.3.1 Polymers in Zeolites 9.3.2 Preparation of Porous Carbon using Zeolites 9.3.3 Fullerenes Assembled in Zeolites 9.3.4 Carbon Nanotube Growth in Zeolites 9.4 Semiconductor Nanoparticles in Zeolites 9.5 Metal Complexes in Molecular Sieves 9.5.1 Incorporation of Metal–Pyridine Ligand Complexes 9.5.2 Incorporation of Metal–Schiff Base Complexes 9.5.3 Incorporation of Porphyrin and Phthalocyanine Complexes 9.5.4 Incorporation of Other Metal Complexes 9.6 Metal–Organic Porous Coordination Polymers 9.6.1 Transition Metal–Multicarboxylate Coordination Polymers 9.6.2 Coordination Polymers with N-containing Multidentate Aromatic Ligands 9.6.3 Coordination Polymers with N- and O-containing Multidentate Ligands 9.6.4 Zinc-containing Porous Coordination Polymers 9.6.5 Adsorption Properties and H2 Storage of MOFs

603 604 604 605 607 612

Further Reading

667

Index

673

613 614 615 616 621 621 623 624 625 631 636 636 640 642 644 647 647 648 650 651 652

Preface Our book ‘Zeolite Molecular Sieves: Structure and Synthesis’ (in Chinese) was first published in 1987. Substantial progress has been made in these 19 years in developing new molecular sieves with microporous structures such as zeolite and aluminophosphate molecular sieves and many new families of molecular sieves with much diversified structural features and compositional elements. Up until 2006, at least 167 types of molecular sieves with unique framework structures had been reported. More then 30 compositional elements have been incorporated into the frameworks. In 1992, scientists at Mobil Corporation for the first time reported the development of a new family of materials (named M41S) characterized by their unique mesoporous structures (diameter ranging from 2 to 50 nm), which instantly became headline news in science. This new discovery has clearly marked a major milestone in this field, opening the door for developing many new types of molecular sieves and porous materials. In 1998, Wijnhoven and Vos reported the successful synthesis of macroporous material TiO2. Since then a number of other new macroporous materials (diameter ranging from 50 to 2000 nm) such as SiO2, ZrO2, etc., have been synthesized. Parallel to these developments is the emergence of another research area focused on development of porous coordination polymers and hybrid solids with metal–organic frameworks (MOFs). The advent of this family of MOFs has substantially expanded the pool of porous materials that traditionally have their frameworks made of inorganic elements. In addition, the MOF materials with their unique structural and functional characteristics have greatly diversified the existing porous materials. Clearly, the rapid development of microporous compounds and the advent of mesoporous, macroporous, and MOF materials have expanded the already rich and complex molecular sieves and porous materials chemistry, leading to the emergence of a brand new scientific discipline namely the porous materials chemistry. Thanks to these new developments and the progress in related theoretical studies, research methodology, and techniques, as well as the expansion in the scope of applications from the traditional areas such as adsorption separation, catalysis and ion-exchange to the making of new and more advanced materials, our understanding about the governing principles and mechanisms and the observations made about molecular sieves and porous material chemistry has improved significantly in the past decade; in particular, our understanding about the relationships of ‘function–structure–synthesis’ of zeolites and

xii

Preface

other porous materials has reached a new level. The idea of this book was conceived and carefully planned in this general context, to which we give a new name ‘Chemistry of Zeolites and Related Porous Materials - Synthesis and Structure’. This book will be published in English by John Wiley & Sons, (Asia) Pte Ltd by the time of the 15th International Zeolite Conference (Beijing, 2007). The present book consists of nine chapters, with the synthetic and structural chemistry of microporous and mesoporous materials as the core. Five chapters (Chapters 3, 4, 5, 6, and 8) are allocated to cover the synthetic aspects of the topic. Chapter 3 introduces the synthesis and related fundamental principles, synthetic strategies, and techniques for the major microporous materials such as zeolites and microporous aluminophosphates. This Chapter serves as Part I of the synthetic aspects of the microporous compounds. A large number of new microporous materials have emerged in the past decade, with (a) specially interesting structures such as extra-large microporous channels, interconnecting 2- and 3-dimensional channel systems, chiral channels, and various cage structures, (b) special types such as the M(III)X(V)O4-type, oxide-, sulfide-, and aluminoborate-type, and (c) specially interesting aggregated states such as nano-size and ultra-fine particles, perfect crystals, and single crystals, microsphere, coating, film, membrane, and special crystal morphologies, etc. All these new developments, along with their increasingly wider range of applications, have motivated us to write a chapter (Chapter 4) about the synthetic chemistry of the microporous materials with special structures, types, and aggregated states. And this chapter serves as Part II of the synthetic aspects of the microporous compounds. Currently, most molecular sieves and porous materials are synthesized through hydrothermal or solvothermal crystallization. Hence it was considered essential to include a chapter addressing the crystallization process and related chemistry problems, to help the reader better understand the formation of microporous compounds, and their channel–framework structure, and the theory of crystallization, which should provide useful guidance for exploring and developing new synthetic strategies, methodologies, and techniques. This is the core of Chapter 5 (Crystallization of Microporous Compounds), which is focused on three key chemistry issues relevant to crystallization, i.e., (a) the aggregated states and polymerization reactions of the source materials at the precrystallization stage; (b) the crystallization mechanism of porous compounds and the templating or structure-directing effects during nucleation and crystallization; (c) crystallization kinetics and the mechanisms of crystal growth. It should be noted that some of the mechanistic issues relevant to crystallization are still not well understood or only partially understood, some of which are still debatable, due to the high complexity of the crystallization processes and the lack of effective techniques for probing them scientifically. So we have honestly presented our current understanding (or lack of it) of these complex scientific issues, and let our readers fully appreciate the complexity of studying the chemistry problems involved in crystallization of porous compounds and understand the feasibility in tackling these problems. The preparation, secondary synthesis, and modification of molecular sieves represent a unique set of problems, different from the issues we have discussed related to crystallization of microporous compounds under hydrothermal (or solvothermal) conditions. These deal with issues related to modifying and refining the crystallized products of microporous compounds and hence their unique process pathways and related mechanistic issues. Chapter 6 is designed to cover such

Preface

xiii

problems. Mesoporous materials have their unique characteristics from the viewpoint of structural chemistry and their synthesis, different from those of microporous materials though some commonalities exist between the two from the viewpoint of studying porous materials in general. This represents a new and extremely rich research field, playing increasingly important roles in expanding the applications of porous materials. Hence we have included one chapter (Chapter 8) focusing on mesoporous materials. Microporous materials with regular pore architectures comprise wonderfully complex structures and compositions. Their fascinating properties, such as ion-exchange, separation, and catalysis, and their roles as hosts in nanocomposite materials, are essentially determined by their unique structural characters, such as the size of the pore window, the accessible void space, the dimensionality of the channel system, and the numbers and sites of cations, etc. Traditionally, the term ‘zeolite’ refers to a crystalline aluminosilicate or silica polymorph based on corner-sharing TO4 (T ¼ Si and Al) tetrahedra forming a three-dimensional four-connected framework with uniformly sized pores of molecular dimensions. Nowadays, a diverse range of zeolite-related microporous materials with novel open-framework structures have been discovered. The framework atoms of microporous materials have expanded to cover most of the elements in the periodic table. For the structural chemistry aspect of our discussions, the second key component of the book, we have a chapter (Chapter 2) to introduce the structural characteristics of zeolites and related microporous materials. In addition to a systematic and in-depth coverage of the above material, we have allocated two chapters (Chapters 7 and 9) to discussion of the cutting-edge research issues in the chemistry of molecular sieves and porous materials, two of the most important growing areas of this field. Chapter 7 focuses on molecular design and rational synthesis of microporous molecular sieves, mainly based on the results of our own research and the knowledge we have gained in the past two decades in the area of molecular engineering of microporous compounds as well as the state-of-the-art research results by other research groups in the world. Both of these areas clearly represent where the science is going in regard to the chemistry of molecular sieves and porous materials. They also demonstrate the ultimate goal that many scientists in different branches of chemistry, such as solid-state chemists, material chemists, and synthesis chemists, have been working diligently to accomplish. Microporous molecular sieves represent one of the most important classes of target systems for molecular engineering studies in recent years, because of the regularity of their framework structures and the large amount of knowledge that scientists have gained about their key structural characteristics and the mechanisms of their formation. Hence we have devoted one chapter (Chapter 7) to presentation of the cutting-edge research issues in molecular engineering of molecular sieves. Chapter 9 focuses on the development of another important area of porous materials, i.e., porous host–guest advanced materials and MOF materials, which represents one of the most promising directions in finding new applications of porous materials in the high-tech materials. Chemistry of molecular sieves and porous materials has increasingly attracted wider attention in the past decade because of the interesting scientific issues that they raise and the prospect of their wide range of applications. This new branch of chemistry is clearly emerging as an exciting new science by itself at the interaction of various scientific disciplines.

xiv

Preface

While writing this book, we have paid special attention to make sure that the most recent and key developments at the forefront of the field are well covered in the book so that the reader gets a good exposure to the true state-of-the-art of this new field. In addition, we have tried to incorporate as many key research results and applications as possible, wherever appropriate, that have been achieved in the field of molecular sieves and porous materials. The overall design of the book’s structure and major content was done by me and Professor Wenqin Pang. The writing of the book was done mainly by Professor Wenqin Pang (Chapter 6), Professor Jihong Yu (Chapters 2 and 7), Professor Jiesheng Chen (Chapter 9) and me (Chapters 1, 3, 4, and 5). Dr Qisheng Huo of the USA, one of the pioneer researchers in the syntheses of mesoporous materials, wrote Chapter 8. The publication of this book is the result of the hard work by the authors of this book including Prof. Ruren Xu, Prof. Wenqin Pang, Prof. Jihong Yu, Dr Qisheng Huo, and Prof. Jiesheng Chen along with the long-term research experience and accumulation of knowledge of many colleagues of the State Key Laboratory of Inorganic Synthesis and Preparative Chemistry in Jilin University. Particularly, we would like to thank Dr Wenfu Yan, Dr Jiyang Li, Dr Yi Li, and Mrs Fengjuan Zhang for their contribution to the preparation of this book. In addition, we invited Prof. Yushan Yan at the University of California, Riverside, USA, to write a section on ‘Preparation and Application of Zeolite Membranes’, and Prof. Zi Gao at Fudan University, Shanghai, to write a section on ‘Channel and External Surface Modification’. Here we would like to express our heartfelt gratitude for their contribution to this book. Finally, we would like to dedicate this book to the 15th International Zeolite Conference (Beijing, 2007) and colleagues from different parts of the world. Ruren Xu Chairman of 15th IZC Professor of Chemistry Jilin University P. R. China November 2006, Changchun

1 Introduction Natural zeolites were first discovered in 1756. During the 19th century, the microporous properties of natural zeolites and their usefulness in adsorption and ion exchange were gradually recognized. However, it was not until the 1940s that a series of zeolites with low Si/Al ratios were hydrothermally synthesized through mimicking of the geothermal formation of natural zeolites. The successful synthesis of zeolites laid the foundation for rapid development of zeolite industry in the 20th and 21st centuries. Porous compounds or porous materials share the common feature of regular and uniform porous structures. To describe a porous structure, several parameters may be used and these include pore size and shape, channel dimensionality and direction, composition and features of channel walls, etc. Among these parameters, pore size and pore shape are the most important. According to the aperture size of pores, porous compounds can be classified as microporous (aperture diameter less than 2 nm), mesoporous (aperture diameter of 2–50 nm), and macroporous (aperture diameters larger than 50 nm) materials, respectively.[1] The International Zeolite Association (IZA) database shows that the number of structural types of unique microporous frameworks has been growing rapidly, from 27 in 1970, to 38 in 1978, to 64 in 1988, to 98 in 1996, and to 133 in 2001,[2] whereas currently (Feb. 2007), this number has reached 174. In fact, during the past half century, a great many microporous compounds with diverse compositional elements and primary building units have been synthesized thanks to the development of synthetic techniques. However, because of a shortage of more powerful characterization techniques, the framework structures of many novel zeolites could not be determined. It has been reported that over 20 elements may be introduced into zeolite frameworks, and taking into account the diversity of zeolite compositions, the number of unique zeolites might be enormous. The announcement of M41S compounds in 1992 by Mobil scientists has stimulated rapid growth of mesoporous materials, whereas the study of macroporous materials has just begun to burgeon, and their special structural features and properties

Chemistry of Zeolites and Related Porous Material – Synthesis and Structure Ruren Xu, Wenqin Pang, Jihong Yu, Qisheng Huo and Jiesheng Chen # 2007 John Wiley & Sons, (Asia) Pte Ltd

2

Chemistry of Zeolites and Related Porous Materials

are very attractive. From microporous to mesoporous to macroporous, the conventional framework compositions of molecular sieves and porous materials are purely inorganic. However, in recent years, the appearance of porous metal-organic frameworks (MOFs) has greatly enhanced the diversity and compositional complexity of porous materials, and has offered further possibilities for the development of porous materials.

1.1 The Evolution and Development of Porous Materials 1.1.1

From Natural Zeolites to Synthesized Zeolites

The first natural microporous aluminosilicate, i.e., natural zeolite, was discovered more than 200 years ago, and after long-term practical applications, the intrinsic properties of natural zeolites such as reversible water-adsorption capacity were fully recognized.[3,4] By the end of the 19th century, during exploitation of ion-exchange capacity of some soils, it was found that natural zeolites exhibited similar properties: some cations in natural zeolites could be ion-exchanged by other metal cations. Meanwhile, natural chabazite could adsorb water, methanol, ethanol, and formic acid vapor, but could hardly adsorb acetone, diethyl ether, or benzene. Soon afterwards, scientists began to realize the importance of such features, and use these materials as adsorbents and desiccants. Later, natural zeolites were also used widely in the field of separation and purification of air. Natural zeolites were first discovered in cavities and vugs of basalts. At the end of the 19th century, they were also found in sedimentary rocks. As a result of many geological explorations, zeolite formation was considered to include the following genetic types:[3] 1. Crystals resulting from hydrothermal or hot-spring activity involving reaction between solutions and basaltic lava flows. 2. Deposits formed from volcanic sediments in closed alkaline and saline lake-systems. 3. Similar formations from open freshwater-lake or groundwater systems acting on volcanic sediments. 4. Deposits formed from volcanic materials in alkaline soils. 5. Deposits resulting from hydrothermal or low-temperature alteration of marine sediments. 6. Formations which are the result of low-grade burial metamorphism. With geological exploration and study on minerals, more and more natural zeolites have been discovered. Up to now, over 40 types of natural zeolites have been found, but fewer than 30 of them have had their structures solved. Recently, many natural zeolite resources have been discovered around the world, and the applications of these natural species are drawing increasing attention. At present, natural zeolites are widely used in the fields of drying and separation of gases and liquids, softening of hard water, treatment of sewage, and melioration of soils. Some well selected or modified natural zeolites are also used as catalysts or supports of catalysts in industry. Zeolite science and technology in China has been in great progress as well in the past several decades. According to incomplete statistics, there are many types of zeolite resources in China, and among the natural zeolites discovered in China are mordenite, clinoptilolite, analcime, heulandite, natrolite, thomsonite, stilbite, and laumontite. With further exploration, it is believed that many more zeolite resources will be

Introduction

3

discovered in China. As research work on natural zeolites deepens, they will be applied more broadly. Because natural zeolites cannot meet the huge demands in industry, it becomes an urgent necessity to use synthesized zeolites besides the natural ones. Synthesis of zeolites was first conducted at the end of the 19th century through mimicking of the geothermal conditions for natural zeolite formation, i.e., high-temperature hydrothermal reactions. By the end of the 1940s, a number of scientists started to carry out research on massive synthesis of zeolites. Abundant natural zeolites were found later in sedimentary rocks. Since these zeolite deposits were usually located near the surface of the earth, it was concluded that they had been produced at temperatures and pressures which were not very high. During a study on strata of Triassic rocks, it was found that zeolites were somehow in a chemicalequilibrium state when they were formed. This state was metastable and was known as the zeolite phase. The equilibrium process for zeolite phases was very similar to that of low-temperature hydrothermal synthesis reactions. Therefore, researchers tried to synthesize zeolites using hydrothermal synthesis techniques at temperatures of around 25–150
C (usually 100
C). In the 1940s, low-silica zeolites were first synthesized. The application of low-temperature hydrothermal techniques facilitated the extensive industrial production of zeolites. By the end of 1954, zeolites A and X began to be produced industrially. Following this, a number of companies in the United States, such as Linde, UCC, Mobil, and Exxon, imitated the formation of natural zeolites and produced a series of synthesized zeolites with an intermediate Si/Al ratio (Si/Al ¼ 2–5), including NaY, mordenite, zeolite L, erionite, chabazite, clinoptilolite, and so on. These zeolites were widely applied in the fields of gas purification and separation, catalytic processes of petroleum refining and petrochemistry, and ion exchange. In China, zeolites A and X were first synthesized in 1959, followed by the industrial production of zeolite Y and mordenite. With the development of the zeolite industry, zeolites were applied in many fields as well in China. In the 1950s, zeolites were mainly used in drying, separation, and purification of gases. Since the 1960s, zeolites have been widely used as catalysts and catalyst supports in petroleum refining. At present, zeolites have become the most important adsorbents and catalysts in the petroleum industry. Although, compared with natural zeolites, synthesized zeolites have many advantages such as high purity, uniform pore size, and better ion-exchange abilities, natural zeolites are more applicable when there are huge demands and fewer quality requirements. The reason is that natural zeolites are often located near the surface of the earth and can be easily exploited and used after some simple treatments, which lead to lower costs and hence lower prices. Therefore, natural zeolites have a good prospect of application especially in the fields of agriculture and environmental protection. 1.1.2

From Low-silica to High-silica Zeolites

The period from 1954 to the early 1980s is the golden age for the development of zeolites. Zeolites with low, medium, and high Si/Al ratios were extensively explored, and this greatly facilitated the applications of zeolites and stimulated industrial progress.[5] In order to increase the thermal stability and acidity of zeolites, Breck et al. synthesized zeolite Y (Si/Al ¼ 1.53.0), which played an extremely important role in the catalysis of

4

Chemistry of Zeolites and Related Porous Materials

hydrocarbon conversion. From then on, a variety of zeolites with an Si/Al ratio of 25, i.e., ‘intermediate silica’ zeolites which include mordenite, zeolite L, erionite, chabazite, clinoptilolite, zeolite
, etc, have been synthesized. At the beginning of the 1960s, scientists at Mobil Corporation started to use organic amines and quaternary alkylammonium cations as templates in the hydrothermal synthesis of high-silica zeolites, and this is considered a milestone in the progress of zeolite synthesis. In 1972, Argauer and Landelt synthesized the first important member of the pentasil family, ZSM-5, using Pr4NCl or Pr4NOH as the template at 120
C, whereas in 1973, Chu synthesized ZSM-11 using Bu4Nþ as the template. In 1974, Rosinski and Rubin prepared ZSM-12 using Et4Nþ as the template, followed by the syntheses of ZSM-21 and ZSM-34 in 1977 and 1978; later on, Wadlinger and Kerr synthesized high-silica zeolite beta (BEA). The pentasil family, which includes high-silica zeolites with hydrophobic surfaces and interconnected two-dimensional (2-D) 10-membered-ring channels, has played an important role in shape-selective catalysis since its inception. In 1970, Flanigen at UCC first synthesized pure-silica forms of ZSM-5 (silicalite-I) and ZSM-11 (silicalite-II), which were the end members of the pentasil family. Meanwhile, the rapid progress in synthesis of high-silica zeolites facilitated the study of the secondary synthesis of zeolites. Some high-silica zeolites such as zeolite Y (Si/Al > 3), which were difficult to synthesize directly, could be prepared from zeolites with medium Si/Al ratios through steam treatment or de-alumination in framework by reaction with Si. For instance, ultra-stable zeolite Y (USY), high-silica mordenite, erionite, BEA, and clinoptilolite were all successfully synthesized in this way. In the past 25 years, the emergence of zeolites with low (Si/Al ¼ 1.01.5), medium (Si/Al ¼ 2.05.0), and high Si/Al ratios (Si/Al ¼ 10100), as well as pure-silica zeolites, facilitated the study of both the structure and property of molecular sieves and porous compounds, and promoted their applications. The increase in type and structural diversity of zeolites, as well as deep insight into zeolite properties such as thermal stability, acidity, hydrophobicity/hydrophilicity of surfaces, and ion-exchange capacity, has led to application of a series of zeolites in industry. These zeolites include synthesized ones such as zeolite A (Na, Ca, K), zeolite X (Na, K, Ba), zeolite Y (Na, Ca, NH4), zeolite L (K, NH4), zeolite
(Na, H), zeolon (MOR-H, Na), ZSM-5, zeolite F (K) and zeolite W (K), and natural ones such as mordenite, chabazite, erionite and clinoptilolite. These materials have been widely used as commercial adsorbents for drying and purification of gases and for bulk separation of, for example, normal-/iso-paraffins, isomers of xylenes and olefins, and O2 from air, as catalysts for petroleum refining and petrochemistry, and as ion exchangers. Because of their excellent ion-exchange capacities, zeolites A and X can be used as auxiliary agents in the detergent industry, in radioactive waste treatment and storage, and in the treatment of industrial liquid wastes. 1.1.3

From Zeolites to Aluminophosphate Molecular Sieves and Other Microporous Phosphates

In 1982, Wilson, Lok, and Flanigen et al. successfully synthesized a novel family of molecular sieves, that is, microporous aluminophosphates AlPO4-n.[6] The discovery of AlPO4-n is regarded as a milestone in the development of porous materials. Not only

Introduction

5

were large-, medium-, and small-pore AlPO4-n molecular sieves prepared, but also SAPO-n (S ¼ Si), MeAPO-n (Me ¼ Fe, Mg, Mn, Zn, Co, etc), MeASO-n, ElAPO-n (El ¼ Ba, Ga, Ge, Li, As, etc) and ElAPSO-n could be obtained through introduction of elements other than Al and P into the microporous frameworks of AlPO4-n. At present, the aluminophosphate-based family of microporous compounds has over 200 members. These compounds were synthesized through the crystallization of Al, P, and other element sources together under hydrothermal or solvothermal conditions. Differing from the aluminosilicate molecular sieves, normally the AlPO4-based compounds must crystallize in the presence of templates or structure-directing agents. There are a large number of structure types for AlPO4-based microporous materials and the compositions of these materials also vary to a considerable degree.[7] Except for a few members which are isostructural with zeolites, most aluminophosphate molecular sieve structures are novel, and their elementary compositions are quite different from those of conventional zeolites containing only silicon and aluminum. By 1986, 16 elements had been successfully incorporated into frameworks of aluminophosphate molecular sieves. The incorporation of heteroatoms into aluminophosphates has played an important role in enhancing the diversity of structures and compositions of microporous compounds and molecular sieves. Since 1982, two major accomplishments have been achieved for aluminophosphatebased molecular sieves. One is the discovery of various aluminophosphate microporous compounds with an Al/P ratio less than unity.[8] For instance, JDF-20 ([Et3NH]2 [Al5P6O24H]2H2O) is a microporous aluminophosphate with the largest aperture size ˚ ); AlPO-CJB1 ([(CH2)6N4H3][Al12P13O52]) is the first (20-membered ring, 14.5  6.2A microporous aluminophosphate with Bro¨nsted acidity. These 3-D microporous aluminophosphates with anionic frameworks are different from AlPO4-n with a neutral framework constructed by the alternation of AlO4 and PO4 tetrahedra. The anionic frameworks are constructed by Al-centered units (AlO4, AlO5, AlO6), and P(Ob)n(Ot)4n tetrahedra (b ¼ bridging, t ¼ terminal, n ¼ 14), and this construction manner results in rich  O groups structural chemistry. The existence of terminal oxygen of P OH and P  strengthens the nonbonding interaction between the framework and template molecules, rendering the templates hard to remove. The other accomplishment is the synthesis of other families of metal phosphates, including zinc, gallium, titanium, iron, cobalt, nickel, vanadium, and molybdenum phosphates.[9] The compositional and structural diversity of aluminophosphates and their derivatives leads to potential applications in the fields of adsorption, separation, formation of host–guest advanced materials, redox catalysis, chiral catalysis, and macromolecular catalysis. 1.1.4

From 12-Membered-ring Micropores to Extra-large Micropores

For nearly 50 years, chemists failed to synthesize molecular sieves with channels larger than 12-membered rings. It was not until 1988 that Davis et al. successfully synthesized the first aluminophosphate molecular sieve, VPI-5 ((H2O)42[Al18P18O72]), with 18˚ ).[10] The synthesis of VPI-5 is another milestone membered-ring apertures (12.7  12.7 A in the development of microporous materials. It has been found that, except for a few silica or germanium oxide porous compounds, most of the microporous molecular sieves with a large aperture are metal phosphates with

6

Chemistry of Zeolites and Related Porous Materials

1-D channels. The structures of large-pore microporous materials share the following common features: 1. The frameworks are constructed by metal-centered primary building units with various coordination states, such as [AlO4], [AlO6], [GaO4], and [GaO4(OH)2];  O, P-OH, and Al 2. There are terminal groups in the frameworks, such as P OH,  which make the structures less stable than zeolites and aluminophosphate molecular sieves with (4,2) networks. These terminal groups also favor the formation of interrupted frameworks, such as cloverite and JDF-20; 3. The structure-directing agents used in the synthesis of these compounds usually possess multiple amino groups, long chains, or large molecular weights, and occasionally the synthesis also involves F ions. Usually, F ions exist in the open frameworks and are located between two metal centers as bridging atoms or inside the double 4-ring (D4R) cages. On the other hand, the oxygen atoms in the terminal groups normally have strong non-bonding interactions with structure directing agents. On the basis of these structural features, it is easy to understand why zeolites constructed by Si and Al cannot have extra-large pores. Nevertheless, pure-silica zeolites with 14-membered rings, i.e. CIT-5 and UTD-1, have been synthesized recently, and further investigation into crystallization mechanisms in combination with the vast experimental data available and with theoretical simulation and computation may help us to rationally design and synthesize extra-large microporous aluminosilicate molecular sieves with special channels such as multidimensionally interconnected and chiral ones. The discovery of extra-large microporous materials facilitates research on the catalytic reaction of large and medium molecules, and also promotes host–guest chemistry and related advanced materials. 1.1.5

From Extra-large Micropores to Mesopores

The discovery of mesoporous materials, which usually refer to materials with ordered pores of diameter size 250 nm, is another leap in the development of molecular sieves and porous materials. In fact, the synthesis of ordered mesoporous materials began as early as 1971. Kuroda et al. also started to synthesize mesoporous materials before 1990. However, it was not until 1992, when Kresge et al. reported the discovery of M41S materials, that mesoporous compounds started to attract real increasing attention.[11,12] Using surfactants as templates, scientists at Mobil synthesized a series of mesoporous compounds, the M41S family, including MCM-41 (hexagonal), MCM-48 (cubic), and MCM-50 (layered). This discovery is comparable with the other great accomplishments in the history of zeolite science and technology; for instance, the synthesis of ZSM-5 also by Mobil scientists. For microporous zeolites used as catalysts, the reactants in their pores and/or channels are ˚ due to the microporous features of the catalysts, even after usually smaller than 10 A modification of the channels. However, the successful synthesis of mesoporous materials with channels of 250 nm might break this limitation. Mesoporous materials have the advantages of ordered mesoporous channels with size of 250 nm, as well as very large specific surfaces and pore volumes. However, since the

Introduction

7

channels in these materials are surrounded by amorphous walls, mesoporous materials have less thermal and hydrothermal stability than do microporous molecular sieves. Recently, the synthesis of SBA-15, MAS-7, and MAS-9 showed that the stabilities of mesoporous materials could be enhanced. Another advantage of mesoporous materials is that there are far fewer restrictions on their composition. Theoretically, any oxides, oxide composites, inorganic compounds, or even metals could form mesoporous materials. In fact, many oxides, such as TiO2, ZrO2, Al2O3, Ga2O3, MnO2, and other non-silicon oxides, have been successfully synthesized in a mesoporous form. Recently, many highly ordered mesoporous materials have been obtained, and these include MCM-41 (P6m), MCM-48 (Ia3d), MCM-50 (layered), FSM-16, SBA-1, SBA-6 (Pm3n), SBA-2, SBA-12 (P63/mmc), SBA-11 (Pm3m), and SBA-16 (Im3m). Low-ordered ones such as HMS, MSU-n, and KIT-1 have also been reported. According to their compositions and structures, the periodic mesoporous materials can be divided into 6 categories: 1. 2. 3. 4. 5. 6.

Mesoporous silicon oxides with different channel networks, sizes, and shapes; Mesoporous silicon oxides with modified surfaces; Mesoporous silicon oxides with organic compositions; Mesoporous silicon oxides with other metal atoms on their channel walls; Inorganic mesoporous materials without silicon;[13] Mesoporous materials without oxygen.

There will be many more categories if we consider specific polymorphs. The rapid development and constant improvement of mesoporous materials as well as the progress in related research areas will render mesoporous materials more widely applicable. 1.1.6

Emergence of Macroporous Materials

Ordered macroporous materials have special optical features due to their pore diameters. Since the synthesis of macroporous materials has just started, there are no general synthetic strategies for this type of materials at present, and hence only a few examples will be mentioned here. By using modified colloidal particles as templates, silicon oxide macroporous materials with uniform submicrometer-sized pores can be synthesized.[14] Modified polystyrene emulsion microspheres (2001000 nm) can be electronegative (sulfates) or electropositive (amidines). After these microspheres are packed in an orderly fashion, they can interact with surfactants and silicon oxides to form macroporous solid composites, and further to form macroporous materials after the removal of the templates by calcination. The sizes of the macropores in the products range from 150 to 1000 nm. Macroporous TiO2 can also be prepared in a similar way. Mineralization on hyphae can also generate macroporous materials.[15] Using this method in the synthesis of mesoporous materials, mesoporous and macroporous composites can be obtained. The long channels in these composites are parallel to each other. The pores are at a micron level, and the thickness of the walls ranges from 50 to 200 nm. By using colloid as the template, inorganic oxides can be deposited on the outer surface of the colloidal droplet to form macroporous materials with apertures of 50 nm

8

Chemistry of Zeolites and Related Porous Materials

to several microns in size.[16] Oil can form uniform droplets in formamide colloid and can further be used as the template. Polymers, such as the triblock copolymer formed by ethylene glycol and propylene glycol, can stabilize this colloid. Many macroporous materials have been synthesized using this method, such as macroporous titanium oxides, silicon oxides, and zirconium oxides. 1.1.7

From Inorganic Porous Frameworks to Porous Metal-organic Frameworks (MOFs)

From natural zeolites to the recently discovered meso- and macro-porous materials, the ordered porous frameworks are all constructed by inorganic species. However, in the past ten years, a new family of porous compounds composed of metal-organic frameworks (MOFs) has attracted enormous attention. The main reason is that the poor thermal and chemical stability of MOFs has been somewhat improved. In addition, the discovery of some advantages of MOFs that are lacking in molecular sieves and mesoporous materials has also stimulated the research on MOFs. In 2001, Chen et al. synthesized a coordination polymer, Cu3(BTB)2(H2O)(DMF)9(H2O)2 (MOF-14) (BTB-4,40 ,400 -benzene-1,3,5-triyltribenzoic acid), from which the DMF could be removed by heating at 250
C under inert gas flow.[17] The N2 and Ar adsorption isotherms of MOF-14 are of type-I, confirming its microporous structure. The adsorption isotherms of MOF-5 are also characteristic of type-I. Adsorptions of CO, CH4, CH2Cl2, CCl4, C6H6, C6H12 and m-xylene in these materials are all reversible, as in zeolites. However, the pore volume for MOF-14 is 0.53 cm3/g whereas the specific surface area is 1502 cm2/g, and these two values are distinctly higher than the corresponding ones for inorganic microporous compounds. In 2002, Yaghi and coworkers reported the synthesis of a microporous compound (MOF-5), Zn4O(R1-BDC)3 (R1 ¼ H), by the crystallization of Zn(NO3)24H2O and 1,4-benzenedicarboxylate (terephthalate (BDC) in N,N-diethylformamide (DEF) solvent at 85105
C.[18] The microporous framework of this compound is constructed by the primary building unit of the [Zn4O(CO2)6] octahedron and bridging R groups. Yaghi and coworkers used different BDC derivatives and related naphthalene -2,6-dicarboxylic acid (2,6-NDC) and triphenyldicarboxylate (TpDC) compounds to obtain a series of microporous compounds with various pore ˚ ), and they found that the pore diameter varies with R. The free diameters (3.828.8 A porous volume increases remarkably from C5H11O-BDC (55.8%) to TpDC (91.1%), both of which are much larger than the free volume of the zeolite FAU. The adsorption properties of the compound are similar to those of zeolites. MOF-6 has a great adsorption capacity for CH4 (240 cm3/g; 36 atm, 298 K), which could be exploited for storage and transportation of CH4. In addition, it has been demonstrated that a number of MOF compounds exhibit promising H2-storage capacities. Furthermore, other groups, such as -Br, -NH2, -OC3H7, -OC5H11, -C2H4, and -C4H4, could be added into the R groups. Therefore, the MOFs may be functionalized to meet special catalysis or adsorption demands. Conventional inorganic porous compounds have no such advantages, and therefore, in a sense, the emergence of MOFs has broadened the applications of porous materials and facilitated their development.

Introduction

9

1.2 Main Applications and Prospects As mentioned earlier in this chapter, it is the social demands and wide applications of porous materials that keep them under continuous exploration. From natural zeolites to synthesized ones, from low-silica zeolites to high-silica ones, from aluminosilicate molecular sieves to aluminophosphate-based ones, from extra-large microporous materials to mesoporous materials, and from inorganic porous frameworks to MOFs, together with newly emerging macroporous materials, all these porous materials have ordered and uniform porous systems. Here, we would like to take ZSM-5 as an example to illustrate the relationship between structure and function. ZSM-5 has an interconnected 2-D 10-membered-ring channel system ([100] 10 5.1  5.5* $ [010] 10 5.3  5.6*). Since the Si/Al ratio of ZSM-5 can be varied from 10 to infinity as found in pure-silica silicalite-I, the type, acidity, and distribution of acidic sites can also be controlled accordingly. Furthermore, because of its special channel system, ZSM-5 may function very differently for different molecules. For example, the diffusion, the adsorption/desorption, the reaction rate, and the formation of intermediate and final product of molecules may vary to a great extent. ZSM-5 has been widely used in petroleum refining as a catalyst with good shape-selectivity. Since 1950s, there have been three traditional fields of application for molecular sieves and porous materials: 1) separation, purification, drying and environment treatment process; 2) petroleum refining, petrochemical, coal and fine chemical industries; 3) ionexchange, detergent industry, radioactive waste storage, and treatment of liquid waste. In addition to the traditional application fields, zeolites and related porous materials may also find applications in new areas such as microelectronics and molecular device manufacture. 1.2.1

The Traditional Fields of Application and Prospects of Microporous Molecular Sieves

Since the first application of NaA in the separation of normal and isoalkanes by the Linde company in the 1950s, and X- and Y- zeolites as catalysts for cracking reactions of hydrocarbon conversion in the 1960s, NaA, NaX, and NaY have been widely used in the petroleum industry in reactions such as cracking, alkylation, isomerization, shapeselective reforming, hydrogenation and dehydrogenation, methanol-to-gasoline conversion (MTG), etc. These porous materials have also been extensively used in the detergent industry and in a variety of adsorption and separation processes such as the drying, the removal of CO2 from, and the desulfurization for natural gas, and the separation of xylene isomers, of alkenes, and of O2/N2 from air.[5] In the past half century, molecular sieves have played increasingly important roles as catalysts in the petroleum refining, petrochemical, and other chemical industries. According to the statistics studies conducted by Marcilly in 2001, the annual output of synthesized molecular sieves exceeded 1.6 million tons, and the annual output of natural zeolites rose to 0.3 million tons (about 18% of the total output).[19] The value of the annual gross product of synthesized molecular sieves exceeded 2.0 G$. Furthermore, the value of annual gross product of other catalysts, adsorbents, and ion-exchangers related to molecular sieves and their

10

Chemistry of Zeolites and Related Porous Materials

derivatives greatly exceeded the values of molecular sieves themselves.[5] Despite this, there are still many prospects for development of molecular sieves in the above three main traditional fields. First, there are 174 known molecular sieve frameworks. Considering the differences in their composition, there should be more space for further development. However, currently only a few frameworks, including LTA, FAU, MOR, LTL, MFI, BEA, MTW, CHA, FER, AEL, and TON, have been widely used in industry. Second, at present, molecular sieves are mainly used in petroleum related industries and intermediary chemistry processes. It is believed that, in the next 20 years, molecular sieves will be more widely used in catalysis, adsorption, and separation, with the development of petroleum refining, petrochemical, intermediary chemical, and fine chemical industries. According to Marcilly’s proposal in 2001, in the next 20 years, there will be several new application fields in petroleum refining and petrochemical industries:[19]  FCC (fluid catalytic cracking): to develop novel molecular sieves which are comparable with or better than ZSM-5 in shape-selectivity of light olefins (C3¼–C5¼).  HDC (hydrocracking): to develop novel zeolitic catalysts dedicated to the production of middle distillates, integrating both the activity and stability of zeolites.  Aliphatic alkylation: to develop novel molecular sieves with a three-dimensional open framework and catalytic activity higher than BEA.  Alkane isomerization of paraffins: to develop novel molecular sieves with high selectivities (2 branches or more) for isomerizations of C7–C9 middle paraffins in gasoline (petrol). In addition, in the field of dewaxing (gas oils, HDC residues, lubricating oil, etc.), synthesis of novel molecular sieves with better adsorption and separation abilities is highly desired. In the past 20 years, thanks to the discovery of many molecular sieves with new compositions and structural features [secondary building units (SBUs) and pores], there have appeared a number of new application fields for molecular sieves, such as basic catalysis, extra-large microporous molecular sieve catalysis, redox catalysis, asymmetric catalysis, and dual- and multi-functional catalysis.[20] All of these will lay a further solid foundation for the development of molecular sieves in catalysis, adsorption, and separation. 1.2.2

Prospects in the Application Fields of Novel, High-tech, and Advanced Materials

In molecular sieves and microporous crystalline compounds, there exist channels with apertures of 12-, 14-, 16-, 18-, 20-, or 24-membered rings, and cages or cavities constructed ˚ ) is constructed by interconnected 2- or 3-D channels. For example, the FAU cavity (11.8 A ˚ ) in LTA by the by the intersection of three 12-membered-ring channels; the a cage (11.4 A ˚ ) in EMC-2 by the intersection of three 8-membered-ring channels; the EMT cage (13.5 A ˚ ) in MAPSO-46 by intersection of three 12-membered-ring channels; the AFS cage (14.0 A the intersection of a 12-membered-ring channel and two 8-membered-ring channels, the ˚ ) in DAF-1 by the intersection of 12-, 8-, and 10-membered-ring DFO cage (15.5 A ˚ ) by the intersection of 20- and 8-membered-ring channels. channels; the CLO cavity (30 A These large cages or cavities can act as favorable reaction venues. For example, through the

Introduction

11

so-called ‘ship-in-bottle’ synthetic strategy,[21] a dye composite can be prepared in the cavities of FAU or channels of AlPO4-5,[22,23] and through using nanoscale chemical synthesis techniques, Cd4S4 semiconductive nanometer-sized clusters can be obtained in the FAU cages.[24] The overall process takes two steps:  Zn; CdÞ Step 1: H44 Na11 Y þ 44ðCH3 Þ2 M ! ðCH3 MÞ44 Na11 Y þ 44 CH4" ðM   Step 2: ðCH3 MÞ44 Na11 Y þ 29:84 H2 X !ðM5:5 X3:73 Þ8 H15:64 Na11 Y þ 44 CH4 ðX  S; SeÞ Another approach to the preparation of zeolite composite materials is to add on some complicated molecules, complexes, metal-organic compounds, supermolecules, clusters, or polymers with specific functions in the nanometer-sized cages or channels in molecular sieves through grafting or other reaction routes. As Pool mentioned in 1994, ‘zeolites – crystalline materials riddled with nanometer-sized cavities – can exert exquisite control over chemical reactions and produce devices on the smallest scale’.[25] In the mid -1990s, Ozin, Herron, Bein,[26] and others extensively studied the preparation of quantum dot arrays, molecular wires, and magnetons inside porous materials. They also carried out a variety of basic research on microdevices, molecular circuits, molecular switchs, sensors, and optical memory. In the past decade, with the development of meso- and macro-porous materials and the successful preparation of molecular sieve membranes and millimeter- to centimeter-sized single crystals, the application of novel advanced materials based on porous materials has undergone great progress. The following are several examples of progress achieved in recent years. With the aid of poly-(propylene glycol), Fan et al. synthesized porous materials with low dielectric constant (k ¼ 1:3),[27] which are promising for commercial use,[1] whereas gadolinium zeolite has been used as a radiography reagent for magnetic resonance imaging (MRI).[1] Another new field of application for microporous materials is the utilization of zeolite-dye composites as microlasing materials.[1,23] In a word, microporous materials have promising prospects, but there is still a long way to go before the application potential of these materials is fully realized. 1.2.3

The Main Application Fields and Prospects for Mesoporous Materials

Since the ordered mesoporous material MCM-41 was reported in 1992,[1–3] comprehensive research on the potential applications of mesoporous materials has been carried out, with focus on their catalysis, adsorption, and the preparation of novel advanced materials. Their applications in catalysis have attracted the most intense attention. The unique properties of mesoporous materials arise from their high specific surface areas (>1000 m2/g) and their uniform mesopores (diameters range from 2 to 50 nm).[11,28,29] In the past decade, mesoporous materials have been widely used in the field of catalysis, such as in petroleum processing, the fine-chemicals industry, and in reactions involving large molecules. For petroleum processing, the conventional catalysts are usually microporous zeolites, such as zeolite Y and ZSM-5. However, with the decrease of petroleum resources in the world and the increase of heavy components in crude oil, the applications of conventional zeolites are more and more restricted due to their small pores. Mesoporous materials have ordered mesopores which might have potential applications in the catalysis of heavy oil processing.[29] For example, Al-MCM-41 has

12

Chemistry of Zeolites and Related Porous Materials

shown better catalysis performance in hydrocracking, hydrodesulfurization, and hydrodenitrogenation reactions than do traditional microporous materials.[30] In green oxidation reactions, zeolite TS-1 is the typical catalyst. Since the size of its ˚ , TS-1 can be used as the catalyst only for benzene and channels ranges from 5 to 6 A phenol conversion. However, ordered mesoporous titanium silicate materials have pores large enough for the catalytic reactions of bulkier molecules, and this is very important for the production of fine chemicals. For example, for the oxidation reaction of terpineol, Ti-MCM-41 performs much better than do microporous titanium silicate molecular sieves as a catalyst.[29] However, on the other hand, the hydrothermal stability and catalytic activity of ordered mesoporous materials are still lower than those of conventional microporous molecular sieves. In recent years, many measures have been taken to solve this problem, such as adding inorganic salts during the synthesis of mesoporous materials,[31] intensifying the post treatment,[32,33] using triblock copolymers as templates to obtain thicker channel walls of mesoporous materials,[28] using neutral surfactants to synthesize mesoporous materials,[34] using mixed templates,[35–37] and synthesizing mesoporous materials at high temperatures.[38] Although these methods more or less help enhance the hydrothermal stability of mesoporous materials, their catalytically active centers are still not comparable with those of conventional microporous molecular sieves. In recent years, scientists have tried to prepare novel ordered mesoporous materials through the selfassembly of nanoparticles consisting of microporous building units and surfactant micelles. Using this approach, both the hydrothermal stability and the catalytic activity of mesoporous materials have been enhanced.[39–41] For instance, the novel mesoporous titanium silicate material, MTS-9, has shown better catalytic activity than have Ti-MCM41 and TS-1 in the synthesis of an intermediate product of vitamin E.[41] Mesoporous materials have great application potential in novel and high-technology areas as well. They can be used for the stabilization or separation of enzymes and proteins, the degradation of organic wastes, the purification of water, and the transformation of exhaust gas. They can also be used for energy storage. Many functional materials are able to be assembled into mesoporous materials. For example, advanced mesoporous optical materials may be prepared through assembly of laser-generating species or materials with optical activities.[42–44] Ordered mesoporous conducting polymers may form through polymerization in ordered mesopores followed by chemical removal of the inorganic host.[45] Ordered mesoporous carbon materials can be obtained through complete mixing of mesoporous materials and a glucoside followed by carbonization and dissolution of the inorganic species.[46] It has been demonstrated that the mesoporous carbon thus formed exhibits better performance than do conventional carbon materials when used as electrodes of fuel cells.[47] Through using the ordered channels in mesoporous materials as micro-reactors, fine nanoparticles and other quantum composite materials can be synthesized. Because of small-size or quantum-size effects arising from the confinement of ordered channels, these composite materials exhibit unique optical, electrical, and magnetic properties. For example, it has been demonstrated that modified mesoporous zirconium oxides show unusual photoluminescence at room temperature. In contrast with carbon nanotubes, mesoporous materials composed of silica and nonsilica species exhibit rich surface chemical activity. The ordered channels in mesoporous materials may act as micro-reactors to assemble nanometer-sized homogeneous guest

Introduction

13

materials, and, as a result, the application fields of mesoporous materials can be further broadened on the basis of the host–guest effects. Through using stable mesoporous materials as hosts, a variety of inorganic photoelectric nano-sized materials such as Si, BN, SiC, AgI, and AlN, and giant magneto-resistant transition metals such as Ni, Cu, and Co can be prepared. Assembly of some semiconductor clusters with a wide band-gap such as ZnO, ZnS, and CdS into mesoporous materials may greatly increase the fluorescence intensities of the former due to the host–guest interactions and quantumsize effects, implying promising applications of these composites in the field of optoelectronics. In view of the many applications in the fields of separation, purification, biology, medicine, chemical industry, catalysis, information, environment, energy, and advanced composite materials, it is believed that mesoporous materials will play more important roles in the 21st century as an increasing number of mesoporous materials with advanced functions are designed and synthesized.

1.3 The Development of Chemistry for Molecular Sieves and Porous Materials In the past half century, with the expansion of structure types and compositions of porous materials, the number of application fields and the total demand for these materials have been continuously growing, and meanwhile, the characterization techniques and instrumentation have been greatly improved. As a result, our comprehension of the chemistry of molecular sieves and porous materials has been deepened to a great extent. Here, we take two main branches in the chemistry of molecular sieves and porous materials as examples to illustrate how zeolite science has been developed. 1.3.1

The Development from Synthesis Chemistry to Molecular Engineering of Porous Materials

In 1968, the first International Zeolite Conference (IZC) was held in London. This was the first international conference focusing on zeolites and microporous aluminosilicates, and various issues related to zeolite research were addressed. Because only a few natural zeolites had been discovered and about 20 synthesized at that time, all the scientific topics about the synthesis of zeolites were focused on the formation of aluminosilicate microporous materials, and the influence of synthetic conditions on reactions and products (for example, crystallization zone diagrams, and crystallization kinetic curves, etc). In the past 30 years, the compositional elements have increased from 2 to over 30, and the framework types have increased to 174 (Feb. 2007). Therefore, it is important to summarize the synthetic chemistry for pore construction, and to conduct an in-depth study on related scientific issues, such as the structures of intermediates and products, the polymerization of reactants, the structures and transformation of sols and gels, nucleation and crystallization, the templating and structure-directing effects, the metastable state and crystal transformation, and the growth of crystals and their aggregation. Inorganic synthesis and preparative chemistry, hydrothermal and solvothermal chemistry, sol–gel chemistry, crystallization and

14

Chemistry of Zeolites and Related Porous Materials

crystal–growth, host–guest chemistry, and combinatorial chemistry all help to paved the way for the progress of the synthetic chemistry of porous materials or the so-called ‘pore-construction’ synthetic chemistry. On the other hand, the most important goal for chemistry is to create new materials. Synthesis and preparative chemistry is the core of chemistry, and it is always on the frontier of development. During the process of development, the research mode of ‘synthesis–structure–function’ is formed. With the progress of science and technology, it has become a key issue to explore ways to avoid creating new materials without clear goals and to develop rational, effective, and environment-friendly synthetic routes in the 21st century. As chemistry and related disciplines have gained deep insight into and reasonable control over molecules, a new research field, that of molecular design and molecular engineering, has emerged. In recent years, molecular design and engineering has attracted increasing attention in chemistry, materials science, and life sciences, leading the development of chemistry into the age of molecular engineering. Differing from traditional chemistry, molecular engineering involves the design of structures based on their required function. Molecular engineering focuses on the formation and assembly of primary building units, and, with the aid of computational simulations, gradually realizes the rational synthesis of compounds with specific functions and structures. In some sense, molecular engineering is the chemistry of rational design and synthesis. The key impact of molecular engineering on chemistry is that it broadens the perspectives on function, structure, and synthesis, draws more attention to ‘function– structure–synthesis’, and promotes a better understanding of structure types and levels beyond molecular structures, rather than excessively focusing on the synthesis of individual compounds. The channels in porous molecular sieves are rather regular and uniform. The framework features, the secondary building units (SBUs), and the interactions between building units and structure-directing agents for porous materials have been thoroughly investigated. Furthermore, the formation behavior, the crystallization mechanism, and the movement and reactions of reactant molecules in the channels have also been elucidated for over half a century. Therefore, in contrast with other materials, porous materials, with molecular sieves as their representatives, have been well studied in terms of the relations among function, structure, and synthesis. With the aid of computers, ideal porous structure models can be designed to meet specific function requests. Then feasible structures and corresponding synthesis conditions can be selected under the guidance of structure and synthesis databases. Finally, rational synthesis can be achieved using combinatorial chemistry. At present, several research groups, including the authors’ own group, in the world have been engaged in this work, and satisfactory results have been obtained in some aspects. Although there is still a long way to go, molecular engineering has pushed the chemistry of porous materials to a new level, and more challenging research directions and scientific issues will come up in this emerging field. 1.3.2

Developments in the Catalysis Study of Porous Materials

The first use of molecular sieves in catalysis occurred in 1959 when zeolite Y was used as a catalyst for isomerization reactions. In 1962, the Mobil Company used zeolite X in the

Introduction

15

catalysis of cracking reactions. In 1969, Grace developed ultra-stable zeolite Y (USY) as a catalyst. At that time, besides the catalysis of cracking and hydrocracking reactions, molecular sieves were also used for the isomerization of normal alkanes, C8 aromatics, and the disproportionation of toluene in industry.[48] With increasing applications of molecular sieves in industry, the theories of acid catalysis, of B- and L-acids,[49] and on carbonium ion reaction mechanisms were established in the field of molecular sieve catalysis. At the same time, study on another important feature of molecular sieves, that is, catalytic shape-selectivity, was also carried out. This study was initiated in 1960 by Weisz and Frilette, followed by investigation in shape-selective catalysis of zeolites erionite,[50] ZSM-5, mordenite, and CaA were studied until the early 1980s.[51] Naccache et al. summarized the shape-selective issues in several different catalytic processes involving molecular sieves, such as the diffusion and adsorption of reactants, the generation of active intermediates, and the desorption and diffusion of intermediates and final products.[52] They believed that shape selectivity depended mainly on the sieving effect, the reverse molecular-size selectivity, and the selectivity of intermediate products. Since the beginning of the 1980s, applications of molecular sieves and porous materials have been continuously increasing due to several reasons: 1) the demands of industry, such as the transformation of hydrocarbons in petroleum processing, the catalysis of intermediates in the fine chemicals and medicine industries,[52,53] and the use of catalysis in the treatment of environmental pollutants;[54,55] 2) the development of secondary synthesis, modification, and treatment of zeolites, including ion-exchange, dealumination of framework, isomorphous substitution, and assembly techniques in channels or cavities, etc.; and 3) the emergence of new molecular sieves and porous materials, such as extra-large microporous molecular sieves and mesoporous materials. On the basis of solid acid and shape-selective catalysis theories, new catalytic processes have been developed, and these include metal–molecular sieve dual functional catalysis,[51,56] redox catalysis,[55] alkali catalysis,[57] catalysis in the channels of extra-large microporous and mesoporous materials,[57] and chiral catalysis of molecular sieves.[57] Furthermore, the enormous amount of experimental data and the rapid development in theory allow for a deep understanding of the relationship between the catalytic functions and the structures of molecular sieves. Our extensive knowledge of the relationships between catalytic functions, structures, and synthesis enables molecular sieves and porous materials to enter the era of molecular engineering ahead of many other types of catalyst. All the above reasons contribute to the prominent position of research on the catalysis of molecular sieves and porous materials in the whole field of catalysis. According to some internationally renowned experts in catalysis research and development, ‘the grand challenge for catalysis science in the 21st century is to understand how to design catalyst structure to control catalytic activity and selectivity’. This preludes a very prosperous future for molecular sieves and porous materials as catalysts. In the past decades, great progress has been made in various fields related to molecular sieves and porous materials, such as synthesis and catalysis, structural chemistry, adsorption and diffusion, characterization, and porous composite chemistry. In particular, the overlap of molecular sieve science with other related sciences, including physics, mathematics, computer science, materials science, and biology, has promoted the in-depth development of the chemistry of molecular sieves and porous materials.

16

Chemistry of Zeolites and Related Porous Materials

References [1] M.E. Davis, Ordered Porous Materials for Emerging Applications. Nature (London), 2002, 417, 813–821. [2] C.H. Baerlocher, W.M. Meier, and D.H. Olson, Atlas of Zeolite Framework Types, 5th Edn. Elsevier, Amsterdam, 2001. [3] R.M. Barrer, Hydrothermal Chemistry of Zeolites Academic Press, London, 1982. [4] Dalian Institute of Chemical Physics, Chinese Academy of Sciences. Zeolite Molecular Sieves. Chemistry of Zeolites and Related Porous Materials – Synthesis and Structure Rureh Xu, Wenqin Pang, Jihong Yu, Qisheng Huo and Jiesheng Chen Science Press, Beijing, 1978. [5] E.M. Flanigen, Molecular Sieve Zeolite Technology - The First Twenty-five Years. Proceedings of the Fifth International Conference on Zeolites, ed. L.V.C. Rees, Heyden, London, 1980, 760–780. [6] S.T. Wilson, B.M. Lok, and E.M. Flanigen, US Patent 4,310,440 (1982). [7] E.M. Flanigen, B.M. Lok, R.L. Patton, and S.T. Willison, Aluminophosphate Molecular Sieves and the Periodic Table. In ‘New Developments in Zeolite Science and Technology.’ Proceedings of the 7th International Zeolite Conference, ed. Y. Murakam, A. Lijima, and J.W. Ward, Kodansha - Elsevier, Tokyo, 1986, 103–112. [8] J.H. Yu and R.R. Xu, Rich Structure Chemistry in the Aluminophosphate Family. Acc. Chem. Res. 2003, 36, 481–490. [9] A.K. Cheetham, G. Fe´rey, and T. Loiseau, Open-Framework Inorganic Materials. Angew. Chem., Int. Ed., 1999, 38, 3268–3292. [10] M.E. Davis, C. Saldarriaga, C. Montes, J.M. Garces, and C. Crowder, A molecular sieve with Eighteen - membered Rings. Nature (London), 1988, 331, 698–699. [11] C.T. Kresge, M.E. Leonowicz, W.J. Roth, J.C. Vartuli, and J.S. Beck, Ordered Mesoporous Molecular Sieves Synthesized by a Liquid-crystal Template Mechanism. Nature (London), 1992, 359, 710–712. [12] J.S. Beck, J.C. Varuli, W.J. Roth, M.E. Leonowicz, C.T. Kresge, K.D. Schmitt, C.T.W. Chu, D.H. Olson, E.W. Sheppard, S.B. McCullen, J.B. Higgins, and J.L. Schlenker, A New family of Mesoporous Molecular Sieves prepared with Liquid Crystal Templates. J. Am. Chem. Soc., 1992 114, 10834–10843. [13] T. Maschmeyer, Derivatised Mesoporous Solids. Curr. Opin. Solid State Mater. Sci., 1998, 3, 71–78. [14] O.D. Velev, T.A. Jede, R.F. Lobo, and A.M. Lenhoff, Microstructured Porous Silica Obained via Colloidal Crystal Template. Chem. Mater., 1998, 10, 3597–3602. [15] S.A. Davis, S.L. Burkett, N.H. Mendelson, and S. Mann, Bacterial Templating of Ordered Macrostructures in Silica and Silica-Surfactant Mesophases. Nature (London), 1997, 385, 420–423. [16] A. Imhof and D.J. Pine, Ordered Macroporous Materials by Emulsion Templating. Nature (London), 1997, 389, 948–951. [17] B. Chen, M. Eddaoudi, S.T. Hyde, M. O’Keeffe, and O.M. Yaghi, Interwoven Metal-organic Framework on a Periodic Minimal Surface with Extra-large Pores. Science, 2001, 291, 1021–1023. [18] M. Eddaoudi, J. Kin, N. Rosi, D. Vodak, J. Wachter, M. O’Keeffe, and O.M. Yaghi, Systematic Design of Pore Size and Functionality in Isoreticular MOFs and their Application in Methane Storage. Science, 2002, 295, 469–472. [19] C. Marcilly. Evolution of Refining and Petrochemicals, What in the Place of Zeolites. Stud. Surf. Sci. Catal., 2001, 135, 37–60. [20] M.E. Davis, New Vistas in Zeolite and Molecular Sieve Catalysis. Acc. Chem. Res., 1993, 26, 111–115.

Introduction

17

[21] J. Weitkamp, Host/Guest Chemistry and Catalysis in Zeolites. Proceedings of the 9th International Zeolite Conference, ed. R. Von Ballmoos, J.B. Higgins, and M.M.J. Treacy, Butterworth-Heinemann, Montreal, 1992, 13–46. [22] M. Wark, M. Ganschow, Y. Rohlfing, G. Schulz-Ekloff, and D. Wo¨hrle, Methods of Synthesis for the Encapsulation of Dye Molecules in Molecular Sieves. Stud. Surf. Sci. Catal., 2001, 135, 180. ¨ . Weiss, F. Schu¨th, L. Benmohammadi, and F. Laeri, Potential Microlasers Based on ALPO4[23] O 5/DCM Composites. Stud. Surf. Sci. Catal., 2001, 135, 161. [24] M.R. Steele, A.I. Holms, and G.A. Ozin, Stepwise Synthesis of II-IV Nanoclusters inside Zeolite Y Supercages using MOCVD type Precursors. Proceeding of the 9th International Zeolite Conference, Part II, ed. R. Von Ballmoos, J.B. Higgins and M.M.J. Treacy, ButterworthHeinemann, Montreal, 1992, 185–192. [25] R. Pool. Science, 1994, 263, 1698–1699. [26] C.G. Wu and T. Bein, Conducting Polyaniline Filaments in a Mesoporous Channel Host. Science, 1994, 264, 1757–1759. [27] H. Fan, H.R. Beutley, K.R. Kathan, P. Clem, Y. Lu, and C.J. Brinker, Self-assembled Aerosollike low Dielectric Constant Films. J. Non-Cryst. Solids, 2001, 285, 79–83. [28] D. Zhao, J. Feng, Q. Huo, N. Melosh, G.H. Fredrickson, B.F. Chmelka, and G.D. Stucky, Triblock Copolymer Syntheses of Mesoporous Silica with Periodic 50 to 300 Angstrom Pores. Science, 1998 279; 548–552. [29] A. Corma, From Microporous to Mesoporous Molecular Sieve Materials and their Use in Catalysis. Chem. Rev., 1997, 97, 2373–2419. [30] P.J. Nat, Wogt E T C. WO 94/26,847 (1994). [31] R. Ryoo, J.M. Kim, and C.H. Shin, Disordered Molecular Sieve with Branched Mesoporous Channel Network. J. Phys. Chem., 1996, 100, 17718–17723. [32] R. Mokaya, Ultrastable Mesoporous Aluminosilicates by Grafting Routes. Angew. Chem., Int. Ed., 1999, 38, 2930–2934. [33] R. Ryoo, S. Jun, J.M. Kim, and M.J. Kim, Generalised Route to the Preparation of Mesoporous Metallosilicates via Post-synthetic Metal Implantation. Chem. Commun., 1997, 2225–2226. [34] S.S. Kim, W. Zhang, and T.J. Pinnavaia, Ultrastable Mesostructured Silica Vesicles. Science, 1998, 282, 1032. [35] A. Karlsson, M. Stocker, and R. Schmidt, Composites of Micro- and Mesoporous Materials: Simultaneous Syntheses of MFI/MCM-41-like Phases by a Mixed Template Approach. Microporous Mesoporous Mater., 1999, 27, 181–192. [36] L. Huang, W. Guo, P. Deng, Z. Xue, and Q. Li, Investigation of Synthesizing MCM-41/ZSM-5 Composites. J. Phys. Chem. B, 2000, 104, 2817–2823. [37] K.R. Kloetstra, H. Van Bekkum, and R.J. Jansen, Mesoporous Material containing Framework Tectosilicate by Pore-wall Recrystallization. Chem. Commun., 1999, 2281–2282. [38] Y. Han, D. Li, L. Zhao, J. Song, X. Yang, N. Li, Y. Di, C. Li, S. Wu, X. Xu, X. Meng, K. Lin, and F. Xiao, High-temperature Generalized Synthesis of Stable Ordered Mesoporous Silicabased Materials using Fluorocarbon-Hydrocarbon Mixtures. Angew. Chem., Int. Ed., 2003, 42, 3633–3637. [39] Y. Liu, W. Zhang, and T.J. Pinnavaia, Steam-stable Aluminosilicate Mesostructures assembled from Zeolite Type Y Seeds. J. Am. Chem. Soc., 2000, 122, 8791–8792. [40] Z. Zhang, Y. Han, F. -S, Xiao, S. Qiu, L. Zhu, R. Wang, Y. Yu, Z. Zhang, B. Zou, Y. Wang, H. Sun, D. Zhao, and Y. Wei, Strongly Acidic and High-temperature Hydrothermally Stable Mesoporous Aluminosilicates with Ordered Hexagonal Structure. Angew. Chem., Int. Ed., 2001, 40, 1256–1258. [41] Z. Zhang, Y. Han, F. -S, Xiao, S. Qiu, L. Zhu, R. Wang, Y. Yu, Z. Zhang, B. Zou, Y. Wang, H. Sun, D. Zhao, and Y. Wei, Mesoporous Aluminosilicates with Ordered Hexagonal Structure,

18

[42]

[43] [44]

[45] [46] [47] [48]

[49] [50] [51]

[52] [53] [54] [55] [56] [57]

Chemistry of Zeolites and Related Porous Materials Strong Acidity, and Extraordinary Hydrothermal Stability at High Temperatures. J. Am. Chem. Soc., 2001, 123, 5014–5021. F.S. Xiao, Y. Han, Y. Yu, X. Meng, M. Yang, and S. Wu, Hydrothermally Stable Ordered Mesoporous Titanosilicates with Highly Active Catalytic Sites. J. Am. Chem. Soc., 2002, 124, 888–889. F. Marlow, M.D. McGehee, D.Y. Zhao, B.F. Chmelka, and G.D. Stucky, Doped Mesoporous Silica Fibers: A New Laser Material. Adv. Mater., 1999, 11, 632–636. B.J. Scott, G. Wirnsberger, M.D. McGehee, B.F. Chmelka, and G.D. Stucky, Dye-doped Mesostructured Silica as a Distributed Feedback Laser Fabricated by Soft Lithography. Adv. Mater., 2001, 13, 1231–1234. B.J. Scott, G. Wirnsberger, and G.D. Stucky, Mesoprous and Mesostructured Materials for Optical Applications. Chem. Mater., 2001, 13, 3140–3150. J. Jang, J.H. Oh, and G.D. Stucky, Fabrication of Ultrafine Conducting Polymer and Graphite Nanoparticles. Angew. Chem., Int. Ed., 2002, 41, 4016–4019. R. Ryoo, S.H. Joo, and S. Jun, Synthesis of Highly Ordered Carbon Molecular Sieves via Template-mediated Structural Transformation. J. Phys. Chem. B, 1999, 103, 7743–7746. S.H. Joo, S.J. Choi, I. Oh, J. Kwak, Z. Liu, O. Terasaki, and R. Ryoo, Ordered Nanoporous Arrays of Carbon Supporting High Dispersions of Platinium Nanoparticles. Nature (London), 2001, 414, 470. Kh.M. Minachev and Ya.I. Isakov, Catalytic Properties of Zeolites – A General in Review, Molecular Sieves- ed. W.M. Meier and J.B. Uytlerhoeven, ACS Symp. Ser. 121, 1973, 451–460. W.O. Haag, Catalysis by Zeolites-Science and Technology. Stud. Surf. Sci. Catal., 1994, 84B, 1375–1394. N.Y. Chen and W.E. Garwood, Molecular Shape-Selective Hydrocarbon Conversion over Erionite. In ‘Molecular Sieves’, ed. W.M. Meier and J.B. Uytlerhoeven, ACS Symp. Ser. 121, 1973, 575–582. C. Naccache and Y B. Tarrit, Recent Developments in Catalysis by Zeolites. Proceedings of the 5th International Conference on Zeolites, ed. L.V.C. Rees, Heyden, London, 1980, 592–606. W.F. Holderich, Zeolites: Catalysts for the Synthesis of Organic Compounds. Stud. Surf. Sci. Catal., 1988, 49A, 69–94. P.B. Venuto, Organic Catalysis over Zeolites: A Perspective on Reaction Paths within Micropores. Microponus Mater., 1994, 2, 297–411. M. Iwamoto, Zeolites in Environmental Catalysis. Stud. Surf. Sci. Catal., 1994 84B, 1395–1410. T. Inui, High Potential of Novel Zeolitic Materials as Catalysts for Solving Energy and Environmental Problems. Stud. Surf. Sci. Catal., 1997, 105B, 1441–1468. B. Wichterlova´, J. Deˇdea˘k, and Z. Sobalik, Redox Catalysis over Molecular Sieves. Structure and Function-active Site. Proceedings of the 12th International Zeolite Conference, Part II, ed. M.M.J. Treacy, B.K. Marcus, M.E. Bisher, and J.B., Higgins MRS, Warrendale, PA, 1998 941–973.

2 Structural Chemistry of Microporous Materials 2.1 Introduction Microporous materials with regular pore architectures comprise wonderfully complex structures and compositions.[1,2] Their fascinating properties, such as ion-exchange, separation, catalysis, and their roles as hosts in nanocomposite materials, are essentially determined by their unique structural characters, such as the size of the pore window, the accessible void space, the dimensionality of the channel system, and the numbers and sites of cations, etc. Zeolites constitute the most important family in microporous materials. Traditionally, the term ‘zeolite’ refers to a crystalline aluminosilicate or silica polymorph based on corner-sharing TO4 (T ¼ Si and Al) tetrahedra forming a three-dimensional fourconnected framework with uniformly sized pores of molecular dimensions. Nowadays, the term ‘zeolite framework’ generally refers to a corner sharing network of tetrahedrally coordinated atoms. The 5th Edition of the Atlas of Zeolite Framework Types published by Baerlocher, Meier, and Olson on behalf of the Structure Commission of the International Zeolite Association in 2001 comprised 133 zeolite structure types.[3] Up to October 2006, the number of entries had risen to 167.[4] Table 2.1 summarizes all zeolite-type materials including silicates (germanates), phosphates (arsenates), and both silicates and phosphates. Each framework type is assigned a three-capital-letter code in alphabetical order. The codes are generally derived from the names of the type materials. They only describe and define the network of corner sharing tetrahedrally coordinated framework atoms. Framework types do not depend on composition, distribution of the T-atoms (Si, Al, P, Ga, Ge, B, Be, etc.), cell dimensions, or symmetry. Table 2.1 also lists some classical type materials.

Chemistry of Zeolites and Related Porous Material – Synthesis and Structure Ruren Xu, Wenqin Pang, Jihong Yu, Qisheng Huo and Jiesheng Chen # 2007 John Wiley & Sons, (Asia) Pte Ltd

20

Chemistry of Zeolites and Related Porous Materials

Table 2.1 Structure types of zeolites Structure-type code

Type material

ABW ACO AEI AEL AEN AET AFG AFI AFN AFO AFR AFS AFT AFX AFY AHT ANA

Li-A(BW) ACP-1 AlPO-18 AlPO-11 AlPO-EN3 AlPO-8 Afghanite AlPO-5 AlPO-14 AlPO-41 SAPO-40 MAPSO-46 AlPO-52 SAPO-56 CoAPO-50 AlPO-H2 Analcime

APC APD AST ASV ATN ATO ATS ATT ATV AWO AWW BCT *BEA BEC BIK BOG BPH BRE CAN

AlPO-C AlPO-D AlPO-16 ASU-7 MAPO-39 AlPO-31 MAPO-36 AlPO-12-TAMU AlPO-25 AlPO-21 AlPO-22 Mg-BCTT b FOS-5 Bikitaite Boggsite Beryllopho-sphate-H Brewsterite Cancrinite

CAS CDO CFI CGF CGS

Cs Aluminosilicate(A) CDS-1 CIT-5 Co-Ga-phosphate-5 Co-Ga-phosphate-6

CHA

Chabazite

-CHI -CLO

Chiavennite Cloverite

Selected isotypes BePO-ABW, ZnASO-ABW MnAPO-11 AlPO-53, CFSAPO-1A, JDF-2, MSC-1 MCM-37 SSZ-24, CoAPO-5, SAPO-5 GaPO-14 CoAPSO-40, ZnAPSO-40 SSZ-16 MgAPO-50 Leucite, Wairakite, AlPO-24, GaGeO-ANA AlPO-H3 Octadecasil SAPO-31 AlPO-33 GaPO-ATV GaPO-AWO BSiO-BEA, GaSiO-BEA, CIT-6 ITQ-17, ITQ-14 overgrowth CsASiO-BIK Linde Q, STA-5 CIT-4 ECR-5, AlGeO-CAN, GaSiO-CAN, ZnPO-CAN MCM-65, UZM-25 ZnGaPO-CGS, TNU-1, GaSiO-CSG, TsG-1 AlPO-34, GaPO-3 MeAPO-47, CoAPO-44, CoAPO-47

S/Pa S/P P P P P P S S/P P P P P P P P P S/P P P S/P S P P P P P P P S S S S S S/P S S S S S P S/P S/P S P

Structural Chemistry of Microporous Materials Table 2.1 (Continued ) Structure-type code CON CZP

Type material

DAC DDR DFO DFT

CIT-1 ‘Chiral’ Zn phosphate Dachiardite Decadodecasil-3R DAF-1 DAF-2

DOH DON EAB EDI EMT EON EPI ERI ESV ETR EUO FAU

Dodecasil-1H UTD-1 TMA-E(AB) Edingtonite EMC-2 ECR-1 Epistilbite Erionite ERS-7 ECR-34 EU-1 Faujasite

FER FRA GIS

Ferrierite Franzinite Gismondine

GIU GME GON GOO HEU IFR IHW ISV ITE ITH ITW IWR IWW JBW KFI LAU LEV LIO -LIT LOS LOV LTA LTL LTN MAR

Giuseppettite Gmelinite GUS-1 Goosecreekite Heulandite ITQ-4 ITQ-32 ITQ-7 ITQ-3 ITQ-13 ITQ-12 ITQ-24 ITQ-22 Na-J(BW) ZK-5 Laumontite Levyne Liottite Lithosite Losod Lovdarite Linde Type A Linde Type L Linde Type N Marinellite

Selected isotypes

S/Pa

SSZ-26/SSZ-33

S P

Svetlozarite s, ZSM-58

S S P S/P

ACP-3, UCSB-3GaGe, UCSB-3ZnAs, UiO-20 UTD-1F Bellbergite K-F, CoAPO-EDI, CoGaPO-EDI CSZ-1, ECR-30, ZSM-20 TNU-7 AlPO-17, LZ220 TPZ-3, ZSM-50 X, Y, SAPO-37, ZnPO-X, AlGeO-FAU, CoAlPO-FAU, GaBeO-FAU Sr-D, FU-9, NU-23, ZSM-35 Na-P, MAPO-43, ZnGaPO-GIS (NH4)4[Zn4B4P8O32]GIS

Clinoptilolite, LZ219 SSZ-42, MCM-58

P, Q Leonhardite, CoGaPO-LAU LZ-132, NU-3, SAPO-35, CoDAF-4 LiBePO-LOS, AlGeO-LOS SAPO-42, ZK-4, GaPO-LTA (K, Ba)-G, L, Perlialite NaZ-21

S S S S/P S S S S/P S S S S/P S S S/P S S S S S S S S S S S S S S S/P S/P S S S/P S S/P S S S

21

22

Chemistry of Zeolites and Related Porous Materials

Table 2.1 (Continued) Structure-type code

Type material

MAZ MEI MEL MEP MER MFI MFS MON MOR

Mazzite ZSM-18 ZSM-11 Melanophlogite Merlinoite ZSM-5 ZSM-57 Montesommaite Mordenite

MOZ MSO MTF MTN MTT MTW MWW NAB NAT NES NON NPO NSI OBW OFF OSI OSO OWE -PAR PAU PHI PON RHO -RON RRO RSN RTE RTH RUT RWR RWY SAO SAS SAT SAV SBE SBS SBT SFE SFF

ZSM-10 MCM-61 MCM-35 ZSM-39 ZSM-23 ZSM-12 MCM-22 Nabesite Natrolite NU-87 Nonasil Nitridophosphate-1 Nu-6(2) OSB-2 Offretite UiO-6 OSB-1 UiO-28 Partheite Paulingite Phillipsite IST-1 r Roggianite RUB-41 RUB-17 RUB-3 RUB-13 RUB-10 RUB-24 UCR-20 STA-1 STA-6 STA-2 Mg-STA-7 UCSB-8Co UCSB-6GaCo UCSB-10GaZn SSZ-48 SSZ-44

Selected isotypes
, ZSM-4 Silicalite-2, SSZ-46, TS-2 K-M, W, AlCoPO-MER Silicalite-1, TS-1 Zeolon, [Ga-Si-O]-MOR, Ca-Q LZ-211, Na-D

S/Pa S S S S S/P S S S S

S S UTM-1 S Dodecasil 3C, CF-4 S EU-13, KZ-1 S CZH-5, VS-12 S ERB-1, ITQ-1, PSH-3, SSZ-25 S S Scolecite, AlGeO-NAT, GaSiO-NAT S Gottardiite S ZSM-51, Borosilicate NON S P S S TMA-O, LZ-217 S S S ACP-2 P S ECR-18 S Harmotome, ZK-19, AlCoPO-PHI S/P P LZ-214, CoAlPO-RHO, BeAsO-RHO S/P S S S S S NU-1, TMA-silicate-RUT S S chalcogenide P P P CoSTA-7, ZnSTA-7 P UCSB-8Mg, UCSB-8Mn, UCSB-8Zn P UCSB-6Co, UCSB-6GaMg, UCSB-6Mn P UCSB-10Co, UCSB-10Mg, UCSB-10Zn P S S

Structural Chemistry of Microporous Materials

23

Table 2.1 (Continued) Structure-type code

Type material

SFG SFH SFN SFO SGT SOD

SSZ-58 SSZ-53 SSZ-59 SSZ-51 s Sodalite

SOS SSY STF STI STT TER THO TON TSC UEI UFI UOZ USI UTL VET VFI VNI VSV WEI -WEN YUG ZON

SU-16 SSZ-60 SSZ-35 Stilbite SSZ-23 Terranovaite Thomsonite y-1 Tscho¨rtnerite Mu-18 UZM-5 IM-10 IM-6 IM-12 VPI-8 VPI-5 VPI-9 VPI-7 Weinebeneite Wenkite Yugawaralite ZAPO-M1

a

Selected isotypes

AlPO-20, AlCoPO-SOD, AlGeO-SOD GaCoPO-SOD, ZnPO-SOD FJ-17 ITQ-9 Barrerite, Stellerite AlCoPO-THO, GaCoPO-THO ISI-1, KZ-2, NU-10, ZSM-22

ITQ-15 AlPO-54, MCM-9, H1 Gaultite Sr-Q GaPO-DAB, UiO-7

S/Pa S S S P S S/P P S S S S S S/P S S P S S P S S P S S S S S P

S: Silicate (germanate); P: phosphate (arsenate).

Besides zeolites, a diverse range of microporous materials with novel open-framework structures have been discovered. The framework atoms of microporous materials have expanded to include most of the elements in the periodic table.[5] The framework elements are not limited to Al and Si atoms alone, and the primary building units are not only confined to tetrahedra. This chapter will mainly describe the structural characteristics of zeolites and some zeolitic open-framework materials.

2.2 Structural Building Units of Zeolites 2.2.1

Primary Building Units

Zeolite comprises of TO4 tetrahedra through corner sharing giving rise to a threedimensional four-connected framework. Framework T atoms generally refer to Si, Al, or P atoms. In some cases, other atoms such as B, Ga, Be, and Ge, etc., are also involved.

24

Chemistry of Zeolites and Related Porous Materials

Figure 2.1 (a) TO4 tetrahedron; (b) TO4 tetrahedra sharing a common oxygen vertex

These [SiO4], [AlO4], or [PO4] tetrahedra are the basic structural building units of a zeolite framework. The primary building units are TO4 tetrahedra. In a zeolite, each T atom is coordinated to four oxygen atoms [Figure 2.1(a)], with each oxygen atom bridging two T atoms [Figure 2.1(b)], so the structure type of a zeolite can be described as the (4;2)-connection. However some zeolites, such as AlPO4-21 and VPI-5, contain five- or six-coordinated Al atoms with one or two extra-framework oxygen species (OH or H2O), as well as four-coordinated Al atoms. By omitting the OH and H2O species, these frameworks have a hypothetical (4;2)-connected framework. The aluminosilicate zeolites constructed from SiO4 tetrahedra and AlO4 tetrahedra possess an anionic framework, the negative charge of which is compensated by extraframework cations. The empirical formula of an aluminosilicate zeolite can be expressed as Ax/n[Si1
xAlxO2].mH2O, where A is a metal cation of valence n. The cations and adsorbed water molecules are located in the channels or cages. The aluminophosphate molecular sieves built up from strict alternation of AlO4 and PO4 tetrahedra through corner sharing possess a neutral framework,[6] in which no extra metal cations exist but only adsorbed water molecules or templating molecules accommodate in the channels. The structures of aluminosilicate zeolites obey the Lo¨wenstein’s Rule[7] with an avoidance of Al

O

Al linkages. Similarly, the linkages P

O

P, P

O

Si, Al

O

Al, Me

O

Al, and Me

O

Me (where Me ¼ metal) appear to be unlikely in aluminophosphate-based molecular sieves.[8] 2.2.2

Secondary Building Units (SBUs)

The framework of a zeolite can be thought of as being made of finite component units or infinite component unit-like chains and layers. The concept of infinite component units, such as secondary building units (SBUs), was introduced by Meier[9,10]and Smith.[11] 18 kinds of SBUs that have been found to occur in tetrahedral frameworks[3] are shown in Figure 2.2. These SBUs, which contain up to 16 tetrahedrally coordinated atoms (Tatoms) are derived by assuming that the entire framework is made up of one type of SBU only. It should be noted that SBUs are invariably nonchiral. A unit cell always contains an integral number of SBUs.[12] The SBUs for various structure types of zeolites given in the database of zeolite structures[4] are summarized in Table 2.2. One type of framework can comprise several SBUs. For example, the LTA framework contains five types of SBUs, including 4, 8, 4-2, 4-4, and 6-2 units, any of which can be used to describe its framework structure. In some instances, combinations of SBUs have been encountered, i.e., the framework cannot be generated by only one type of SBU. Examples include LOV, MEP, and other clathrasiltype frameworks.

Structural Chemistry of Microporous Materials

25

Figure 2.2 Secondary building units (SBUs). The SBU codes are given below the figures. Reproduced with permission from [3]. Copyright (2001) Elsevier

It should be noted that the SBUs are only theoretical topological building units and should not be considered to be or equated with species that may be in the solution/gel during the crystallization of a zeolitic material. 2.2.3

Characteristic Cage-building Units

There are some characteristic cage-building units in zeolite frameworks. Cages are generally described in terms of the n-rings defining their faces. For example, a truncated octahedron (sodalite unit), whose surface is defined by six 4-rings and eight 6-rings, would be designated a [4668] cage. Smith defined the names of parts of cages.[11] It should be noted that a polyhedral pore, which has at least one face defined by a ring large enough to be penetrated by guest species, but which is not infinitely extended (i.e., not a channel), is called a cavity[13] according to IUPAC recommendations. For example, the [4126886] polyhedron in zeolite LTA, traditionally called an a cage, is actually a cavity. In this book the polyhedra are described in the traditional way.

26

Chemistry of Zeolites and Related Porous Materials

Table 2.2 SBUs found in zeolite structures Code

SBU

Code

SBU

Code

ABW ACO AEI AEL AEN AET AFG AFI AFN AFO AFR AFS AFT AFX AFY AHT ANA APC APD AST ASV ATN ATO ATS ATT ATV AWO AWW BCT *BEA BEC BIK BOG BPH BRE CAN CAS CDO CFI CGF

4, 8 4, 4-4, 8 4, 6, 4-2, 6-6 10, 4-1 4, 6 6 4, 6 4, 6, 12 4, 8 4-1, 2-6-2 4, 6-2, 4-4 6*1 4, 6, 6-6, 4-2 4, 6, 6-6, 4-2 4, 4-4 4-2, 6 4, 6, 6-2, 4-[1,1], 1-4-1 4, 8 4, 8, 6-2 4-1 4-1 4, 8 4, 6, 12 4, 6, 12 4-2, 6 4-[1,1], 6 6, 4-2, 4 4, 6 4, 8 combination 6-2 5-1 4, 6, 5-1 6*1 4 4, 6, 12 5-1 5-1 5-[1,1,1] 4-1-1

ERI ESV ETR EUO FAU FER FRA GIS GIU GME GON GOO HEU IFR IHW ISV ITE ITH ITW IWR IWW JBW KFI LAU LEV LIO -LIT LOS LOV LTA LTL LTN MAR MAZ MEI MEL MEP MER MFI MFS

4, 6 5-1 4 1-5-1 4, 6, 6-6, 6-2, 4-2, 1-4-1 5-1 6, 4 4, 8 4, 6 4, 6, 8, 12, 4-2, 6-6 5-3 -4-44-4¼1 6-2 combination 6-2 4 combination 1-4-1, 4-[1,1] 1-5-1 1-5-1 6 4,6, 8, 6-6, 6-2, 4-2 6, 1-4-1 6 6, 4 6, 4-[1,1], 4-2 4, 6, 6-2 combination 4, 6, 8, 1-4-1, 4-4, 6-2 6, 4-2 6, 4-2 6, 4 4-2, 5-1 combination 5-1 combination 4, 8, 8-8 5-1 combination

OSI OSO OWE -PAR PAU PHI PON RHO -RON RRO RSN RTE RTH RUT RWR RWY SAO SAS SAT SAV SBE SBS SBT SFE SFF SFG SFH SFN SFO SGT SOD SOS SSY STF STI STT TER THO TON TSC

CGS CHA -CHI -CLO CON CZP DAC DDR DFO DFT DOH DON EAB EDI EMT EON EPI

4 4, 6, 6-6, 4-2 5-[1,1] 4, 4-4 5-2 4, 4-[1,1] 5-1 combination combination 4 combination 5-3 4, 6 4¼1 4,6, 6-6, 6-2, 4-2, 1-4-1 5-1 5-1

MON MOR MOZ MSO MTF MTN MTT MTW MWW NAB NAT NES NON NPO NSI OBW OFF

4 5-1 combination 2-6-2, 4-1 5-5¼1 combination 5-1 5-[1,1] combination 4-1, 1-3-1 4¼1 combination combination 3 5-1 combination 6, 4-2

UEI UFI UOZ USI UTL VET VFI VNI VSV WEI -WEN YUG ZON

SBU 6-2 combination 4, 4-44 4, 8 4, 8 4-2 4, 6, 8, 8-8 combination 4-4¼1 combination 6, 5-1 4 6 6-2 8, 3*1 4 4, 6-2 6, 4 4, 6, 4-2, 6-6 4, 8 4, 8 4 combination 5-3 combination 5-3 5-3 6-2, 4-4-, 4 5-3 6 4-2 combination 5-3 4-4¼1 5-3 2-6-2, 4-1 4¼1 5-1 4, 6, 8, 4-2, 6-6, 8-8 4, 6, 4-2 8 4-1 4-1 combination combination 18, 6, 4-2 combination combination spiro-5 combination 4, 8 4, 6-2, 4-4-

Structural Chemistry of Microporous Materials

Figure 2.3 Cage-building units in known zeolites

27

28

Chemistry of Zeolites and Related Porous Materials

Figure 2.3 (Continued)

Structural Chemistry of Microporous Materials

29

Table 2.3 Cage-building units in zeolites Zeolite ACO AEI AFG AFS AFT AFX AFY AST ASV ATN AWW BEC BPH CAN CHA -CLO DDR DFO DOH EAB EMT ERI ESV FAU FRA GIU GME HEU ISV ITE ITH ITW IWW KFI LEV

Cage number 2 5, 39 13, 49 51 5, 17, 35, 56 5, 17, 56 2 2, 31 2 20 10, 30 2 8 13 5, 35 2, 30, 54, 61 11, 16, 33 2, 34, 59 15, 16, 36, 5, 17, 40 5, 19, 38, 57 5, 13, 48 14, 27 5, 19, 55 13, 19, 26 13, 19, 62 5, 17 2 2 6, 45 2 2 2 5, 32, 54 5, 25

Zeolite LIO LOS LTA LTL LTN MAR MAZ MEI MEP MER MSO MTN MWW OBW OFF PAU RHO RTE RTH RUT RWY SAO SAS SAT SAV SBE SBS SBT SFG SGT SOD TSC UEI UFI -WEN

Cage number 13, 26, 49 13, 26 2, 19, 54 5, 13, 53 13, 19, 24, 54 13, 19, 49 17, 28 4, 7, 52 16, 18 9, 32 5, 50 16, 23 5 63, 64 5, 13, 17 9, 32, 54 9, 54 12, 41 6, 46 12, 22 58 10 5, 43 5, 13, 47 5, 29, 44 9, 20 5, 13 5, 13 3 7, 37 19 5, 9, 19, 60 42 2, 6, 21, 54 5, 13

The cages occuring in the known zeolite frameworks are shown in Figure 2.3. Table 2.3 summarizes the cage-building units in the zeolite frameworks. Different zeolite frameworks may feature the same cage building unit, that is to say, the same cage-building unit may construct different framework types via different linkages. For example, starting from the SOD cage, the SOD structure is obtained when b-cages are linked by sharing 4-rings; the LTA structure is obtained when b-cages are linked through double 4-rings; FAU and EMT structures are obtained when b-cages are linked through double 6-rings (Figure 2.4).[14] 2.2.4

Characteristic Chain- and Layer-building Units

Some characteristic chain building units frequently occur in the zeolite frameworks. Figure 2.5 shows five types of chain units, which are the double zig-zag chain, double

30

Chemistry of Zeolites and Related Porous Materials

Image Not Available In The Electronic Edition

sawtooth chain, double crankshaft chain, narsarsukite chain, and pentasil chain, respectively.[15] The three double chains are all composed of edge-sharing 4-rings but with different orientations (up or down) of the fourth connection of a tetrahedron. The narsarsukite chain is found more often in AlPO4-based structures, whereas the pentasil chain, composed of edge-sharing [58] cages, is characteristic of the family of high-silica zeolites such as MFI and MEL. The zeolite structures may also be described by two dimensional (2-D) 3-connected nets.[11] The most obvious way to generate the four-connected 3-D net from a 2-D threeconnected net is to arrange one type of 2-D three-connected net into a parallel stack and to link each vertex to only one other vertex. To form a 3-D net, some of the new edges must point upward and some downward from each 2-D net. Figure 2.6 presents the 4.82

Figure 2.5 Some chains that occur in several zeolite framework types. (a) Double zig-zag chain; (b) double sawtooth chain; (c) double crankshaft chain; (d) narsarsukite chain; (e) Pentasil chain. Reproduced with permission from [15]. Copyright (2001) Elsevier

Structural Chemistry of Microporous Materials

31

Figure 2.6 (a) GIS framework; (b) 4.82 net

2-D net found in the GIS framework type, where the connections from half of each 8-ring point up (U) and the other half point down (D). A 2-D net is denoted by the size of three circuits that meet at the different types of nodes. For instance, as shown in Figure 2.6(b), the 4.82 2-D net in the GIS framework associates with one 4-ring and two 8-rings at each node. Figure 2.7 shows several typical 2-D nets occurring in the zeolite frameworks. Many 3-D frameworks can be described in terms of the same 2-D 3-connected net. For example, the 4.82 net is found not only in the GIS framework type, but also in ABW, BRE, MER, PHI, ATT, and APC framework types. It is noted that the orientation of tetrahedra around the 8-rings in the 4.82 net in each framework is different. For example, the ABW framework has a UUDUDDUD orientation of tetrahedra around the 8-rings,

Figure 2.7 Several common 2-D 3-connected nets in zeolite frameworks. (a) 4.82; (b) (4.6.8)1(6.8.8)1; (c) 4.6.10; (d) 4.6.12; (e) (4.6.8)1(4.8.12)1

32

Chemistry of Zeolites and Related Porous Materials

Figure 2.8 PBU in AFI, seen along c, (a) constructed from six crankshaft chains and (b) from 6-ring bands, which can be built using two 6-fold-connected 12-rings. Reproduced from www.iza-structure.org

whereas the GIS framework has a UUUUDDDD orientation of tetrahedra around the 8-rings. This variation explains the diversity of structures constructed from a stacking of one type of 2-D net. The 4.6.12 net shown in Figure 2.7(d) is one of the most common sheets in the 2-D 3-connected nets that can be found in many zeolite framework types, for instance, AFG, CAN, CHA, ERI, GME, LEV, LIO, OFF, EAB, LOS, AFI, AFS, and AFY, and so forth. 2.2.5

Periodic Building Units (PBUs)

The PBUs are built from smaller units composed of a limited number of T-atoms, by applying a simple operation to the smaller unit, e.g. translation, rotation.[4] The zeolite framework types can be analysed in terms of these component PBUs. The infinite PBUs, such as (multiple) chains, tubes, and layers, and the finite PBUs, such as (double) 4-rings, (double) 6-rings, and cages are far from unique. However, they are common to several zeolite framework types and allow an easy description of the frameworks to be made. Infinite PBUs and finite PBUs can be used to build the zeolite frameworks (see details in the database of zeolite structures: Schemes for Building Zeolite Structure Models).[4] Here only an example from AFI is presented to show the building of zeolites. AFI can be built using the crankshaft chain [bold in Figure 2.8(a)] running along c. The repeat unit consists of 4 T atoms. Six such chains are connected around a 6-fold axis forming a periodic building unit (PBU). In Figure 2.8(b), the PBU is a cylinder surrounded by 12 T atoms, and its wall is solely composed of T6-rings. The repeat unit of the PBU is a cylindrical T6-ring band made of 24 T atoms. As shown in

Figure 2.9 Connection mode and unit-cell content in AFI along c in perspective view (left) and in parallel projection (top right). For clarity, only 1½ repeat units of the PBUs along c are drawn. A double crankshaft chain (consisting of 2-fold-connected double 4-rings), and 3-foldconnected double 6-rings are indicated in bold. Reproduced from www.iza-structure.org

Structural Chemistry of Microporous Materials

33

Figure 2.9, adjacent PBUs are connected around a 3-fold axis through T4-rings, giving rise to the framework of AFI.

2.3 Composition of Zeolites 2.3.1

Framework Composition

The aluminosilicate zeolite framework comprises SiO4 and AlO4 tetrahedra. It is difficult to determine the distribution or arrangement of Si and Al atoms in the zeolite framework by using conventional structural characterization methods due to their similar ionic radii and electronic shell arrangement of ions. However, one criterion which should be obeyed is that the linkage of two tetrahedral Al atoms is forbidden due to the restriction of Lo¨wenstein’s Rule.[7] Therefore, one Al atom can only connect with four adjacent Si atoms, denoted as Al(4Si). A Si atom can link with either a Si atom or an Al atom, so the coordination enviroments of Si involve Si(0Al, 4Si), Si(1Al, 3Si), Si(2Al, 2Si), Si(3Al, 1Si), and Si(4Al, 0Si). Consequently, the Si/Al ratio of zeolite can vary from 1 to 1. In the framework of LTA, the arrangement of SiO4 and AlO4 tetrahedra is ordered and strictly alternated, resulting in a Si/Al ratio of unity. However, in most of the zeolite frameworks the distribution of Si and Al atoms is disordered; in other words, Si and Al atoms randomly distribute on the T-sites. Generally only T (T ¼ Si, Al) sites are refined in the framework structure of zeolites. In the family of aluminophosphate molecular sieves, AlO4 and PO4 tetrahedra strictly alternate in the framework.[16] The Al and P atoms in the framework can be substituted by other elements, such as Li, Be, B, Ge, As, Si, Mg, Cr, Mn, Fe, and Co, etc., namely isomorphic substitutions.[17,18] Flanigen et al. elucidated the bonding concepts in AlPO4based molecular sieves.8 The linkages Al–O–P, Si–O–Si, Si–O–Al, Me–O–P, and Me–O–P–O–Me have been observed, whilst the linkages P–O–P, P–O–Si, Al–O–Al, Me–O–Al, and Me–O–Me appear to be unlikely. Based on these patterns, they proposed rules for framework cation siting of metal and Si in AlPO frameworks: i) Me, incorporation into a hypothetical Al site, ii) Si into a hypothetical P site, and iii) 2 Si for Al þ P. Martens and Jacobs further elaborated the types of isomorphic substitution according to various substitution mechanisms.[17] Figure 2.10 shows different types of substitutions. The isomorphic substitution mechanism (SM) can be classified as: i) SM I - substitution of Al atoms. SM Ia, SM Ib, and SM Ic refer to monovalent, divalent and trivalent element substitutions of Al atoms, respectively, thus resulting in an M

O

P bond; ii) SM II substitution of P atoms. SM IIa and SM IIb refer to tetravalent and pentavalent element substitutions, respectively, thus resulting in an M

O

Al bond; iii) SM III substitution of pairs of adjacent Al and P atoms. Si is the only element exhibiting SM III substitution. Various possibilities to generate silicoaluminophosphate (SAPO) frameworks by substitution of P and Al atoms with Si are shown in Figure 2.11.[19] The substitution is conveniently explained by using a twodimensional grid representation of T-atom configurations. In SAPOs, the formation of Si

O

P linkages is unlikely. The substitution of an isolated pair of adjacent Al and P atoms in an AlPO4 framework with two Si atoms inevitably would generate Si

O

P linkages (SM III ho), but this does not occur. One way to avoid this linkage is by applying the SM III substitution in a heterogeneous instead of a homogeneous way (SM III he). As shown in Figure 2.11(d), such a substitution results in an electron

34

Chemistry of Zeolites and Related Porous Materials

Figure 2.10 Isomorphic substitution mechanisms in AlPO4-based frameworks. Reproduced with permission from [17]. Copyright (1994) Elsevier

neutral framework comprising AlPO4 layer(s) and topotactic SiO2 overlayer(s). At the boundary of the two crystal domains, Si(3Si, 1Al) and Si (1Si, 3Al) environments are present. For most reported SAPO materials, Si is incorporated according to a combination of SM IIa and SM III mechanisms. The substitution is sensitive to many synthetic parameters, such as the Si content, the nature of the organic template, the amine/ Al2O3 and P2O5/Al2O3 ratios in the reaction mixture, pH, and the crystallization time and temperature. In general, the first Si atom is incorporated into the framework according to SM IIa. Beyond a critical Si concentration, which may be far below the stoichiometric SM IIa substitution level, SM IIa and SM III start to occur simultaneously, and extensive regions in the individual crystals become siliceous. Situations may be encountered where some Si atoms of the silicon patches are replaced with Al atoms, thus generating negative framework charges [Figure 2.11(e)]. There are many reports of Si substitutions in SAPO-5, SAPO-11, SAPO-31, SAPO-34, and SAPO-37 in the literature.[20–24] 2.3.2

Distribution and Position of Cations in the Structure

The cations balancing the negative charge of the framework locate in the channels and cages of a zeolite structure. The number and sites of cations are of interest due to their

Structural Chemistry of Microporous Materials

35

Figure 2.11 Two-dimensional representation of different possible T-atom configurations arising from Si substitution in an AlPO4 framework. The exponents indicate the substitution of individual Si atoms. Reproduced with permission from [17]. Copyright (1994) Elsevier

effects on the performance of a zeolite such as ion-exchange and catalytic properties. Modern crystallographic techniques have been applied to obtain such information from diffraction data. However, there are still some limitations on the identification of the sites of cations. One of the main problems is due to the fact that extra-framework species do not generally follow the high symmetry of the framework, so they are ‘disordered’.[15] For example, in the case of LTA, each 8-ring can accommodate only one Naþ ion. However, because of the 4-fold symmetry of the 8-ring, there are four equivalent positions for the Naþ ion (Figure 2.12). One Naþ ion may hop between the four equivalent positions (dynamic disorder) or it may be stationary but occupy different positions in different 8-rings (static disorder). Conventional X-ray diffraction analysis

36

Chemistry of Zeolites and Related Porous Materials

Figure 2.12 Position (solid circle) of Naþ cation in the 8-ring of zeolite LTA. Three hollow dotted circles represent its symmetrical equivalent positions. Reproduced with permission from [15]. Copyright (2001) Elsevier

cannot distinguish between these two possibilities. An electron-density map generated from the diffraction data will show 1/4 of an Naþ ion (e.g. 10/4 electrons) at each equivalent position rather than one ion (10 electrons) at a single position. There have been many investigations of the cation sites in the FAU framework type.[25–32] The cations in the FAU framework generally locate at the diagonals of cubic unit cells, denoted as I, I0 , II, II0 , II00 , III, III0 , U, and so on.[32] As shown in Figure 2.13, site I is located at the center of the hexagonal column cage; site I0 is away from the center of the ˚ in the b cage; sites II and II00 are located in the hexagonal column cage by about 1 A 0 ˚ ; sites III and III0 faujasite cage; site II is away from the center of the 6-rings by about 1 A

Figure 2.13 Position of Naþ cations in FAU framework. Reproduced with permission from [32]. Copyright (1996) Elsevier

Structural Chemistry of Microporous Materials

37

Table 2.4 The distribution of Naþ ions in various positions of Na88X Experimental Naþ

Site name

No./unit cell

Na1 Na2

I I0

2.9(0.5)a 21.1(1.9)

Na3 Na4

I00 II

8.0(1.9) 31.0(0.3)

Theoretical Site name I I10 I30 I20 II

No./unit cell 2.7 10.4 8.1 8.1 29.3

Site description Normal position Normal position Nearer to U Displacement from site III

Na5 Na6

0

III III0

10.6(1.0) 10.6(1.0)

Na60 Total

III0

8.6(1.0) 92.9(3.9)

0

III1 III30 III20 III200

10.4 8.1 2.7 8.1 87.9

Small Moderate Moderate Large

a

Numbers in parentheses denote experimental uncertainties.

are located somewhere around the wall of the faujasite cage. Site III is in the 4-rings of the b cage, and a cation on site III easily moves to site III0 by even a very small asymmetric perturbation. Site U is located at the center of the b cage. By using single-crystal X-ray diffraction, Olson determined the structure of dehydrated ˚ ],[31] in which all the Naþ cations could NaX [Na88 Al88Si104O384, Fd3, a0 ¼ 25.099(5) A be located. Table 2.4 summarizes the distribution of Naþ cations in Na88X. In each unit cell, the occupancy of Naþ is 2.9 for site I, 21.1 for site I0 , 31.0 for site II, and 29.8 for site III0 . Naþ on site I0 is split into two closely related sites, and on site III0 into three closely related sites. Takaishi successfully explained the distribution of cations in the FAU structure based on an ordered Si-Al distribution model.[32] The theoretical calculation is in excellent agreement with the experiment results. In Takaishi’ s model, D6Rs contain six or three Al atoms, abbreviated as D6Rs-6 and D6Rs-3, respectively, both of which have a 3-fold symmetry. Four D6Rs constitute a tertiary building unit (TBU) containing 48 T sites. The unit cell of FAU contains four TBU including 16 D6Rs with 192 T sites. Figure 2.14 shows the TBU containing 24 Al atoms and the TBU containing 21 Al atoms. Four TBUs connected tetrahedrally result in a new b cage containing 24 T sites in their center as shown by dotted lines in Figure 2.15. This means that it is possible to construct a unit cell with eight b cages with 192 atoms. As shown in Figure 2.16, there are three kinds of cage in Na88X, e.g., cage 1, cage 2, and cage 3, which contain 2.6 (1.3  2) cage 1, 2.7 cage 2, and 2.7 cage 3. Figure 2.16 shows the configuration of Naþ ions in the SOD cage. Each cage 1 contains 12 Al atoms and 12 Naþ ions including 4 Na/I10 , 4 Na/II and 4 Na/III10 . Six III10 sites are randomly occupied by Naþ ions with an occupancy of 4/6. Each cage 2 contains 12 Al atoms, and 12 Naþ ions distributed as Na/I, 3 Na/I20 , 4 Na/II2, 3 Na/III20 , and Na/III20 (occupancy: 1/3). Na/I20 is attracted to the direction of U (cage center) by a dipole interaction of Al3-Na/I in D6R-3. As a consequence, site I20 is located nearer to U rather than to the usual site I0 . Site III20 is located in a strongly asymmetric environment and largely displaced from site III. A Naþ ion on site III20 experiences a

38

Chemistry of Zeolites and Related Porous Materials

Figure 2.14 TBUs containing 24 or 21 Al atoms and SOD cage located at the centre. . indicates Al. (a) TBU (24Al); (b) TBU (21Al); (c) TBU (21Al). Reproduced with permission from [32]. Copyright (1996) Elsevier

weakly asymmetric electric field caused by Na/III200 , which produces a moderate displacement of site III20 from site III. Each cage 3 contains 9 Al atoms and 3 Na/I30 , 3 Na/II, and 3 Na/III30 . Four site IIs contain only 3 Na ions, and then site II at the bottom of the cage becomes vacant to satisfy the symmetry requirement. Site III30 has two neighboring site IIs, one of which is vacant at the bottom. This asymmetric field produces a moderate displacement of site III30 from site III. As seen from Table 2.4, the theoretical model agrees with the experimental results.

Figure 2.15 Connection of four TBUs and a newly formed SOD cage (dotted lines). Reproduced with permission from [32]. Copyright (1996) Elsevier

Structural Chemistry of Microporous Materials

39

Figure 2.16 Configuration of Naþ ions in SOD cages. (a) cage 1; (b) cage 2; (c) cage 3. Reproduced with permission from [32]. Copyright (1996) Elsevier

In the view of structural chemistry, the distribution of cations is also affected by many other factors, such as size of the cage, distribution of the static electric field, radius of the cation, and hydration and dehydration states of the zeolites, etc. 2.3.3

Organic Templates

In the synthesis of high-silica zeolites and phosphate-based molecular sieves, some organic amines are commonly introduced into the reaction system to act as templates or structure-directing agents (SDAs). They are located in the channels or cages of zeolites, playing the following roles in the formation of specific channels and cages: i) Space filling; ii) Structure directing; iii) True templating.[33] In the case that multiple organic guest species can direct the formation of one structure, their lower specificity indicates their space-filling roles. For example, AlPO4-5 (AFI) is much less template specific, and could be synthesized with more than 25 different templates. Tetrapropylammonium (TPA) hydroxide is a typical template for the synthesis of AlPO4-5, which is stacked in a tripod arrangement with the head of one TPA ion suspended between the three feet of the next TPA ion with a hydroxy group neatly suspended between them. As shown in Figure 2.17, although this tripod arrangement is such a good geometrical fit with the cylindrical wall, TPAOH is not a template in the true sense. One of the reasons is the inconsistency in the symmetry of the TPAOH and the channel. TPAOH must be disordered because its molecular symmetry is three-fold whereas the channel symmetry is six-fold; hence, there would be incomplete structural control between the encapsulated TPAOH and the framework. The structure-directing effect means that the specific structure can be directed only by a specific organic species. For instance, crown ether 18-crown-6 in the synthesis of hexagonal faujasite (EMT) plays such a structure-directing role. Figure 2.18 shows a hypothetical model of an 18-crown-6 molecule in the supercage of EMT. [(Crown-, Na)þ, OH
] is combined into the channel during the crystallization process.[34]

40

Chemistry of Zeolites and Related Porous Materials

Figure 2.17 Cylindrical channel in AlPO4-5 and stacking of the encapsulated tetrapropylammonium hydroxide species. Reproduced with permission from [16]. Copyright (1986) Elsevier

A true ‘templating’ occurs only when a zeolite structure adopts the geometric and electronic configuration that is unique to the templating molecule and, upon removal of the organic species, retains the shape of the guest molecules. This is exemplified in the synthesis of ZSM-18 (MEI) using a specific triquaternary amine (tri-quat) C18H36Nþ. As shown in Figure 2.19, the geometric character of the organic amine is consistent with that of the MEI cage.[35] Both the cage and template molecule possess the same 3-fold rotational symmetry. Furthermore, the size of the rigid template molecule matches the

Figure 2.18 Hypothesis on the location and stabilizing interactions of the 18-crown ether-6 in the hexagonal faujasite. Reproduced with permission from [34]. Copyright (1990) Elsevier

Structural Chemistry of Microporous Materials

Figure 2.19

41

Position of tri-quat C18H36Nþ in the cage of ZSM-18

size of the cage, which does not allow the template molecule to rotate freely in the cage. Therefore the C18H36Nþ cation is a template in the true sense. Further discussion of the templating or structure-directing effect of organic amines will be given in Chapter 5.

2.4 Framework Structures of Zeolites 2.4.1

Loop Configuration and Coordination Sequences

The loop configuration is a simple graph showing how many 3- or 4-rings a given T–atom is involved in.[3] Sato used the term ‘second coordination networks’.[36] Figure 2.20 shows all observed loop configurations and their frequency of occurrence.[3]

Figure 2.20 Loop configurations. Number in parenthesis ¼ frequency of occurrence. Reproduced with permission from [3]. Copyright (2001) Elsevier

42

Chemistry of Zeolites and Related Porous Materials

Table 2.5 Topological features of different zeolite-type frameworks Framework composition

Prevailing loop configuration

High-silica zeolites High-alumima zeolites Aluminophosphate

Solid lines represent T

O

T linkages, whereas dotted lines indicate nonconnected T

O bonds found in the interrupted frameworks. Loop configurations serve to characterize the immediate surroundings of the T-atoms in a zeolite framework and provide useful information, particularly in the case of complex networks. They provide information on the smallest rings present, and can also be used for classification purposes and for deducing rules relating to these structures which might be of predictive value. As can be seen from Figure 2.20, some loop configurations frequently occur in zeolite structures, whereas others are rare. Loop configurations can also be correlated with the framework composition. Table 2.5 shows the prevailing loop configurations observed in high-silica zeolites, high-alumina zeolites, and aluminophosphate molecular sieves.[12] The concept of coordination sequences (CSQ) was originally introduced by Brunner and Laves,[37] and first applied to zeolite frameworks by Meier and Moeck.[38] In a typical zeolite framework, each T-atom is connected to N1 ¼ 4 neighboring T-atoms through oxygen bridges. These neighboring T-atoms are then linked in the same manner to N2 T-atoms in the next shell. The latter are connected with N3 T-atoms, and so forth. Each T-atom is counted only once. In this way, a coordination sequence can be determined for each T-atom of the 4-connected net of T-atoms. It follows that: N0 ¼ 1; N1  4; N2  12; N3  36; . . . ; Nk  4  3k
1 Coordination sequences have been listed in the Atlas[3] from N1 up to N10 for each topologically distinct T-atom in every framework structure. For example, the coordination sequences in FAU are as follows: T1ð192; lÞ4 9 16 25 37 53 73 96 120 145 The site multiplicity and the site symmetry are both given in parenthesis. The vertex symbol was first used in connection with zeolite-type networks by O’ Keeffe and Hyde.[39] This symbol indicates the size of the smallest ring associated with each of the 6 angles of a TO4 tetrahedron. For example, the vertex symbol in FAU reads 4 4 4 66 12, indicating that one pair of opposite angles contains two 4-rings, a second pair contains a 4-ring and a 6-ring, and the final pair of opposite angles contains a 6-ring and a 12-ring. The vertex symbol is useful for determining the smallest rings in a framework structure. Sometimes more than one ring of the same size is found at a vertex. This is indicated by a subscript, as in 62 or 82.

Structural Chemistry of Microporous Materials

43

The coordination sequences and the vertex symbol are unique for a particular framework topology, i.e. they can be used to distinguish between different zeolite framework types unambiguously. In this way, frameworks with the same topologies can be easily identified. Currently, it is easier to calculate the coordination sequences and vertex symbol using computer program based on crystallographic data. 2.4.2

Ring Number of Pore Opening and Channel Dimension in Zeolites

The channels of zeolites are delimited by the rings formed by n T-atoms. Besides some small channel systems like 6-rings, the pore openings of zeolites contain 8-, 9-, 10-, 12-, 14-, 18-, and 20-rings. A summary of the largest rings of various zeolite structure types is presented in Table 2.6. In general, zeolites are classified as small-pore, medium-pore, and large-pore systems according to the number of pore openings. The small-pore zeolites such as LTA, SOD, and GIS contain the pore opening enclosed by 8 TO4 tetrahedra, with a diameter of about ˚ . Medium pore zeolites (the typical example being MFI) generally feature a 10-ring 4A ˚ . The large pore zeolites such as pore opening with a diameter of approximately 5.5 A FAU, MOR, and *BEA have pore openings formed by 12 TO4 tetrahedra, with a diameter ˚ . The zeolites with pore openings comprising more than 12 T-atoms are of about 7.5 A called extra-large pore zeolites. It is worth noting that 8-, 10-, and 12-rings are common in zeolites. At present, extra-large pore zeolites are still rare, and the largest ring is limited to a 20-ring system, as observed in gallophosphate cloverite (CLO). The channel system in zeolites can be 1-, 2-, or 3-dimensional; that is, the channels extend along one, two, three-dimensions. Table 2.6 lists the channel dimensions of zeolites ordered by decreasing number of T-atoms in the largest rings.[4] Interconnecting channel systems are separated by a double arrow ( $ ). A vertical bar (j) means that there is no direct access from one channel system to the other. A perpendicular symbol (?) represents a direction that is perpendicular to a crystallographic plane. The number of asterisks (*) indicates the dimensions of the channel system, e.g. 1-, 2-, or 3-dimension. A few examples are selected from Table 2.6 to illustrate the notation for crystallographic characterization of the channels. CAN OFF RHO GIS

[001] [001] h100i {[100]

12 12 8 8

5.9  5.9* 6.7  6.8* $ ? 3.6  3.6* * * 3.1  4.5 $

[001] 8 j h100i 8 [010] 8

3.6  4.9* * 3.6  3.6* * * 2.8  4.8}* * *

CAN is characterized by a 1-dimensional system of channels parallel to [001] with circular 12-ring apertures. In OFF, the main 12-ring channels are interconnected with a 2-dimensional system of 8-ring channels, thus forming a 3-dimensional channel system. RHO is an example of a framework type containing two noninterconnecting 3-dimensional channel systems, while h100i means there are channels parallel to all crystallographically equivalent axes of the cubic structure, i.e., along the x-, y-, and z-directions. In GIS, the channels parallel to [100], together with those parallel to [010], give rise to a 3-dimensional channel system. Table 2.6 also gives the free diameters (effective pore width) of the channels, which means minimum and maximum distances ˚ ). The between the oxygen atoms in the channel (oxygen’s Van der Waals’ radius is 1.35 A

44

Chemistry of Zeolites and Related Porous Materials

Table 2.6 Channel dimensions -CLO ETR VFI AET CFI DON OSO UTL AF1 AFR AFS AFY ASV ATO ATS *BEA BOG BPH CAN CON CZP DFO EMT EON FAU GME GON IFR ISV IWR LTL MAZ MEI MOR MOZ MTW NPO OFF OSI -RON SAO SBE SBS SBT SFE SFO SOS

Clovente ECR-34 VP1-5 AIPO-8 CIT-5 UTD-IF OSB-1 IM-12

20-, 18-, and 14-Ring Structures <100>20 4.0  13.2*** | <100> 8 3.8  3.8*** [001] 18 10.1* $ [001] 8 2.5  6.0** [001] 18 12.7  12.7* [010] 14 7.9  8.7* [010] 14 7.2  7.5* [010] 14 8.1  8.2* [001] 14 5.4  7.3* $ ? [001] 8 2.8  3.3** [001] 14 7.1  9.5* $ [010] 12 5.5  8.5*

12-Ring Structures [001] 12 7.3  7.3* [001] 12 6.7  6.9* $ [010] 8 3.7  3.7* [001] 12 7.0  7.0* $ ? [001] 8 4.0  4.0** [001] 12 6.1  6.1* $ ? [001] 8 4.0  4.3** [001] 12 4.1  4.1* [001] 12 5.4  5.4* [001] 12 6.5  7.5* h100i 12 6.6  6.7* $ [001] 12 5.6  5.6* [100] 12 7.0  7.0* $ [010] 10 5.5  5.8* [001] 12 6.3  6.3* $ ? [001] 8 2.7  3.5** [001] 12 5.9  5.9* [001] 12 6.4  7.0* $ [100] 12 7.0  5.9* $ [010] 10 5.1  4.5* Chiral Zincophosphate [001] 12 3.8  7.2* DAF-1 {[001] 12 7.3  7.3 $ ? [001] 8 3.4  5.6}*** $ {[001] 12 6.2  6.2 $ ? [00l] 10 5.4  6.4}*** EMC-2 [001] 12 7.3  7.3* $ ? [001] 12 6.5  7.5** ECR-1 {[100] 12 6.7  6.8* $ [010] 8 {[001] 3.4  4.9 $ [100] 8 2.9  2.9}*}** Faujasite <111> 12 7.4  7.4*** Gmelinite [001] 12 7.0  7.0* $ ? [001] 8 3.6  3.9** GUS-1 [001] 12 5.4  6.8* ITQ-4 [001] 12 6.2  7.2* ITQ-7 <100> 12 6.1  6.5** $ [001] 12 5.9  6.6* ITQ-24 [001] 12 5.8  6.8* $ [110] 10 4.6  5.3* $ [010] 10 4.6  5.3* Linde Type L [001] 12 7.1  7.1* Mazzite [001] 12 7.4  7.4* | [001] 8 3.1  3.1*** ZSM-18 [001] 12 6.9  6.9* $ ? [001] 7 3.2  3.5** Mordenite [001] 12 6.5  7.0* $ {[010] 8 3.4  4.8 $ [001] 8 2.6  >5.7}* ZSM-10 {[001]12 6.87.0 $ [001]8 3.84.8}***|[001] 12 6.86.8* ZSM-12 [010] 12 5.6  6.0* Nitridophosphate-1 [100] 12 3.3  4.4* Offretite [001] 12 6.7  6.8* $ ? [001] 8 3.6  4.9** UiO-6 [001] 12 5.2  6.0* Roggianite [001] 12 4.3  4.3* STA-1 h100i 12 6.5  7.2** $ [001] 12 7.0  7.0* UCSB-8Co <100> 12 7.2  7.4** $ [001] 8 4.0  4.0* UCSB-6GaCo [001] 12 6.8  6.8* $ ? [001] 12 6.9  7.0** UCSB-10GaZn [001] 12 6.4  7.4* $ ? [001] 12 7.3  7.8** SSZ-48 [010] 12 5.4  7.6* SSZ-51 [001] 12 6.9  7.1* $ [010] 8 3.1  3.9* SU-16 {[100] 12 3.9  9.1 $ [010] 8 3.3  3.3}** AIPO-5 SAPO-40 MAPSO-46 CoAPO-50 ASU-7 AIPO-31 MAPO-36 Beta Boggsite Beryllophosphate-H Canoncrinite CIT-1

Structural Chemistry of Microporous Materials Table 2.6 (Continued ) SSY USI VET

SSZ-60 IM-6 VPI-8

AEL AFO AHT CGF CGS DAC EPI EUO FER HEU LAU MEL MFI MFS MTT MWW NES OBW

AlPO-11 AlPO-41 AlPO-H2 Co-Ga-Phosphate-5 Co-Ga-Phosphate-6 Dachiardite Epistilbite EU-1 Ferrierite Heulandite Laumontite ZSM-11 ZSM-5 ZSM-57 ZSM-23 MCM-22 NU-87 OSB-2

-PAR PON RRO SFF STI TER TON WEI -WEN

Partheite IST-1 RUB-41 SSZ-44 Stilbite Terranovaite Theta-1 Weinebeneite Wenkite

-CHI LOV NAT RSN STT VSV

Chiavennite Lovdarite Natrolite RUB-17 SSZ-23 VPI-7

ABW ACO AEI

Li-A ACP-1 AlPO-18

AEN AFN AFT AFX ANA APC APD

AlPO-EN3 AlPO-14 AlPO-52 SAPO-56 Analcime AlPO-C AlPO-D

[001] 12 5.0  7.6* [100] 12 6.1  6.2* $ [001] 10 3.9  6.4* [001] 12 5.9  5.9* 10-Ring Structures [001] 10 4.0  6.5* [001] 10 4.3  7.0* [001] 10 3.3  6.8* {[100] 10 2.5  9.2* þ 8 2.1  6.7*} $ [001] 8 2.4  4.8* {[001] 10 3.5  8.1 $ [100] 8 2.5  4.6}*** [010] 10 3.4  5.3* $ [001] 8 3.7  4.8* [100] 10 3.4  5.6* $ [00l] 8 3.7  4.5* [100] 10 4.1  5.4* [001] 10 4.2  5.4* $ [010] 8 3.5  4.8* [001] 10 3.1  7.5* þ 8 3.6  4.6*} $ [100] 8 2.3  4.7* [100] 10 4.0  5.3* <100> 10 5.3  5.4*** {[100] 10 5.1  5.5 $ [010] 10 5.3  5.6}*** [100] 10 5.1  5.4* $ [010] 8 3.3  4.8* [001] 10 4.5  5.2* ? [001] 10 4.0  5.5** | ? [001] 10 4.1  5.1** [100] 10 4.8  5.7** {<110> 10 5.0  5.0** $ ([001] 8 3.4  3.4*}| <101> 8 2.8  4.0**) $ <100> 8 3.3  3.4**}*** [001] 10 3.5  6.9* [100] 10 4.4  4.6* [100] 10 4.0  6.5* $ [001] 8 2.7  5.0* [001] 10 5.4  5.7* [100] 10 4.7  5.0* $ [001] 8 2.7  5.6* [100] 10 5.0  5.0* $ [001] 10 4.1  7.0* [001] 10 4.6  5.7* [001] 10 3.1  5.4* $ [100] 8 3.3  5.0* <100> 10 2.5  4.8** $ [001] 8 2.3  2.7* 9-Ring Structures [001] 9 3.9  4.3* [001] 9 3.2  4.5* $ [001] 9 3.0  4.2* $ [100] 8 3.6  3.7* <100> 8 2.6  3.9** $ [001] 9 2.5  4.1* [100] 9 3.3  4.4* $ [001] 9 3.1  4.3* $ [010] 8 3.4  4.1* [101] 9 3.7  5.3* $ [001] 7 2.4  3.5* [01-1] 9 3.3  4.3* $ [011] 9 2.9  4.2* $ [011] 8 2.1  2.7* 8-Ring Structures [001] 8 3.4  3.8* <100> 8 2.8  3.5** $ [001] 8 3.5  3.5* {[100] 8 3.8  3.8 $ [110] 8 3.8  3.8 $ [001] 8 3.8  3.8}*** [100] 8 3.1  4.3* $ [010] 8 2.7  5.0* [100] 8 1.9  4.6* $ [010] 8 2.1  4.9* $ [001] 8 3.3  4.0* ? [001] 8 3.2  3.8*** ? [001] 8 3.4  3.6*** irregular distorted 8-rings [001] 8 3.4  3.7* $ [100] 8 2.0  4.7* [010] 8 2.3  6.0* $ [201] 8 1.3  5 8*

45

46

Chemistry of Zeolites and Related Porous Materials

Table 2.6 (Continued ) ATN ATT ATV AWO AWW BIK BRE CAS CDO CHA DDR DFT EAB EDI ERI ESV GIS GOO IHW ITE JBW KFI LEV -LIT LTA MER

MAPO-39 AlPO-12-TAMU AlPO-25 AlPO-21 AlPO-22 Bikitaite Brewsterite Cesium Aluminosilicate CDS-1 Chabazite Decadodecasil-3R DAF-2 TMA-E Elingtonite Erionite ERS-7 Gismondine Goosecreekite ITQ-32 ITQ-3 NaJ ZK-5 Levype Lithosite Linde Type A Merlinoite

MON MTF NSI OWE PAU PHI RHO RTE RTH RWR SAS SAT SAV THO TSC VNI YUG ZON

Meniesommaite MCM-35 Nu-6(2) UiO-28 Paulingite Phillipsite Rho RUB-3 RUB-13 RUB-24 STA-6 STA-2 Mg-STA-7 Thomsonite Tscho¨rtnerite VPI-9 Yugawaralite ZAPO-M1

[001] 8 4.0  4.0* [100] 8 4.2  4.6* $ [010] 8 3.8  3.8* [001] 8 3.0  4.9* [100] 8 2.7  5.5* [100] 8 3.9  3.9* [001] 8 2.8  3.7* [100] 8 2.3  5.0* $ [001] 8 2.8  4.1* [001] 8 2.4  4.7* [010] 8 3.1  4.7* $ [001] 8 2.5  4.2* ? [001] 8 3.8  3.8*** ? [001] 8 3.6  4.4** [001] 8 4.1  4.1* $ [100] 8 1.8  4.7* $ [010] 8 1.8  4.7* ? [001] 8 3.7  5.1** <110> 8 2.8  3.8** $ [001] 8 2.0  3.1* ? [001] 8 3.6  5.1*** [010] 8 3.5  4.7* {[100] 8 3.1  4.5 $ [010] 8 2.8  4.8}*** [100] 8 2.8  4.0* $ [010] 8 2.7  4.1* $ [001] 8 2.9  4.7* [100] 8 3.54.3** [010] 8 3.8  4.3* $ [001] 8 2.7  5.8* [001] 8 3.7  4.8* <100> 8 3.9  3.9*** | 100> 8 3.9  3.9*** ? [001] 8 3.6  4.8** <100> 8 4.1  4.1*** [100] 8 3.1  3.5* $ [010] 8 2.7  3.6* $ [001] {8 3.4  5.1* þ 8 3.3  3.3*} [100] 8 3.2  4.4* $ [001] 8 3.6  3.6* [001] 8 3.6  3.9* [010] 8 2.6  4.5* | [010] 8 2.4  4.8* [010] 8 3.5  4.0* $ [001] 8 3.2  4.8* <100> 8 3.6  3.6*** | <100> 8 3.6  3.6*** [100] 8 3.8  3.8* $ [010] 8 3.0  4.3* $ [001] 8 3.2  3.3* 8 3.6  3.6*** | <100> 8 3.6  3.6*** [001] 8 3.7  4.4* [100] 8 3.8  4 1* $ [001] 8 2.5  5.6* [100] 8 2.8  5.0* | [010] 8 2.8  5.0* [001] 8 4.2  4.2* ? [001] 3.0  5.5*** <100> 8 3.8  3.8** $ [001] 8 3.9  3.9* [100] 8 2.3  3.9* $ [010] 8 2.2  4.0* $ [001] 8 2.2  3.0* <100>8 4.2  4.2*** $ <110> 8 3.1  5.6*** {<110> 8 3.1  4.0 $ [001] 8 3.5  3.6}*** [100] 8 3.8  3.6* $ [001] 8 3.1  5.0* [100] 8 2.5  5.1* $ [010] 8 3.7  4.4*

free diameters of channels differ from those of the cavities. For example, the 3-dimensional channel system in the LTA framework has 8-ring pore openings with a ˚ and an a-cavity with a maximum diameter of 11.4 A ˚ (Figure 2.21). free diameter of 4.1 A It should be noted that crystallographic free diameters may depend upon the hydration

Structural Chemistry of Microporous Materials

47

Figure 2.21 Features of the pores in LTA: (a) the 3-dimensional channel system; (b) the 8-ring window; (c) the a-cavity [4126886]. Reproduced with permission from [13]. Copyright (2003) Elsevier

state of zeolites, particularly for the more flexible frameworks. On the other hand, the effective free diameters could also be affected by nonframework cations, and temperature, etc. 2.4.3

Framework Densities (FDs)

A straightforward criterion for distinguishing zeolites and zeolitic open-framework materials from denser tectosilicates is based on the framework density (FD), i.e., the ˚ 3. The FD is obviously related to the pore volume but does number of T-atoms per 1000 A not reflect the size of the pore openings. Figure 2.22 shows the relationship of framework density with the smallest ring in loop configuration.[3] A gap is clearly recognizable between zeolite types and dense tetrahedral framework structures (marked by the hatched area in Figure 2.22). For nonzeolitic framework structures, the FD values are generally 20 ˚ 3, while for zeolites with fully cross-linked frameworks the observed to 21 T/1000 A ˚ 3 for structures with the largest pore volume to values range from about 12.1 T/1000 A 3 ˚ around 20.6 T/1000 A . To date, FD values of less than 12 have been encountered only for the interrupted framework cloverite (CLO), and for some hypothetical frameworks. FD values are usually given for typical materials and these may also depend on chemical composition to some extent. As can be seen from Figure 2.22, a majority of the structure types are in category ‘4’. ‘þ’ Indicates that T-atoms are associated with larger rings only. One important conclusion derived from this plot is that low-density frameworks are likely to contain 3-rings. This hypothesis has been tested by hypothetical structures and 3-ring structures having FDs below 10. 2.4.4

Selected Zeolite Framework Structures

Up to now, 167 zeolite framework types have been identified. Here, only some typical framework types with specific structural features will be described.

48

Chemistry of Zeolites and Related Porous Materials

Figure 2.22 Framework density vs. smallest ring in loop configuration. The þ sign indicates that there are some T-positions associated with larger rings only. Reproduced with permission from [3]. Copyright (2001) Elsevier

Sodalite (SOD)[40–51] ˚ .[41] Naþ and Cl
Type material jNa8þCl2
j[Al6Si6O24]-SOD, cubic P-43n, a ¼ 8:870 A ions in the structure both can be exchanged. The composition of synthetic sodalite is jNa6j[Al6Si6O24]. The SOD framework type is best described as a body-centered cubic arrangement of b or sodalite cages jointed through single 4- and 6-rings (Figure 2.23). It also can be viewed as a primitive cubic arrangement of b-cages joined through 4-rings, producing a b-cage

Structural Chemistry of Microporous Materials

49

Figure 2.23 SOD framework type. Reproduced with permission from [15]. Copyright (2001) Elsevier

in the center. In a strict sense, sodalite is not a zeolite, since it has only 6-ring pore openings and thus has very limited sorption capacity. It has a framework density of 17.2 ˚ 3. Sodalite is an important host material for creating simple periodic T-atoms per 1000 A arrays of clusters. LTA[52–57] ˚ [52]. Its Type material jNa12þ(H2O)27j8[Al12Si12O48]8-LTA, cubic Fm-3c, a ¼ 24:61 A 3 ˚ . framework density is 12.9 T/1000 A The LTA framework type is related to SOD. In LTA the sodalite cages are joined via double 4-rings, creating an a-cage in the center of unit cell (Figure 2.24). Alternatively, the framework can be described as a primitive cubic arrangement of a-cages joined through single 8-rings. LTA has a three-dimensional 8-ring channel system running along ˚. the [100], [010], and [001] directions, respectively, with a free aperture of 4.1  4.1 A Typically, zeolite LTA is synthesized with a framework Si/Al ratio of 1. By using tetramethylammonium cation (TMAþ) as the structure-directing agent, the Si/Al ratio of

Figure 2.24 LTA framework type. Reproduced with permission from [15]. Copyright (2001) Elsevier

50

Chemistry of Zeolites and Related Porous Materials

Figure 2.25 The FAU framework type and its supercage. The three different layers of sodalite cagesareindicatedasA,B,andC.Reproducedwithpermissionfrom[15].Copyright(2001)Elsevier

LTA framework could be increased up to about three. By using a supramolecular synthon as organic structure-directing agent, it has been possible to synthesize the Al-free as well as the pure-silica forms of zeolite ITQ-29 with the LTA structure.[58] FAU[31,56,59–61] Type material j(Ca2þ,Mg2þNaþ2)29(H2O)240j[Al58Si134O384]-FAU, cubic Fd-3m, ˚ .[60] Its framework density is 12.7 T/1000 A ˚ 3. a ¼ 24:74 A As with LTA and SOD, the FAU framework type is also featured by sodalite cages. The sodalite cages are arranged in the same way as the carbon atoms in diamond, and are joined to one another via double 6-rings, producing the so-called surpercage with four, tetrahedrally oriented 12-ring pore openings. FAU has a 3-dimensional channel system (Figure 2.25). The ˚ . The framework type of FAU can also be described free aperture of the 12-ring is 7.4  7.4 A as an ABCABC stacking of puckered layers of sodalite cages (shadow shown in the Figure) which are related to one another by inversion in each of the double 6-rings.[15] Both the X- and Y-zeolite have the FAU framework structure. Generally, X-zeolite has a framework SiO2/Al2O3 ratio of 2.23.0 while Y-zeolite has a SiO2/Al2O3 ratio higher than 3.0. The FAU zeolites have important catalytic applications due to their large void volume of  50%, 12-ring pore openings, and 3-dimensional channel system. EMT[62] Type material EMC-2, jNaþ21(C12H24O6)4j[Al21Si75O192]-EMT, hexagonal P63/mmc, ˚ , c ¼ 28:365 A ˚ .[62] Its framework density is 12.9 T/1000 A ˚ 3. a ¼ 17:374 A

Figure 2.26 The EMT framework type showing the medium and larger cages separately. The two different layers of sodalite cages are indicated as A and B. Reproduced with permission from [15]. Copyright (2001) Elsevier

Structural Chemistry of Microporous Materials

51

Figure 2.27 Skeletal diagram of the [001] projection of LTL framework and the -Can-D6R-Can- Column

The EMT framework type is a hexagonal analog of FAU (Figure 2.26). In EMT, the puckered sodalite cage layers are stacked in an ABAB sequence instead of the ABCABC sequence in FAU and the layers are related to one another by a mirror plane. This arrangement creates a medium cage with three 12-ring pore openings and a larger cage with five 12-ring pore openings. As with FAU, EMT has a 3-dimensional channel system with 12-ring pores. Since EMT and FAU are built up from the same sodalite cage layer, the intergrowth of FAU/EMT is easily encountered.[63,64] LTL[65–68] ˚, Type material jK6þNa3þ(H2O)21j[Al9Si27O72]-LTL, hexagonal P6/mmm, a ¼ 18:40 A [66] 3 ˚ ˚ c ¼ 7:52 A. Its framework density is 16.3 T/1000 A . The LTL framework is based upon the cancrinite (can cage) and double 6-ring (D6R). Can cages and D6Rs are alternately connected along [001], forming the -can-D6R-can- columns. Six such columns connect with each other along the c axis, giving rise to a 3-dimensional ˚ (Figure 2.27). framework with 12-ring pore openings of aperture of 7.1  7.1 A Cancrinite (CAN)[69–74] ˚, Type material jNa6þCa2þCO32
(H2O)2j[Al6Si6O24]-CAN, hexagonal P63, a ¼ 12:75 A ˚ .[69] The framework density is 16.6 T/1000 A ˚ 3. c ¼ 5:14 A The framework of CAN is featured by can cages (Figure 2.28). Can cages form the can column along c through the sharing of 6-rings. Six such columns are connected

Figure 2.28 [001] Projection of CAN framework and the column of can cages

52

Chemistry of Zeolites and Related Porous Materials

Figure 2.29 The CHA framework type (AABBCC 6-ring stacking indicated) and its 4126286 cage. Reproduced with permission from [15]. Copyright (2001) Elsevier

through sharing of 4-rings along the c axis, creating the 12-ring channel of aperture of ˚. 5.9  5.9 A Chabazite (CHA)[56,59,75–85] ˚, Type material jCa62þ(H2O)40j[Al12Si24O72]-CHA, rhombohedral R-3m, a ¼ 9:42 A 3  [75] ˚ a ¼ 94:47 . Its framework density is 14.5 T/1000 A . The chabazite framework has a 3-dimensional 8-ring channel system. The free ˚ . Its framework contains columns of aperture of the 8-ring pore opening is 3.8  3.8 A alternating D6R and CHA cages along the c direction. The CHA framework type is another member of the ABC-6 family of zeolite frameworks. As shown in Figure 2.29, CHA has an ABC stacking of double 6-ring arrays or an AABBCC stacking of single arrays. This stacking produces an elongated [4126286] cage. Mordenite (MOR)[86–88] ˚, Type material jNa8þ(H2O)24 j[Al8Si40O96]-MOR, orthorhombic Cmcm, a ¼ 18:1 A ˚ , c ¼ 7:5 A ˚ .[86] Its framework density is 17.2 T/1000 A ˚ 3. b ¼ 20:5 A The framework has 12- and 8-ring channels along the [001] direction (Figure 2.30). ˚ and the aperture of the 8-ring channel The aperture of the elliptical 12-ring is 6.5  7.0 A ˚. is 2.6  5.7 A

Figure 2.30

The MOR framework type and the chain formed by T12 units

Structural Chemistry of Microporous Materials

53

Figure 2.31 Characteristic structural units in ZSM-5. (a) [58] Unit; (b) pentasil chain and its linkage within chains in ZSM-5

The MOR structure is featured by the T12 unit, which is composed of 12 T-atoms. These T12 units form infinite chains along the c axis (Figure 2.30). ZSM-5(MFI)[89–96] Type material ZSM-5, jNanþ(H2O)16j[AlnSi96
nO192]-MFI (n < 27), orthorhombic, ˚ , b ¼ 19:92 A ˚ , c ¼ 13:42 A ˚ .[89] Its framework density is 17.9 T/ Pnma, a ¼ 20:07 A 3 ˚ 1000 A . The number of Al atoms in the unit cell varies from 0 to 27, so the ratios of Si/ Al can be changed within a wide range. The MFI framework contains a characteristic [58] unit with D2d symmetry. These [58] units are linked via edge sharing to form a pentasil chain parallel to the c axis (Figure 2.31). These pentasil chains related by a mirror plane are connected via oxygen bridges to form corrugated sheets with 10-ring holes. The sheet parallel to the (100) plane is shown in Figure 2.32(a). Adjacent sheets that are related by an inversion center are linked by oxygen bridges to the next, forming a 3-dimensional framework. This produces ˚ ) parallel to an intersecting channel system with straight 10-ring channels (5.3  5.6 A ˚ corrugation (along y), and sinusoidal 10-ring channels (5.5  5.1 A) perpendicular to the sheets (along x) with an angle of 150 [Figure 2.32(b)].

Figure 2.32 ZSM-5

(a) Porous sheet parallel to the (100) plane in ZSM-5; (b) the channel structure in

54

Chemistry of Zeolites and Related Porous Materials

Figure 2.33 (a) The porous sheet comprising pentasil chains in MEL framework type; (b) the channel structure in ZSM-11. Reproduced with permission from [97]. Copyright (1978) Nature Publishing

ZSM-11(MEL)[97–100] Type material ZSM-11, jNaþn(H2O)16j[AlnSi96
nO192]-MEL, n < 16, tetragonal I-4m2, ˚ , c ¼ 13:44 A ˚ .[97] Its framework density is 17.6 T/1000 A ˚ 3. a ¼ 20:12 A The MEL framework type is closely related to MFI. The corrugated sheets of pentasil chains that are found in MFI are also present in the MEL framework [Figure 2.33(a)]. However, in MEL, the adjacent sheets are related to one another by a mirror plane instead of an inversion center as in MFI. This produces straight 10-ring channels along both ˚ [Figure 2.33(b)]. x- and y-directions with a free aperture of 5.4  5.3 A As expected, intergrowth of the MFI and MEL frameworks often occurs. Zeolite Beta (BEA)[101–104] Zeolite beta (jNanj[AlnSi64
nO128], n < 7) is known as a highly faulted intergrowth of two distinct but closely related structures that have fully three-dimensional pore systems with 12-rings.[101] One end member, polymorph A, forms an enantiomorphic pair (space ˚ , b ¼ 26:6 A ˚ ). Polymorph B is achiral (space group groups P4122 and P4322, a ¼ 12:5 A ˚ , b ¼ 17:8 A ˚ , c ¼ 14:4 A ˚ , b ¼ 114:5 ). Both structures are constructed C2/c, a ¼ 17:6 A from the same centrosymmetric tertiary building unit (TBU) arranged in layers that are successively interconnected in either a left- (L) or a right-handed (R) fashion. Polymorph A represents an uninterrupted sequence of RRRR   (or LLLL    ) stacking. Polymorph B has an alternating RLRL    stacking sequence. The TBU has no intrinsic preference for either mode of connection, enabling both to occur with almost equal probability in zeolite beta, giving rise to a near random extent of interplanar stacking faults and, to a lesser extent, intraplanar defects terminated by hydroxy groups. The faulting does not significantly affect the accessible pore volume, but influences the tortuosity of the pore connectivity along the c direction. The high stacking fault density gives rise to complex powder X-ray diffraction patterns for zeolite beta materials that comprise both sharp and broad features. In both polymorphs A and B (Figures 2.34 and 2.35), the pore systems are threedimensional, with straight 12-ring channels parallel to a and b, and with somewhat more tortuous 12-ring pore paths along the respective c-directions. The free aperture dimen˚ (for the straight channels along a and sions for the polymorph A structure are 7.3  6.0 A ˚ b), and 5.6  5.6 A (along c). The 12-ring intersections between the straight channels

Structural Chemistry of Microporous Materials

55

Figure 2.34 The framework structure of polymorph A viewed along [0–10]. Reproduced with permission from [101]. Copyright (1988) Royal Society of Chemistry

follow the four-fold screw operation along c for the chiral polymorph A, thus defining a helix of right-handedness (P4122) or left-handedness (P4322). As shown in Figure 2.36, the helix comprises double 6-rings, each of which is made of two 4-rings and four 5rings. The chiral character of the polymorph A implies that a pure material with this structure might have potential for effective enantio-(chiral) catalysis or chiral separations. Since no ordered material has yet been produced, the chiral polymorph A (*BEA) is preceded by an asterisk to indicate that the framework type described in the Atlas is an idealized end member of a series.[3] The type material is jNaþ7j[Al7Si57O128]-*BEA with ˚ , and c ¼ 26:406 A ˚ .[101] Its framework tetragonal space group P4122, a ¼ 12:661 A 3 ˚ density is 15.1 T/1000 A .

Figure 2.35 The framework structure of polymorph B viewed approximately along [110]. Reproduced with permission from [101]. Copyright (1988) Royal Society of Chemistry

56

Chemistry of Zeolites and Related Porous Materials

Figure 2.36 The double 6-rings in polymorph A and the helices formed by double 6-rings

BEC[105–107] Type material FOS-5, j(C3H9N)48(H2O)36j[Ge256O512]-BEC, tetragonal I41/amd, ˚ , b ¼ 25:990 A ˚ , c ¼ 27:271 A ˚ .[105] Its framework density is 13.9 T/ a ¼ 25:990 A 3 ˚ . 1000 A The structure of polymorph C of zeolite beta has been proposed by Newsam et al.[101] The structure is closely related to those of polymorphs A and B. It has a space group P42/ mmc. The hypothetical structure of polymorph C has a three-dimensional channel system, in which all 12-ring channels are linear while in the case of the other two polymorphs one of the channels is sinusoidal. An additional important structural difference between the different polymorphs is that polymorph C contains double 4-ring (D4R) cages as secondary building units, while polymorphs A and B do not contain such secondary building units. It should be noted that the D4R cages are under high tension in this structure. The pure polymorph C of zeolite beta (BEC) has been successfully synthesized by Zou and coworkers with GeO2 composition (denoted as FOS-5).[105] Corma et al. obtained silicogermanate ITQ-17 with BEC structure by using framework isomorphous substitution as a structure-directing mechanism.[107] Germanium has a greater tendency to occupy the positions in the D4R cages which may be responsible for stabilizing this structure. The structure of BEC is consistent with the hypothetical structure proposed by Newsan et al. (Figure 2.37). Figure 2.38 shows the arrangement of the two structural building units of polymorph C, i.e., [54] unit formed by four 5-rings and D4R unit formed by double 4-rings, along the 42 screw axis. The mirror-related [54] units are connected to form 4-rings, which are linked to the [54] units of the neighboring unit cells through the D4R units. Structural analysis indicates that the D4Rs are preferentially occupied by Ge atoms. CFI[108,109] ˚ , b ¼ 5:0213 A ˚, Type material CIT-5, jSi32O64j-CFI, orthorhombic Pmn21, a ¼ 13:694 A [109] 3 ˚ ˚ and c ¼ 25:4970 A. Its framework density is 18.3 T/1000 A .

Structural Chemistry of Microporous Materials

57

Figure 2.37 The polymorph C framework structure

CIT-5 is a high-silica zeolite with extra-large pores. Its structure viewed down the b-axis is shown in Figure 2.39(a). It is composed of one-dimensional, extra-large pores of nearly circular cross-section, circumscribed by 14 T-atoms, with an aperture of ˚. 7.2  7.5 A Figure 2.39(b) shows the columns of cage units along b in CIT-5. Each cage is composed of two 4-rings, two 5-rings, and four 6-rings. The connection of such columns of cage units via 5- and 6-rings results in the 3-dimensional framework. The fundamental building unit of CIT-5 can also be described on the basis of a zig-zag ladder of 4-rings with pendant 5-rings. These units are interconnected through single zig-zag chains forming the three-dimensional framework structure. STT[110] Type material SSZ-23, j(C13H24Nþ)4.1F
3.3(OH)
0.8j[Si64O128]-STT, monoclinic P21/n, ˚ , b ¼ 21:792 A ˚ , c ¼ 13:598 A ˚ , b ¼ 101:85 .[110] Its framework density is a ¼ 12:959 A 3 ˚ 17.0 T/1000 A . SSZ-23 is the first aluminosilicate zeolite known with 9-ring pore openings.

Figure 2.38 Arrangement of the two structural building units of polymorph C along the 42 screw axis, cage [54] formed by four 5-rings and D4R cage formed by double 4-rings. Reproduced with permission from [107]. Copyright (2001) Wiley-VCH

58

Chemistry of Zeolites and Related Porous Materials

Figure 2.39 along b

(a) The CIT-5 framework structure viewed along b; (b) the column of cage units

The smallest characteristic subunit of STT is a small cage composed of three 4-rings and four 5-rings. Figure 2.40 shows two characteristic cage units. There are three fluorine ions in the cage. Apart from coordinating with oxygen atoms, some Si atoms also coordinate with F
ions to complete five-coordination. These units are linked together, through 4-rings and 6-rings, resulting in slightly puckered layers parallel to the (101) plane. These layers are then linked together via 4-rings, giving rise to the threedimensional structure as shown in Figure 2.41(a). The ‘pillaring’ of layers produces channels running parallel to the [001] direction that are bounded at their narrowest points ˚ . Parallel to the [101] direction are by 7-ring windows with a free aperture of 2.4  3.5 A ˚ , slightly offset the channels bounded by 9-ring windows with free aperture of 3.7  5.3 A against each other [Figure 2.41(b)]. Cages are located at the intersection of 7-ring and 9-ring channels. The N, N, N-trimethyl-l-adamantammonium (TMAdaþ) template cations are located inside the cages.

Figure 2.40 Two of the characteristic subunits of SSZ-23 and the fluorines inside the small cages. Reproduced with permission from [110]. Copyright (1998) Wiley-VCH

Structural Chemistry of Microporous Materials

59

Figure 2.41 (a) The structure of SSZ-23 viewed along the [001] direction, showing the 7-ring windows and the position of the TMAdaþ cations; (b) viewed along the [101] direction, showing the channels bound by the 9-ring windows. Reproduced with permission from [110]. Copyright (1998) Wiley-VCH

Among the known zeolite framework types, rings with an odd number of tetrahedra are rare, with the exception of the more frequent 5-rings, while even-numbered rings are ubiquitous, such as 4-, 6-, 8-, 10-, and 12-rings. In the very few cases are 3-, 7-, or 9-rings known, and their presence is associated with either Zn2þ or Be2þ sites in the tetrahedral building units or to a very specific templating effect. The LOV, VSV, and RSN structure types represent the zeolite-type materials possessing 3- and 9-rings, and all of them have tetrahedral Zn2þ or Be2þ sites in the structure. SSZ-23 (STT) is the first zeolite containing 7- and 9-rings. DON[111,112] Type material UTD-1F, j[(Cp*)2Co]þ2F
1.5(OH)
0.5j[Si64O128]-DON), monoclinic Pc, ˚ , b ¼ 8:476 A ˚ , c ¼ 30:028 A ˚ , b ¼ 102:65 .[111] Its framework density a ¼ 14:970 A 3 ˚ . UTD-1F is an extra-large-pore silicate-based zeolite with oneis 17.2 T/1000 A ˚. dimensional 14-ring channels with a pore aperture of  8.1  8.2 A The framework topology of DON can be described in terms of layers of ‘butterfly’ units (a 6-ring ‘body’ with two 5-rings on each side forming the ‘wings’) arranged in a centered manner (see Figure 2.42). The 5-rings of these units are linked via oxygen bridges to form a layer with 4-rings and oval 14-rings. Adjacent layers are related by a mirror plane. A strict UD (up-down) alternation of the orientation of the Si tetrahedra in the 14-rings creates a channel wall consisting of 6-rings only. The 4-rings have a UUDD arrangement of Si tetrahedra, and form double crankshafts between parallel 14-ring channels. The [(Cp*)2Co]þ ion lies in the plane of one of the two symmetrically independent 14-rings with the Cp* rings lying above and below it (between adjacent 14-rings). The ordered arrangement of the [(Cp*)2Co]þ ions described above causes the ideally B-centered orthorhombic framework to become primitive monoclinic. The relationship between the two unit cells is shown in Figure 2.42. The ‘orthorhombic’ setting of the unit ˚ , b ¼ 8:470 A ˚ , c ¼ 23:414 cell for UTD-1F has the lattice parameters with a ¼ 18:736 A  ˚ A, and b ¼ 90:2 . In contrast to the structure analysis of calcined UTD-1 reported by

60

Chemistry of Zeolites and Related Porous Materials

Figure 2.42 Framework structure of UTD-1F showing the relationship between the topological orthorhombic unit cell (dashed line) and the observed monoclinic one (full line). Filled open circles indicate Si tetrahedra pointing down into the page and up out of the page, respectively. A ‘butterfly’ unit in the upper left-hand corner is highlighted. Reproduced with permission from [111]. Copyright (1999) American Chemical Society

Lobo et al.,[112] this refinement clearly showed that there was no significant disorder or faulting in the UTD-1F framework. IWW[113] ˚, Type material ITQ-22, [Si89.8Ge22.2O224]-IWW, orthorhombic Pbam, a ¼ 42:1326 A [113] 3 ˚ ˚ ˚ b ¼ 12:9885 A, c ¼ 12:6814 A, framework density, 16.1 T/1000 A . It contains interconnected 8-, 10-, and 12-ring pores synthesized by combining the structuredirecting effect of the organic 1,5-bis-(methylpyrrolidinium)-pentane and the framework isomorphic substitution of germanium for silicon. As shown in Figure 2.43(a), its structure contains [4458612] cages that are stacked along the c axis, forming D4R units and giving rise to columns. Each column is linked to the nearest ones either directly or through an additional bridging T atom. ITQ-22 contains ˚ ) and 12-ring a three-dimensional channel system. It contains 8-ring (3.3  4.6 A ˚ (6.0  6.4 A) straight channels along the c axis [Figure 2.43(b)]. There is a sinusoidal ˚ ) channel system normal to the c axis [represented by the black 10-ring (4.9  4.9 A ribbon in Figure 2.43(c)] intersecting the 8- and 12-ring channels. The maximum Ge incorporation was found for a Si/Ge ratio of 3.2, which closely corresponds to the half-occupancy of the D4R sites by Ge atoms, leaving the remaining positions as purely siliceous. The Ge distribution indicates that the stabilization of the D4R units by Ge is an important parameter for the formation of ITQ-22 structure. CON[114,115] ˚, Type material CIT-1, jHþ2j[B2Si54O112]-CON, monoclinic C2/m, a ¼ 22:624 A [114] 3 ˚ ˚ ˚ b ¼ 13:350 A, c ¼ 12:364 A, framework density, 16.1 T/1000 A . The structure of CIT-1 is the pure polymorph B of zeolites SSZ-33 and SSZ-26.[115] The structures of SSZ-33 and SSZ-26 are intergrowths of two different polymorphs A and B with fault probabilities of approximately 15% and 30% (polymorph B is predominant), respectively. Polymorph A is formed by the stacking of layers in an ˚ , b ¼ 12:33 A ˚ , c ¼ 21:08 A ˚ ), ABAB. . . sequence (orthorhombic, Pmna, a ¼ 13:26 A

Structural Chemistry of Microporous Materials

61

Figure 2.43 Structure of ITQ-22. (a) [4458612] cage; (b) framework viewed along c axis; (c) three-dimensional structure model showing the 8- and 12-ring channels that are intersected by the sinusoidal 10-ring pore (black ribbon). Reproduced with permission from [113]. Copyright (2003) Nature Publishing

62

Chemistry of Zeolites and Related Porous Materials

Figure 2.44 Framework structure of polymorph A. (a) viewed along the 12-ring pores; (b) along the 10-ring pores. Framework of polymorph B (c) viewed along the 12-ring pores; (d) along the 10-ring pores. Reproduced with permission from [114]. Copyright (1995) American Chemical Society

while polymorph B is formed by an ABCABC. . . stacking sequence (monoclinic, B2=m, ˚ , b ¼ 12:33 A ˚ , c ¼ 22:62 A ˚ , a ¼ 68:7 ). Figure 2.44 shows the framework a ¼ 13:26 A structures of polymorphs A and B. Both of them contain a three-dimensional pore system with intersecting 10- and 12-ring pores. CIT-1 contains intersecting 10- and 12-ring pores but is not an intergrowth of two ˚ different polymorphs. The free diameters for the 10- and 12-ring channels are 5.1  5.1 A ˚ , respectively. However, the results of the structural refinement also and 6.8  6.4 A indicate that CIT-1 is not completely fault-free. The overall stacking sequence of CIT-1 crystal contains a small probability of faulting, probably close to 1%. ETR[116] Type material ECR-34, jH1.2K6.3Na4.4j[Ga11.6Al0.3Si36.1O96]-ETR, hexagonal P63mc, ˚ , b ¼ 21:030 A ˚ , c ¼ 21:030 A ˚ ,[116] framework density, 14.7 T/1000A ˚ 3. a ¼ 21:030 A 6 2 6 As shown in Figure 2.45, the structure of ECR-34 is featured by a 4 6 8 polyhedral unit (plg cage). It is built up by open hexagonal prisms to form one-dimensional puckered ˚ (Figure 2.46). The channel walls are composed 18-ring channels with aperture of 10.1 A of 4-ring ribbons having 8-ring windows to the pgl cages and shallow side pockets of 10ring windows at the channel. The cations are located in the 8-ring window of the open hexagonal prism and the 8-ring window of the pgl polyhedra in the 18-ring channel. The combination of Na, K, and TEA cations (tetraethylammonium) is suggested to help stabilize the pgl polyhedral unit in the ECR-34 framework. SFH and SFN[117] ˚, Type material SSZ-53, [Si62.4B1.6O128]-SFH, monoclinic C2/c, a ¼ 5:0192 A 3  [117] ˚ ˚ ˚ b ¼ 33:7437 A, c ¼ 21:1653 A, b ¼ 90:485 , framework density, 17.9 T/1000 A .

Structural Chemistry of Microporous Materials

63

Figure 2.45 Polyhedra found in ECR-34: left, plg cage; right, open hexagonal prisms right. Reproduced with permission from [116]. Copyright (2003) American Chemical Society

˚ , b ¼ 12:735 Type material SSZ-59, [Si15.65B0.35O32]-SFN, triclinic P1, a ¼ 5:023 A    [117] ˚ ˚ A, c ¼ 14:722 A, a ¼ 103:440 , b ¼ 90:51 , g ¼ 100:880 , framework density, ˚ 3. 17.8 T/1000 A The framework structures of SSZ-53 and SSZ-59 are very similar, both of which ˚. possess elliptical 14-ring pores with dimensions of approximately 8.5  6.5 A Figure 2.47 shows the framework structures of SSZ-53 and SSZ-59. The topological structures can be derived from different symmetry operations performed on the same subunit. The layers in SSZ-53 are related by a two-fold rotation about an axis within the plane of the Figure, while the layers in SSZ-59 are related by a two-fold rotation about an axis perpendicular to the Figure. The subtle differences between SSZ-53 and SSZ-59 are reflected by their pore structures. Figure 2.48 shows the pore structures of SSZ-53, SSZ-59, and UTD-1

Figure 2.46 Framework structure of ECR-34 viewed down the c direction, showing the 18-ring channel. Framework oxygen atoms and nonframework cations are omitted for clarity. Reproduced with permission from [116]. Copyright (2003) American Chemical Society

64

Chemistry of Zeolites and Related Porous Materials

Figure 2.47 Structural relationship between (a) SSZ-53 and (b) SSZ-59. The layers in SSZ-53 are related by rotations about two-fold axes within the plane of the Figure. The layers in SSZ-59 are related by rotations about two-fold axes perpendicular to the plane of the Figure. Reproduced with permission from [117]. Copyright (2003) Wiley-VCH

Figure 2.48 Comparison of the pore structures in SSZ-53, SSZ-59, and UTD-1. Reproduced with permission from [117]. Copyright (2003) Wiley-VCH

Structural Chemistry of Microporous Materials

65

Figure 2.49 Projections of the ITQ-24 structure along the crystallographic axes. Reproduced with permission from [118]. Copyright (2003) American Chemical Society

(DON), all of them containing a 14-ring channel system. While UTD-1 has pores bound by smooth 6-ring nets, both SSZ-53 and SSZ-59 possess pores with corrugated surfaces. In SSZ-53, these corrugations are centered about the vertices (of the major axis) of the elliptical pore, while in SSZ-59 the corrugations are positioned on the sides (i.e., near the vertices of the minor axis) of the elliptical pore. IWR[118] Type material ITQ-24, [Si48.3Ge5.1Al2.6O224]-IWR, orthorhombic Cmmm, a ¼ 21:2549 ˚ , b ¼ 13:5210 A ˚ , c ¼ 12:6095 A ˚ ,[118] framework density, 15.5 T/1000 A ˚ 3. It is a D4RA containing silicogermanate zeolite synthesized by isomorphic incorporation of Ge atoms into the siliceous zeolite frameworks. Its structure possesses a three-dimensional channel system, which includes a 12-ring straight channel running perpendicular to the ab plane with a pore aperture of ˚ , a 12-ring sinusoidal channel (7.7  6.2 A ˚ ) along the a axis, approximately 7.7  5.6 A ˚ and a 10-ring channel (5.7  4.8 A) intersecting perpendicularly both the 12-ring channel systems (Figure 2.49). The sinuous 12-ring channel is surrounded by D4R units. As with other Ge containing zeolites, there is a preferential occupation of Ge atoms in the D4R units. UTL[119,120] ˚, Type material IM-12, [Ge13.8Si62.2O152]-UTL, monoclinic C2/m, a ¼ 29:8004 A ˚ , c ¼ 12:3926 A ˚ , b ¼ 105:185 ,[119] framework density, 15.2 T/1000 A ˚ 3. b ¼ 13:9926 A It is an extra-large pore zeolite with a two-dimensional channel system formed by 14- and 12-rings. The structure of IM-12 can be described as layers stacked along the [100] direction, which are connected to each other by their four-rings and thereby forming D4R units. It ˚ ) and 12-ring (5.5  8.5 A ˚) contains a 2-D channel system formed by 14-ring (7.1  9.5 A intersecting channels parallel to the c and b axes, respectively (Figure 2.50). The Ge atoms in the D4R units are highlighted by dark gray circles. According to chemical analysis, the average Si/Ge ratio in the D4R units is 7. The framework density of 15.2 T ˚ 3 is lower than those of other zeolite materials containing 14-ring atoms per 1000 A channels, such as UTD-1 (DON, 17.2), CIT-5 (CFI, 18.3), SSZ-53 (SFH, 17.9), and SSZ59 (SFN, 17.8), but higher than those for ECR-34 (ETR, 14.7) with an 18-ring pore and OSB-1 (OSO, 13.4) with 14- and 8-ring intersecting channels.

66

Chemistry of Zeolites and Related Porous Materials

Image Not Available In The Electronic Edition

AFI[121–126] Type material AlPO4-5, j(C12H28Nþ)(OH
)(H2O)xj[Al12P12O48]-AFI, hexagonal P6cc, ˚ , c ¼ 8:484 A ˚ .[121] Its framework density is 17.3 T/1000 A ˚ 3. a ¼ 13:726 A AlPO4-5 is the most well known member in the family of aluminophosphate molecular sieves. The PO4 and AlO4 tetrahedra are strictly alternating via bridging oxygen atoms. The AFI framework structure is based on the (4.6.12) 2-D net shown in Figure 2.51(a). The fourth oxygen atom of each TO4 tetrahedron points upward and downward alternately from the 2-D net. A parallel stack of these 3-connected 2-D nets along the c axis produces 1-dimensional 12-ring channels with a free pore aperture ˚ [Figure 2.51(b)]. The 12-ring channel wall in AFI is lined solely with of  7.3  7.3 A 6-rings.

Figure 2.51 (a) The 4.6.12 net in AlPO4-5; (b) framework structure of AlPO4-5

Structural Chemistry of Microporous Materials

67

Figure 2.52 Framework topology of AlPO4-5 and VPI-5 (bridging oxygen atoms are omitted)

VPI-5 (VFI)[127–132] ˚, Type material VPI-5, j(H2O)42j[Al18P18O72]-VFI, hexagonal P63, a ¼ 18:975 A ˚ .[128] The framework density is 14.2 T/1000 A ˚ 3. VPI-5 is the first molecular c ¼ 8:104 A sieve containing pore openings larger than 12 T-atoms. It has 18-ring channels with free ˚. aperture of 12.7  12.7 A The framework of VFI is closely related to that of AlPO4-5 (Figure 2.52). In AFI the 12-ring is enclosed by six 4-rings and six 6-rings. Instead of being linked via 4-rings, the 6-rings in the VFI framework type are linked via two 4-rings sharing a common edge, i.e., fused 4-rings. This produces an 18-ring in VFI instead of the 12-ring found in AFI. The structure of VPI-5 is made of AlO4, AlO4(H2O)2, and PO4 units through Al-O-P linkages (Figure 2.53). Two water molecules are bonded to the Al atom at the center of the fused 4-rings of trans conformation. A chain of hydrogen-bonded water molecules following the 63 screw axis links the octahedrally coordinated Al atoms and forms a triple helix of water molecules inside the 18-ring channel of VPI-5 (Figure 2.54). In contrast to vast majority of hydrated zeolite structures, the water molecules in this structure have the same symmetry as the framework. Under carefully controlled conditions, VPI-5 can be dehydrated and retain its framework structure. All the Al atoms are tetrahedrally coordinated in the dehydrated VPI-5. To accommodate tetrahedral angles at the Al site between the fused 4-rings, some of the P

O

Al linkages have been forced to adopt extremely small angles, less than 130 . All

Figure 2.53 The framework structure of VPI-5 (bridging oxygen atoms are omitted)

68

Chemistry of Zeolites and Related Porous Materials

Figure 2.54 A representation of the triple helix of water molecules within the 18-ring channel. The dotted lines indicate bonds to octahedrally coordinated Al in the framework. Reproduced with permission from [128]. Copyright (1991) Elsevier

of these strained linkages are located in the 18-rings and are associated with the fused 4-rings. This distortion causes a symmetry reduction from P63 in the as-synthesized form to Cm in the dehydrated one.[132] VPI-5 can transform into the 14-ring molecular sieve AlPO4-8 after dehydration.[133–135] AlPO4-8 (AET)[136,137] ˚, Type material AlPO4-8, [Al36P36O144]-AET, orthorhombic Cmc21, a ¼ 33:29 A [136] 3 ˚ ˚ ˚ b ¼ 14:76 A, c ¼ 8:257 A. The framework density is 17.7 T/1000 A . It is the first molecular sieve with 14-ring channels. The AET framework structure is shown in Figure 2.55. The AlO4 and PO4 tetrahedra alternate strictly in the framework to form a one-dimensional 14-ring channel along the ˚. [001] direction with a free aperture of 7.9  8.7 A The AET framework is based on the (4.6.14)2(4.4.14)1(4.6.14)2(4.6.6)2(6.6.14)2 2-D net, on which the vertices of the tetrahedra point up and down to the layer alternately. The layers are stacked along the [001] direction to form the AET framework. Under appropriate conditions the 18-ring molecular sieve VPI-5 is transformed into AlPO4-8 with 14-rings. As shown in Figure 2.56, the VPI-5 framework is based on the (4.4.18)1(4.6.18)2 2-D net with 6-rings separated by fused 4-rings.[17] In the

Figure 2.55 Projection of AlPO4-8 along [001]

Structural Chemistry of Microporous Materials

69

Figure 2.56 Schematic showing of the phase transformation of VPI-5 into AlPO4-8. Reproduced with permission from [17]. Copyright (1994) Elsevier

transformation of VPI-5 into AlPO4-8, 4/6 of the fused 4-rings are transformed into 6-rings, with the ‘dangling’ Als and Ps reconnecting to form new 4-rings; thus, the largest channel opening is reduced from an 18- to a 14-ring. It should be noted that Figure 2.56 shows only the (4;2)-connected topology of VPI-5 and AlPO4-8. In fact, the AlO4(H2O)2 coordination is maintained in the transformation of VPI-5 into AlPO4-8. -CLO[138] Type material cloverite, j(C7H14Nþ)24j8[F24Ga96P96O372(OH)24]8-CLO, cubic Fm-3c, ˚ ,[138] and the framework density is 11.1 T/1000 A ˚ 3. a ¼ 52:712 A Cloverite is a gallophosphate molecular sieve with 20-ring channels. The GaO4 tetrahedra and PO4 tetrahedra with terminal -OH groups alternate strictly to produce an interrupted three-dimensional open framework. Figure 2.57 shows a projection of cloverite along the [100] direction with a pore opening in the shape of a four-leafed

Figure 2.57 Projection of cloverite along [100] showing the cloverleaf-shaped 20-ring windows. The nodes represent either Ga or P; the filled circles, terminal hydroxy groups. Reproduced with permission from [138]. Copyright (1991) Nature Publishing

70

Chemistry of Zeolites and Related Porous Materials

Figure 2.58 The framework topology of the CLO showing the cubic arrangement of a- and rpa-cages. Reproduced with permission from [138]. Copyright (1991) Nature Publishing

clover. The terminal -OH groups protrude into the channel and the free aperture of the ˚. pore is 6.0  13.2 A Figure 2.58 shows the framework topology of cloverite. Its framework can be described as a cubic arrangement of a-cages connected via two rpa-cages. It is worth noting that all the 4-rings of these cages are involved in double 4-rings (D4Rs). In fact, the entire structure can be built from D4Rs. Half of these D4Rs are not fully connected, and the coordination spheres of a Ga and a neighboring P in these cubes are completed by terminal hydroxy groups. Thus, this material has an interrupted framework. The framework has two nonintersecting three-dimensional channel systems. One of these passes through the a- and rpa-cages and has 8-ring pore openings, and the other passes through the face of the cube described by 20 tetrahedral atoms (Ga and P) and 24 oxygen atoms. At the intersection of these channels is a large cubic supercage with pockets at the corners (Figure 2.59). The body diagonal from pocket to pocket is

Figure 2.59 The supercage of cloverite, showing the cloverleaf-shaped windows and the pockets at the corners of the cage

Structural Chemistry of Microporous Materials

Figure 2.60

71

(a) The CZP framework along [001]; (b) the helix formed by 4-ring ‘squares’

˚ . Cloverite has an extremely open structure, and its framework density is one of 29–30 A the lowest among the known zeolite structures. All the D4Rs in the structure have a fluoride ion incorporated in the centre. The D4Rs are distorted in such a way that the F
ion can approach all four Ga atoms at distances ˚ . The Ga atoms, in turn, become five-coordinated with ranging from 2.30 to 2.66 A ˚, distorted trigonal bipyramid coordination. The P

F distances are all greater than 2.78 A and the P atoms retain the tetrahedral geometry. CZP[139–142] ˚, Type material jNaþ12(H2O)12j[Zn12P12O48]-CZP, hexagonal P6122, a ¼ 10:480 A ˚ .[140] Its framework density is 16.7 T/1000 A ˚ 3. CZP is a chiral zinc c ¼ 15:089 A phosphate. The framework of CZP along the [001] direction is shown in Figure 2.60(a). The alternation of vertex-sharing ZnO4 and PO4 tetrahedra gives rise to a framework which includes novel 4-ring ‘squares’ and interesting edge-sharing helices of 4-rings. The helices form a highly tortuous 1-dimensional, 12-ring channel with an aperture of ˚ along the [001] direction [Figure 2.60(b)]. 3.8  7.2 A SBS[143] Type material UCSB-6GaCo, j(C9H24N2)2þ12j[Ga24Co24P48O192]-SBS, trigonal P-31c, ˚ , c ¼ 27:182 A ˚ ,[143] framework density, 12.8 T/1000 A ˚ 3. It is a cobalt a ¼ 17:836 A gallophosphate zeolite with a multidirectional 12-ring channel system. A fundamental structural unit in SBS is the cancrinite (can) cage. Two can cages join through double 6-ring units (D6Rs) to form the can-D6R-can unit. The can-D6R-can unit is symmetrically capped with six additional T-atoms at each end to form the capped unit [Figure 2.61(a)]. The structure of SBS consists of columns of capped can-D6R-can units cross-linked to each other by oxygen bridges. Figure 2.61(b) shows the framework of SBS. It has 12-ring channel systems along c-direction and perpendicular to c-direction. The framework can also be considered as a hexagonal packing of capped can-D6R-can units with their long axes aligned along the hexagonal c axis.

72

Chemistry of Zeolites and Related Porous Materials

Image Not Available In The Electronic Edition

Figure 2.61(c) shows two different 12-ring large cages in SBS denoted as a and b: a is bound with five 12-rings, and b with three. The 3-D 12-ring channel system of UCSB-6 is similar to that found in EMT. It can be derived from the EMT type structure by simply replacing the sodalite cage with a one-side capped cancrinite cage. It is noted that the sodalite cage and the one-side capped cancrinite cage have the same number of T-atoms (24 T-atoms), and thus the resulting framework has a similar T-atom density ˚ 3 for SBS and 129 T/1000 A ˚ 3 for EMT). (12.8 T/1000 A

2.5 Zeolitic Open-framework Structures We have discussed above the tetrahedrally connected zeolite frameworks. Following the discovery of aluminophosphate molecular sieves in 1982, the emergence of zeolite-like inorganic open-framework compounds has greatly enriched the compositional and structural chemistry of microporus materials. To date, the structural types of openframework microporous compounds have rapidly increased, and the framework elements have expanded to include the majority of elements in the periodic table. In contrast to aluminosilicate and aluminophosphate zeolites built up from TO4 tetrahedra, their frameworks are constructed from TOn polyhedra (n ¼ 3, 4, 5, 6, etc). Templates, water molecules and other guest species are typically trapped in the channels or cages. As compared with zeolites, most zeolite-like open-framework compounds have lower thermal stability. In this section, we will discuss mainly zeolite-like inorganic openframework compounds with interesting structural features. 2.5.1

Anionic Framework Aluminophosphates with Al/P  1

Since the discovery of neutral framework aluminophosphate molecular sieves AlPO4-n, a large number of anionic framework aluminophosphates have been reported. Jilin group has carried out systematic research in this field.[144–147] Recently, they have built up a

Structural Chemistry of Microporous Materials

73

structure database containing detailed structural information on open-framework aluminophosphates reported in the literature.[148] To date, more than 200 structure-types of open-framework aluminophosphates (hereafter designated AlPOs) have been identified. In contrast to AlPO4-n with Al/P ¼ 1, the Al/P ratios in anionic framework AlPOs are typically less than unity (in some cases Al/P ¼ 1). Their frameworks are made up of strictly alternating Al-centered polyhedra (AlO4, AlO5, AlO6) and P-centered tetrahedra P(Ob)n(Ot)4
n(b, bridging; t, terminal; n ¼ 1, 2, 3, 4) sharing four, three, two, or one oxygen atoms with adjacent Al atoms. The anionic framework aluminophosphates exhibit diverse stoichiometries and fascinating structural architectures. Their stoichiometries include AlPO4(OH)
,[149–52] AlP4O169
,[153] AlP2O83
,[154–164] Al2P3O123
,[165–171] Al3P4O163
,[155,172–185]Al3P5O206
,[168,182] Al4P5O203
,[186–189] Al5P6O243
,[190,191] Al11P12O483
,[192,193] Al12P13O523
,[194] Al13P18O7215
,[195] and so forth. Table 2.7 summarizes the structural features of the anionic framework aluminophosphates with various stoichiometries. These anionic open frameworks comprise threedimensional (3-D) framework, 2-D layer, 1-D chain, and 0-D cluster structures. Notable examples are JDF-20[190,191] with the largest channel ring size of 20 among openframework AlPOs, and AlPO-CJB1 which is the first aluminophosphate moleculer sieve with Bro¨nsted acidity.[194] Anionic framework AlPOs have also been prepared with diverse low-dimensional framework structures, which provide important information for the understanding of the formation mechanism of microporous materials. In this chapter, we will discuss the structural chemistry of anionic framework aluminophosphates ranging from 3-dimensional frameworks to low dimensional frameworks. Table 2.7 Anionic aluminophosphates with various dimensionalities, stoichiometries, and coordinations (Al/P  1)a Dimensionality

Stoichiometry

Al and P coordination

3-D

Al12P13O523
Al11P12O483
A5P6O243
Al4P5O203
Al3P4O163
Al2P3O123
AlP2O83
Al13P18O7215
Al4P5O203
Al3P4O163
Al3P4O163
Al2P3O123
Al2P3O123
AlP2O83
AlP2O83
AlPO4(OH)
Al3P5O206
AlP2O83
AlP2O83
AlP4O169


AlO4b, AlO5b, PO4b AlO4b, AlO6b, PO4b AlO4b, PO4b, PO3bOt AlO4b, AlO5b, PO4b, PO2bO2t AlO4b, PO3bOt AlO4b, PO3bOt, PO2bO2t AlO6b, PO3bOt AlO6b, AlO4b, PO3bOt AlO4b, AlO5b, PO4b, PO3bOt, PO2bO2t AlO4b, AlO5b, PO4b, PO3bOt, PO2bO2t AlO4b, PO3bOt AlO4b, AlO5b, PO4b, PO3bOt, PO2bO2t AlO4b, PO 3bOt, PO2bO2t AlO4b(H2O)2, PO2bO2t AlO4b, PO2bO2t AlO3b(OH), PO3bOt AlO4b, PO3bOt, PO2bO2t, PObO3t AlO4b, PO2bO2t AlO4b, PO3bOt, PObO3t AlO4b, PObO3t

2-D

1-D 0-D a

b, bridging oxygen; t, terminal oxygen.

74

Chemistry of Zeolites and Related Porous Materials

Figure 2.62 A propeller-like chiral motif formed by three cyclic 4-rings. Reproduced with permission from [154]. Copyright (2000) Royal Society Chemistry

Anionic 3-D Open-framework Aluminophosphates To date, the Al/P ratios of anionic 3-D open-framework AlPOs less than unity are found to be 1/2, 2/3, 3/4, 4/5, 5/6, 11/12, and 12/13. AlPO-CJ4 ([H3O][AlP2O6(OH)2])[154] is the first anionic 3-D open-framework aluminophosphate with an Al/P ratio of 1/2. It crystallizes in the triclinic space group P-1 with ˚ , b ¼ 8:6972ð2Þ A ˚ , c ¼ 9:2200ð3Þ A ˚ , a ¼ 65:108ð2Þ , b ¼ 70:521ð1Þ , a ¼ 7:1177ð2Þ A  and g ¼ 68:504ð2Þ . Its framework is constructed from alternation of AlO6b octahedra and PO3b(OH) tetrahedra via vertex oxygen atoms. In this structure, the primary Al building units are solely made up of AlO6 octahedra with all six oxygen vertices being shared by adjacent P atoms. Each asymmetric unit contains two crystallographically distinct Al atoms and three crystallographically distinct P atoms. Interestingly, the structure of AlPO-CJ4 features chiral propeller-like motifs as found for Co(en)33þ. As can be seen from Figure 2.62, this chiral motif is formed by Al(2)-centered octahedron with three cyclic 4-rings. These chiral motifs are connected with each other to form a puckered 2-D layer parallel to the ab plane, which contains double crown-like 12-rings as shown in Figure 2.63. Each 12-ring is surrounded by six chiral motifs, three of - and three of

Figure 2.63 2-D layer formed by connection of the chiral motifs. Reproduced with permission from [154]. Copyright (2000) Royal Society of Chemistry

Structural Chemistry of Microporous Materials

75

Figure 2.64 3-D open framework formed by connection of the 2D layers via Al(1) atoms (viewed along the [100] direction). Reproduced with permission from [154]. Copyright (2000) Royal Society of Chemistry

-configuration. The 2-D layers are connected via Al(1) atoms lying at inversion centers to form the 3-D open framework of AlPO-CJ4, containing interconnected 8-ring channels along the [100], [010], and [001] directions. Figure 2.64 shows the framework of AlPOCJ4 viewed along the [100] direction. The protonated water molecules are trapped in the channels, and interact with terminal oxygen atoms attached to P atoms through H-bonds ˚. with OW. . .Oterminal distances in the range of 2.853–3.086 A [165] AlPO-DETA ([C4H16N3][Al2P3O12]) has an Al/P ratio of 2/3. It crystallizes in the ˚ , b ¼ 8:537ð2Þ A ˚ , c ¼ 10:252ð3Þ A ˚, monoclinic space group C2/c with a ¼ 17:669ð4Þ A  and b ¼ 103:42ð3Þ . Its structure is constructed from alternation of AlO4b tetrahedra and

O) and PO2b(


O)2] via vertex oxygen atoms (b, bridging). PO4 tetrahedra [PO3b(


Figure 2.65(a) shows the framework of AlPO-DETA viewed along the [001] direction. It consists of parallel 12- and 8-ring channels. The triprotonated DETA molecules þ H3N(CH2)2NH2þ(CH2)2NH3þ are trapped in the 12-ring channel, and interact with

Figure 2.65 Framework of AlPO-DETA along [001] (a) and [010] (b) (its SBU and SBU chain are shown). Reproduced with permission from [146]. Copyright (2003) American Chemical Society

76

Chemistry of Zeolites and Related Porous Materials

Figure 2.66 (a) The framework of [C24H91N16][Al9(PO4)12].17H2O; (b) the cubic structural unit formed by four 4-rings; (c) the supercage with 12-ring windows. Reproduced with permission from [172]. Copyright (1999) Elsevier

the terminal oxygen atoms attached to P atoms through H-bonds. Along the [010] direction, it contains 8-ring channels. The framework of AlPO-DETA can also be viewed as built up from connection of the 4.8-net sheets parallel to the bc plane through
PO2(


O)2 tetrahedra. Figure 2.65(b) shows the SBU of AlPO-DETA and a chain along a built up from SBUs. [C24H91N16][Al9(PO4)12].17H2O[172] has an Al/P ratio of 3/4. It crystallizes in the ˚ . The 3-D framework is constructed cubic space group I-43m with a ¼ 16:7963ð13Þ A
from alternating AlO4b tetrahedra and PO3b(
O)

tetrahedra via vertex oxygen atoms forming 12-ring channels. Figure 2.66(a) shows its framework. The structure contains a cubic structural unit made up of four 4-rings [Figure 2.66(b)], and a supercage with 12-ring windows [Figure 2.66(c)]. The protonated tris(l-aminoethyl)amine (TREN)
molecules are located in the 12-ring windows, and interact with P


O groups protruding into the channel through H-bonds. AlPO-HDA ([(C6H18N2)4][Al16P20O80H4])[186] has an Al/P ratio of 4/5. It crystallizes ˚ , b ¼ 5:108ð1Þ A ˚ , c ¼ 25:488ð1Þ in the monoclinic space group Cc with a ¼ 17:682ð1Þ A ˚ , and b ¼ 103:07ð1Þ . The structure is made up of alternation of Al-centered units A
(AlO4b, AlO5b), and P-centered units [PO4b, PO2b(OH)(


O)] in which all the vertices are shared except for the terminal oxygen atoms. Figure 2.67 shows the framework of AlPO-HDA along the [010] direction. The 12-ring channels along the [010] direction are intersected with 8-ring channels along the [100] direction. The diprotonated 1,6-hexanediamine (HDA) molecules are trapped in the main 12-ring channels. The framework of AlPO-HDA is composed of a series of D6R columns stacked in the [010] direction. These columns of D6Rs interconnect through 4-rings to

O) tetrahedra to form layers, which are further connected together through PO2b(OH)(

build up the 3-D open framework. JDF-20 ([(Et3NH)2][Al5P6O24H].2H2O)[190,191] has an Al/P ratio of 5/6. To date, JDF20 has the largest pore opening in microporous aluminophosphates. It crystallizes in the ˚ , b ¼ 14:308 A ˚ , c ¼ 8:852 A ˚ , and monoclinic space group C2/c with a ¼ 32:035 A b ¼ 104:65 . The structure is constructed from alternation of AlO4 tetrahedra and PO4 tetrahedra in which all the Al vertices are shared by adjacent P vertices; one third of P

Structural Chemistry of Microporous Materials

77

Figure 2.67 Structure of AlPO-HDA viewed along b axis (Its SBU and characteristic columns of D6Rs are shown.) Reproduced with permission from [146]. Copyright (2003) American Chemical Society

atoms are connected to Al atoms via four vertex oxygen atoms, and two thirds of P atoms have a terminal P

O bond. Figure 2.68 shows the framework of JDF-20 which contains ˚ ) intersected by 20-ring channels with elliptical apertures (free diameter of 6.2  7.9 A smaller 10- and 8-ring channels. Four (protonated) triethylammonium Et3NHþmoieties are trapped in the large 20-ring channel and interact with terminal P

O groups through

Figure 2.68 Framework of JDF-20 [Its SBU and the linear corner sharing 4-ring chain (1) and zig-zag edge-sharing 4-ring chain (2) are shown.] Reproduced with permission from [146]. Copyright (2003) American Chemical Society

78

Chemistry of Zeolites and Related Porous Materials

Figure 2.69 View of cage 1 (4366 86 cage) and cage 2 (612 cage) in AlPO-CJ11. Reproduced with permission from [192]. Copyright (2001) Royal Society of Chemistry

H-bonds. The structure of JDF-20 also features a series of corner-sharing and edgesharing 4-ring chains. AlPO-CJ11 ([C4H12N2][C4H11N2][Al11P12O48])[192] has an Al/P ratio of 11/12. It ˚ , and c ¼ 42:091ð6Þ A ˚. crystallizes in the trigonal space group R-3c, with a ¼ 14:045ð2Þ A In contrast to most of the anionic frameworks containing terminal oxygen atoms, the framework of AlPO-CJ11 is constructed from alternation of AlOn polyhedra (n ¼ 4 and 6) and PO4 tetrahedra via oxygen bridges in which all the vertex oxygen atoms are shared by adjacent Al and P atoms. The framework of AlPO-CJ11 features two new types of cages, as shown in Figure 2.69, that is, cage 1, which is composed of three 4-rings, six 6-rings, and six 8-rings (denoted as 43 66 86 cage), and cage 2 composed of 12 6-rings (denoted as 612 cage). A column is formed through alternation of cages 1 and 2 that share a common snowflake-like motif along the c direction (Figure 2.70). Such columns connect with each other to form the 3-D framework of AlPO-CJ11 (Figure 2.71). AlPO-CJB1 ([(CH2)6N4H3]3þ[Al12P13O52]3
)[194] has an Al/P ratio of 12/13. It is the first anionic aluminophosphate molecular sieve possessing Bro¨nsted acidity. It crystallizes

Figure 2.70 (a) A column formed by alternation of cage 1 and cage 2 along the c direction; (b) A chiral motif which is composed of Al(1)-centered six 4-rings. Reproduced with permission from [192]. Copyright (2001) Royal Society of Chemistry

Structural Chemistry of Microporous Materials

79

Figure 2.71 (a) The cage column viewed along c direction; (b) the framework of AlPO-CJ11 formed via connection of cage columns. Reproduced with permission from [192]. Copyright (2001) Royal Society of Chemistry

˚ and c ¼ 15:547ð2Þ A ˚ . Its in the tetragonal space group P-421c with a ¼ 13:610ð1Þ A framework is constructed by strict alternation of Al-centered polyhedra (AlO4 and AlO5) and P-centered tetrahedra (PO4) by sharing all oxygen atoms at vertexes. Its structure is analogous to that of AlPO4-22 (AWW zeotype). The open framework of AlPO-CJB1, viewed along [001] direction, is shown in Figure 2.72. In AlPO-CJB1, three kinds of cages exist, i.e., an aww cage, a P-incorporated aww cage, and an rpa cage. The aww cages and P-incorporated aww cages alternate along the [001] direction to form a column. Four such columns are connected with each other through 4-rings and 6-rings to form the rpa cages sharing 8-rings, giving rise to an 8-ring channel with a free diameter ˚ along the [001] direction. The maximum cavity diameter measured of 3.9  3.9 A

Figure 2.72 (a) Framework of AlPO-CJB1 viewed along c; (b) a side view of the structure showing the connection of three types of cages. Reproduced with permission from [146]. Copyright (2003) American Chemical Society

80

Chemistry of Zeolites and Related Porous Materials

˚ . Differently from AlPO4-22, between the oxygen atoms of the rpa cage is about 10.7 A the oxygen atoms of each PO4 group in the P-incorporated aww cage form four covalent bonds with adjacent Al atoms, and the Al atoms, in turn, become five-coordinated with distorted trigonal bipyramid geometry. The existence of AlO5 species confers the negative charge on the framework. In contrast to other anionic open-frameworks with interrupted structures, the framework of AlPO-CJB1 is stable up to 550 C upon removal of occluded protonated template molecules trapped in the rpa cages. Protons are left to balance the negatively charged framework and as a consequence act as Bro¨nsted acid centers. IR spectroscopic and NH3-TPD (temperature program desorption) studies have confirmed the existence of Bro¨nsted acid centers. AlPO-CJ19 [(NH4)2Al4(PO4)4(HPO4).H2O][188] has an Al/P ratio of 4/5. It crystallizes ˚ , b ¼ 21:6211ð18Þ A ˚, in the monoclinic space group P21 with a ¼ 5:0568ð3Þ A  ˚ c ¼ 8:1724ð4Þ A, and b ¼ 91:361ð4Þ . Alternation of the Al-centered polyhedra (including AlO4, AlO5, and AlO6) and the P-centered tetrahedra (including PO4 and PO3OH) results in an interrupted open-framework structure. It is the first aluminophosphate containing three kinds of Al coordination (AlO4, AlO5, and AlO6) with all oxygen vertices connected to framework P atoms. The open framework of AlPO-CJ19 has 8-ring channels along the [100] direction. The H2O and protonated NH3 molecules reside in the channels [Figure 2.73(a)]. Interestingly, the structure features a series of puckered continuous networks constructed from the ordered connection of Al2P2 4-rings [Figure 2.73(b)]. Two types of one-dimensional (1-D) chain, edge-sharing 4-rings (AlPO-ESC) and corner-sharing 4-rings (AlPO-CSC), are present in the net and crosslinked at the node of Al(3) and P(2). The protonated NH3 molecules, which balance the negative charge of the framework, are trapped in the channels. The H2O molecules interact with the terminal oxygen atoms [attached to P(5) atoms] through H-bonding. Two-dimensional Layered Structures 1. AlP2O83
layers (Al/P ¼ 1/2) There are three kinds of layered aluminophosphates with AlP2O83
stoichiometry. [NH4]3[Co(NH3)6]3[Al2(PO4)4]2[156] crystallizes in the orthorhombic space group Aba2,

Figure 2.73 (a) Framework of AlPO-CJ19 viewed along the [100] direction [The dotted frame is shown in (b)]; (b) characteristic network structure in AlPO-CJ19 viewed along [001] direction and the SBU. Reproduced with permission from [188]. Copyright (2005) American Chemical Society

Structural Chemistry of Microporous Materials

81

Figure 2.74 [NH4]3[Co(NH3)6]3[Al2(PO4)4]2. (a) The structure of the inorganic layers; (b) the stacking of the inorganic layers. Reproduced with permission from [156]. Copyright (1997) Royal Society of Chemistry

˚ , b ¼ 29:699ð6Þ A ˚ , c ¼ 17:210ð3Þ A ˚ . The anionic layers are conswith a ¼ 9:502ð2Þ A

O)2 tetrahedra, resulting in an tructed by alternation of AlO4 tetrahedra and PO2(

AlP2O83
stoichiometry. Figure 2.74(a) shows the structure of the inorganic layer containing 4- and 20-rings. The [Co(NH3)6]3þ cations reside in the 20-ring windows. Figure 2.74(b) shows the stacking of inorganic layers along the c direction. The NH4þ ions and the water molecules are trapped between the inorganic layers. There exists extensive H bonding

O groups, [Co(NH3)6]3þ, NH4þ, and H2O species. among the terminal P

[C3H5N2][AlP2O8H2(H2O)2][155] crystallizes in the monoclinic space group C2/c with ˚ , b ¼ 7:188ð2Þ A ˚ , c ¼ 6:990ð2Þ A ˚ , b ¼ 103:77ð2Þ . The inorganic layer a ¼ 21:854ð4Þ A

O)(OH) tetrahedra. is constructed by alternation of AlO4(H2O)2 octahedra and PO2(

The two H2O molecules of AlO4(OH2)2 octahedra are located in trans positions. The layers feature a highly puckered sheet structure containing interconnected Al4P4 8-rings (Figure 2.75). The template imidazole molecules reside between the inorganic layers, and form extensive H-bonds with oxygen atoms of the layers. [C6H22N4][C2H10N2][Al2P4O16][162] crystallizes in the monoclinic space group P21/n ˚ , b ¼ 8:143ð8Þ A ˚ , c ¼ 13:770ð1Þ A ˚ , b ¼ 95:104ð2Þ . The inorganic with a ¼ 10:826ð1Þ A layer is constructed from alternation of AlO4 tetrahedra and PO4 tetrahedra, in which all AlO4 tetrahedra are vertex linked with four PO4 tetrahedra, but only two vertices of the

Figure 2.75 The inorganic layer of [C3H5N2][AlP2O8H2(H2O)2]. Reproduced with permission from [155]. Copyright (1998) Elsevier

82

Chemistry of Zeolites and Related Porous Materials

Figure 2.76 The 4.12-net porous sheet parallel to the (-101) plane, with ethylenediammonium cations located inside the 12-ring openings (H-bonds are represented by dotted lines). Reproduced with permission from [162]. Copyright (2000) Royal Society of Chemistry

PO4 tetrahedra are linked with adjacent AlO4 tetrahedra, with the remaining vertices

O groups. being terminal P

As shown in Figure 2.76, [C6H22N4][C2H10N2][Al2P4O16] features a novel 4.12-net sheet containing 4-rings and 12-rings. The structure consists of macroanion [Al2P4O16]6
, which is charge-balanced by tetraprotonated TETA molecule þH3N(CH2)2NH2þ(CH2)2NH2þ(CH2)2NH3þ, and diprotonated ethylenediamine þH3N(CH2)2NH3þ. The ethylenediamine molecules are believed to be generated during the process of solvothermal synthesis through fragmentation of the TETA molecules. Diprotonated ethylenediamine
cations þH3N(CH2)2NH3þ existing in the 12-ring openings interact with terminal P


O groups through strong H-bonding. Figure 2.77 shows the stacking of the layers in an

Figure 2.77 Packing of the sheets along the [-101] direction in an AAAA sequence, with TETA molecules intercalated in the interlayer region. Reproduced with permission from [162]. Copyright (2000) Royal Society of Chemistry

Structural Chemistry of Microporous Materials

83

Figure 2.78 Structuralrelationshipbetweenthe1-DAlP2O83
chainandthe 2-DAlP2O83
sheet. (a) 1-D AlP2O83
chain with corner-sharing 4-rings; (b) breaking the P

O    Al bond in the 1-D chain; (c) relinking the P

O    Al bond with adjacent chains to form a 4.12-net sheet. Reproduced with permission from [162]. Copyright (2000) Royal Society of Chemistry

AAAA sequence along the [
1 0 1] direction. TETAH44þ cations reside in the interlayer
region, and each cation provides a total of ten H-bonds to terminal P


O groups protruding into the interlayer region (H3Nþgroups at the two ends each providing three H-bonds, and the middle H2Nþgroups each providing two H-bonds). It is believed that H-bonds between the organic amine templates and the inorganic framework play an important role in the stabilization of the sheet structure. The sheet structure of [C6H22N4][C2H10N2][Al2P4O16] is closely related to the onedimensional AlP2O83
chain made up of vertex-sharing Al2P2 4-rings. This 1-D AlP2O83
chain is proposed as a parent chain for complex structures of aluminophosphates. As shown in Figure 2.78, through breaking the P–O   Al bonds in the 1-D chain, and relinking the Al

O group with the dangling P atoms in adjacent chain, the 4.12-net sheet can be transformed from the 1-D AlP2O83
chain. However, there has been no evidence yet that this process is related to the formation mechanism of [C6H22N4][C2H10N2][Al2P4O16]. 2. Al2P3O123
layers (Al/P ¼ 2/3) All the layered compounds: [(BuNH3)2][Al2(HPO4)(PO4)2][167] (1); UT-4: [(C6H14N)2] [Al2(HPO4)(PO4)2][169] (2); UT-5: [(C6H14N)2][Al2(HPO4)(PO4)2][169] (3); [(pyH)] [Al2(HPO4)2(PO4)][167] (4); [(C6H8N)][Al2(HPO4)2(PO4)][166] (5); and [(C9H20N)] [Al2(HPO4)2(PO4)] [170] (6), have an Al/P ratio of 2/3. Their chemical formulae can be expressed as [(TH)3
x][Al2(HPO4)x(PO4)3
x] (x ¼ 12).[170] The sheets of compounds (1), (2), (3), and (6) are constructed by the alternation of AlO4 (AlO4b) tetrahedra and PO4 (PO3bOt, PO2bO2t) tetrahedra. All AlO4 tetrahedra share oxygen atoms with adjacent PO4 tetrahedra, but PO4 tetrahedra share only three or two oxygen atoms with adjacent Al atoms. The anionic sheets of compounds (4) and (5) are made up of alternating Al-centered polyhedra (AlO4b and AlO5b) and PO4 tetrahedra (PO4b, PO3bOt, PO2bO2t). Six sheet topologies with the stoichiometry Al2P3O123
are shown in Figure 2.79. As can be seen, the structures of (4) and (6) are related to each other, and those of (5) and (3) are related to each other. When breaking the connection between AlO5b and PO4b in (4) and (5), the structure of (4) can be transformed into (6), and (5) similarly into (3). If we

84

Chemistry of Zeolites and Related Porous Materials

Figure 2.79 Six sheet structures with stoichiometry Al2P3O123
and their SBU. Reproduced with permission from [166]. Copyright (1998) American Chemical Society

ignore the coordinations of AlO5b and PO4b (breaking the connection between AlO5b and PO4b), these sheets with the stoichiometry Al2P3O123
have the same secondary building unit (SBU) which is made up of two AlO4 tetrahedra, two PO3b tetrahedra, and one PO2b tetrahedron (Figure 2.79). Based on such an SBU, many hypothetical inorganic sheets with the stoichiometry Al2P3O123
can be designed. Recently, a new 2-D layered compound [C2H8N]2.[Al2(HPO4)(PO4)2] has been synthesized with Al2P3O123
stoichiometry.[171] The macroanionic [Al2(HPO4)(PO4)2]2
sheet
is based on strictly alternating AlO4 tetrahedra and PO4 tetrahedra [PO2(OH)(


O) and
PO3(
O)] via vertex oxygen atoms. Its structure contains distinctive H-bonded helices

formed by the organic amine templates and the inorganic network. 3. Al3P4O163
layers (Al/P ¼ 3/4) A large number of layered aluminophosphates possess the stoichiometry Al3P4O163
. Except for [(C3N2H5)2][Al3P4O16],[155] the anionic sheets of Al3P4O163
are exclusively

O) tetrahedra. Different linkages of these constructed from alternating AlO4 and PO3(

tetrahedra lead to various sheet topologies. Figure 2.80 shows eight distinct 2-D sheet structures, including 4.6.12-, 4.6.8(I)-, 4.6.8(II)-, 4.6(I)-, 4.6(II)-, 4.6(III)-, 4.6(IV)-, and 4.8-net. Five types of SBUs for constructing the sheets are shown in the Figure. Interestingly, the 2-D sheets exhibit various stacking sequences such as AAAA, ABAB, ABCABC, and ABCDEF and so on

Structural Chemistry of Microporous Materials

85

Figure 2.80 Eight distinct 2-D sheet structures. SBUs for constructing the sheets are also shown. Reproduced with permission from [146]. Copyright (2003) American Chemical Society

(Figure 2.81). The style of stacking is manipulated by the structure-directing agent in the interlayer region, which interacts with the inorganic frameworks through extensive H-bonding. In contrast to the above layered structures with stoichiometry Al3P4O163
, [(C3N2H5)2][Al3P4O16H] contains AlO5 units.[155] It crystallizes in the triclinic space ˚ , b ¼ 9:36ð2Þ A ˚ , c ¼ 11:721ð2Þ A ˚ , a ¼ 97:10ð2Þ , group P-1 with a ¼ 8:940ð2Þ A   b ¼ 95:10ð1Þ , and g ¼ 91:91ð2Þ . The inorganic sheets are constructed from alternating Al-centered polyhedra (AlO4b and AlO5b) and PO4 tetrahedra (PO4b , PO3bOt, PO2bO2t). Its 2-D layered structure is shown in Figure 2.82. Distinct from the 2-D layered AlPOs with a mono sheet structure, this compound contains double sheets with double 6-rings (D6Rs) that are commonly featured in zeolite materials. Four D6Rs are connected with each other through 4-rings to form an 8-ring pore opening within the layer. One fraction of the imidazole molecules reside in the cavity, and can be removed upon heating of the sample at 250  C, whereas the other fraction of the protonated imidazole molecules are located in the interlayer region, and interact with the layers through H-bonds. Thus, the structure of [(C3N2H5)2][Al3P4O16H] might be regarded as an intermediate between a 2-D sheet structure and a 3-D open framework.

86

Chemistry of Zeolites and Related Porous Materials

Figure 2.81 Different stacking sequences of 2-D layers with stoichiometry Al3P4O163
. Reproduced with permission from [146]. Copyright (2003) American Chemical Society

Figure 2.82 The sheet structure of [(C3N2H5)2][Al3P4O16H]. Reproduced with permission from [155]. Copyright (1998) Elsevier

Structural Chemistry of Microporous Materials

Figure 2.83

87

The sheet structure of [(C2H5)2NH2]4[Al8P10O40H2][H2O]2.5

4. Al4P5O203
layers (Al/P ¼ 4/5) [(C2H5)2NH2]4[Al8P10O40H2][H2O]2.5[187] crystallizes in the triclinic space group P-1 ˚ , b ¼ 9:267ð7Þ A ˚ , c ¼ 17:461ð10Þ A ˚ , a ¼ 86:66ð5Þ , b ¼ 82:20ð4Þ , with a ¼ 8:632ð4Þ A  g ¼ 89:28ð5Þ . The sheet is constructed from alternating Al-centered polyhedra (AlO4b
and AlO5b) and PO4 tetrahedra [PO4b, PO2(OH)(


O)]. Similarly to [(C3N2H5)2][Al3 P4O16H] described above, this layered compound also contains double sheets with double 4-rings (D4Rs), a common SBU in zeolite materials. D4Rs are connected with each other through 4-rings along the c direction and through 6- and 8-rings along the b direction to form the 2-D network (Figure 2.83). One part of the diethylamine molecules, residing in the 8-ring window in the layer without H-bonding with in the sheet, plays a structure-filling role, while the other part, located in the interlayer region and forming H-bonds with the sheet, plays an important role in stabilizing the inorganic network. One-dimensional Chain Structures The stoichiometries of 1-D aluminophosphate chains have been found as AlP2O83
[158–160] and Al3P5O206
[168,182] with Al/P ratios of 1/2 and 3/5, respectively. The corner-sharing 4-ring chains (denoted as AlPO-CSC) and edge-sharing 4-ring chains (denoted as AlPOESC) have the simplest chain structures, which are found to exist in many 2-D layer and 3-D open-framework structures. [Et3NH][AlP2O8H2][158] crystallizes in the monoclinic space group P2l/n with ˚ , b ¼ 13:201 A ˚ , c ¼ 8:522 A ˚ , and b ¼ 97:2 . It has the AlPO-CSC chain a ¼ 12:073 A structure as shown in Figure 2.84(a). All AlO4 tetrahedra share four oxygen atoms with adjacent P atoms, whereas all PO4 tetrahedra share only two oxygen atoms with adjacent Al atoms, leaving the other two oxygen atoms terminal. The alternation of AlO4 tetrahedra and PO4 tetrahedra gives a 1-D chain with corner-sharing 4-rings. The macroanionic chains along the a direction are hydrogen-bonded to form layers parallel to the ab plane. The layers stacked along the [001] direction are intercalated by the Et3NHþcations. [H3NCH2CH2NH3][AlP2O8H][159] crystallizes in the triclinic space group P-1 with ˚ , b ¼ 9:032ð1Þ A ˚ , c ¼ 11:691ð1Þ A ˚ , a ¼ 81:38ð1Þ , b ¼ 82:27ð1Þ , and a ¼ 4:901ð1Þ A  g ¼ 75:83ð1Þ . It has the AlPO-ESC chain structure as shown in Figure 2.84(b).

88

Chemistry of Zeolites and Related Porous Materials

Figure 2.84 Two 1-D chain structures with stoichiometry AlP2O8.3
(a) AlPO-CSC; (b) AlPO-ESC

The structure is composed of AlO4 units and PO4 units (PO3bOt and PObO3t) forming an edge-sharing 4-ring ladder chain with pendant PO4H side groups. The existence of terminal P

O groups in AlPO-CSC and AlPO-ESC suggests that these 1-D chains might have a potential to further condense to give rise to more complex frameworks.[196] Experimentally, 3-D open frameworks have been successfully built up through the coordination of transition metal ions to the terminal oxygen atoms of 1-D AlPO-CSC. The AlPO-CSC structure is kept intact in the 3-D open frameworks.[197,198] We have discussed anionic framework AlPOs with an Al/P ratio less than 1. Two possibilities may lead to the Al/P ratio of less than unity: one is the existence of terminal P

O groups and the other is the existence of AlO5b and AlO6b. It is found that the Al and P coordinations satisfy Equation (1):[146] X X mAlOib  iAlOib ¼ nPOjb  jPOjb ð1Þ i

j

where iðjÞ is the number of bridging oxygen atomsPcoordinated to Al (P), mðnÞ is the P mAlOib = nPOjb ¼ Al=P, i ¼ 3, 4, 5, and 6 number of AlOib (POjb) coordination, corresponding to AlO3b, AlO4b, AlO5b, and AlO6b units, respectively, and j ¼ 1, 2, 3, and 4 corresponding to PO4 units with one, two, three, and four bridging oxygen atoms, respectively. Based on Equation (1), the detailed Al and P coordinations for a given stoichiometry can be enumerated. By applying advanced solid-state NMR techniques, the chemical environments of Al and P atoms can be determined. Yu and coworkers have recently developed a new method to determine the Al/P ratio of open-framework AlPOs based on Equation (1) as well as NMR studies.[199] By using various solid-state NMR techniques, including 27Al, 31P magic-angle spinning (MAS), 27Al! 31P cross-polarization (CP), 27Al{31P} rotational echo double resonance (REDOR), and 31P{27Al} transfer-of-population double resonance (TRAPDOR), different Al coordinations (AlO4b, AlO5b, and AlO6b) and P coordinations (PO4b, PO3bOt, PO2bO2t, and PO6O3t), where b represents bridging oxygens and t represents terminal oxygens, can be unambiguously distinguished. 2.5.2

Open-framework Gallophosphates with Extra-large Pores

Since the discovery of gallophosphate cloverite with 20-ring pore openings, a variety of open-framework gallophosphates have been reported with extra-large pores, such as

Structural Chemistry of Microporous Materials

89

Figure 2.85 (a) Three basic building units in the framework of ULM-5 and (b) perspective view of ULM-5 close to [100]

ULM-5[200] and ULM-16[201] with 16-rings, MIL-31[202], MIL-46,[203] and MIL-50[204] with 18-rings, ICL-1[205] and [H3N(CH2)4NH3]2[Ga4(HPO4)2(PO4)3(OH)3].5.4H2O[206] with 20-rings, and NTHU-1[207] with 24-rings. ULM-5 ([H3N(CH2)6NH3]4[Ga16(PO4)14(HPO4)2(OH)2F7].6H2O)[200] possesses 16-ring channels. It crystallizes in the orthorhombic space group P22121 with ˚ , b ¼ 18:409ð4Þ A ˚ , c ¼ 20:639ð7Þ A ˚ . As shown in Figure 2.85(a), its a ¼ 10:252ð2Þ A three-dimensional framework is built up from three different types of basic building units: I consists of corner-linked [Ga3(PO4)2(HPO4)F2] hexameric units composed of two PO4 tetrahedra, one HPO4 tetrahedron, two GaO4F trigonal bipyramids, and one GaO4F2 octahedron, with fluorine atoms shared between the gallium polyhedra; II is very similar to I except that one of the trigonal bipyramids is replaced by a GaO3(OH) tetrahedron; III is octameric Ga4(PO4)4 which can be considered as a cube of corner-linked GaO4 and PO4 tetrahedra in which a fluorine atom resides and is bonded to two of the four gallium atoms of this cube. The three-dimensional framework of ULM-5 is built up from the linkage of these units via oxygen bridges shared by both gallium and phosphorus atoms. The framework contains 16- and 6-ring channels along [100] [Figure 2.85(b)] and 8-ring channels along [010]. The diprotonated amines are accommodated in the 16-ring void ˚ . The water molecules are in the 6-ring channels whose free aperture is 12.20  8.34 A tunnels. MIL-31 ([C20H52N4][Ga9(PO4)9F3(OH)2(H2O)].2H2O)[202] contains 18-ring channels. ˚, It crystallizes in the orthorhombic space group Pca21 with a ¼ 17:4941ð1Þ A ˚ , and c ¼ 10:0749ð2Þ A ˚ . As shown in Figure 2.86(a), the structure b ¼ 32:3930ð4Þ A of MIL-31 consists of the connection of three crystallographically distinct building units (BUs) composed of three phosphate groups and three gallium polyhedra. Two of them are hexameric units (denoted as type I) built up from three PO4 tetrahedra corner-sharing with one GaO4(OH,F)2 octahedron and two GaO4(OH,F) trigonal bipyramids. Within these hexameric units, the octahedron is in a central position and linked to the two GaO4(OH, F) trigonal bipyramids by corner-sharing via fluorine or hydroxy groups. The

90

Chemistry of Zeolites and Related Porous Materials

Figure 2.86 (a) Hexameric building units of MIL-31; (b) a polyhedral view of the structure of MIL-31 along the [001] direction. Reproduced with permission from [202]. Copyright (2000) Royal Society of Chemistry

second type of BU (type II) is composed of three phosphate groups sharing corners with three gallium polyhedra which exhibit three different coordinations: GaO4 tetrahedra, GaO4(OH) trigonal bipyramids, and GaO4(OH)(H2O) octahedra. The gallium bipyramidal unit is in the central position and has one common hydroxy group with the gallium octahedral unit. The connection of these different building units generates a threedimensional framework composed of large hexagon-shaped tunnels with 18- and 6-rings running along the c axis [Figure 2.86(b)]. Two diprotonated organic templates are accommodated in the 18-ring channels, whereas two water molecules are trapped in the 6-ring channels. [NH3(CH2)4NH3]2[Ga4(HPO4)2(PO4)3(OH)3].5.4H2O[206] possesses 20-ring channels. ˚ , and It crystallizes in the tetragonal space group I41/a with a ¼ 15:261ð1Þ A ˚ . The structure is constructed from GaO6 octahedra and PO4 tetrahedra. c ¼ 28:898ð2Þ A The basic building unit is a Ga4O20 tetramer. As shown in Figure 2.87(a), a central pair of edge-sharing GaO6 octahedra is connected with additional two GaO6 octahedra by corner sharing. Figure 2.87(b) shows the pore system containing zig-zag tunnels running parallel to the a and b axes and intersecting to form a three-dimensional framework. Figure 2.88 shows the tunnels that are bounded by 20-ring windows along the a direction in which diprotonated 1,4-diaminobutane dications reside. The framework density is 10.7 M/ ˚ 3 (M ¼ Ga, P), which is comparable to that of 11.1 M/1000 A ˚ 3 for cloverite. 1000 A [207] NTHU-1 ([Ga2(DETA)(PO4)2].2H2O, DETA ¼ diethylenetriamine) contains ˚, 24-ring channels. It crystallizes in the trigonal space group R-3 with a ¼ 23:781ð1Þ A ˚ . Its three-dimensional framework structure consists of channels and c ¼ 13:466ð1Þ A close-packed in hexagonal, honeycomb arrays, which are oriented parallel to the c-axis (Figure 2.89). The walls of the channels are made up of corner-sharing GaO4 tetrahedra, GaO3N3 octahedra, and PO4 tetrahedra. DETA binds to Ga atoms as a terminal tridentate ligand. Within the unidimensional channels are puckered 24-ring windows formed by the edges of 12 GaO4 tetrahedra and 12 PO4 tetrahedra in an alternating manner (the shortest ˚ ). Each 24-ring channel contains six lateral O   O distance across the 24-rings is 10.4 A 12-ring windows delimited by two GaO3N3 octahedra, four GaO4 tetrahedra, and 6 PO4

Structural Chemistry of Microporous Materials

91

Figure 2.87 [NH3(CH2)4NH3]2[Ga4(HPO4)2(PO4)3(OH)3].5.4H2O. (a) The tetrameric building unit Ga4O20; (b) schematic view of the pore system. Zig-zag tunnels run parallel to the a and b axes and intersect to form a three-dimensional pore network. Reproduced with permission from [206]. Copyright (1999) Elsevier

Figure 2.88 Perspective view of [NH3(CH2)4NH3]2[Ga4(HPO4)2(PO4)3(OH)3]5.4H2O along the a-axis. The TMDA dications and water molecules reside in the channels. Reproduced with permission from [206]. Copyright (1999) Elsevier

Figure 2.89

The structural view of NTHU-1 along c-axis

92

Chemistry of Zeolites and Related Porous Materials


tetrahedra, through which they connect neighboring channels. Some P


O groups point ˚ 3 (M ¼ Ga, P). towards the channels. For NTHU-1, the framework density is 10.9 M/1000 A It is one of the lowest-framework-density materials known. 2.5.3

Indium Phosphates with Extra-large Pores and Chiral Open Frameworks

Although Al atom and Ga atom in Group 13 in the periodic table can adopt four, five, and/or six coordinations in their phosphates, In atom all adopts 6-coordination in the open-framework indium phosphates. Dhingra and Haushalter reported the first openframework indium phosphate [H3NCH2CH2NH3][In2(HPO4)4][208] using organic amine as the template. Since then a large number of indium phosphates have been prepared with novel framework topologies, including 4[NH3(CH2)3NH3].3[H3O].[In9(PO4)6(HPO4)2F16].3H2O[209] with 14-rings, [NH3(CH2)2NH2(CH2)2NH3]2[NH2(CH2)2NH2(CH2)2NH2][In6.8F8(H2O)2(PO4)4(HPO4)4].2H2O with 16-rings[210], and [In(OH)PO4][H3O] with spiral channels.[211] 4[NH3(CH2)3NH3].3[H3O].[In9(PO4)6(HPO4)2F16].3H2O[209] possesses 14-ring ˚, channels. It crystallizes in the monoclinic space group P21/n with a ¼ 13:616ð2Þ A  ˚ ˚ b ¼ 9:372ð2Þ A, c ¼ 23:293ð4Þ A, and b ¼ 99:44ð2Þ . The structure is composed of In(O,F)6 octahedra and PO4 tetrahedra. Fluorine atoms act as terminal groups or bridging atoms to connect two indium centers. Figure 2.90 shows its framework containing 8- and 14-ring channels running parallel to the [010] direction. The 14-ring pore opening is ˚ . These channels highly elliptical with approximate maximum dimensions of 14.0  6.5 A are intersected by other 8-ring channels running along the [110] and [1
10] directions. [H3O][In(OH)PO4][211] crystallizes in enantiomorphic space group P41212 with ˚ , c ¼ 11:156ð4Þ A ˚ , and therefore the unit-cell contents are chiral. The a ¼ 9:412ð3Þ A structure contains helical chains constructed from InO6 octahedra sharing cis-corners. Neighboring InO6 octahedra are bridged by OH groups. As shown in Figure 2.91(a), the central axis of the helical chains is located at a 21 screw axis running along the crystallographic c axis. The helical chains are linked into a three-dimensional framework through PO4 tetrahedra. Each PO4 tetrahedron shares corner with two neighboring InO6 octahedra from the same helical chain and two InO6 octahedra from the neighboring chain. Four such helical chains related by a 41 screw axis are connected together to form a square-window-shaped channel [Figure 2.91(b)].

Figure 2.90 A polyhedral view of the framework of 4[NH3(CH2)3NH3].3[H3O]. [In9 (PO4)6(HPO4)2F16].3H2O viewed along the [010] direction. Reproduced with permission from [209]. Copyright (1998) American Chemical Society

Structural Chemistry of Microporous Materials

93

Figure 2.91 (a) The helical chain constructed from InO6 octahedra sharing cis-corners in [H3O][In(OH)PO4]; (b) a polyhedral view of the framework along c axis

2.5.4

Zinc Phosphates with Extra-large Pores and Chiral Open Frameworks

Among open-framework metal phosphates, zincophosphates exhibit diverse compositions and fascinating structural topologies.[212] Notable examples are zincophosphates with extra-large pores bound by 16-, 20-, and 24-rings, and helical channels. ND-1 [Zn3(PO4)2(PO3OH)(H2DACH)].2H2O(DACH ¼ 1,2-diaminocyclohexane) contains 24-ring channels.[213] It crystallizes in the rhombohedral space group R-3 with ˚ , and c ¼ 9:241ð4Þ A ˚ . The structure is made of ZnO4 tetrahedra and PO4 a ¼ 33:401ð7Þ A tetrahedra via corner-sharing of oxygen atoms. There exist m3-O atoms in the structure which connect two Zn atoms and one P atom. As shown in Figure 2.92, the

Figure 2.92 Structure of ND-1 viewed along c axis. Reproduced with permission from [213]. Copyright (1999) American Chemical Society

94

Chemistry of Zeolites and Related Porous Materials

Figure 2.93 The polyhedral view of [H3N(CH2)6NH3][Zn4(PO4)2(HPO4)2].3H2O along the [100] direction. Reproduced with permission from [214]. Copyright (2000) Royal Society of Chemistry

three-dimensional framework of ND-1 comprises 24-ring channels close-packed in hexagonal, honeycomb arrays, which are oriented parallel to the c-axis. P

OH groups are pointing toward the center of channels. Trans-1,2-DACH and the water molecules are found near the walls of channels. Although a mixture of the cis- and trans- isomers of 1,2-DACH was used as the template, ND-1 contains only trans-1,2-DACH. Since the structure is centrosymmetric, both (1S; 2S) and (1R; 2R) enantiomers of trans-1,2-DACH are present in the channels. A unique structural feature of ND-1 is that it contains sizable ˚ is measured between openings, even with the templates present. A diameter of 8.6 A the end carbon atoms of two cyclohexane across the channel. The framework density of ˚ 3. ND-1 is 12.1 T per 1000 A [H3N(CH2)6NH3][Zn4(PO4)2(HPO4)2].3H2O contains 20-ring channels.[214] It crystal˚ , b ¼ 13:6024ð11Þ A ˚, lizes in the triclinic space group P-1 with a ¼ 5:2016ð4Þ A ˚ , a ¼ 97:869ð2Þ , b ¼ 93:302ð2Þ , and g ¼ 91:828ð2Þ . The structure c ¼ 17:2394ð13Þ A is constructed from ZnO4, PO4, and HPO4 tetrahedra. Besides the Zn

O

P linkages, Zn

O

Zn linkages are also present in this structure via the three-coordinated oxygen atoms. This structure contains 3-, 4-, 5-, 6-, and 20-rings. Figure 2.93 shows the framework representing the 20-ring pores along the [100] direction with maximum ˚ . The diprotonated organic cations and water atom-to-atom dimensions of 8.0  16.0 A molecules are located in the channels. [H3N(CH2)2NH2(CH2)2NH3][Zn4(PO4)3(HPO4)].H2O possesses helical channels.[215] ˚, It crystallizes in the enantiomorphic space group P21 with a ¼ 10:021ð4Þ A ˚ , c ¼ 11:856ð7Þ A ˚ , and b ¼ 103:13ð1Þ . The structure is built from b ¼ 8:286ð3Þ A ZnO4, PO4, and HPO4 tetrahedral units. Zn

O

P and Zn

O

Zn linkages are present. The entire framework can be considered to be built from the network of 3-, 4-, 6-, and 8-rings. The 3- and 4-rings are connected together via edge sharing to form onedimensional helical columns along the b axis. Figure 2.94(a) shows how these columns are interconnected via HPO4 groups forming an 8-ring channel system along the a axis. The 8-ring channel along the a axis is connected to another 8-ring channel along the b axis, forming a helical interconnected one-dimensional channel system. Figure 2.94(b) shows the three-dimensional framework along the b axis.

Structural Chemistry of Microporous Materials

95

Figure 2.94 [NH3(CH2)2NH2(CH2)2NH3][Zn4(PO4)3(HPO4)].H2O. (a) Helical columns are interconnected via the HPO4 groups to form the 8-ring cavities and the helical channels; (b) a view of the 8-ring channels along the [010] direction. Reproduced with permission from [215]. Copyright (1999) Royal Society of Chemistry

2.5.5

Iron and Nickel Phosphates with Extra-large Pores

The discovery of the mineral phosphate, cacoxenite with extra-large pores in 1983 has aroused considerable interest in the search for zeolite-type materials with giant cavities.[216] Cacoxenite, [AlFe24O6(OH)12(PO4)17(H2O)24].51H2O, crystallizes in the hex˚ , and c ¼ 10:550 A ˚ . The structure is agonal space group P63/m with a ¼ 27:559 A composed of FeO6, AlO5 and AlO6 polyhedra, and PO4 tetrahedra. The most intriguing feature of the structure is the enormous free diameter of the pore or channel that is ˚ ). Figure 2.95 oriented parallel to the c axis (the calculated free pore diameter is 14.2 A shows the framework of cacoxenite. Until now, synthetic cacoxenite has not been produced. However, some zeolitic porous materials have been reported to have substantially larger pores than that found in cacoxenite.

Figure 2.95

The polyhedral view of cacoxenite

96

Chemistry of Zeolites and Related Porous Materials

Figure 2.96 The framework of [(C4N3H16)(C4N3H15)][Fe5F4(H2PO4)(HPO4)3(PO4)3].H2O viewed along the [100] direction. Reproduced with permission from [217]. Copyright (1999) Royal Society of Chemistry

Another open-framework iron phosphate with an extra-large pore is [(C4H16N3) (C4H15N3)][Fe5F4(H2PO4)(HPO4)3(PO4)3].H2O.[217] It crystallizes in the monoclinic ˚ , b ¼ 15:618ð1Þ A ˚ , c ¼ 22:563ð1Þ A ˚ , and space group P21/n with a ¼ 9:670ð1Þ A  b ¼ 90:82ð1Þ . The structure consists of FeO6, FeO5F, and FeO6F2 octahedra and PO4 tetrahedra. As shown in Figure 2.96, the structure has one-dimensional channels bound by 24 M (M ¼ Fe, P) atoms along the [010] direction, within which the DETA and water ˚ . Along the [100] direction, molecules reside. The width of the channels is 15.3  4.5 A the structure has another narrow channel, bound by 16 M atoms, and the pendent H2PO4 and HPO4 moieties protrude into this channel. VSB-5 (Ni20[(OH)12(H2O)6][(HPO4)8(PO4)4].12H2O)[218] is an open-framework nickel phosphate with gigantic pores of 24-ring size. As compared with other openframeworks with extra-large pores, VSB-5 has higher thermal stability and better magnetic and catalytic properties. It crystallizes in the hexagonal space group P63/m, with ˚ , and c ¼ 6:3898ð7Þ A ˚ . Figure 2.97 shows the framework of VSB-5, a ¼ 18:209ð1Þ A presenting a one-dimensional 24-ring system running along the [001] direction. These

Figure 2.97 The polyhedral view of VSB-5 down c axis

Structural Chemistry of Microporous Materials

97

channels are delimited by 24 NiO6 octahedra, which are connected by sharing faces, edges, and corners. The 12 PO3(OH) tetrahedra decorate the surface of the channel and ˚. extend into the channel. The free diameter of the pore is about 10.2 A 2.5.6

Vanadium Phosphates with Extra-large Pores and Chiral Open Frameworks

Since vanadium atom possesses diverse valence states (III, IV, V) and coordination geometries (tetrahedron, square pyramid, distorted and regular octahedron), vanadium phosphates display rich structural chemistry. Intriguing examples of the open-framework vanadium phosphates are [(CH3)2NH2]K4[V10O10(H2O)2(OH)4(PO4)7].4H2O[219] with inorganic double-helical chains and [HN(CH2CH2)3NH]K1.35[V5O9(PO4)2].xH2O[220] with giant cavities. [(CH3)2NH2]K4[V10O10(H2O)2(OH)4(PO4)7].4H2O[219] contains chiral double helices formed by interpenetrating spirals of vanadium phosphate units. It crystallizes in the ˚ , and c ¼ 30:555 A ˚ . The structure is enantiomorphic space group P43 with a ¼ 12:130 A built up from VO6 octahedra, VO5 square pyramids, and PO4 tetrahedra. The fundamental building blocks are vanadium oxo pentamers. As shown in Figure 2.98, the ˚ ) and pentamer has four V

O

V and two V

OH

V linkages. The long V

O (2.4 A ˚ ) bonds alternate along the central V
short V

O (1.7 A
O backbone of the pentamer. These pentamers are arranged so as to form spirals along [001]. The spirals in turn are intertwined to give two strands of a double helix as shown in Figure 2.99. Some P atoms join the pentamers and some connect the strands to one another to form the helix whereas others bond one double helix to another. These strands and double helices intergrow with one another in an extremely complicated fashion to give rise to a three-dimensional open framework (Figures 2.100 and 2.101). The Kþ and protonated (CH3)2NH2þ cations reside in the tunnels. [HN(CH2CH2)3NH]K1.35[V5O9(PO4)2].xH2O[220] crystallizes in the cubic space group ˚ . The framework density is about 9.3 M atoms (M ¼ V, P) per I-43m with a ¼ 26:247 A 3 ˚ 1000 A . It is an open-framework vanadium phosphate containing enormous 32-ring voids. The structure is composed of VO5 square pyramids and PO4 tetrahedra. All Vatoms possess terminal V

O groups and are present in the form of unusual cross-shaped V5 pentamers.

Image Not Available In The Electronic Edition

98

Chemistry of Zeolites and Related Porous Materials

Image Not Available In The Electronic Edition

The most prominent feature of this structure is the extremely large voids centered about (0,0,0) and (½,½,½), which have  43m site symmetry. Figure 2.102 shows the framework structure as a projection of the unit-cell contents down [001]. Each cavity is filled by 12 diprotonated 1,5-diazabicyclooctane (H2DABCO2þ) dications and 32 Kþ cations. The large cavity at (0,0,0) has six rectangular 16-ring windows, through which it communicates with other voids via intervening smaller cavities. The array of voids centered at (0,0,0) and tunnels centered at (½,½,½) interpenetrates, but never intersects (Figure 2.103).

Image Not Available In The Electronic Edition

Structural Chemistry of Microporous Materials

99

Image Not Available In The Electronic Edition

Figure 2.102 The framework of [HN(CH2CH2)3NH]K1.35[V5O9(PO4)2].xH2O. Reproduced with permission from [220]. Copyright (1996) American Chemical Society

Figure 2.103 (a) A view of [HN(CH2CH2)3NH]K1.35[V5O9(PO4)2].xH2O down [100] of the isosurface representing the tunnel structure along with a schematic representation (b). There are two crystallographically identical sets of tunnels that interpenetrate one another but never intersect. Reprinted from [220]. Copyright (1996) American Chemical Society

100

2.5.7

Chemistry of Zeolites and Related Porous Materials

Germanates with Extra-large Pores

The open-framework metal phosphates described above are generally constructed from M-centered polyhedra (MO4, MO5, or MO6) and PO4 tetrahedra. In contrast to the relative few germanate zeolites comprised of GeO4 tetrahedra, there are a number of openframework germanates containing GeO4, GeO5, and/or GeO6 units.[221] FDU-4[222] and ASU-16[223] are two of the well known germanates with an extra-large micropore structure. In particularly, Zou et al. have reported a mesoporous germanium oxide with crystalline pore walls. It has the largest primitive cell and lowest framework density of any inorganic material, and channels that are defined by 30-rings.[224] FDU-4 ([N(CH2CH2NH3)3]2/3[Ge9O17(OH)4][HCON(CH3)2]1/6(H2O)11/3) has extralarge 24-ring channels.[222] It crystallizes in the hexagonal space group P63cm with ˚ , and c ¼ 9:798ð2Þ A ˚ . Its SBU is a cluster composed of nine germanium a ¼ 23:941ð3Þ A centers: one germanium atom is bonded to five oxygen atoms to yield a distorted square pyramidal coordination geometry; four germanium atoms each are bonded to five oxygen atoms to display a trigonal bipyramidal geometry; the remaining four germanium atoms have distorted tetrahedral coordination geometries [Figure 2.104(a)]. Neighboring SBUs are linked by bridging oxygen atoms to yield an open framework with intersecting 3-D channel system. Two kinds of channels with 12- and 24-rings extending along the crystallographic c axis are arranged in close-packed hexagonal honeycomb arrays. Each 24-ring channel is surrounded by six 12-ring channels. The approximate diameters of the ˚ . The wall of the narrowest cross-section of the 24-ring channel are 12.65  9.52 A 24-ring channels contains 12-ring windows, resulting in an unprecedented 3-D intersecting channel framework. The framework density of FDU-4 measured by the number of ˚ 3 is 11.1, which is one of the lowest framework densities known. polyhedra per 1000 A The organic amine molecules in FDU-4 are located within the 12-ring channels. Some solvent species such as dimethylformamide (DMF) and water molecules are disordered and located in the 24-ring channels. ASU-16 ([H2DAB]3[Ge14O29F4][DAB]0.5.16H2O, DAB: 1,4-butanediamine) contains 24-ring channels.[223] It crystallizes in the orthorhombic space group I 222, with ˚ , b ¼ 24:267ð2Þ A ˚ , and c ¼ 30:210ð3Þ A ˚ . The structure is made up a ¼ 16:9109ð8Þ A of two crystallographically independent clusters [Figure 2.105(a)], identical in composition. Each cluster is composed of seven germanium atoms with mixed coordination

Figure 2.104

FDU-4. (a) SBU; (b) projection of the 3-D framework down the c axis

Structural Chemistry of Microporous Materials

101

Figure 2.105 ASU-16. (a) SBU; (b) the polyhedral view of ASU-16 along the a axis. Reproduced with permission from [223]. Copyright (2001) American Chemical Society

states: Four GeO4 tetrahedra, two GeO4F trigonal bipyramids, and one GeO5F octahedron. The clusters are linked to each other through five bonds by corner-sharing of oxygen atoms of the four GeO4 tetrahedra and one GeO4F unit. ASU-16 has a 3-D framework characterized by 1-D 24-ring channels. As shown in Figure 2.105(b), the channels are elliptical in shape and the largest free diameters of the pore aperture are ˚ . The framework density of ASU-16 measured by the number of approximately 8.5  15 A ˚ 3 is very low, about 8.6. Two organic molecules are located in the polyhedra per 1000 A 10- and 12-ring windows. Remaining species are disordered in the channels. 2.5.8

Indium Sulfides with Extra-large-pore Open Frameworks

Although aluminosilicates and metal phosphates can form zeolitic open-framework materials, sulfide analogs usually form high-density phases because of the relatively small T

S

T angles as compared with the T

O

T angles in zeolite frameworks. One strategy to overcome this limitation, proposed by O’ Keeffe and Yaghi and coworkers,[225] is to use tetrahedral clusters called supertetrahedra as the building blocks, instead of simple TS4 tetrahedra, to construct porous sulfide-based networks. Two indium sulfide open frameworks, ASU-31 and ASU-32,[225] have been successfully prepared with giant cavities and channels. Figure 2.106 shows the first three of the supertetrahedral family Tn. ASU-31 ([In10S18(HPP)6(H2O)15, HPP ¼ hexahydro-2H-pyrimido[1,2-a]pyrimidi˚ .[225] The nium]) crystallizes in the cubic space group I-43m with a ¼ 34:0802ð7Þ A structure of ASU-31 is based on the sodalite cage (Figure 2.107, top left). In the sodalite structure with regular TX4 tetrahedra, the maximum T

X

T angle is 161 (Figure 2.107, top middle) but this value can be decreased by concerted rotations of the tetrahedra (Figure 2.107, top right). ASU-31 is constructed from T3 supertetrahedra containing a very open structure with large cavities centered at the corners and body center of the body-centered cubic cell (Figure 2.107, bottom). The fixed diameter (the maximum size ˚. of a sphere that can fit inside the largest cavities) is about 25.6 A ˚ , and ASU-32 crystallizes in the tetragonal space group I-4m2 with a ¼ 35:452ð7Þ A [225] [226] ˚ c ¼ 17:3304ð1Þ A. Its structure is based on the tetragonal CrB4 net. Figure 2.108

102

Chemistry of Zeolites and Related Porous Materials

Image Not Available In The Electronic Edition

Image Not Available In The Electronic Edition

Structural Chemistry of Microporous Materials

103

Image Not Available In The Electronic Edition

shows the framework of ASU-32 with one-dimensional channels parallel to the crystallographic c axis. In ASU-32, the tetrahedral units are again T3 supertetrahedra. It should be noted that cations in the cavities of ASU-31 and ASU-32 can be exchanged by Naþ. O’ Keeffe and Yaghi and coworkers[225] compared the sizes of such spheres for ASU-31 and ASU-32 with those of the most open zeolites with larger cavities. From Table 2.8, it can be seen that ASU-31 and ASU-32 exhibit the largest size of the free diameter of the cavities.

Table 2.8 Sizes of fixed and free spheres in some structures Structure Framework Pore Free Free Fixed Fixed name type dimensionality diametera (A˚) volume (A˚) diameterb (A˚) volume (A˚) ASU-31 Faujasite Cloverite ASU-32 VPI-5 a

SOD FAU CLO CrB4 VFI

3 3 3 1 1

11.2 7.1 6.0 14.4 11.7

736 187 113 1563 839

25.6 11.1 15.5 17.2 12.2

8785 720 1950 2664 951

Free diameter: the maximum size of a sphere that can be freely moved along channels through the structure. Fixed diameter: the maximum size of a sphere that can fit inside the largest cavity.

b

104

Chemistry of Zeolites and Related Porous Materials

2.6 Summary In this chapter, the structural characteristics of zeolites and zeolitic open-framework materials have been described. Zeolites that are built up by corner-sharing of TO4 tetrahedra with regular pore architectures are important in industrial technologies such as catalysis, sorption, and ion-exchange; those zeolites with extra-large pores, helical channels, or intersecting channels are especially highly desirable. In recent decades, the discovery of various kinds of open-framework compounds has been promoting enormous growth in the chemical diversity of inorganic porous materials. However, strictly speaking, most of the zeolitic open-framework structures are not porous because their channels or cavities are normally filled by guest templating molecules. The removal of the occluded guest molecules by calcination usually results in collapse of the framework. Despite the limited number of framework types reported, the number of hypothetical topologies that can be designed by computational methods is infinite (see Chapter 7). It is believed that advances in synthetic chemistry, especially the use of hydrothermal/ solvothermal combinatorial techniques, will significantly accelerate the rapid discovery of new open-framework materials with diverse framework structures.

References [1] D.W. Breck, Zeolite Molecular Sieves, Structure, Chemistry and Use. John Wiley & Sons, Inc., New York, 1974. [2] H. van Bekkum, E.M. Flanigen, P.A. Jacobs, and J.C. Jansen (Editors), Introduction to Zeolite Science and Practice, Stud. Surf. Sci. Catal., 2001, 137. [3] Ch. Baerlocher, W.M. Meier, and D.H. Olson, Atlas of Zeolite Framework Types. 5th Edn., Elsevier, Amsterdam, 2001. [4] C.H. Baerlocher, and L.B. McCusker, http://www.iza-structure.org/databases. [5] A.K. Cheetham, G. Fe´rey, and T. Loiseau, Open-framework Inorganic Materials. Angew. Chem. Int. Ed. 1999, 38, 3268–3292. [6] S.T. Wilson, B.M. Lok, C.A. Messian, T.R. Cannan, and E.M. Flanigen. Aluminophosphate Molecular Sieves- a New Class of Microporous Crystalline Inorganic Solids. J. Am. Chem. Soc., 1982, 104, 1146–1147. [7] W. Lo¨wenstein, Am. Mineral., 1954, 39, 92–96. [8] E.M. Flanigen, R.L. Patton, and S.T. Wilson, Innovation in Zeolite Material Science. Stud. Surf. Sci. Catal., 1988, 37, 13–27. [9] W.M. Meier. Molecular Sieves. Society of Chemical Industry, London, 1968, 10–27. [10] W.M. Meier, and D.H. Olson, Atlas of Zeolite Structure Types. Butterworths, London, 1987. [11] J.V. Smith, Topochemistry of Zeolites and Related Materials. l. Topology and Geometry. Chem. Rev., 1988, 88, 149–182. [12] W.M. Meier, and C.H. Baerlocher, Zeolite Type Framework: Connectivities, Configurations and Conformations. Mol. Sieves, 1999, 2, 141–161. [13] L.B. McCusker, F. Liebau, and G. Engelhardt, Nomenclature of Structural and Compositional Characteristics of Ordered Microporous and Mesoporous Materials with Inorganic Hosts (IUPAC Recommendations 2001). Microporous Mesoporous Mater., 2003, 58, 3–13. [14] J.M. Newsam, The Zeolite Cage Structure. Science, 1986, 231, 1093–1099.

Structural Chemistry of Microporous Materials

105

[15] L.B. McCusker, and Ch. Baerlocher, Zeolite Structures. In ‘Introduction to Zeolite Science and Practice,’ Stud. Surf. Sci. Catal., 2001, 137, 37–56. [16] J.M. Bennett, W.J. Dytrych, J.J. Pluth, J.W. Jr. Richardson, and J.V. Smith, Structural Features of Aluminophosphate Materials with Al/P ¼ 1. Zeolites, 1986, 6, 349-361. [17] J.A. Martens, and P.A. Jacobs, Crystalline Microporous Phosphates: a Family of Versatile Catalysts and Adsorbents. Elsevier Sci. B, 1994, 85, 653–685. [18] E.M. Flanigen, B.M. Lok, R.L. Patton, and S.T. Wilson, New Developments Zeolite Science and Technology. Proceeding of the 7th International Zeolite Conference, Ed. Murakami, A. Lijima, and J.W. Ward Kodansha, Tokyo, 1986, 103–112. [19] M. Mertens, J.A. Martens, P.J. Grobet, and P.A. Jacobs, Guidelines for Mastering the Properties of Molecular Sieves – Relationship between the Physicochemical Properties of Zeolite Systems and their Low Dimensionality. NATO ASI, Ser. B, ed. D. Barthomeuf, E.G. Derouane, and W. Ho¨ldrich, Plenum Press, New York, London, 1990, 221, 1. [20] J.A. Martens, M. Mertens, P.J. Grobet, and P.A. Jacobs, Innovation in Zeolite Materials Science. Stud. Surf. Sci. Catal., 1988, 37, 97–105. [21] Y. Liu, A. Yan, and Q. Xu, Acidity, Diffusion and Catalytic Properties of the Silicoaluminophosphate SAPO-11. Appl. Catal., 1991, 67, 169–177. [22] H.-L. Zubowa, E. Alsdorf, R. Fricke, F. Neissendorfer, J. Richter-Mendau, E. Schreier, D. Zeigan, and B. Zibrowius, Synthesis and Properties of the Silicoaluminophosphate Molecular Sieve SAPO-31. J. Chem. Soc., Faraday Trans., 1990, 86, 2307–2312. [23] Y. Xu, P.J. Maddox, and J.W. Couves, The Synthesis of SAPO-34 and CoSAPO-34 from a Triethylamine-Hydrofluoric Acid-Water System. J. Chem. Soc., Faraday Trans., 1990, 86, 425–429. [24] L. Maistriau, N. Dumont, J.B. Nagy, Z. Gabelica, and E.G. Derouane, Single- and Two - liquid - phase Synthetic Routes to SAPO-37. Zeolites, 1990, 10, 243–250. [25] M. Feuerstein and R.F. Lobo, Characterization of Li Cations in Zeolite LiX by Solidstate NMR Spectroscopy and Neutron Diffraction. Chem. Mater., 1998, 10, 2197– 2204. [26] M.L. Costenoble, W.J. Mortier, and J.B. Uytterhoeven, Location of Cations in Synthetic Zeolites X and Y. Part 4. Exchange Limiting Factors for Ca2þ in Zeolite Y. J. Chem. Soc., Faraday Trans. 1, 1976, 72, 1877–1883. [27] G. Vitale, C.F. Mellot, L.M. Bull, and A.K. Cheetham, Neutron Diffraction and Computational Study of Zeolite NaX: Influence of SIII’ Cations on its Complex with Benzene. J. Phys. Chem. B, 1997, 101, 4559–4567. [28] Y. Kim, Y.M. Han, and K. Seff, Crystal Structure of Fully Dehydrated Fully Tiþ-exchanged Zeolite X. Zeolites, 1997, 18, 325–333. [29] W. Shibata, and K. Seff, Terahertz Laser Vibration-rotation-tunneling Spectroscopy of the Water Tetramer. J. Phys. Chem. B, 1997, 101, 9022–9031. [30] F. Porcher, M. Souhassou, Y. Dusausoy, and C. Lecomte, The Crystal Structure of a Low Silica Dehydrated NaX Zeolite. Eur. J. Mineral., 1999, 11, 333–343. [31] D.H. Olson, The Crystal Structure of Dehydrated NaX. Zeolites, 1995, 15, 439–443. [32] T. Takaishi, Ordered Distribution of Na Ions in Dehydrated NaX Zeolite. Zeolites, 1996, 17, 389–392. [33] M.E. Davis, and R.F. Lobo, Zeolite and Molecular Sieve Synthesis. Chem. Mater., 1992, 4, 756–768. [34] F. Delprato, L. Delmotte, J.L. Guth, and L. Huve, Synthesis of New Silica-rich Cubic and Hexagonal Faujasites using Crown-ether-based Supramolecules as Templates. Zeolites, 1990, 10, 546–552. [35] R.F. Lobo, S.I. Zones, and M.E. Davis, Structure-direction in Zeolite Synthesis. J. Inclusion Phenom. Mol. Recognit. Chem., 1995, 21, 47–78.

106

Chemistry of Zeolites and Related Porous Materials

[36] M. Sato, Proceedings of the 6th International Conference on Zeolites, Reno, Butterworths Guildford, 1984, 851–857. [37] G.O. Brunner, F. Laves, and Z. Wiss, Techn. Univers. Dresden, 1971, 20, 387–390. [38] W.M. Meier and H.J. Moeck, The Topology of Three-dimensional 4-Connected Nets: Classification of Zeolite Framework Types using Coordination Sequences. J. Solid State Chem., 1979, 27, 349–355. [39] M. O’Keeffe and S.T. Hyde, Vertex Symbols for Zeolite Nets. Zeolites, 1997, 19, 370–374. [40] L. Pauling, The Structure of Sodalite and Helvite. Z. Kristallogr. Kristallgeom. Kristall phys. Kristallchem., 1930, 74, 213–225. [41] J. Loens and H. Schulz, Strukturverfeinerung von Sodalith, Na8Si6Al6O24Cl2, Acta Crystallogr., 1967, 23, 434–436. [42] I. Hassan, and H.D. Grundy, Structure of Basic Sodalite, Na8Al6Si6O24(OH)2.2H2O. Acta Crystallogr., Sect. C: Cryst. Struct. Commun., 1983, 39, 3–5. [43] J. Felsche, S. Luger, and Ch. Baerlocher, Crystal Structures of the Hydro-sodalite Na6 [AlSiO4]6.8H2O and of the Anhydrous Sodalite Na6[AlSiO4]6. Zeolites, 1986, 6, 367–372. [44] L.B. McCusker, W.M. Meier, K. Suzuki, and S. Shin, The Crystal Structure of a Sodium Gallosilicate Sodalite. Zeolites, 1986, 6, 388–391. [45] T.M. Nenoff, W.T.A. Harrison, T.E. Gier, and G.D. Stucky, Room-temperature Synthesis and Characterization of New ZnPO and ZnAsO Sodalite Open Frameworks. J. Am. Chem. Soc., 1991, 113, 378–379. [46] T.E. Gier, W.T.A. Harrison, and G.D. Stucky, The Synthesis and Structure of some New Sodalites: the Lithium Haloberyllophosphates and -Arsenates. Angew. Chem., Int. Ed., 1991, 30, 1169–1171. [47] M. Wiebcke, P. Sieger, J. Felsche, G. Engelhardt, P. Behrens, and J. Schefer, Sodium Aluminogermanat Hydroxosodalite Hydrate Na6þx[Al6Ge6O24](OH)x.nH2O (x-Approximate-to-1.6, n-Approximate-to-3.0) - Synthesis, Phase Transitions and Dynamical Disorder of the Hydrogen Dihydroxide Anion, H3O2
, in the Cubic High-Temperature Form. Z. Anorg. Allg. Chem., 1993, 619, 1321–1329. [48] M.A. Camblor, R.F. Lobo, H. Koller, and M.E. Davis, Synthesis and Characterization of Zincosilicates with the SOD Topology. Chem. Mater., 1994, 6, 2193–2199. [49] S.E. Dann and M.T. Weller, Synthesis and Structure of Cadmium Chalcogenide Beryllosilicate Sodalites. Inorg. Chem., 1996, 35, 555–558. [50] X.H. Bu, P.Y. Feng, T.E. Gier, D.Y. Zhao, and G.D. Stucky, Hydrothermal Synthesis and Structural Characterization of Zeolite-like Structures based on Gallium and Aluminum Germinates. J. Am. Chem. Soc., 1998, 120, 13389–13397. [51] X. Bu, T.E. Gier, P. Feng, and G.D. Stucky, Template Control of Framework Topology and Charge in New Phosphate- and Arsenate-based Sodalite Analogs. Microporous Mesoporous Mater., 1998, 20, 371–379. [52] V. Gramlich and W.M. Meier, The Crystal Structure of Hydrated NaA: a Detailed Refinement of a Pseudosymmetric Zeolite Structure. Z. Kristallogr. Kristallgeom. Kristallphys. Kristallchem., 1971, 133, 134–149. [53] T.B. Reed and D.W. Breck, Crystalline Zeolites. II. Crystal Structures of Synthetic Zeolite, Type A. J. Am. Chem. Soc., 1956, 78, 5972–5977. [54] G.T. Kerr, Chemistry of Crystalline Aluminosilicates. II. The Synthesis and Properties of Zeolite ZK-4. Inorg. Chem., 1966, 5, 1537–1539. [55] G.H. Kuehl, High-Silica Analogs of Zeolite A containing Intercalated Phosphate. Inorg. Chem., 1971, 10, 2488–2495. [56] B.M. Lok, C.A. Messina, R.L. Patton, R.T. Gajek, T.R. Cannan, and E.M. Flanigen, Silicoaluminophosphate Molecular Sieves: Another new Class of Microporous Crystalline Inorganic Solids. J. Am. Chem. Soc., 1984, 106, 6092–6093.

Structural Chemistry of Microporous Materials

107

[57] A. Simmen, J. Patarin, and Ch. Baerlocher, Rietveld Refinement of F
-containing GaPO4LTA. Proceedings of the 9th International Zeolite Conference Veienna, 1993, 433–440. [58] A. Corma, F. Rey, J. Rius, M.J. Sabater, and S. Valencia, Supramolecular Self-assembled Molecules as Organic Directing Agent for Synthesis of Zeolites. Nature (London), 2004, 431, 287–290. [59] P. Feng, X. Bu, and G.D. Stucky, Hydrothermal Syntheses and Structural Characterization of Zeolite Analogue Compounds based on Cobalt Phosphate. Nature (London), 1997, 388, 735–741. [60] W.H. Baur, On the Cation and Water Positions in Faujasite. Am. Mineral., 1964, 49, 697–704. [61] W.T.A. Harrison, T.E. Gier, K.L. Moran, J.M. Nicol, H. Eckert, and G.D. Stucky, Structures and Properties of New Zeolite X-type Zincophosphate and Beryllophosphate Molecular Sieves. Chem. Mater., 1991, 3, 27–29. [62] Ch. Baerlocher, L.B. McCusker, and A.M. Chippindale, Location of the 18-Crown-6 Template in EMC-2 (EMT). Rietveld Refinement of the Calcined and as-Synthesized Forms. Microporous Mater., 1994, 2, 269–280. [63] G.T. Kokotailo and J. Ciric, Synthesis and Structural Features of Zeolite ZSM-3. Adv. Chem. Ser., 1971, 101, 109–121. [64] J.M. Newsam, M.M.J. Treacy, D.E.W. Vaughan, K.G. Strohmaier, and W.J. Mortier, The Structure of Zeolite ZSM-20: Mixed Cubic and Hexagonal Stackings of Faujasite Sheets. J. Chem. Soc., Chem. Commun., 1989, 493–495. [65] R.M. Barrer and H. Villiger, The Crystal Structure of the Synthetic Zeolite L. Z. Kristallogr. Kristallgeom. Kristallphys. Kristallchem., 1969, 128, 352–370. [66] Ch. Baerlocher, and R.M. Barrer, The Structure of the Synthetic Zeolite (K,Ba)-G,L. Z. Kristallogr. Kristallgeom. Kristallphys. Kristallchem., 1972, 136, 245–254. [67] J.M. Newsam, Structural Characterization of Dehydrated Gallium Zeolite L. Mater. Res. Bull., 1986, 21, 661–672. [68] N. Venkatathri, Synthesis and Characterization of AlPO4-n Molecular Sieves from Hexamethyleneimine Template. Indian J. Chem. Sect. A, 2002, 41, 2223–2230. [69] O. Jarchow, Atomordnung und Strukturverfeinerung von Cancrinit. Z. Kristallogr. Kristallgeom. Kristallphys. Kristallchem., 1965, 122, 407–422. [70] P.N. Bresciana, M. Calligaris, G. Nardin, and L. Randaccio, Structure of a Basic Cancrinite. Acta Crystallogr., Sect. B: Struct Crystallogr. Crist. Chem., 1982, 38, 893–895. [71] E.L. Belokoneva, T.G. Uvarova, and L.N. Dem’yanets, Crystal Structure of Synthetic Ge-Cancrinite Na8[Al6Ge6O24]Ge(OH)6.2H2O. Sov. Phys. Crystallogr. (Engl. Transl.), 1986, 31, 516–519. [72] D.R. Peacor, R.C. Rouse, and J.H. Ahn, Crystal Structure of Tiptopite, a Framework Beryllophosphate Isotypic with Basic Cancrinite. Am. Mineral., 1987, 72, 816–820. [73] O.V. Yakubovich, O.V. Karimova, and O.K. Mel’nikov, A New Representative of the Cancrinite Family (Cs,K)0.33[Na0.18Fe0.16(H2O)1.05]{ZnPO4}: Preparation and Crystal Structure. Crystallogr. Rep., 1994, 39, 630–634. [74] Y. Lee, J.B. Parise, A. Tripathi, S.J. Kim, and T. Vogt, Synthesis and Crystal Structures of Gallium and Germanium Variants of Cancrinite. Microporous Mesoporous Mater., 2000, 39, 445–455. [75] J.V. Smith, R. Rinaldi, and L.S.D. Glasser, Crystal Structures with a Chabazite Framework. II. Hydrated Ca-chabazite at Room Temperature. Acta Crystallogr., 1963, 16, 45–53. [76] L.S. Dent and J.V. Smith, Crystal Structure of Chabazite, a Molecular Sieve. Nature (London), 1958, 181, 1794–1796. [77] M. Ito, Y. Shimoyama, Y. Saito, Y. Tsurita, and M. Otake, Structure of an Aluminosilicophosphate. Acta Crystallogr., Sect. C: Cryst. Struct. Commun, 1985, 41, 1698–1700.

108

Chemistry of Zeolites and Related Porous Materials

[78] J.M. Bennett and B.K. Marcus, The Crystal Structures of Several Metal Aluminophosphate Molecular Sieves. Stud. Surf. Sci. Catal., 1988, 37, 269–279. [79] J.J. Pluth and J.V. Smith, Silicoaluminophosphate with Encapsulated Methylbutylamine Species: Chabazite Structure, Charge Coupling between Framework and Inferred Ammonium Species, and Severe Molecular Disorder. J. Phys. Chem., 1989, 93, 6516– 6520. [80] M.M. Harding and B.M. Kariuki, Microcrystal Structure Determination of AlPO4-CHA using Synchrotronradiation. Acta Crystallogr., Sect. C: Cryst. Struct. Commun., 1994, 50, 852–854. [81] C. Schott-Darie, H. Kessler, M. Soulard, V. Gramlich, and E. Benazzi, Diversity of the System Ga2O3-P2O5-H2O-HF in the Presence of Organic Species. Stud. Surf. Sci. Catal., 1994, 84, 101–108. [82] P. Feng, X. Bu, T.E. Gier and G.D. Stucky, Amine-directed Syntheses and Crystal Structures of Phosphate-based Zeolite Analogs. Microporous Mesoporous Mater., 1998, 23, 221–229. ` az-Cabanas, P.A. Barrett, and M.A. Camblor, Synthesis and Structure of Pure SiO2 [83] M.-J. DA Chabazite: the SiO2 Polymorph with the Lowest Framework Density. Chem. Commun., 1998, 1881–1882. [84] K.O. Kongshaug, H. Fjellvag, and K.P. Lillerud, The Synthesis and Crystal Structure of Two Novel 3D Open-framework Zinc Phosphates UiO-21 and UiO-22. Microporous Mesoporous Mater., 2000, 39, 341–350. [85] H. Zhang, L. Weng, Y. Zhou, Z. Chen, J. Sun, and D. Zhao, [C6N4H24]CoBe6P6O24.3H2O: a Novel 3-Dimensional Beryllophosphate Zeolite-like Structure Encapsulating CoII ions J. Mater. Chem., 2002, 12, 658–662. [86] W.M. Meier, The Crystal Structure of Mordenite (Ptilolite). Z. Kristallogr. Kristallgeom. Kristallphys. Kristallchem., 1961, 115, 439–450. [87] L.B. Sand, Synthesis of Large-port and Small-port Mordenites. Mol. Sieves, 1968, 71–77. [88] M.J. Eapen, K.S.N. Reddy, P.N. Joshi, and V.P. Shiralkar, Synthesis of a Gallosilicate Analogue of High Silica, Large Port Mordenite. J. Inclusion Phenom. Mol. Recognit. Chem., 1992, 14, 119–129. [89] D.H. Olson, G.T. Kokotailo, S.L. Lawton, and W.M. Meier, Crystal Structure and Structurerelated Properties of ZSM-5. J. Phys. Chem., 1981, 85, 2238–2243. [90] E.M. Flanigen, J.M. Bennett, R.W. Grose, J.P. Cohen, R.L. Patton., R.M. Kirchner, and J.V. Smith, Silicalite, a new Hydrophobic Crystalline Silica Molecular Sieve. Nature (London), 1978, 271, 512–516. [91] G.T. Kokotailo, S.L. Lawton, D.H. Olson, and W.M. Meier, Structure of Synthetic Zeolite ZSM-5. Nature (London), 1978, 272, 437–438. [92] M. Taramasso, G. Perego, and B. Notari, Molecular Sieve Borosilicates. Proceedings of the 5th International Zeolite Conference, Heyden and son, London, 1980, 40–48. [93] H. van Koningsveld, H. van Bekkum, and J.C. Jansen, On the Location and Disorder of the Tetrapropylammonium (TPA) Ion in Zeolite ZSM-5 with Improved Framework Accuracy. Acta Crystallogr., Sect. B: Struct. Sci., 1987, 43, 127–132. [94] J. Patarin, H. Kessler, and J.L. Guth, Iron Distribution in Iron MFI-type Zeolite Samples Synthesized in Fluoride Medium - Influence of the Synthesis Procedure. Zeolites, 1990, 10, 674–679. [95] H. van Koningsveld, J.C. Jansen, and H. van Bekkum, The Monoclinic Framework Structure of Zeolite H-ZSM-5. Comparison with the Orthorhombic Framework of as-Synthesized ZSM-5. Zeolites, 1990, 10, 235–242. [96] S.V. Awate, P.N. Joshi, V.P. Shiralkar, and A.N. Kotasthane, Synthesis and Characterization of Gallosilicate Pentasil (MFI) Framework Zeolites. J. Inclusion Phenom. Mol. Recognit. Chem., 1992, 13, 207–218.

Structural Chemistry of Microporous Materials

109

[97] G.T. Kokotailo, P. Chu, S.L. Lawton, and W.M. Meier, Synthesis and Structure of Synthetic Zeolite ZSM-11. Nature (London), 1978, 275, 119–120. [98] D.M. Bibby, N.B. Milestone, and L.P. Aldridge, Silicalite-2, a Silica Analogue of the Aluminosilicate Zeolite ZSM-11. Nature (London), 1979, 280, 664–665. [99] C.A. Fyfe, H. Gies, G.T. Kokotailo, C. Pasztor, H. Strobl, and D.E. Cox, Detailed Investigation of the Lattice Structure of Zeolite ZSM-11 by a Combination of Solid-state NMR and Synchrotron X-Ray Diffraction Techniques. J. Am. Chem. Soc., 1989, 111, 2470–2474. [100] H. van Koningsveld, M.J. den Exter, J.H. Koegler, C.D. Njo, S.L. Laman, and H. Graafsma, Single Crystal Structure Analysis of a Microcrystal of ZSM-11 using Synchrotron X-Ray Data. Proceedings of the 12th International Zeolite Conference, MRS, Wattendale, PA, 1999, IV, 2419–2424. [101] J.M. Newsam, M.M.J. Treacy, W.T. Koetsier, and C.B. de Gruyter, Structural Characterization of Zeolite Beta. Proc. R. Soc. London, Sec. A, 1988, 420, 375–405. [102] J.B. Higgins, R.B. Lapierre, J.L. Schlenker, A.C. Rohrman, J.D. Wood, G.T. Kerr, and W.J. Rohrbaugh, The Framework Topology of Zeolite Beta. Zeolites, 1988, 8, 446–452. [103] B. Marler, R.H. Bo¨hme, and H. Gies, Single Crystal Structure Analysis of Zeolite Beta: the Superposition Structure. Proceedings of the 9th International Zeolite Conference, ed. R. von Ballmoos, J.B. Higgins, and M.M.J. Treacy, Butternorth-Heinemann, Boston, 1993, 425–432. [104] T. Takewaki, L.W. Beck, and M.E. Davis, Synthesis of CIT-6, a Zincosilicate with the *BEA Topology. Top. Catal., 1999, 9, 35–42. [105] T. Conradsson, M.S. Adachov, and X. Zou, Synthesis and Structure of (Me3N)6 [Ge32O64](H2O)4.5, a Thermally Stable Novel Zeotype with 3D Interconnected 12-ring Channels. Microporous Mesoporous Mater., 2000, 41, 183–191. [106] Z. Liu, T. Ohsuna, O. Terasaki, M.A. Camblor, M.-J. Diaz-Caban˜as, and K. Hiraga, The First Zeolite with Three Dimensional Intersecting Straight Channel System of Twelve Membered Rings. J. Am. Chem. Soc., 2001, 123, 5370–5371. [107] A. Corma, M.T. Navarro, F. Rey, J. Rius, and S. Valencia, Pure Polymorph C of Zeolite Beta Synthesized by using Framework Isomorphous Substitution as a Structure-directing Mechanism. Angew. Chem., Int. Ed., 2001, 40, 2277–2280. [108] P. Wagner, M. Yoshikawa, M. Lovallo, K. Tsuji, M. Tsapatsis, and M.E. Davis, CIT-5: a High-silica Zeolite with 14-ring Pores. Chem. Commun., 1997, 2179–2180. [109] M. Yoshikawa, P. Wagner, M. Lovallo, K. Tsuji, T. Takewaki, C. Chen, L.W. Beck, C. Jones, M. Tsapatsis, S.I. Zones, and M.E. Davis, Synthesis, Characterization, and Structure Solution of CIT-5, a New, High-silica, Extra-large-pore Molecular Sieve. J. Phys. Chem. B, 1998, 102, 7139–7147. [110] M.A. Camblor, M.J. Diaz-Cabanas, J. Perez-Pariente, S.J. Teat, W. Clegg, I.J. Shannon, P. Lightfoot, P.A. Wright, and R.E. Morris, SSZ-23: An Odd Zeolite with Pore Openings of Seven and Nine Tetrahedral Atoms. Angew. Chem., Int. Ed., 1998, 37, 2122–2126. [111] T. Wessels, Ch. Baerlocher, L.B. McCusker, and E.J. Creyghton, An Ordered Form of the Extra-large-pore Zeolite UTD-1: Synthesis and Structure Analysis from Powder Diffraction Data. J. Am. Chem. Soc., 1999, 121, 6242–6247. [112] R.F. Lobo, M. Tsapatsis, C.C. Freyhardt, S. Khodabandeh, P. Wagner, C. Chen, K.J. Balkus, S.I. Zones, and M.E. Davis, Characterization of the Extra-large-pore Zeolite UTD-1. J. Am. Chem. Soc., 1997, 119, 8474–8484. [113] A. Corma, F. Rey, S. Valencia, J.L. Jorda, and J. Rius, A Zeolite with Interconnected 8-, 10and 12-Ring Pores and its Unique Catalytic Selectivity. Nature Mater. (London), 2003, 2, 493–497. [114] R.F. Lobo and M.E. Davis, CIT-1: A New Molecular Sieve with Intersecting Pores Bounded by 10- and 12-rings. J. Am. Chem. Soc., 1995, 117, 3764–3779.

110

Chemistry of Zeolites and Related Porous Materials

[115] R.F. Lobo, M. Pan, I. Chan, H. Li, R.C. Medrud, S.I. Zones, P.A. Crozier, and M.E. Davis, SSZ-26 and SSZ-33 – 2 Molecular Sieves with Intersecting 10-Ring and 12-Ring Pores, Science, 1993, 262, 1543–1546. [116] K.G. Strohmaier and D.E.W. Vaughan, Structure of the First Silicate Molecular Sieve with 18-Ring Pore Openings, ECR-34. J. Am. Chem. Soc., 2003, 125, 16035–16039. [117] A. Burton, S. Elomari, C. Chen, R.C. Medrud, I.Y. Chan, L.M. Bull, C. Kibby, T.V. Harris, S.I. Zones, and E.S. Vittoratos, SSZ-53 and SSZ-59: Two Novel Extra-large Pore Zeolites. Chem. Eur. J. 2003, 9, 5737–5748. [118] R. Castaneda, A. Corma, V. Fornes, F. Rey, and J. Rius, Synthesis of a New Zeolite Structure ITQ-24, with Intersecting 10- and 12-Membered Ring Pores. J. Am. Chem. Soc., 2003, 125, 7820–7821. [119] B. Harbuzaru, J.L. Paillaud, J. Patarin, and N. Bats, Extra-large-pore Zeolites with Twodimensional Channels formed by 14- and 12-rings. Science, 2004, 304, 990–992. [120] A. Corma, M.J. DIˆaz-Caban˜as, F. Rey, S. Nicolopoulus, and K. Boulahya, ITQ-15: The First Ultralarge Pore Zeolite with a Bi-directional Pore System formed by Intersecting 14- and 1-ring Channels, and its Catalytic Implications. Chem. Commun., 2004, 1356–1357. [121] J.M. Bennett, J.P. Cohen, E.M. Flanigen, J.J. Pluth, and J.V. Smith, Crystal Structure of Tetrapropylammonium Hydroxide-Aluminum Phosphate Number 5. ACS Symp. Ser., 1983, 218, 109–118. [122] E.M. Flanigen, B.M. Lok, R.L. Patton, and S.T. Wilson, Aluminophosphate Molecular Sieves and the Periodic Table. Pure Appl. Chem., 1986, 58, 1351–1358. [123] S. Qiu, W. Pang, H. Kessler, and J.L. Guth, Synthesis and Structure of the [AlPO4]12.Pr4NF Molecular Sieve with AFI Structure. Zeolites, 1989, 9, 440–444. [124] R. Bialek, W.M. Meier, M.E. Davis, and M.J. Annen, The Synthesis and Structure of SSZ-24, the Silica Analog of AlPO4-5. Zeolites, 1991, 11, 438–442. [125] K.J. Chao, S.P. Sheu, and H.S. Sheu, Structure and Chemistry of Cobalt in CoAPO-5 Molecular Sieve. J. Chem. Soc., Faraday Trans., 1992, 88, 2949–2954. [126] S.F. Radaev, W. Joswig, and W.H. Baur, Synthesis and Precise Characterization of Large Single Crystals of as-Synthesized CrAPO-5, H2.1[Cr0.08Al11.62P12O48]F1.2.1.6TEA. J. Mater. Chem., 1996, 6, 1413–1418. [127] M.E. Davis, C. Saldarriaga, C. Montes, J. Garces, and C. Crowder, A Molecular Sieve with Eighteen-membered Rings. Nature (London), 1988, 331, 698–699. [128] L.B. McCusker, Ch. Baerlocher, E. Jahn, and M. Bu¨low, The Triple Helix inside the Largepore Aluminophosphate Molecular Sieve VPI-5. Zeolites, 1991, 11, 308–313. [129] F. d’Yvoire, Etude des Phosphates d’aluminium et de Fer Trivalent. I. L’orthophosphate Neutre d’aluminium. Bull. Soc. Chim. Fr., 1961, 1762–1776. [130] E.G. Derouane, L. Maistreiau, Z. Gabelica, A. Tuel, J.B. Nagy, and R. von Ballmoos, Synthesis and Characterization of the Very Large Pore Molecular Sieve MCM-9. Appl. Catal., 1989, 51, 13–20. [131] P.S. Singh, R.A. Shaikh, R. Bandyopadhyay, and B.S. Rao, Synthesis of CoVPI-5 with Bifunctional Catalytic Activity. J. Chem. Soc., Chem. Commun., 1995, 2255–2256. [132] J.D. Martı´nez, L.B. McCusker, and Ch. Baerlocher, Characterization and Structural Analysis of Differently Prepared Samples of Dehydrated VPI-5. Microporous Mesoporous Mater., 2000, 34, 99–113. [133] E.T.C. Vogt and J.W. Richardson, The Reversible Transition of the Molecular Sieve VPI-5 into AlPO4-8 and the Structure of AlPO4-8. J. Solid State Chem., 1990, 87, 469–471. [134] J.A. Martens, H. Geerts, P.J. Grobet, and P.A. Jacobs, in Zeolite Microporous Solids: Synthesis, Structure and Reactivity, ed. E.G. Derouane, F. Lemos, C. Naccache, and F.R. Ribeiro, NATO ASI Ser. C, Kluwer Academic, Dordrecht, Boston, London, 1992, 352, 477.

Structural Chemistry of Microporous Materials

111

[135] D.E. Akporiaye and M. Sto¨ker, Solid State MAS NMR Characterisation of the Transformation of VPI-5 to AIP04-8. Proceedings of the 9th International Zeolite Conference, ed. R. von Ballmoos, J.B. Higgins, and M.M.J. Treacy, Butterworth-Heinemann, Boston, 1993, 563. [136] R.M. Dessau, J.L. Schlenker, and J.B. Higgins, Framework Topology of AlPO4-8: the First 14-ring Molecular Sieve. Zeolites, 1990, 10, 522–524. [137] J.W. Richardson and E.T.C. Vogt, Structure Determination and Rietveld Refinement of Aluminophosphate Molecular Sieve AlPO4-8. Zeolites, 1992, 12, 13–19. [138] M. Estermann, L.B. McCusker, Ch. Baerlocher, A. Merrouche, and H. Kessler, A Synthetic Gallophosphate Molecular Sieve with a 20-Tetrahedral-atom Pore Opening. Nature (London), 1991, 352, 320–332. [139] N. Rajic, N.Z. Logar, and V. Kaucic, A Novel Open Framework Zincophosphate: Synthesis and Characterization. Zeolites, 1995, 15, 672–678. [140] W.T.A. Harrison, T.E. Gier, G.D. Stucky, R.W. Broach, and R.A. Bedard, NaZnPO4.H2O, an Open-framework Sodium Zincophosphate with a New Chiral Tetrahedral Framework Topology. Chem. Mater., 1996, 8, 145–151. [141] M. Helliwell, J.R. Helliwell, V. Kaucic, N.Z. Logar, L. Barba, E. Busetto, and A. Lausi, Determination of the Site of Incorporation of Cobalt in CoZnPO-CZP by Multiplewavelength Anomalous-dispersion Crystallography. Acta Crystallogr., Sect. B: Struct. Sci., 1999, B55, 327–332. [142] C. Lin and S. Wang, Chiral Metal Gallophosphates Templated by Achiral Triamine: Syntheses and Characterizations of A[Mn(H2O)2Ga(PO4)2]3 and A[Zn3Ga(PO4)4].H2O (A ¼ H3DETA). Chem. Mater., 2002, 14, 96–102. [143] X. Bu, P. Feng, and G.D. Stucky, Large-cage Zeolite Structures with Multidimensional 12-Ring Channels. Science, 1997, 278, 2080–2085. [144] J. Chen, W. Pang, and R. Xu, Mixed-bonded Open-framework Aluminophosphates and Related Layered Materials. Top. Catal., 1999, 9, 93–103. [145] J. Yu, J. Li, and R. Xu, Structural Diversity of a Family of Aluminophosphates with Al/P Ratio of Non-unity. Solid State Sci., 2000, 2, 181–192. [146] J. Yu and R. Xu, Rich Structure Chemistry in the Aluminophosphate Family. Acc. Chem. Res., 2003, 36, 481–490. [147] J. Yu and R. Xu, Insight into the Construction of Open-framework Aluminophosphates. Chem. Soc. Rev., 2006, 35, 593–604. [148] Y. Li, J. Yu, and R. Xu, http://mezeopor.jlu.edu.cn/alpo and references therein. [149] K. Wang, J. Yu, P. Miao, Y. Song, J. Li, Z. Shi, and R. Xu, A New Layered Aluminophosphate [C4H12N2][Al2P2O8(OH)2] Templated by Piperazine. J. Mater. Chem., 2001, 11, 1898– 1902. [150] K.O. Kongshaug, H. Fjellvag, and K.P. Lillerud, Layered Aluminophosphates I. Crystal Structure of Two Novel Layered Aluminophosphates Solved ab initio from Powder Diffraction Data. Microporous Mesoporous Mater. 1999, 32, 17–28. [151] K.O. Kongshaug, H. Fjellvag, and K.P. Lillerud, Layered Aluminophosphates II. Crystal Structure and Thermal Behaviour of the Layered Aluminophosphate UiO-15 and its High Temperature Variants. J. Mater. Chem., 1999, 9, 1591–1598. [152] K. Maeda, A. Tuel, and Ch. Baerlocher, Synthesis and Characterization of a New Layered Aluminophosphate Templated with 1,3-Diaminopropane: [H3N(CH2)3NH3]0.5[AlPO4 (OH)(OH2)]H2O. J. Chem. Soc., Dalton Trans., 2000, 2457–2462. [153] D. Riou, T. Loiseau, and G. Fe´rey, Structure Determination of (N2C2H10)4(NH4)AlP4O16: A New Aluminophosphate Templated by Ethylenediamine. J. Solid State Chem., 1992, 99, 414–418. [154] W. Yan, J. Yu, Z. Shi, and R. Xu, A Novel Open-framework Aluminophosphate [AlP2O6(OH)2][H3O] containing Propeller-like Chiral Motifs. Chem. Commun., 2000, 1431–1432.

112

Chemistry of Zeolites and Related Porous Materials

[155] J. Yu and I.D. Williams, Two Unusual Layer Aluminophosphates Templated by Imidazolium Ions:[N2C3H5][AlP2O8H2.2H2O] and 2[N2C3H5][Al3P4O16H]. J. Solid State Chem., 1998, 136, 141–144. [156] K.R. Morgan, G.J. Gainsford, and N.B. Milestone, A New Type of Layered Aluminium phosphate [NH4]3[Co(NH3)6]3[Al2(PO4)4]2 Assembled about a Cobalt(III) Hexammine Complex. Chem. Commun., 1997, 61–62. [157] M.A. Leech, A.R. Cowley, K. Prout, and A.M. Chippindale, Ambient-temperature Synthesis of New Layered AlPOs and GaPOs in Silica Gels. Chem. Mater., 1998, 10, 451–456. [158] R.H. Jones, J.M. Thomas, R. Xu, Q. Huo, Y. Xu, A.K. Cheetham, and D. Bieber, Synthesis and Structure of a One-dimensionally Extended Aluminum Phosphate Et3NHþ(H2AlP2O8)
. J. Chem. Soc., Chem. Commun., 1990, 1170–1172. [159] I.D. Williams, J. Yu, Q. Gao, J. Chen, and R. Xu, New Chain Architecture for a Onedimensional Aluminophosphate, [H3NCH2CH2NH3][AlP2O8H]. Chem. Commun., 1997, 1273–1274. [160] Q. Gao, J. Chen, S. Li, and R. Xu, Synthesis and Structure of a Chain Aluminophosphate Filled with [NH4]þ and [H3NCH2CH2NH3]2þ Cations. J. Solid State Chem., 1996, 127, 145–150. [161] A.A. Ayi, A. Choudhury, and S. Natarajan. Linear-chain AlPOs Obtained by the Reaction of Amine Phosphates with Al3þ ions. J. Solid State Chem., 2001, 156, 185–193. [162] B. Wei, J. Yu, Z. Shi, S. Qiu, and J. Li, A New Layered Aluminophosphate [Al2P4O16][C6H22N4][C2H10N2] with 4.12-Net Porous Sheets. J. Chem. Soc., Dalton Trans., 2000, 1979–1980. [163] Z. Bircsak, and W.T.A. Harrison, Template Cooperation Effect Leading to the New Layered Aluminophosphate CN3H6.Al(HPO4)2.2H2O. Chem. Mater., 1998, 10, 3016–3019. [164] K. Sugiyama, K. Hiraga, J. Yu, S. Zheng, S. Qiu, R. Xu, and O. Terasaki, A Onedimensionally Extended Chain Aluminophosphate. Acta Crystallogr., Sect. C: Cryst. Struct. Commun., 1999, 55, 1615–1617. [165] B. Wei, G. Zhu, J. Yu, S. Qiu, F. Xiao, and O. Terasaki, Solvothermal Synthesis and Characterization of a New 3-D Open-framework Aluminophosphate [Al2P3O12][C4N3H16]. Chem. Mater., 1999, 11, 3417–3419. [166] J. Yu, K. Sugiyama, K. Hiraga, N. Togashi, O. Terasaki, Y. Tanaka, S. Nakata, S. Qiu, and R. Xu, Synthesis and Characterization of a New 2-D Aluminophosphate Layer [Al2P3O10(OH)2][C6NH8] and Structural Diversity in Anionic Aluminophosphates with Al2P3O123
Stoichiometry. Chem. Mater., 1998, 10, 3636–3642. [167] A.M. Chippindale, A.V. Powell, L.M. Bull, R.H. Jones, A.K. Cheetham, J.M. Thomas, and R. Xu, Synthesis and Characterization of Two Layered Aluminophosphates, (T)2HAl2P3O12 (T ¼ 2-BuNH3þ) and (T)H2Al2P3O12 (T ¼ PyHþ). J. Solid State Chem., 1992, 96, 199–210. [168] S. Oliver, A. Kuperman, A. Lough, and G.A. Ozin, Aluminophosphate Chain-to-layer Transformation. Chem. Mater., 1996, 8, 2391–2398. [169] S. Oliver, A. Kuperman, A. Lough, and G.A. Ozin, The Synthesis and Structure of Two Novel Layered Aluminophosphates containing Interlamellar Cyclohexylammonium. Chem. Commun., 1996, 1761–1762. [170] A.M. Chippindale, and R.I. Walton, [C9H20N][Al2(HPO4)2(PO4)]: An Aluminium Phosphate with a New Layer Topology. J. Solid State Chem., 1999, 145, 731–738. [171] Y. Song, J. Yu, Y. Li, G. Li, and R. Xu, Hydrogen-bonded Helices in the Layered Aluminophosphate (C2H8N)2[Al2(HPO4)(PO4)2]. Angew. Chem., Int. Ed., 2004, 43, 2399–2402. [172] Y. Xu, B. Zhang, X. Chen, S. Liu, C. Duan, and X. You, An Open Framework Aluminophosphate with Unique 12-Membered Ring Channels: Al9(PO4)12(C24H91N16)17H2O. J. Solid State Chem., 1999, 145, 220–226.

Structural Chemistry of Microporous Materials

113

[173] R.H. Jones, J.M. Thomas, R. Xu, Q. Huo, A.K. Cheetham, and A.V.J. Powell, Synthesis and Structure of a Novel Aluminum Phosphate Anion - (Al3P4O16)3
. J. Chem. Soc., Chem. Commun., 1991, 1266–1268. [174] Q. Gao, B. Li, J. Chen, S. Li, R. Xu, I.D. Williams, J. Zheng, and D. Barber, Nonaqueous Synthesis and Characterization of a New 2-Dimensional Layered Aluminophosphate [Al3P4O16].3[CH3CH2NH3]. J. Solid State Chem., 1997, 129, 37–44. [175] J.M. Thomas, R.H. Jones, R. Xu, J. Chen, A.M. Chippindale, S. Natarajan, and A.K. Cheetham, A Novel Porous Sheet Aluminophosphate Al3P4O163
1.5[NH3(CH2)4NH3]2þ. J. Chem. Soc., Chem. Commun., 1992, 929–931. [176] A.M. Chippindale, A.R. Cowley, Q. Huo, R.H. Jones, A.D. Law, J.M. Thomas, and R. Xu, Synthesis and Structure of a New Layered Aluminium Phosphate: [BuNH3]3[Al3P4O16]. J. Chem. Soc., Dalton. Trans., 1997, 2639–2643. [177] I.D. Williams, Q. Gao, J. Chen, L. Ngai, Z. Lin, and R. Xu, Organic-template Control of Inorganic Structure: a Low-symmetry Two-dimensional Sheet Aluminophosphate 3[NH3CHMeCH2NH3][Al6P8O32].H2O. Chem. Commun., 1996, 1781–1782. [178] J. Yu, J. Li, K. Sugiyama, N. Togashi, O. Terasaki, K. Hiraga, B. Zhou, S. Qiu, and R. Xu, Formation of a New Layered Aluminophosphate [Al3P4O16][C5N2H9]2[NH4]. Chem. Mater., 1999, 11, 1727–1732. [179] R.H. Jones, A.M. Chippindale, S. Natarajan, and J.M. Thomas, A Reactive Template in the Synthesis of a Novel Layered Aluminum Phosphate (Al3P4O16)3
[NH3(CH2)5NH3]2þ (C5H10NH2)þ. J. Chem. Soc., Chem. Commun., 1994, 5, 565–566. [180] K. Morgan, G. Gainsford, and N. Milestone, A Novel Layered Aluminum Phosphate [Co(en)3Al3P4O163H2O] Assembled about a Chiral Metal-Complex. J. Chem. Soc., Chem. Commun., 1995, 425–426. [181] P.A. Barrett and R.H. Jones, The Pore-size Modification of a Layered Aluminophosphate [Al3P4O16C9H24N3] by Rational Selection of the Intercalated Template Cation. J. Chem. Soc., Chem. Commun., 1995, 1979–1980. [182] S. Oliver, A. Kuperman, A. Lough, and G.A. Ozin, Synthesis and Crystal Structures of Two Novel Anionic Aluminophosphates: a One-dimensional Chain, UT-7 ([Al3P5O20H]5
[C7H13NH3þ]5), and a Layer Containing Two Cyclic Amines, UT-8 ([Al3P4O16]3
[C4H7 NH3þ]2[C5H10NH2þ]). Inorg. Chem., 1996, 35, 6373–6380. [183] N. Togashi, J. Yu, S. Zheng, K. Sugiyama, K. Hiraga, O. Terasaki, W. Yan, S. Qiu, and R. Xu, Synthesis and Structure of a 2-D Layer Al3P4O16.3CH3CH2CH2NH3. J. Mater. Chem., 1998, 8, 2827–2831. [184] Y. Yao, S. Natarajan, J. Chen, and W. Pang, Synthesis and Structural Characterization of a New Layered Aluminophosphate Intercalated with Triply-protonated Triethylenetetramine [C6H21N4][Al3P4O16]. J. Solid State Chem., 1999, 146, 458–463. [185] L. Vidal, V. Gramlich, J. Patarin, and Z. Gabelica, Synthesis, Characterization and Structure of Mu-7, a New Layered Aluminophosphate with a 4  8 network. Chem. Lett, 1999, 201–202. [186] J. Yu, K. Sugiyama, S. Zheng, S. Qiu, J. Chen, R. Xu, Y. Sakamoto, O. Terasaki, K. Hiraga, M. Light, M.B. Hursthouse, and J.M. Thomas, Al16P20O80H4.4C6H18N2: A New Microporous Aluminophosphate Containing Intersecting 12- and 8-Membered Ring Channels. Chem. Mater., 1998, 10, 1208–1211. [187] L. Vidal, V. Gramlich, J. Patarin, and Z. Gabelica, Synthesis and Structure of Mu-4, the New Layered Aluminophosphate [(C2H5)2NH2]4[Al8P10O40H2][H2O]2.5. Eur. J. Solid. State Inorg. Chem., 1998, 35, 545–563. [188] D. Zhou, L. Chen, J. Yu, Y. Li, W. Yan, F. Deng, and R. Xu, Synthesis, Crystal Structure, and Solid-state NMR Spectroscopy of a New open-framework Aluminophosphate (NH4)2Al4 (PO4)4(HPO4).H2O Inorg. Chem., 2005, 44, 4391–4397.

114

Chemistry of Zeolites and Related Porous Materials

[189] E.R. Cooper, C.D. Andrews, P.S. Wheatley, P.B. Webb, P. Wormald, and R.E. Morris, Ionic Liquids and Eutectic Muctuses as Solvent and Template in Synthesis of zeolite Analogues. Nature (London), 2004, 430, 1012–1016. [190] Q. Huo, R. Xu, S. Li, Z. Ma, J.M. Thomas, R.H. Jones, and A.M. Chippindale, Synthesis and Characterization of a Novel Extra Large Ring of Aluminophosphate JDF-20. J. Chem. Soc., Chem. Commun., 1992, 875–876. [191] R.H. Jones, J.M. Thomas, J. Chen, R. Xu, Q. Huo, S. Li, Z. Ma, and A.M. Chippindale, Structure of an Unusual Aluminium Phosphate ([Al5P6O24H]2
2[N(C2H5)3H]þ2H2O) JDF20 with Large Elliptical Apertures. J. Solid State Chem., 1993, 102, 204–208. [192] K. Wang, J. Yu, Z. Shi, P. Miao, W. Yan, and R. Xu, Synthesis and Characterization of a New Three-dimensional Aluminophosphate [Al11P12O48][C4H12N2][C4H11N2] with an Al/P Ratio of 11/12. J. Chem. Soc. Dalton Trans., 2001, 1809–1812. [193] W. Yan, J. Yu, Z. Shi, P. Miao, K. Wang, Y. Wang, and R. Xu, An Anionic Framework Aluminophosphate j(CH2)6N4H3.H2Oj[Al11P12O48] and Computer Simulation of the Template Positions. Microporous Mesoporous Mater., 2001, 50, 151–158. [194] W. Yan, J. Yu, R. Xu, G. Zhu, F. Xiao, Y. Han, K. Sugiyama, and O. Terasaki, An Anionic Aluminophosphate Molecular Sievewith Bro¨nstedAcidity. Chem. Mater., 2000, 12, 2517–2519. [195] P. Feng, X. Bu, and G.D. Stucky, Communications - Control of Structural Ordering in ˚ . Inorg. Chem., Crystalline Lamellar Aluminophosphates with Periodicity from 51 to 62 A 2000, 39, 2–3. [196] S. Oliver, A. Kuperman, and G.A. Ozin, A New Model for Aluminophosphate Formation: Transformation of a Linear Chain Aluminophosphate to Chain, Layer, and Framework Structure. Angew. Chem. Int. Ed., 1998, 37, 46–62. [197] B. Wei, J. Yu, Z. Shi, S. Qiu, W. Yan, and O. Terasaki, A 3-D Open-framework Nickel Aluminophosphate [NiAlP2O8][C2N2H9]: Assembly of 1-D AlP2O83
Chains through [NiO5N] Octahedra. Chem. Mater. 2000, 12, 2065–2067. [198] K. Wang, J. Yu, Y. Song, and R. Xu, Assembly of one-dimensional AlP2O83
Chains to Three-dimensional Framework MAlP2O8.C2N2H9 through Transition Metal Cations

Ni2þ, Co2þ and Fe2þ). Dalton Trans., 2003, 99–103. (M

[199] D. Zhou, J. Xu, J. Yu, L. Chen, F. Deng, and R. Xu, Solid-state NMR spectroscopy of Anionic Framework Aluminophosphates: A New Method to Determine the Al/P Ratio. J. Phys. Chem. B, 2006, 110, 2131–2137. [200] T. Loiseau and G. Fe´rey, Oxyfluorinated Microporous Compounds, VII. Synthesis and Crystal Structure of ULM-5, a New Fluorinated Gallophosphate Ga16(PO4)14 (HPO4)2 (OH)2F7 [H3N(CH2)6NH3]4.6H2O with 16-Membered Rings and both Bonding and Encapsulated F
. J. Solid State Chem., 1994, 111, 403–415. [201] T. Loiseau and G. Fe´rey, Synthesis and Crystal Structure of ULM-16, a New Openframework Fluorinated Gallium Phosphate with 16-Ring Channels: Ga4(PO4)4F2.1. 5NC6H14.0.5H2O.0.5H3O. J. Mater. Chem., 1996, 6, 1073–1074. [202] C. Sassoye, T. Loiseau, F. Taulelle, and G. Fe´rey, A New Open-framework Fluorinated Gallium Phosphate with Large 18-Ring Channels (MIL-31). Chem. Commun., 2000, 943–944. [203] C. Sassoye, J. Marrot, T. Loiseau, and G. Ferey, Utilization of Cyclopentylamine as Structure-directing Agent for the Formation of Fluorinated Gallium Phosphates Exhibiting Extra- large-pore Open Frameworks with 16-Ring (ULM-16) and 18-Ring Channels (MIL-46). Chem. Mater., 2002, 14, 1340–1347. [204] L. Beitone, J. Marrot, T. Loiseau, G. Fe´rey, M. Henry, C. Huguenard, A. Gansmuller, and F. Taulelle, MIL-50, an Open-framework GaPO with a Periodic Pattern of Small Water Ponds and Dry Rubidium Atoms: a Combined XRD, NMR, and Computational Study. J. Am. Chem. Soc., 2003, 125, 1912–1922.

Structural Chemistry of Microporous Materials

115

[205] R.I. Walton, F. Millange, T. Loiseau, D. O’Hare, and G. Ferey, Crystallization of a Large-Pore Three-Dimensional Gallium Fluorophosphate under Mild Conditions. Angew. Chem., Int. Ed., 2000, 39, 4552–4555 [206] A.M. Chippindale, K.J. Peacock, and A.R. Cowley, Synthesis and Characterization of a Large-pore, Open-framework Gallium Phosphate, [NH3(CH2)4NH3]2[Ga4(HPO4)2(PO4)3 (OH)3].yH2O (y  5.4), and its Vanadium-Gallium Phosphate Analogue, [NH3(CH2)4NH3]2 [Ga44
xVx(HPO4)2(PO4)3(OH)3].yH2O (x  0.4, y  0.6). J. Solid State Chem., 1999, 145, 379–386. [207] C. Lin, S. Wang, and K.H. Lii, [Ga2(DETA)(PO4)2].2H2O (DETA ¼ diethylenetriamine): A Novel Porous Gallium Phosphate containing 24-Ring Channels. J. Am. Chem. Soc., 2001, 123, 4649–4650. [208] S.S. Dhingra and R.C. Haushalter, Hydrothermal Synthesis and Crystal Structure of H3NCH2CH2NH3[In2(HPO 4)4]. A Novel Octahedral-Tetrahedral Framework Indium Phosphate with Occluded Organic Cations. J. Chem. Soc., Chem. Commun., 1993, 1665– 1667. [209] I.D. Williams, J. Yu, H. Du, J. Chen, and W. Pang. A Metal-rich Fluorinated Indium Phosphate, 4[NH3(CH2)3NH3].3[H3O][In9(PO4)6(HPO4)2F16].3H2O with 14-Membered Ring Cavities. Chem. Mater., 1998, 10, 773–776. [210] A. Thirumurugan and S. Natarajan, Synthesis and Structure of a New Three-dimensional Indium Phosphate with 16-Membered One-dimensional Channels. Dalton Trans., 2003, 3387–3391. [211] H. Du, Synthesis and Preparation of New Inorganic Microporous Crystals. PhD Thesis, Jilin University, China (1997). [212] C.N.R. Rao, S. Natarajan, A. Choudhury, S. Neeraj, and A.A. Ayi, Aufbau Principle of Complex Open-framework Structures of Metal Phosphates with Different Dimensionalities. Acc. Chem. Res., 2001, 34, 80–87. [213] G. Yang and S.C. Sevov, Zinc Phosphate with Gigantic Pores of 24 Tetrahedra. J. Am. Chem. Soc., 1999, 121, 8389–8390. [214] J.A. Rodgers and W.T.A. Harrison, H3N(CH2)6NH3.Zn4(PO4)2(HPO4)2.3H2O: A Novel Three-dimensional Zinc Phosphate Framework Containing 5- and 20-Rings. J. Mater. Chem., 2000, 10, 2853–2856. [215] S. Neeraj, S. Natarajan, and C.N.R. Rao, A Novel Open-framework Zinc Phosphate with Intersecting Helical Channels. Chem. Commun., 1999, 165–166. [216] P.B. Moore and J. Shen, An X-ray Structural Study of Cacoxenite, a Mineral Phosphate. Nature (London), 1983, 306, 356–358. [217] A. Choudhury, S. Natarajan, and C.N.R. Rao, An Open-framework Iron Phosphate with Large Voids, Exhibiting Spin-crossover. Chem. Commun., 1999, 1305–1306. [218] N. Guillou, Q. Gao, P.M. Forster, J. Chang, M. Nogue´s, S.E. Park, G. Fe´rey, and A.K. Cheetham, Nickel(II) Phosphate VSB-5: a Magnetic Nanoporous Hydrogenation Catalyst with 24-Ring Tunnels. Angew. Chem., Int. Ed., 2001, 40, 2831–2834. [219] V. Soghomonian, Q. Chen, R.C. Haushalter, J. Zubieta, and C.J. O’Connor, An Inorganic Double Helix: Hydrothermal Synthesis, Structure, and Magnetism of Chiral [(CH3)2NH2]K4{V10O10(H2O)2(OH)4(PO4)7}.4H2O. Science, 1993, 259, 1596–1599. [220] M.I. Khan, L.M. Meyer, R.C. Haushalter, A.L. Schweitzer, J. Zubieta, and J.K. Dye, Giant Voids in the Hydrothermally Synthesized Microporous Square Pyramidal-Tetrahedral Framework Vanadium Phosphates [HN(CH2CH2)3NH]K1.35[V5O9(PO4)2].xH2O and Cs3[V5O9(PO4)2].xH2O. Chem. Mater., 1996, 8, 43–53. [221] M. O’Keeffe and O.M. Yaghi, Germanate Zeolites: Contrasting the Behavior of Germanate and Silicate Structures Built from Cubic T8O20 Units (T ¼ Ge or Si). Chem. Eur. J., 1999, 5, 2796–2801.

116

Chemistry of Zeolites and Related Porous Materials

[222] Y. Zhou, H. Zhu, Z. Chen, M. Chen, Y. Xu, H. Zhang, and D. Zhao, A Large 24-Memberedring Germanate Zeolite-type Open-framework Structure with Three-dimensional Intersecting Channels. Angew. Chem., Int. Ed., 2001, 40, 2166–2168. [223] J. Ple´vert, T.M. Gentz, A. Laine, H. Li, V.G. Young, O.M. Yaghi, and M. O’Keeffe, A Flexible Germanate Structure Containing 24-Ring Channels and with Very Low Framework Density. J. Am. Chem. Soc., 2001, 123, 12706–12707. [224] X. Zou, T. Conradsson, M. Klingstedt, M.S. Dadachov, and M. O’Keeffe, A Mesoporous Germanium Oxide with Crystalline Pore Walls and its Chiral Derivative. Nature (London), 2005, 437, 716–719. [225] H. Li, A. Laine, M. O’Keeffe, and O.M. Yaghi, Supertetrahedral Sulfide Crystals with Giant Cavities and Channels. Science, 1999, 283, 1145–1147. [226] M. O’Keeffe, Crystal Structures I: Patterns and Symmetry. Mineralogical Society of America, Washington DC, 1996.

3 Synthetic Chemistry of Microporous Compounds (I) – Fundamentals and Synthetic Routes The term molecular sieve refers to microporous framework compounds consisting of regular channels generated by the removal of templates via calcination, chemical treatment, extraction, or microwave assistance routes (methods), or by secondary synthesis approaches such as modification of framework, surface, or channel, ion-exchange, or isomorphous heteroatom substitution. Crystallization of microporous compounds is the core of the synthetic chemistry of molecular sieves. Most of the microporous compounds such as zeolites, microporous aluminophosphates, metal phosphates, oxides, and sulfides have been synthesized via hydrothermal synthetic reactions. In the early 1980s, D.M. Bibby first reported the solvothermal synthesis of zeolite sodalite in the medium of ethylene glycol.[1] In the middle of the 1980s, Ruren Xu and coworkers developed the solvothermal synthesis route for microporous AlPO4s and GaPO4s (reviewed in Morris and Weigel[2]). In their studies, many alcohols and amines with different structures and properties were used as solvents and/or templates. Hydrothermal and solvothermal synthesis reactions have become the basis and core of the synthetic chemistry of microporous crystals, and have been widely used in the preparation and modification of porous materials. The basic concepts of hydro(solvo)thermal chemistry will be introduced in the first section of this chapter.

3.1 Introduction to Hydro(solvo)thermal Synthesis 3.1.1

Features of Hydro(solvo)thermal Synthetic Reactions

Hydro(solvo)thermal synthesis refers to the synthetic reactions conducted at appropriate temperature (100
1000  C) and pressure (1
100 MPa) in aqueous or organic solvents

Chemistry of Zeolites and Related Porous Material – Synthesis and Structure Ruren Xu, Wenqin Pang, Jihong Yu, Qisheng Huo and Jiesheng Chen # 2007 John Wiley & Sons, (Asia) Pte Ltd

118

Chemistry of Zeolites and Related Porous Materials

within a specially sealed container or high-pressure autoclave under subcritical or supercritical conditions. The studies on hydro(solvo)thermal synthesis have mainly focused on the reactivity of raw materials, regularities of synthetic reactions and conditions, and their relationship with the structures and properties of products. Because of the nonideal and nonequilibrium features of the hydrothermal and solvothermal systems, non equilibrium thermodynamics have been applied to obtain the detailed information during the reaction.[3] It is found that water or other solvents can be activated under high temperature and pressure conditions due to the subcritical or supercritical environment. Under hydrothermal and solvothermal conditions, the physical and chemical properties of reactants can be significantly changed. Hydrothermal syntheses and related hydrothermal reactions have become increasingly important routes for the preparation of most inorganic functional materials such as microporous and porous materials, inorganic compounds with special compositions, structures, and condensed states and special morphology such as nano and ultra-fine powders, sol–gels, noncrystalline states, inorganic membranes, and single crystals. Another feature of hydro(solvo)thermal synthesis is the operability and tunability of hydrothermal and solvothermal chemistry, which bridges the synthetic chemistry and physical properties of synthesized materials. With deepening studies on hydrothermal and solvothermal synthesis chemistry, more and more reaction types have been discovered. Compared with other synthesis and preparation techniques, hydro(solvo)thermal synthesis methodology and techniques have irreplaceable advantages. So far, a variety of materials and crystals used in many fields could be hydrothermally or solvothermally synthesized, and the quality and properties of the resulting products are often much better than those prepared by other methods. The main difference between the solid-state reaction synthesis route and hydro(solvo)thermal synthesis route lies in ‘reactivity’, which is reflected in reaction mechanisms. Reactions in solid-state synthesis depend on the diffusion of the reactants at the interface, whereas individual reactant molecules existing in the liquid phase can react with each other in hydro(solvo)thermal synthesis. Variation in the reaction mechanism leads to the formation of different structures from the same or similar starting materials. In addition, even the same material that can be obtained by both preparation routes can have totally different morphology and properties due to different formation mechanisms. For instance, perfect single crystals can usually be formed from liquid-phase synthesis, while being very difficult to obtain in solid-state synthesis. Hydro(solvo)thermal synthesis chemistry focuses on the chemistry in preparation, synthesis, and assembly of special compounds or materials under hydro(solvo)thermal conditions. More importantly, hydrothermal or solvothermal synthesis routes can be used to prepare materials with special structures and properties, or phases, types, and morphologies which cannot be obtained by using solid-state reactions. In some cases, the materials can be obtained under mild conditions by using hydrothermal and solvothermal synthesis instead of under critical conditions by using a solid-phase reaction synthesis route. Features of hydrothermal and solvothermal synthesis chemistry: (1) To conduct synthetic reactions which cannot be carried out in solid-phase synthesis; (2) To prepare new compounds or phases with special valence states, metastable structures, or aggregation states;

Synthetic Chemistry of Microporous Compounds (I)-Fundamentals and Synthetic Routes

119

(3) To crystallize materials with a low melting point, high vapor pressure, or low thermal stability, which cannot be obtained by using traditional solid-phase synthesis; (4) To grow perfect and large single crystals; (5) To control the morphology and particle size of the products; (6) To prepare compounds with special low-valence states, intermediate states, or special valence states. (7) To uniformly dope heteroatoms to the crystalline structures.

3.1.2

Basic Types of Hydro(solvo)thermal Reactions

The following reactions could be conducted under hydro(solvo)thermal conditions: (1) Synthetic reactions. Polycrystals or single-crystal materials could be obtained by the chemical reaction of multicomponent reactants or via their intermediates under hydro(solvo)thermal synthesis. (2) Crystallization reactions. Sol, gel, and other amorphous species could be crystallized under hydro(solvo)thermal conditions (e.g., the crystallization of zeolites and microporous crystals). (3) Hydrolysation reactions. Compounds like alkoxides could hydrolyse under hydro(solvo)thermal conditions. (4) Growth of a large single crystal in the presence of seed crystals from the mother liquid. (5) Heat treatment. Crystals could be hydrothermally or solvothermally treated to change or modify their properties. (6) Phase transition. The difference in the thermodynamic or dynamic stability of chemical species could lead to phase transition under hydrothermal or solvothermal conditions. Examples include numerous phase transformations among metastable microporous crystals. (7) Ion-exchange reaction. For instance, properties of zeolites are modified at elevated temperature and pressure. (8) Condensation reactions. (9) Extraction reactions. Valuable components such as metals could be extracted from compounds or minerals under hydrothermal or solvothermal conditions. For example, potassium and tungsten components could be extracted from mother liquids of potassium minerals and tungsten minerals (Scheelite, CaWO4), respectively, under hydrothermal conditions. (10) Redox reactions. With the help of special redox agents, new compounds, complexes, or metallorganic compounds with special valence states could be prepared under hydro(solvo)thermal conditions. Complete oxidation of organic compounds could be conducted under supercritical conditions as well. (11) Precipitation reactions. (12) Agglomeration reactions. Agglomeration reactions among multicomponents to prepare special composites could be conducted under hydrothermal or solvothermal conditions. Examples include the preparation of multi-oxide composite materials and ceramic materials containing volatile OH, F, and S2 components. (13) Hydrothermocompression reactions. Formation and solidification of materials or composites such as in the treatment of radioactive waste, solidification of special

120

Chemistry of Zeolites and Related Porous Materials

materials, and the preparation of special materials could be conducted under hydrothermal and solvothermal conditions. According to the reaction temperature, hydrothermal and solvothermal synthesis can be classified into subcritical and supercritical synthesis reactions. In subcritical synthesis, the temperature is in the range of 100 to 240  C, while in supercritical synthesis, the temperature could reach 1000  C and the pressure could reach 0.3 GPa. By using the special properties of solvent water and other reactants under supercritical high temperature and pressure, various syntheses with specific features could be conducted, resulting in the formation of numerous crystal materials with simple to very complex structures. In addition, it should be pointed out that some crystal materials cannot be obtained by using other preparation approaches except for using hydrothermal or solvothermal synthesis routes. So far, hydrothermal and solvothermal synthesis have been widely used in the 1) modification, 2) crystal growth and morphology control, 3) phase-transition study, and 4) discovery of new species of zeolites and porous materials. 3.1.3

Properties of Reaction Media

Water 1. Changes of the properties of water under high temperature and pressure conditions Compared with normal conditions, the properties of water under high temperature and pressure hydrothermal conditions will be significantly changed. For example, the vapor pressure and ion product will be higher, and the density, surface tension, and viscosity of water will be lower. 2. Features of the hydrothermal system High temperature and pressure hydrothermal conditions can accelerate the reaction rate among the complex ions, intensify the hydrolysation reaction, and significantly change the redox potential of the reactants. Normally, there are two types of basic chemical reactions. One is ionic reaction, including metathetical reaction of inorganic salts which could be instantly finished at ambient temperature; the other is free radical reaction, including explosive reaction of organic compounds. Other chemical reactions may possess some of the above reaction properties. According to electron theory, the reactions of organic compounds with polar bonds usually have some characteristics of ionic reactions. Therefore, when the medium is water and the system is heated above its boiling point in a sealed container, the ionic reaction rate will be certainly accelerated, which is consistent with the Arrhenius equation: d ln k=dT ¼ E=RT 2 , i.e., the reaction rate constant k will exponentially increase with the increase of temperature. Therefore, the hydrothermal (high temperature and pressure) conditions can promote the ionic or hydrolysis reaction for those indissoluble minerals at ambient temperature or organic compounds because of the increased ionization constant of water caused by the increased temperature. The change in redox potential of compounds under hydrothermal conditions was reviewed by A. Rabenau,[4] and readers can get more information about this topic from this literature source. Quantitation of the Properties of Organic Solvents Solvothermal synthesis can be conducted in various organic solvents that have different properties, which offer researchers many chances to obtain new structures. So far, several

Synthetic Chemistry of Microporous Compounds (I)-Fundamentals and Synthetic Routes

121

dozens of alcohols have been used in solvothermal synthesis. The first consideration in choosing an alcohol as solvent is the role it will play in the synthesis. The reason is that the solvent not only supplies a medium but also dissolves or partially dissolves the reactants to form a solvent–reactant complex, which will affect the chemical reaction rate. The solvent can also affect the concentration and state of the active species of the reactants, which could finally change the reaction process. The classification of organic solvent is usually based on the macroscopic and microscopic molecular parameters and empirical solvent polar parameters of compounds such as molecular weight (Mr), density (d), melting point (mp), boiling point (bp), molecular volume, heat of evaporation, dielectric constant (e), dipole moment (m), and solvent polarity (ENT ). Among these parameters, solvent polarity is the main one used to describe the solvation properties of a solvent, defined as the sum of the interactions of solvent and solute including Coulombic force, induction force, dispersion force, H-bond, and charge-transport force. Table 3.1 lists the solvent polarity for commonly used solvents in solvothermal synthesis. Other parameters related to the solvent are included in Table 3.1 as well. The choice and application of the solvent will be discussed in detail in the following sections.

Table 3.1 Parameters for the solvents used in solvothermal synthesis (units: omitted) Solvent Tetradecanol 2-Methyl-2-hexanol 2-Methyl-2-butanol 2-Methyl-2-propanol 2-Pentanol Cyclohexanol 2-Butanol 2-Propanol 1-Heptanol 2-Methyl-1-propanol 1-Hexanol 3-Methyl-1-butanol 1-Pentanol 1-Butanol Benzyl alcohol 1-Propanol Ethanol Tetraethylene glycol 1,3-Butanediol Triethylene glycol 1,4-Butanediol Diethylene glycol 1,2-Propanediol 1,3-Propanediol Methanol 3,30 -Oxybis-1,2-propanediol Ethylene glycol Glycerol Water

Mr 214.39 116.20 88.15 74.12 88.15 100.16 74.12 60.10 116.20 74.12 102.18 88.15 88.15 74.12 108.14 60.10 46.07 194.23 90.12 150.18 90.12 106.12 76.10 76.10 32.04 166.18 62.07 92.09 18.01

d

mp

0.823 39 0.8119 0.805 12 0.786 25 0.809 0.963 21 0.807 115 0.785 90 0.822 36 0.802 10 0.814 52 0.809 11 0.811 78 0.810 90 1.045 15 0.804 127 0.785 130 1.125 6 1.004 50 1.123 7 1.017 16 1.118 10 1.036 60 1.053 27 0.791 98 1.300 1.109 11 1.261 20 1.000 0

bp 289 139.4 102 83 120 160 98 82 176 108 157 130 137 118 205 97 78 314 207 287 230 245 187 214 65 199 180 100

e

m

T EN

7.0

1.70

0.321 0.389

13.8 15.0 15.8 18.3 12.1 17.7 13.3 14.7 13.9 17.1 13.1 20.1 24.3

1.66 1.90

23.7 31.1

5.58 2.40

32.0 35.0 32.6

2.25 2.50 1.70

37.7 42.5 80.4

2.28

1.66 1.64 1.82 1.80 1.66 1.70 1.66 1.69

1.94

0.500 0.506 0.546 0.549 0.552 0.559 0.565 0.568 0.602 0.608 0.602 0.654 0.664 0.682 0.704 0.704 0.713 0.722 0.747 0.762 0.790 0.812 1.000

122

Chemistry of Zeolites and Related Porous Materials

Figure 3.1 Commonly used autoclaves in the laboratory for the synthesis of microporous compounds

3.1.4

Hydro(solvo)thermal Synthesis Techniques

Hydrothermal and solvothermal synthesis involves a reaction container (e.g., autoclave), reaction-control system, characterization techniques (in situ), hydrothermal and solvothermal synthesis process, and intermediate species. Because of space limitations only commonly used autoclaves will be introduced here. Detailed information can be found in the literature.[5] Figures 3.1 shows 4 types of commonly used autoclaves in the laboratory for the hydro(solvo)thermal synthesis of microporous compounds: (1) Teflon-lined stainless steel autoclave (up to 1000 mL in volume); (2) 25 mL stainless steel high-pressure autoclave; (3) carbon or glass fiber enhanced Arlon PEEK highpressure autoclave; (4) quartz-lined high-pressure autoclave (minitype high-pressure autoclave made by Parr Inc. The quartz lining could be replaced by a Teflon lining, Figure 3.2).

Figure 3.2 Autoclave made by Parr Inc.[6]

Synthetic Chemistry of Microporous Compounds (I)-Fundamentals and Synthetic Routes

3.1.5

123

Survey of the Applications of Hydro(solvo)thermal Synthetic Routes in the Synthesis of Microporous Crystals and the Preparation of Porous Materials

Hydro(solvo)thermal synthesis has been widely used in the following areas: (1) Synthesis and preparation of microporous, mesoporous, and macroporous compounds and materials; (2) Ion-exchange, framework modification, and secondary synthesis of microporous and porous materials; (3) Assembly of host–guest materials and preparation and modification of porous material composites; (4) Preparation of microporous and porous materials with special states of aggregation such as nano and ultra-fine particles, large single crystals, perfect crystals, molecular sieve membranes, and materials with various morphologies; (5) Preparation of porous materials with special defects and heteroatoms; (6) Preparation of composite materials and solidification of special materials.

3.2 Synthetic Approaches and Basic Synthetic Laws for Microporous Compounds The synthesis of zeolites can be traced back to the middle of the 19th century. The earliest synthesis of zeolites was performed in part by imitating the natural conditions of the formation of natural zeolites, i.e., high temperature and pressure (greater than 200  C and 10 Mpa); however, the efforts were not very successful. The real success for the synthesis of zeolite molecular sieves happened in the 1940s when R.M. Barrer and J. Sameshima started to study the synthesis of zeolites. Later, chemists at Union Carbide Corporation (UCC), R.M. Milton and D.W. Breck, employed mild hydrothermal synthesis (at about 100  C and under self-generated pressure) and achieved a great success in developing a synthetic approach for the synthesis of zeolite. By using hydrothermal synthesis approach, they successfully synthesized unnatural zeolites of types A and X, followed by zeolite Y. Another milestone for the synthesis of zeolites was the introduction of organic quaternary ammonium salt cations to the synthetic system by R.M. Barrer and P.J. Denny in 1961, which allowed the synthesis of high-Si/Al-ratio zeolites, and even pure silica zeolite molecular sieves. After that, a large number of new zeolites and microporous compounds have been successfully synthesized from a solvothermal synthetic system. Another great improvement in the synthesis of microporous materials is the successful synthesis of the microporous aluminophosphate molecular sieve family (including AlPOn, SAPO-n, MeAPO-n, and MeAPSO-n) by S.T. Wilson and E.M. Flanigen in 1986.[7] Remarkable examples for microporous aluminophosphates include AlPO-5 (AFI), AlPO11 (AEL), MeAPO-5 (AFI), MeAlPO-11 (AEL), SAPO-34 (CHA), and SAPO-37 (FAU). Owing to the high moldability of the aluminophosphate framework, it is not very difficult to introduce various elements into the framework of aluminophosphates, which would significantly change their physical chemistry and catalytic properties. Moreover, the introduction of two or more metal elements into the framework of aluminophosphates is

124

Chemistry of Zeolites and Related Porous Materials

also possible. Distinct from the strong basic conditions necessary for the synthesis of traditional aluminosilicate zeolite molecular sieves, microporous aluminophosphates were synthesized under slightly acidic or nearly neutral conditions. The other distinct feature of the synthesis of microporous aluminophosphates is the wide use of various amines as a template or structure-directing agent (SDA) in their hydro(solvo)thermal synthesis, which leads to the formation of a large number of microporous aluminophosphates of new types and structures. 3.2.1

Hydrothermal Synthesis Approach to Zeolites

So far, the hydrothermal synthetic approach is the best way to synthesize a large number of zeolites and microporous materials. Hydrothermal synthetic conditions can enhance the effective solvation ability of water, increase the solubility of the reactants, and activate the reactivity of the source materials, leading to the rearrangement and dissolution of the primary gel formed in the first stage and resulting in an increased nucleation and crystallization rate. Basically, the hydrothermal synthesis process of a zeolite consists of two stages: the initial formation of the hydrated aluminosilicate gel and the following crystallization process of the gel. In fact, the crystallization process of the hydrated aluminosilicate gel is very complicated. No decisive conclusions have been reached for this complicated crystallization process so far. However, regardless of the liquid- or solid-phase transformation mechanism proposed before, it is commonly accepted that the crystallization process consists of four steps: (1) condensation of polysilicate and aluminate anions; (2) nucleation of zeolites; (3) growth of nuclei; and (4) crystal growth of zeolites which sometimes results in secondary nucleation. It is still very difficult to achive a deep understanding of the formation mechanism and the detailed crystallization process of zeolites because: a) the whole crystallization process involves very complicated chemical reactions; b) the nucleation and growth of crystals are performed under heterogeneous conditions; and c) the whole process keeps changing with time. Another difficulty is the lack of effective in situ measurement tools for the structure of both gel and solution. The crystallization mechanism and the chemistry involved in the formation of zeolites will be discussed in Chapter 6 in detail. Some synthetic regularities and features of the hydrothermal synthesis of zeolites could be illustrated by the synthesis of zeolites in presence of sodium ions from Na2O–Al2O3– SiO2–H2O hydrothermal system which will be discussed in the following sections. The typical process for the synthesis of zeolites containing sodium ions includes the mixing of source materials such as sodium silicate (Na2OxSiO2) and sodium aluminate [NaAl(OH)4] in a strongly basic medium with stirring to form a homogeneous gel, aging of the resulting gel under certain conditions, crystallization in a sealed autoclave at an elevated temperature for certain time, recovery of the resulting zeolites crystals with washing, drying, and calcination to form the final molecular sieve product, which could be schematically represented by Equation (3.1): T1

Na2 O  SiO2 ðaqÞ þ NaAlðOHÞ4 ðaqÞ þ NaOHðaqÞ!hydrated aluminosilicate T2

gel!sodium-type zeolite molecular sieve where T1 is the aging temperature, and T2 is the crystallization temperature.

ð3:1Þ

Synthetic Chemistry of Microporous Compounds (I)-Fundamentals and Synthetic Routes

125

The reaction represented by Equation (3.1) looks like a very simple synthetic reaction at first glance. In fact, it is a very complicated process. General synthetic regularities of zeolites will be discussed in the following sections by using the crystallization of sodiumtype zeolites as an example. So far, the lack of a comprehensive understanding of the formation mechanism of zeolites is still a main barrier to any in-depth study of the hydrothermal synthesis of zeolites. The kind of zeolites finally crystallized from the parent gel depends on many factors in the hydrothermal synthesis process and it is not yet clear how these factors work. However, years of studies by chemists working in this field indicate that factors such as composition of the reactants, type and properties of the reactants, aging conditions, crystallization temperature and time, pH of the gel, inorganic or organic cations present in the synthetic system, and reaction container, play very important roles in the hydrothermal synthesis of zeolites. Sometimes one factor can affect other factors. Thus, it is difficult to study the influence of one single factor on the whole synthetic reaction. However, some general synthetic regularities that will be described in the following sections still could be obtained based on the results of plenty of experiments, even though exceptions could happen due to the complicated nature of the hydrothermal synthetic system. Reactants and Batch Composition Basic reactants used in the synthesis of zeolites include silicon source, aluminum source, metal ions, base, mineralizer, and water. Some additives such as organic template or inorganic salts could be critical for the successful crystallization of a specific zeolite. Among them, silicon source and aluminum source are the two most important reactants. Frequently used silicon source and aluminum source reactants are listed below: Silicon source: Water glass: (Na2OxSiO2), where x is modulus; sodium silicate: Na2SiO39 H2O; silica gel: Ludox-AS-40 colloidal sol: SiO2 40 wt%, NH4+ (counter ion); Ludox-HS-40 colloidal sol: SiO2 40 wt%, Na+ (counter ion); Fumed silica: Aerosil-200, Cab-O-Sil M-5; ethyl orthosilicate (TEOS): Si(OC2H5)4; methyl orthosilicate (TMOS): Si(OCH3)4. Aluminum source: Sodium aluminate: NaAlO2; Boehmite (pseudo-boehmite): AlOOH, Al2O3 70%, H2O 30%; aluminum hydroxide (Gibbsite), Al(OH)3; aluminum isopropoxide: Al(O-iC3H7)3; aluminum nitrate: Al(NO3)39 H2O; metallic aluminum. Clay can be used as both silicon and aluminum source with or without pre-treatment. The type of zeolite that can be crystallized from a synthetic system depends on the reactant composition (i.e., batch composition) of the parent gel, which is called the crystallization field. For example, Figure 3.3 shows a set of typical crystallization fields.[8] Results in Figure 3.3(a), (c), and (e) indicate that different batch compositions can lead to the formation of different products even under the same crystallization conditions. By varying the batch compositions, four synthetic zeolites [i.e., X, Y (FAU), B (ANA), and A] were successfully synthesized from the Na2O–Al2O3–SiO2–H2O system [Figure 3.3(a)] and two synthetic zeolites (i.e., W and H) were successfully crystallized from the K2O–Al2O3–SiO2–H2O system [Figure 3.3(e)]. Therefore, it can be concluded that a specific crystallization field can result in the formation of a specific

126

Chemistry of Zeolites and Related Porous Materials

zeolite. Further investigation of the effect of the amount of aluminum and base in the parent mixture on the crystallization of zeolites is shown in Figure 3.4. The variation of the ratio of SiO2/Al2O3 or OH/SiO2 leads to the formation of ZSM-39, ZSM-5, and ZSM-35 molecular sieves.[9] Moreover, it should be further pointed out that the amount of water in the parent mixture has a significant effect on the crystallization of zeolites as well [Figure 3.3(a) and (b), Figure 3.3(c) and (d)].

Figure 3.3 Crystallization field. (a) Na2O–Al2O3–SiO2–H2O, 100  C, H2O content of gels is 90–98 mole%; silicon source: sodium silicate; (b) Same as (a) with 60–85 mole% H2O in the gel; (c)Same as (a), silicon source: colloidal silica; (d) Effect of water content in gel on synthesis of zeolites Y and S, silicon source: colloidal silica; (e) K2O–Al2O3–SiO2–H2O, 100  C, H2O content of gels is 95–98 mole%; (f) K2O–Al2O3–SiO2–H2O, 100  C, H2O content of gels is 80–92 mole%. Reproduced with permission from [8]. Copyright (1974) John Wiley & Sons, Inc.

Synthetic Chemistry of Microporous Compounds (I)-Fundamentals and Synthetic Routes

127

Figure 3.4 Effect of the amount of base and aluminum on the crystallization of zeolites (H2O/ SiO2 ¼ 40, template/SiO2 ¼ 0.68, 150  C, 40 h). Reproduced with permission from [9]. Copyright (1986) Elsevier

For the K2O–Al2O3–SiO2–H2O (100  C) synthetic system, water content in the parent mixture has a significant effect on the crystallization field as well [Figure 3.3(e) and (f)]. When the water content in the parent mixture is between 95–98 mole% [Figure 3.3(e)], there are only two potassium-type zeolites (W and H) formed, whereas when the water content is decreased to 80–90 mole% [Figure 3.3(f)], three new phases, i.e., zeolites K, L, and M, appear in this synthetic system. This phenomenon is rare in other synthetic systems. Further studies reveal another unique feature of zeolite synthesis, i.e., the types and properties of the reactants can affect the crystallization field of zeolites. For example, the points marked with ‘þ’ in Figure 3.3(a), (b), and (c) show typical composition of zeolite phase A (LTA), X (low-silica FAU), Y (high-silica FAU), B (analcime ANA), HS (SOD) and S (GME), and R (CHA). The results in Figure 3.3(a) and (c) indicate that the silicon source (sodium silicate and colloidal silica) has significant influence on the crystallization of the final product. The variation of the silicon source from sodium silicate to colloidal silica results in a modified crystallization field of zeolites X, Y, and A as well as the crystallized phases. When sodium silicate was used as the silicon source, zeolites HS and B were formed, whereas zeolites R and S crystallized when colloidal silica was used as the silicon source instead of sodium silicate. This phenomenon is quite common in the synthesis of zeolites and microporous aluminophosphates. Owing to the complexity of the hydro(solvo)thermal synthesis system, it is believed that the reasons for this phenomenon are very complicated. One possibility is that the solubility, the diversity of the polysilicate existing states, and their distribution are different from silicon source to silicon source, and play an important role in the dynamics of nucleation and crystallization. Even for the same silicon source, the effect of property variation can affect the crystallization process of zeolites as well. For example, it is found that the silicon source of fumed silica with different surface areas can affect the crystallization

128

Chemistry of Zeolites and Related Porous Materials

Figure 3.5 Effect of fumed SiO2 with different surface areas on the crystallization rate and average particle size of zeolite A. Reproduced with permission from [13]. Copyright (1973) American Chemical Society

rate, the crystal size, and their particle-size distribution of zeolite A (Figure 3.5). The fumed silica with high surface area is easier to dissolve in base than that with low surface area. The former leads to a quicker nucleation and a higher supersaturation in basic solution, which is favorable for the formation of small crystals, while a silicon source with low solubility is favorable for the formation of large crystals. Si/Al Ratio The Si/Al ratio in the parent mixture plays an important role in determining the structure and composition of the final product. However, there is no quantitative correlation between the Si/Al ratio in the product and that in the batch composition. Roughly, zeolites with low Si/Al ratios such as zeolite A (LTA), hydroxysodalite (SOD), KH, and KJ can be crystallized from a parent mixture (precursor gel) with low Si/Al ratio and strong alkalinity, whereas high-silica zeolites (high Si/Al ratio) such as mordenite (MOR, Si/Al ¼ 5–9) and beta (BEA, Si/Al > 8) can be crystallized from a gel with high Si/Al ratio and weak alkalinity. Usually, the Si/Al ratio in the precursor gel is always higher than that in the crystallized product. The excess of silicon is left in the mother liquid. For some zeolite structures, the Si/Al ratio in the crystallized product can be tuned without the loss of structure. For instance, some, but not all, zeolites can have both high- and lowsilica forms. Furthermore, some zeolites such as low-silica FAU (X, Si/Al ¼ 1–1.5), highsilica FAU (Y, Si/Al ¼ 1.5–3), GME (Si/Al ¼ 2.3–2.95), zeolite Pt (Si/Al ¼ 1.6–2.65), ANA (Si/Al ¼ 1.4–4.1), and MOR (Si/Al ¼ 4.5–9.75) can only be crystallized within a narrow Si/Al ratio from a specified batch composition with a narrow Si/Al ratio. Those zeolites whose Si/Al ratio seriously deviates from the normal values, such as FAU-type zeolites with an Si/Al > 3, high-aluminum ZSM-5, ZSM-11, and BEA, and other highsilica natrolites, are very difficult to synthesize. They cannot be synthesized by simply varying the Si/Al ratio of the parent mixture. Special synthetic conditions, such as a special SDA or secondary synthesis approach, are needed for their preparation. For example, X is a low-silica faujasite zeolite with a typical Si/Al of 1–1.5, while Y is a high-silica faujasite zeolite with a typical Si/Al of 1.5–3. X-type zeolite with an Si/Al

Synthetic Chemistry of Microporous Compounds (I)-Fundamentals and Synthetic Routes

129

Figure 3.6 Correlation between the Si/Al ratio in faujasite-type zeolite and the reaction rate constant k (crystallization temperature: 88  C).[10,11] Reproduced with permission from [10]. Copyright (1975) American Chemical Society

ratio of 1–1.5 and Y-type zeolite with an Si/Al ratio of 1.5–3 can easily be synthesized from the Na2O–SiO2–Al2O3–H2O system by varying the batch composition. However, little success has been reported for the synthesis of Y zeolite with Si/Al > 3 from this synthetic system. H. Lechert has carefully studied this case and concluded that the crystallization process of high-silica Y zeolite was controlled by reaction dynamics.[10,11] He further pointed out that the activation energy of the polymerization of polysilicate and aluminate anions in the solution is very high, resulting in a low reaction rate constant k. Thus, the structure of high-silica Y zeolite is very difficult to crystallize under normal conditions. Figure 3.6 shows the Si/Al ratio of high-silica Y zeolite as a function of its reaction rate constant k. The correlation between Si/Al ratio in the parent gel and that in the crystallized product is very complicated. Besides the above discussed dynamics factor, thermodynamics and crystal structure of the resulting zeolites can also affect the Si/Al ratio in the final product. Alkalinity Zeolite synthesis is usually performed under basic or strongly basic conditions. Many zeolites can be crystallized from the basic Na2O–Al2O3–SiO2–H2O system. For this specific system, the alkalinity is defined as the OH/Si ratio or the concentration of base (H2O/Na2O). Basically, increasing the ratio of OH/Si leads to a higher solubility of silicon and aluminum sources, which will alter the polymerization state and their distribution. Moreover, a higher alkalinity can decrease the polymerization degree of the silicate anions and speed up the polymerization of the polysilicate and aluminate anions. Thus, increasing alkalinity will shorten the induction period and nucleation time and speed up the crystallization of zeolites. For instance, increasing the pH of the parent mixture (from 10.2 to 12.85) decreases the induction period and accelerates the crystallization of mordenite (Figure 3.7).[12]

130

Chemistry of Zeolites and Related Porous Materials

Figure 3.7 Effect of pH on the crystallization rate for the synthesis of mordenite at 300  C. Reproduced with permission from [3]. Copyright (1982) Academic Press, New York

Another effect of alkalinity on the crystallization of zeolites is that increasing the alkalinity of the system favors the formation of aluminum-rich zeolites. The most remarkable example is faujasite zeolite. High-silica Y zeolite can only be synthesized from the precursor gel with a low alkalinity (OH/Si), whereas low-silica X zeolite can be crystallized from that with high alkalinity. The typical batch composition for the synthesis of Y zeolite is 8 Na2OAl2O320 SiO2320 H2O with an OH/Si ¼ 0.8. The composition of the resulting zeolite Y is Na2OAl2O3(3.6–6.0) SiO29 H2O. When the alkalinity of the precursor gel increases to OH/Si ¼ 2.4 and the batch composition reaches 3.6 Na2OAl2O33 SiO2144 H2O, aluminum-rich X zeolite crystallizes with a composition of Na2OAl2O3(2–3) SiO26 H2O. Alkalinity also has an important influence on the crystallization rate of zeolites. A remarkable example is the crystallization of zeolite A (LTA) from the precursor gel with a batch composition of 5 Na2OAl2O32 SiO2(100–200) H2O.[13] Figure 3.8 shows the effect of different alkalinity (H2O/Na2O¼20, 30, 40) on the crystallization rate (including induction period and growth rate) and particle size of the product. Clearly, with an increase of alkalinity the crystallization process is speeded up, the particle size is decreased, and the distribution of the particle size is narrowed due to an increased nucleation rate and an increased polymerization rate between polysilicate and aluminate anions. Aging The period between the formation of a homogeneous gel and the start of crystallization is named ‘aging’. The aluminosilicate gel and liquid phase within the gel are formed at this stage. The composition, structure, and properties of the aluminosilicate gel were studied in detail between the mid-1960s and late 1970s. It was found that the composition and structure of aluminosilicate gels were metastable and changed over time, and the primary and secondary gel could be formed and aged in this period as well. Nucleation happening at this stage was significantly affected by the composition and structure of the gel. Thus, the purpose of aging is to adjust synthetic conditions such as temperature and time to assist the transformation of gel to the zeolite structure and to speed up nucleation. This

Synthetic Chemistry of Microporous Compounds (I)-Fundamentals and Synthetic Routes

131

Figure 3.8 (a) Effect of alkalinity on the crystallization of Zeolite A [5 Na2OAl2O32 SiO2 (100–200)H2O, 70  C]; (b) Effect of alkalinity on the particle-size distribution of Zeolite A[5 Na2OAl2O32 SiO2(100–200)H2O, 70  C]. Reproduced with permission from [13]. Copyright (1973) American Chemical Society

could be illustrated by the study of C.L. Angell[14] on the pre-crystallization stage of zeolite A (LTA). 1. The formation of primary aluminosilicate gel The batch composition for the synthesis of zeolite A was 1.98 Na2OAl2O3 1.96 SiO233 H2O. The primary aluminosilicate gel formed immediately when the reactants were mixed together. In the first hour, the system was stable and the composition of the gel and solid content stayed almost constant as shown in Figures 3.9–3.11. 2. The formation of secondary aluminosilicate gel The changes in the composition of the solid and liquid components (Figures 3.9 and 3.11), solid component percentage (Figure 3.10), and average particle size (Figure 3.12) strongly suggested that the composition and structure of the primary aluminosilicate gel

132

Chemistry of Zeolites and Related Porous Materials

Figure 3.9 Variation of the solid and liquid component in the crystallization of zeolite A. *: SiO2/Al2O3 ratio in solid component; : SiO2/Al2O3 ratio in liquid component; ~: Na2O(liquid component)/Na2O(solid component) ; : Al2O3 (liquid component)/Al2O3 (solid component)

changed when the temperature reached 96  C. Almost all of the Al(OH) 4 existing in the liquid component polymerized with silicate anions to form a structurally more compacted secondary gel. At this stage, the percentage of solid component increased, the average particle size decreased, and some other structurally different polysilicates entered the liquid component. Although the primary gel changed in both composition and structure

Figure 3.10

Variation of percentage of solid component in the crystallization of zeolite A

Synthetic Chemistry of Microporous Compounds (I)-Fundamentals and Synthetic Routes

133

Figure 3.11 Variation of solid component in the crystallization of zeolite A. j: parent mixture; k: after mixing; l: after aging 1 hour; m: crystallization starting; n: after crystallization 4 hours

Figure 3.12

Variation of the average particle size in the formation of zeolite A

134

Chemistry of Zeolites and Related Porous Materials

Figure 3.13 Phase diagram of the Na2O–Al2O3–SiO2–H2O system. Composition of the parent mixture; * Composition of water-washed dry aluminosilicate gel;  Composition of the liquid phase in the parent mixture Reproduced with permission from [8]. Copyright (1974) John Wiley & Sons, Inc.

due to heating, it is unlikely that there existed a direct process between the solid components of the gel and the crystallized zeolites. The composition change of the liquid component revealed that the transformation of primary gel to secondary gel was achieved via either an overall or a local re-dissolution and gelation process. S.P. Zhdanov studied the phase diagram of aluminosilicate gel formed from the Na2O– Al2O3–SiO2–H2O system and reached some conclusions about gel formation and transformation of the gel in the early stages of the aging period.[15] These regularities are very useful in helping us to understand the effect of aging. This topic will be more deeply discussed in Section 5.2 of Chapter 5, under the heading of ‘Crystallization Process and Formation Mechanism of Zeolites’. Zhdanov further investigated the sodium aluminosilicate gel and liquid phase within the gel formed from a four-component system of Na2O–Al2O3–SiO2–H2O with a water content of 85 mole%.[15] Figure 3.13, a projection of this four-component system on a Na2O–Al2O3–SiO2 triangle, shows the crystallization field of the gel and the correlation between the liquid phase and the aluminosilicate gel solid. The content of Na2O, SiO2, and Al2O3 shown in Figure 3.13 is expressed as the mole percentage in the anhydrous gel. At ambient temperature, the time needed for the formation of a stable gel strongly depends on the parent mixture. When the Si/Al ratio is close to 1, the gel can reach equilibrium in a short time, whereas the gel needs a longer time to reach equilibrium when the Si/Al ratio is either increased or decreased. The resulting aluminosilicate gel has a typical colloid structure and contains excess of base. Most of the excessive base can

Synthetic Chemistry of Microporous Compounds (I)-Fundamentals and Synthetic Routes

135

be washed out with water, while a small amount of base still remains, and is believed to exist on the surface or in the framework of the gel. Results in Figure 3.13 indicate that the aluminosilicate gel could only be formed within a limited composition (marked as a dashed line). When the Si/Al ratio of the parent mixture is too high or too low, or the content of the base is lower, the gel cannot form. In Figure 3.13, four lines between the vertex of Na2O and the bottom line of SiO2– Al2O3 stand for Si/Al ratios of 1/3, 1, 2, and 5, respectively. The Na2O/Al2O3 ratio of the points on the line through the vertex of SiO2 and the line of Na2O–Al2O3 is always 1. Apparently, the parent mixture could have a very wide range of compositions. However, the resulting gel, after thorough washing with water, has a very narrow composition and is limited about a line with an Na2O/Al2O3 ratio of 1 (Figure 3.13, * points). Zeolite molecular sieves can only be crystallized from parent mixtures whose compositions are limited about the line with an Na2O/Al2O3 ratio of 1. Evidently, the composition of the liquid phase within the gel skeleton depends on that of the corresponding parent mixture. The variation of the concentration of the components in the parent mixtures and the corresponding liquid phases within the gel is listed in Table 3.2. According to the results in Figure 3.13 and Table 3.2, the following conclusions can be reached: (1) The SiO2/Al2O3 ratio of the water-washed dry gel is always greater than 2 and is limited within a narrow range (from 6.6 to 2.2) even though that of the parent mixture varys over a very wide range (from 36.8 to 0.333). (2) The SiO2/Al2O3 ratio of the resulting zeolites is always close to 1 no matter what the composition of the parent mixture is. (3) The Na2O/Al2O3/SiO2 molar properties of the liquid phase within the parent mixture are always beyond those of the field of the water-washed dry gel. Table 3.2 Parent mixture and liquid phase within the gel Concentration of the components in the liquid phase within the gel (mol/L)

Concentration of the components in parent mixture (mol/L) Gel samples Na2O

Al2O3

SiO2

SiO2/ Al2O3

Na2O

Al2O3

SiO2

V VIII 965 967 463 963 VII 467 III 966 IX 196 660

0.390 0.560 0.720 0.270 0.510 0.620 0.350 0.380 0.094 0.185 0.228 0.278 0.033

0.134 0.184 0.239 0.270 0.510 0.620 0.700 0.760 0.470 0.925 1.140 1.340 1.210

0.333 0.333 0.333 1.00 1.00 1.00 2.00 2.00 5.00 5.00 5.00 5.00 36.8

1.72 1.40 1.20 1.76 1.17 0.77 1.11 0.71 1.90 1.27 1.03 0.72 0.49

0.234 0.400 0.550 0.064 0.217 0.285 0.033 0.032 0.026 0.017 0.015 0.010 0.004

0.030 0.127 0.027 0.067 0.027 0.050 0.061 0.95 0.018 0.08 0.010 0.035 0.062 1.88 0.023 0.71 0.315 12.2 0.530 31.4 0.680 44.4 0.547 54.7 1.030 258.9

2.07 1.74 1.44 2.15 1.50 1.23 1.58 1.16 2.24 1.65 1.37 1.08 0.57

SiO2/ Resulting Al2O3 Zeolite A A A A A A A A X X X X Y

136

Chemistry of Zeolites and Related Porous Materials

Crystallization and Aging Temperature Temperature is a very important parameter in the synthesis of zeolites. The self-generated pressure of water (solvent) which will directly affect the crystallization and structure of zeolites formed from the synthetic system strongly depends on the crystallization temperature. Moreover, the variation of crystallization temperature can affect many other factors during the synthesis, such as the polymerization reaction between the polysilicate and aluminate anions contained in the liquid phase of the gel; the polymeric state of the silicates; the formation, dissolution, and transformation of the gel; the nucleation and crystal growth; and the phase transition of metastable phases resulting in the formation of zeolites with different pore structures from a single synthetic system. The formation of various zeolites in the Na2O–SiO2–Al2O3–H2O system at different temperatures will serve as an example blow. Studies by many zeolite chemists since the 1950s indicate that zeolites A, Pc, X, Y, chabazite (CHA), and gmelinite could be crystallized from the synthetic system Na2O– SiO2–Al2O3–H2O when the crystallization temperature is in the range of 100 to 150  C; zeolite sodalite (SOD) and mordenite (MOR) with a small pore system could be formed when the crystallization temperature reaches 200–300  C; and when the crystallization temperature is higher than 300  C, very-small-pore zeolites such as analcime (ANA) (AC) and natrolite (NAT) and nonporous albite (AB) and nepheline hydrate (NH) are the main products. The crystallization field of this synthetic system at 300  C is shown in Figure 3.14. The pore diameter, pore volume, framework density, and secondary building units (SBU) of the typical microporous zeolites crystallized from this system are listed in Table 3.3 for further study of the correlation between crystallization temperature and the structures of the corresponding microporous zeolitic crystallines. The data in Table 3.3 show that the pore diameter and volume of the zeolites crystallized from this synthetic system dramatically decreased and the framework density

Figure 3.14 Crystallization field for the Na2O–SiO2–Al2O3–H2O system (300  C, 1 day), where AC is zeolite ANA; NH is nepheline hydrate; and S is zeolite SOD

Synthetic Chemistry of Microporous Compounds (I)-Fundamentals and Synthetic Routes

137

Table 3.3 Structural properties and secondary building units (SBU) of the zeolites formed from the synthetic system Na2O–Al2O3–SiO2–H2O at various temperatures T/ C

Zeolites

25

X or YA A A Pc Chabazite Gmelinite X or Y Pc Chabazite Gmelinite HS Albite Sodalite Analcime Mordenite Sodalite Natrolite Albite Nepheline

90–100

120–200

200–300

300–460

Pore diameter/A˚ 8.4 4.1 4.1 4.0 4.3 4.3 8.4 4.0 4.3 4.3 2.6 nonporous 2.6 2.6 4.0 2.6 2.6 nonporous nonporous

Pore volume/(cm3/g)

Framework density/(g/cm3)

SBU

0.48–0.50 0.47 0.47 0.41 0.47 0.44 0.48–0.50 0.41 0.47 0.47 0.35

1.37 1.27 1.27 1.57 1.45 1.46 1.37 1.57 1.45 1/46 1/72

D6R D4R D4R S4R D6R D6R D6R S4R D6R D6R S6R

0.18 0.35 0.28 0.18 0.23

1.85 1.72 1.70 1.88 1.76

S4R S0R 5-1 S4R 4-1

increased with an increase of crystallization temperature. When the temperature is higher than 300  C, only nonporous albite and nepheline hydrate crystallized from this system. On the other hand, the SBU of the zeolitic structure formed from this system became simpler with an increase of crystallization temperature. For example, the structure of the zeolites crystallized from this system and other cation synthetic systems is usually built up from 4- or 6-membered rings with Si or Al vertices when the crystallization temperature is lower than 150  C. The structures with 5-membered rings such as mordenite and ZSM series zeolite molecular sieves are preferred when the temperature is between 150 and 200  C. Thus, we can infer that there exists a tight correlation between the pore-creation of the inorganic compounds (i.e., aluminosilicate) and the crystallization temperature or the water vapor pressure. Low-temperature (e.g., ambient temperature) aging can increase the rate of nucleation from the parent mixture, which is equivalent to a low-temperature reaction. However, the growth rate of a crystal at ambient temperature is very slow and can be neglected. Previous studies indicate that the aging process is necessary for the synthesis of both high-silica zeolites (e.g., TS-1) and low-silica zeolites (e.g., zeolite A and X). Normally, increase of the crystal growth rate caused by an increase of crystallization temperature is much higher than that caused by an increase of the nucleation rate. Thus, big crystals could be obtained at high temperature in a short crystallization time (e.g., NaX, Silicalite-I). Crystallization temperature can affect the morphology of the crystals as well because the activation energy of the crystal faces is related to crystallization temperature. Besides the above discussed basic synthetic regularities for the hydrothermal synthesis of zeolites, attention should be paid to the influence of temperature on the nucleation rate

138

Chemistry of Zeolites and Related Porous Materials

Figure 3.15 Growth-rate curves of the faujasite zeolite NaX at various temperatures. The temperatures on the curves are in  C. Reproduced with permission from [16]. Copyright (1980) John Wiley & Sons, Ltd

in the induction period and the crystal growth rate. This effect can be addressed by two examples shown below: Example 1: The crystal-growth rate of zeolite NaX at various temperatures in the synthetic system 3.12 Na2O–Al2O3–3.5 SiO2–593 H2O (Figure 3.15).[16] A measured linear rate of 0.5
l=
t for the crystal growth rate of zeolite NaX, where l is the crystal size of zeolite NaX, could be obtained from these curves. Results in Figure 3.15 indicate that the growth rates of the zeolites are significantly accelerated with an increase of crystallization temperature and the values of this rate at 70, 80, 90, and 100  C are 0.0175, 0.0375, 0.0625, and 0.1071 mm/h, respectively. Example 2: The effect of temperature on the crystallization of zeolite mordenite from the synthetic system 8.5 Na2O–Al2O3–35 SiO2–182 H2O (Figure 3.16) It is evident from these crystallization curves showed in Example 2 that the crystal growth was significantly speeded up and the aging induction period was obviously shortened with an increase of crystallization temperature (i.e., 200  C ! 250  C ! 300  C ! 320  C ! 350  C). This phenomenon was also observed in both the crystallization of zeolite MOR[12] at higher temperatures and the synthesis of zeolite A (LTA),[17] faujasite (FAU),[10,11,18] sodalite (SOD), phillipsite (PHI), ZK-5 (KFI),  (MAZ), K-F (EDI), and ferrierite (FER). Inorganic Cations So far, the synthesis of zeolites has been extensively studied in order to discover new zeolites with distinct structures. According to the results obtained, it is found that

Synthetic Chemistry of Microporous Compounds (I)-Fundamentals and Synthetic Routes

139

Figure 3.16 Effect of temperature on the crystallization of MOR. Reproduced with permission from [3]. Copyright (1982) Academic Press

inorganic cations play an important role in the crystallization of zeolites. For instance, the zeolites analcime (ANA), cancrinite (CAN), chabazite (CHA), gmelinite (GME), faujasite (FAU), A (LTA), and phillipsite (PHI) could be crystallized from the aluminosilicate crystallization system (Al2O3–SiO2–H2O) in the presence of sodium-containing species, whereas zeolites KE, KF, KZ, KG, KH, KJ, KL, KM, KQ, and KW could be promoted from the same precursor gel system in the presence of potassium-containing species.[8] Moreover, zeolite L (LTL), offretite (OFF), and erionite (ERI) can be crystallized from this synthetic system in the presence of both sodium- and potassiumcontaining species. Aluminosilicate zeolites are normally synthesized under basic conditions. The introduction of OH ions to the synthetic system will necessarily lead to the introduction of correlated cations. These positively charged cations play an important role in the polymerization of polysilicates and aluminates by affecting the polymeric state and their distribution, and have an important effect on the colloidal chemistry of aluminosilicate as well. In addition, cations existing in the synthetic system also have important effects on the formation of the framework structure of zeolites. For example, plenty of synthetic data indicated that a tight correlation between the formation of the SBU cages of zeolites and the charge and size of the cations existed, and this was named the templating effect of cations by R.M. Barrer.[19] The templating effect of cations can be illustrated by the formation of the zeolites sodalite and . Zeolite sodalite has been successfully synthesized in a TMA+ system and the number of TMA+ cations in each SOD cage has been determined; e.g., each SOD cage contains one TMA+ ion. The framework of zeolite sodalite consists of 14-faced SOD (b) cages (truncated octahedra). Each SOD cage is connected with eight others through common 6-rings. The TMA+ ion is accommodated in the SOD cage. However, ˚ ) is prohibited because the entering and exiting of the SOD cage by the TMA+ ion (6.9 A ˚ ). the maximum opening of the SOD cage is a 6-membered-ring window (3.6 A

140

Chemistry of Zeolites and Related Porous Materials

Experimental results indicate that each SOD cage contains one TMA+ ion.[19–21] Evidently, these TMA+ ions were trapped at the beginning of SOD-cage formation via the polymerization of polysilicate and aluminate during the crystallization process. Zeolite  could be synthesized from this TMA+ system as well.[22] The structure of zeolite  consists of 14-faced gmelinite (gmel) cage. Similarly to the SOD cage, the gmelinite cage contains the maximum opening of a 6-membered-ring window with a size ˚ , which is much smaller than that of the TMA+ ion. After complete ion-exchange of 3.6 A with Na+ or K+, zeolite  synthesized from the TMA+ system still has some TMA+ ions left, e.g., zeolite  contains unexchangeable TMA+ ions in its structure. It is believed that these unexchangeable TMA+ ions were trapped in the gmelinite cage as a template during the crystallization process. The structure of many zeolites contains cages with trapped cation templates inside. E.M. Flanigen summarized the zeolite framework structures with an SBU cage and corresponding cation template. The results are shown in Tables 3.4 and 3.5.[23] The data in Tables 3.4 and 3.5 indicate a good fit between the size of the specific cation or its hydrate and the SBUs of the corresponding zeolite structure, suggesting a templating role for the hydrated cations. The introduction of the second type of cation to a four-component system makes a five-component system such as Na2O–K2O–Al2O3–SiO2–H2O. For convenience in real use, researchers always fix the ratio of part of the components to simplify the

Table 3.4 Correlation between the structure of zeolites and their corresponding synthesis cations Secondary building units Zeolite structure type

Double rings

Polyhedra

LTA

D-4

sodal, a cage

Faujasite ZK-5 ZSM-3 Gmelinite 

D-6 D-6 D-6 D-6

sodal cage a cage sodal cage gmel cage gmel cage

Offretite

D-6

Erionite (with offretite)

D-6

gmel cage canc (can) canc cage (gmel cage)

L

D-6

Chabazite

D-6

canc (can)

Synthesis cation Na, Na-TMA, Na-K, Na-Li Same as above Na-DDO, (Ba salt) Na-Li Na, Na-TMA Na-Li-TMA, Na-TMA, Na-K-TMA K-TMA Na-K-TMA Na-K, Ba-TMA, Ra-Rb, Na-TMA, Na-K-TMA Na-Li-TMA Na-K-Ba-TMA K, K-Na, K-DDO K-Na-TMA, Ba, Ba-TMA Na, K, Na-K, Ba-K, Sr, (K-TM) (K-Na-TMA)

Cation specificity for framework structure Na Na Na-DDO Na-Li Na Na-TMA K-TMA Na-K Na-Rb Na-TMA Ba-TMA K or Ba Na, K or Sr

Synthetic Chemistry of Microporous Compounds (I)-Fundamentals and Synthetic Routes

141

Table 3.5 Cation specific building units in zeolite structures Specific cation Diameter/A˚ Building unit D-4 a cage Sodalite cage Gmelinite cage Cancrinite cage D-6

Free dimensions/A˚ 2.3 11.4 6.6 6:0  7:4 3.5–5.0 3.6

Cation Na Na Na, TMA Na, TMA K, Ba, Rb Na, K, Sr, Ba

Anhydrous 2.0 2.0 2.0, 6.9 2.0, 6.9 2.8, 2.7, 3.0 2.0–2.8

Hydrated 7.2 7.2 7.2, 7.3 7.2, 7.3 6.6, 8.1, 6.6 7.2–8.1

five-component system to a three-component system. The phase diagram (crystallization field) of the simplified system is similar to that of the real three-component system. The introduction of the second type of cation will significantly change the phase diagram (crystallization field) of the original system. For example, the presence of the second cation K+ in the four-component synthetic system Na2O–Al2O3–SiO2–H2O will lead to three main changes in their crystallization fields: (1) The crystallization field or the chemical composition of the zeolites could be significantly changed. When the ratio of K2O/R2O (R ¼ K þ Na) is low, the zeolites crystallized from the original sodium aluminosilicate system can still be crystallized but have a changed chemical composition, such as Si/Al ratio. For instance, NaX zeolite with modified chemical composition can be crystallized from a parent mixture with SiO2/Al2O3 ¼ 3–4, H2O/R2O ¼ 30–45, and K2O/R2O < 0.15 (Table 3.6). The change of the chemical composition of the resulting zeolite may be caused by the change of the framework charge since the Kþ ion cannot enter the b cage. When the ratio of K2O/R2O is greater than 0.15, zeolite NaP is formed instead of NaX, as shown in Figure 3.17. The presence of small amounts of potassium cations in the parent gel which originally produces zeolite NaY will lead to the crystallization of zeolites NaP and W as shown in Figure 3.18. This is one of the reasons why the impurity of zeolite NaP always existed when kaolin or pumice was used as source material to prepare NaX or NaY zeolites. (2) New zeolite structure containing mixed cations could be formed. As shown in Figure 3.19, decreasing the ratio of Na2O/R2O or increasing the ratio of K2O/R2O Table 3.6 Chemical compositions of zeolite X crystallized from K–Na gels K2O/R2O (in parent gel) 0.02 0.05 0.10 0.15

K2O/R2O (in zeolite) 0.04 0.09 0.14 0.17

SiO2/Al2O3 (in zeolite) 2.65 2.58 2.44 2.39

142

Chemistry of Zeolites and Related Porous Materials

Figure 3.17 Effect of Kþ ion on the crystallization field change of zeolite NaX (a) and zeolite NaA (b) at 100  C. Reproduced with permission from [8]. Copyright (1974) John Wiley & Sons, Inc.

Figure 3.18 Effect of Kþ on the crystallization-field change of zeolite NaY. Reproduced with permission from [8]. Copyright (1974) John Wiley & Sons, Inc.

Figure 3.19 Effect of Kþ on the crystallization-field change of zeolite NaY (100  C), where SiO2/Al2O3 ¼ 20–28; H2O/R2O ¼ 25–40. Reproduced with permission from [8]. Copyright (1974) John Wiley & Sons, Inc.

Synthetic Chemistry of Microporous Compounds (I)-Fundamentals and Synthetic Routes

143

results in the formation of zeolites NaY, D, and L from the parent mixture with a SiO2/Al2O3 ratio of 20–28 and H2O/R2O ratio of 25–40. Zeolites D (0.5 Na2O0.5 K2OAl2O34.8 SiO26.7 H2O) and T (0.3 Na2O0.7 K2OAl2O36.9 SiO27.2 H2O) that contain both Na and K cations could be formed between the crystallization field of zeolites NaY and KL. Zeolites D, T, and L are middlesilica zeolites and have totally different structures from that of zeolite NaY. In Figure 3.17(b), zeolites NaA and E (0.4 Na2O0.5 K2OAl2O32.0 SiO23.3 H2O) have the same Si/Al ratio and structure. However, when the ratio of Na2O/R2O is decreased, zeolite KF (K2OAl2O32 SiO22.9 H2O, tetragonal system) is obtained instead of zeolite E. Zeolite KF has the same Si/Al ratio as that of zeolite E but a different structure. (3) Owing to the presence of K+, the crystallization field of potassium-type zeolite is less affected by the variation of Na2O/R2O ratio. For instance, as shown in Figure 3.18, zeolite KW will be crystallized instead NaY from the parent mixture with SiO2/Al2O3 ratio of 10, H2O/R2O ratio of 80, and R2O/SiO2 ratio of 0.8–1.0 when the ratio of Na2O/R2O is decreased from 1.0 to 0.9–0.7 (zeolite NaP impurity exists). Figure 3.20 shows the crystallization field of zeolites T, D, L, and P at 100  C crystallized from the parent mixture of K2O–Na2O–Al2O3–SiO2–H2O. The silica content is 68–72 mole% on a dry basis and the water content is 87–90 mole% of the total composition.

Figure 3.20 Crystallization field of zeolites T, D, L, and P from the K2O–Na2O–Al2O3–SiO2– H2O system. (100  C, SiO2: 68–72 mole%, H2O: 87–90 mole%). Reproduced with permission from [8]. Copyright (1974) John Wiley & Sons, Inc.

144

Chemistry of Zeolites and Related Porous Materials

The addition of Kþ cation to the synthetic system Na2O–Al2O3–SiO2–H2O significantly changed the type and structure of the zeolites crystallized from the parent mixture, which might be caused by the different templating ability of the cations on the SBUs of the zeolites. 3.2.2

Solvothermal Synthesis Approach to Aluminophosphates

Another significant area of progress in the synthesis of microporous materials is the successful synthesis of aluminophosphate molecular sieves and related heteroatomsubstituted derivatives (AlPO4-n, SAPO4-n, and MeAPO-n) in the 1980s by S.T. Wilson and E.M. Flanigen, scientists at Union Carbide Corporation (UCC).[24,25] So far, detailed structure information on more than 60 AlPO4-n and related derivatives has been obtained. Most of them were hydrothermally synthesized. Distinct from the crystallization of traditional aluminosilicate zeolite molecular sieves from strongly basic conditions, most of the microporous aluminophosphates were hydrothermally synthesized in slightly acidic or nearly neutral media. In the 1990s, Ruren Xu and coworkers were the first to explore the solvothermal synthesis of aluminophosphates and they achieved great success. The solvents they used in the solvothermal synthesis included diols and alcohols instead of water.[26,27] In the early stage of the solvothermal synthesis, several known aluminophosphate structures with various pore diameters, such as AlPO4-5, AlPO4-11, and AlPO4-21, were successfully synthesized. Subsequently, a series of new compounds with one-dimensional chain, two-dimensional sheet, and three-dimensional open-framework structures were synthesized. By using the solvothermal synthesis approach, it is possible to obtain large single crystals for most of the compounds, allowing their structure determination via single-crystal X-ray diffraction analysis. Among these new structures, JDF-20, the most remarkable example, has a 20-membered-ring channel system, which is the largest opening in the microporous aluminophosphates discovered so far. The synthesis of JDF-20 involved a very simple template (i.e., triethylamine) and low-polarity solvents such as diethylene glycol, triethylene glycol, tetraethylene glycol, or 1,4-butanediol. The use of high-polarity solvents such as ethylene glycol or ethanol only resulted in the crystallization of AlPO4-5 from the same parent mixture under identical crystallization conditions. It should be pointed out that a solid-phase transition of JDF-20 to AlPO4-5 could happen during the calcination of JDF-20 crystals. Motivated by the successful synthesis of new structures such as JDF-20, Xu and coworkers systematically investigated the solvothermal syntheses of aluminophosphates by using various alcohols as solvent and organic amines as structure-directing agent or template, and successfully synthesized a large number of new aluminophosphates with three-dimensional open-framework, two-dimensional sheet, and one-dimensional chain structures.[27–29] These two-dimensional sheet and one-dimensional chain structures could further form three-dimensional open-framework structures through pillaring and coordination-crosslink processes,[30] providing a new route for the synthesis of microporous compounds. So far, a large number of nondense aluminophosphates have been successfully synthesized via the solvothermal synthesis system. Most of them have anionic frameworks except for those crystallized from hydrothermal synthesis systems, such as AlPO4-5, AlPO4-11, and AlPO4-21. The stoichiometries of these aluminophosphates include

Synthetic Chemistry of Microporous Compounds (I)-Fundamentals and Synthetic Routes

145

[39–45] [46–53] [40,53–63] [64] AlPO4,[25,31–38] AlP2O3 Al2P3O3 Al3P4O3 Al3P5O6 8 , 12 , 16 , 20 , 3 [65, 66] 3 [67] 3 [68, 69] 3 [70] 3 [71] Al4P5O20 , Al5P6O24 , Al11P12O48 , Al12P13O52 , and Al13P18O72 . In contrast to the neutral open-framework aluminophosphates AlPO4-n which are built up by the strict alternation of AlO4 and PO4 tetrahedra, the anionic framework aluminophosphates are made up by the strict alternation of Al-centered polyhedra (AlO4, AlO5, AlO6) and P-centered tetrahedra [PO4, PO3(O), PO3(OH), PO2(O)(OH), PO2(OH)2, and PO(O)(OH)2]. The presence of Al-centered polyhedra instead of tetrahedra and the terminal P-O(H) bonds results in an Al/P ratio of less than 1.

Influence of Starting Gel Composition on the Structure of Aluminophosphates The synthetic regularities in solvothermal synthesis systems are similar to those in hydrothermal synthesis systems, i.e., a structure usually crystallizes from an initial gel with appropriate composition under appropriate crystallization temperature for a period of time. The discovery of these gel compositions for a specific structure, and their plotting on a chart, usually form a self-closed crystallization field diagram. When more than one structure can be synthesized from the same synthetic system, the crystallization fields for these structures can overlap with each other. The overlap of the crystallization fields means that more than one structure will crystallize from the initial gel whose composition falls into the overlap region. It is very common in solvothermal synthesis systems that more than one structure can crystallize from the same system. For example, for the aluminum isopropoxide–phosphoric acid–hexamethylenetetramine–ethylene glycol [Al(OPri)3– H3PO4–HMTA–EG] solvothermal synthesis system, when the molar ratio of Al(OPri)3/ EG was fixed and that of Al(OPri)3/H3PO4 and Al(OPri)3/HMTA was individually varied, two new aluminophosphates with different anionic open-frameworks, AlPO-CJB1 and AlPO-CJB2, crystallized from the initial gel with a batch composition of Al(OPri)3: 2.4 H3PO4: 3. 0 (CH2)6N4: 30 EG and Al(OPri)3: 3. 2 H3PO4: 3.6 (CH2)6N4: 30 EG, respectively, under identical crystallization temperatures and times. AlPO-CJB1 and AlPO-CJB2 have totally different structures and stoichiometries. For the aluminum isopropoxide–phosphoric acid–2-aminopyridine–2-butanol [Al(OPri)3–H3PO4–2-NH2 Py–BusOH] solvothermal synthesis system, two new anionic aluminophosphates with totally different structures and stoichiometries, AlPO-CJ4 and AlPO-CJ5, were successfully synthesized from the gel, with a batch composition of Al(OPri)3: 2.4 H3PO4: 2.0 2-NH2Py: 20 BusOH and Al(OPri)3: 2.4 H3PO4: 4.0 2-NH2Py: 20 BusOH, respectively, under the same crystallization temperature but different times. For the aluminum isopropoxide–phosphoric acid–ethylenediamine–ethylene glycol [Al(OPri)3–H3PO4–en-EG] system, an anionic 2-dimensional sheet (Al3P4O20C6N2H23),[55] an anionic chain with cornershared four-membered ring (AlP2O8[NH3(CH2)2NH3][NH4]),[72] and an anionic chain with edge-shared four-membered ring (AlP2O8H[NH3(CH2)2NH3)][43] crystallized from the gel with the compositions Al(OPri)3 : 1.8 H3PO4: 6.0 en: 20 EG, Al(OPri)3: 3.0 H3PO4: 2.5 en: 45 EG, and Al(OPri)3: 12.5 H3PO4: 5.0 en: 80 EG, respectively. Besides the composition of the initial gel, the type and properties of the alcohol solvent play an important role in the formation of the final structure of aluminophosphates as well. For example, Xu and coworkers studied in detail the correlation between the crystallized products and the properties of more than 20 alcohols for the system 5 Et3N– Al2O3–1.8 P2O5–X ROH at 180  C and the results are summarized in Table 3.7[29] They found that the structure of the product mainly depends on the polarity of solvent. Among

146

Chemistry of Zeolites and Related Porous Materials Table 3.7 Crystallization products from the 5 Et3N–Al2O3–1.8 P2O5–X ROH system with different alcohols ðENT Þ Alcohol Water Glycerol Ethylene glycol Methanol 1,3-Propanediol Diethylene glycol (DEG) Triethylene glycol (TEG) 1,4-Butanediol 1,3-Butanediol Tetraethylene glycol (tEG) 1-Butanol s-Butyl alcohol c-Hexanol t-Amyl alcohol

T EN

1.000 0.812 0.790 0.762 0.747 0.713 0.704 0.704 0.682 0.664 0.602 0.506 0.500 0.321

Products AlPO4-5 AlPO4-5 AlPO4-5 AlPO4-5 AlPO4-5 JDF-20 JDF-20 JDF-20 JDF-20 JDF-20 AlPO-CI AlPO-CI AlPO-CI amorphous

many chemical and physical parameters of solvents, the empirical solvent polarity parameter, ENT ,[73] was selected to represent the solvent polarity. It implied that different precursor species were present in different-polarity solvents as the nutrients for crystallization, as identified by 31P NMR spectroscopy of solutions of AlPO4-5 and AlPO-CI during crystallization. Although an alcohol instead of water was used in the solvothermal system, a small amount of water was still unavoidablly introduced due to the use of 85 wt% of phosphoric acid in water as phosphorus source and boehmite (6-coordinated Al) as Al source. If a fluoride source was used as mineralization agent, a small amount of water could be introduced by the fluorine source as well. Studies indicate that the small amount of water played an important role in the formation of the structures of aluminophosphates from solvothermal systems.[74] Compared with the amount of organic solvent, the amount of water in the solvothermal synthesis system is little. However, it plays an extremely important role in the successful formation of the final structures. Previous studies revealed that the synthetic reaction would not happen without the presence of this small amount of water. Another direct proof showing the importance of this small amount of water is the variation of structure type crystallized from the aluminum isopropoxide–phosphoric acid–triethylamine–tetraethylene glycol–water (extra added) system with the variation of the amount of extra added water, keeping the other parameters constant. By a gradual increase in the amount of the extra added water, one-dimensional aluminophosphate chain, two-dimensional aluminophosphate sheet, three-dimensional aluminophosphate open-framework JDF-20 and AlPO4-5, and dense-phase cristobalite sequentially crystallized from this system under conditions of identical crystallization temperature and time. Because Al and P atoms in the final structure come from the isolated aluminum and phosphorus sources in the parent mixture, respectively, the formation reaction of Al– O–P linkage must have happened during the crystallization process. Considering the importance of the introduction of organic amines to the synthesis system, we can

Synthetic Chemistry of Microporous Compounds (I)-Fundamentals and Synthetic Routes

147

conclude that the following reactions will definitely happen during the crystallization process: (1) The hydrolysis of aluminum isopropoxide (to supply Al atoms); (2) The condensation between phosphoric acid and aluminum isopropoxide or the hydrolysis Al source. Individual Al and P atoms from Al and P sources connect with each other to form framework structure; (3) The reaction between organic amine and phosphoric acid (acid–base reaction). The above reactions usually require proton transportation with the assistance of a water molecule. In a typical synthesis, the phosphoric acid is usually dispersed into the alcohol solvent. The water molecule within the phosphoric acid will carry part of the protons of phosphoric acid to alcohol solvent. After the addition of the organic amine, these protons will react with the added base to form ammonium species, which will then  form H-bonds with the P OH or P  O group of phosphoric acid. In addition, isopropyl oxide coordinated to the aluminum atom in aluminum isopropoxide will react with protons to form stable isopropyl alcohol molecules. The vacancy due to the leaving of one isopropyl alcohol base needs to be occupied by a negative group such as phosphoric acid group with part or total loss of their protons. The amount of the protons in the synthesis system is controlled by the amount of phosphoric acid, while the amount of water in the synthesis system will determine how many protons can be carried to the alcohol solvent. The reactivity to the proton and the amount of added organic amine will affect the amount of protons within the alcohol solvent, which will significantly affect the hydrolysis rate of aluminum isopropoxide and the condensation rate between hydrolysed aluminum species and phosphoric acid, which will determine the formation of the final structures. In addition, the added organic amines can form strength variable H-bonds with P OH groups by donating a proton to a P OH group or accepting a proton from P OH group, which will affect the status of P OH group entering the framework. Influence of Crystallization Temperature and Time on the Structure of Aluminophosphates Similar to hydrothermal synthesis systems, in solvothermal synthesis systems the product could be synthesized after crystallizing the parent mixture for a certain time at an appropriate temperature. Beyond that temperature and time range, the crystallization process may not happen or instead only dense phases will be formed. Within the appropriate temperature range, increasing the crystallization temperature results in a decrease of crystallization time. For some specific solvothermal synthesis systems, different structures could be synthesized from the same batch composition by varying crystallization temperature and time. For example, AlPO-CJB1 could crystallize from the aluminum isopropoxide–phosphoric acid–hexamethylenetetramine–ethylene glycol synthesis system at 195  C for 5 days. When the crystallization temperature was decreased to 180  C, no crystalline product was obtained for a crystallization time as long as 8 days. However, large single-crystals of AlPO-CJB1 could be obtained after 4 days crystallization when the temperature was increased to 200  C. Another example is the Al2O3: 1.8 P2O5: 4.7 Et3N: 18 TEG system. When the parent mixture was heated at 180  C for 5 days, JDF-20 featuring an extra-large 20-membered-ring channel system

148

Chemistry of Zeolites and Related Porous Materials

crystallized, whereas AlPO4-5, featuring a large 12-membered-ring channel system, formed when the crystallization temperature was increased to 200  C. Influence of Structure-directing Agent Type on the Structure of Aluminophosphates In the last 20 years, more than 100 organic amines have been used as templates (structure-directing agents, SDAs) in the synthesis of aluminophosphates. So far, several dozen structurally distinct aluminophosphate molecular sieves and microporous compounds have been synthesized. To study or reveal the structure-directing effect of the templates on the resulting aluminophosphate structures, organic amines have been classified according to various criteria such as single-amine and multi-amine; chainamine, branch-amine, and ring-shape amine; arylamine and alkylamine; primary amine, secondary amine, tertiary amine, and quarternary ammonium salt. Owing to the different protonation abilities (alkalinity, related to the electronic and stereo effects of the amine) and H-bonding abilities to P OH group, the amines have a different effect on the hydrolysis rate of aluminum isopropoxide and the condensation rate between hydrolysed aluminum species and phosphoric acid, which will affect the coordination state of Al and P atoms in the parent mixture, leading to the formation of various structures. According to previously reported results and our recent studies, the following synthetic laws could be addressed: (1) The ring-shape single amine, secondary single amine, or aromatic single amine which can supply an H-atom to form a weak H-bond with phosphoric acid favors the formation of a terminal P OH group in the final aluminophosphate structure, whereas the chain-shape primary amine or multi-amine which can supply an H-atom to form a strong H-bond with phosphoric acid favors the formation of a  O group in the final aluminophosphate structure. For example, most of terminal P  the amines directing the formation of layered aluminophosphates with an Al/P ratio of 2/3 and a terminal P OH group are ring-shape single amines or secondary single amines (Table 3.8), whereas most of the amines directing the formation of layered  O group are chainaluminophosphates with an Al/P ratio of 3/4 and a terminal P  shape primary amines or multi-amines (Table 3.9). (2) A coordination number higher than 4 for framework Al atoms is more likely when strong organic bases (weak H-bonding ability) such as aryl, ring-shape, or cageshape organic amines were used as the template. So far, as listed in Table 3.10, more than 7 organic templates have been reported in the synthesis of aluminophosphates containing 5- and/or 6-coordinated Al atoms in their frameworks. The results summarized in Table 3.10 showed that the aluminophosphates containing 6-coordinated Al atoms were directed by 2-aminopyridine and hexamethylenetetramine, while imidazole, pyridine, 4-methylpyridine, 1,6-hexamethylenediamine, 2-aminopyridine, and hexamethylenetetramine directed the aluminophosphates containing 5-coordinated Al atoms. Except for 1,6-hexamethylenediamine, all the other amines have strong alkalinity. For example, the pKa of pyridine, 4-methylpyridine, and imidazole is 5.25, 6.02, and 6.95, respectively. These results indicated that strong bases favored the formation of 5- and/or 6-coordinated Al atoms in the structure they directed. More examples came from the synthesis of anionic aluminophosphates AlPO-CJ4, AlPO-CJB1,

Synthetic Chemistry of Microporous Compounds (I)-Fundamentals and Synthetic Routes

149

Table 3.8 Templates and solvents for the solvothermal synthesis of layered aluminophosphates with an Al/P ratio of 2/3 No.

Al coordination environment

1 2 3 4 5

2 AlO4 2 AlO4 2AlO4 2AlO4 AlO4, AlO5

6

AlO4, AlO5

7

2AlO4

P coordination environment

Template

 O), PO2(  O)(OH) 2 PO3(  2PO3(  O), PO2(  O)(OH)  O), PO2(  O)(OH) 2PO3(   2PO3(  O), PO2(  O)(OH) PO4, PO3(OH),  PO2(  O)(OH) PO4, PO3(OH),  PO2(  O)(OH)  O), PO3(OH), PO3(   PO2(  O)(OH)

Solvent

cyclopentylamine cyclohexylamine cyclohexylamine s-butylamine pyridine

tetraethylene glycol tetraethylene glycol tetraethylene glycol s-butyl alcohol s-butyl alcohol

4-methylpyridine

s-butyl alcohol

2,2,6,6-tetramethylpiperidine

s-butyl alcohol

Note: The empirical formulas for the above seven layered aluminophosphates are: 1. [Al2P3O12H][C5H9NH3]2; 2. [Al2P3O12H][C6H11NH3]2; 3. [Al2P3O12H][C6H11NH3]2; 4. [Al2P3O12H][2-BuNH3]2; 5. [Al2P3O12H2][PyH]; 6. [Al2P3O12H2][C6H8N]; 7. [Al2P3O12H2] [C9H20N].

Table 3.9 Templates and solvents for the solvothermal synthesis of layered aluminophosphates with an Al/P ratio of 3/4.

No.

Al coordination environment

P coordination environment

1

3AlO4

4PO4(  O)

ethylamine

2 3 4 5 6 7 8 9

3AlO4 3AlO4 3AlO4 3AlO4 3AlO4 3AlO4 3AlO4 2AlO4, AlO5

propylamine butylamine 1,2-propylenediamine cyclobutylamine, piperidine triethylamine isopropanolamine 1,2-dimethylimidazole imidazole

10 11 12 13 14 15 16

3AlO4 3AlO4 3AlO4 3AlO4 3AlO4 3AlO4 3AlO4

4PO4(  O)  O) 4PO4( 4PO4(  O)  O) 4PO4( 4PO4(  O)  O) 4PO4( 4PO4(  O)  O), 2PO4, PO3(   PO3(  O)(OH)  O) 4PO4( 4PO4(  O)  O) 4PO4( 4PO4(  O)  O) 4PO4( 4PO4(  O)  O) 4PO4(

ethylene glycol, s-butyl alcohol s-butyl alcohol 1-butanol ethylene glycol tetraethylene glycol poly(vinyl alcohol) isopropanolamine triethylene glycol s-butyl alcohol

ethylenediamine diethylenetriamine triethylenetetramine 1,5-pentamethylenediamine 2-methyl-1,5-pentamethylenediamine 1,2-hexanediamine tetramethylethylenediamine

ethylene glycol ethylene glycol ethylene glycol triethylene glycol triethylene glycol s-butyl alcohol triethylene glycol

Template

Solvent

Note: The empirical formulas for the above sixteen layered aluminophosphates are: 1. [Al3P4O16][CH3CH2NH3]3; 2. [Al3P4O16][CH3(CH2)2NH3]3; 3. [Al3P4O16][CH3(CH2)3NH3]3; 4. [Al6P8O32][NH3CH(CH3)CH2NH3]3H2O; 5. [Al3P4O16][C4H7NH3]2[C5H10NH2]; 6. [Al3P4O16][Et3NH]3; 7. [Al3P4O16][NH3CH2CH(OH)CH3]3; 8. [Al3P4O16][C5H9N2][NH4]; 9. [Al3P4O16H][C3H5N2]2; 10. [Al3P4O20][H3NCH2CH2NH3][H2OCH2CH2OH][HOCH2CH2OH]; 11. [Al3P4O16][H3N(CH2)2NH(CH2)2NH3]1.5; 12. [Al3P4O16][C6H21N4]; 13. [Al3P4O16][NH3(CH2)5NH3][C5H10NH2]; 14. [Al3P4O16][H3NCH2CH2CH2CH(CH3)CH2NH3]; 15. [Al3P4O16][C6H16N2]1.5; 16. [Al3P4O16][(CH3)2NHCH2CH2NH(CH3)2] [H3O].

150

Chemistry of Zeolites and Related Porous Materials

Table 3.10 Organic templates resulting in the aluminophosphates containing 5- and/or 6-coordinated Al atoms Al coordination No. environment 1 2 3 4 5 6 7 8

2AlO4, AlO5 AlO4, AlO5 AlO4, AlO5 2AlO4, 2AlO5 2AlO5 AlO6 9AlO4, 2AlO6 8AlO4, 4AlO5

P coordination environment  2PO4, PO3(  O), PO2(  O)(OH)  O)(OH) PO4, PO3(OH), PO2( PO4, PO3(OH), PO2(  O)(OH)  O)(OH) 4PO4, PO2(   2PO4, PO2(  O)(OH) 2PO3(OH) 12PO4 12PO4

Template imidazole pyridine 4-methylpyridine 1,6-hexamethylenediamine 2-aminopyridine 2-aminopyridine hexamethylenetetramine hexamethylenetetramine

Solvent s-butyl alcohol s-butyl alcohol s-butyl alcohol ethylene glycol s-butyl alcohol s-butyl alcohol ethylene glycol ethylene glycol

Note: The empirical formulas for the above eight layered aluminophosphates are: 1. [Al3P4O16H][C3N2H5]2; 2. [Al2P3O12H2][PyH]; 3. [Al2P3O12H2][C6NH8]; 4. [Al4P5O20H][H3NCH2(CH2)4CH2NH3]; 5. [Al2P3O12H2][H3O]; 6. [AlP2O6(OH)2][H3O]; 7. [Al11P12O48][(CH2)6N4H3H2O]; 8. [Al12P13O52][(CH2)6N4H3].

and AlPO-CJB2. When 2-aminopyridine was used as the template, a three-dimensional anionic aluminophosphate framework, AlPO-CJ4, containing only 6-coordinated Al atoms formed; when hexamethylenetetramine was used as the organic additive, two threedimensional anionic aluminophosphate frameworks, AlPO-CJB1 and AlPO-CJB2, containing 5-coordinated and 6-coordinated Al atoms, respectively, were synthesized. In an anionic aluminophosphate framework with a stoichiometry of [Al2P3O12H2][H3O] (AlPO-CJ5) recently synthesized by us, all Al atoms are 5-coordinated. This structure was directed by 2-aminopyridine, but the organic template was not encapsulated in the final structure. Both anionic aluminophosphate frameworks AlPO-CJ4 and AlPO-CJ5 were synthesized from the aluminum isopropoxide–phosphoric acid–2-aminopyridine–s-butyl alcohol system. (3) Organic amines with big size are not appropriate templates in weak polar solvent. Numerous aluminophosphates encapsulated big organic amines in their framework and crystallized from a strong polar solvent instead of from a weak polar solvent. For example, when polyethylene multi-amines [H2N(C2H4N)nH, n  5] were used as templates, and weak polar alcohols such as s-butyl alcohol or tetraethylene glycol was used as solvent, only an amorphous gel was obtained after crystallization under normal synthesis conditions, while aluminophosphate powder with an open-framework structure could be obtained when strongly polar alcohols such as ethylene glycol were used as solvent. However, it is very difficult to obtain large single crystals for structural analysis via single-crystal X-ray diffraction. (4) A chain-shape polyamine and ethylene glycol system favors the formation of layered aluminophosphates with an Al/P ratio of 3/4. For instance, ethylenediamine, diethylenetriamine, and triethylenetetramine combining with ethylene glycol solvent lead to the formation of layered aluminophosphates with an Al/P ratio of 3/4 (see Table 3.9). Therefore, it is expected that using tetraethylenepentamine as template and ethylene glycol as solvent will result in the crystallization of a layered aluminophosphate with an Al/P ratio of 3/4. Guided by this prediction, we did

Synthetic Chemistry of Microporous Compounds (I)-Fundamentals and Synthetic Routes

151

Table 3.11 Aluminophosphates crystallized from s-butyl alcohol system

No.

Al coordination environment

1 2 3 4 5 6 7 8 9 10 11

2AlO4, AlO5 AlO4, AlO5 AlO4, AlO5 2AlO5 AlO6 2AlO4 2AlO4 3AlO4 2AlO4, AlO5 3AlO4 AlO4

P coordination environment  O), PO2(  O)(OH) 2PO4, PO3(    PO4, PO3(OH), PO2(  O)(OH)  O)(OH) PO4, PO3(OH), PO2(  2PO4, PO2(OH)2 2PO3(OH)   2PO3(  O), PO2(  O)(OH)  O), PO3(OH), PO2(  O)(OH) PO3(    O) 4PO3(   O), PO2(  O)(OH) 2PO4,PO3(    O) 4PO3(   O)(OH) 2PO2( 

Template imidazole pyridine 4-methylpyridine 2-aminopyridine 2-aminopyridine s-butylamine 2,2,6,6-tetramethylpiperidine propylamine imidazole 1,2-cyclohexanediamine triethylamine

Note: The empirical formulas for the above eleven aluminophosphates synthesized from the s-butyl alcohol system are: 1. [Al3P4O16][C3N2H5]2; 2. [Al2P3O12H2][PyH]; 3. [Al2P3O12H2][C6NH8]; 4. [Al2P3O12H2][H3O]; 5. [AlP2O6(OH)2] [H3O]; 6. [Al2P3O12H][2-BuNH3]2; 7. [Al2P3O12H2][C9H20N]; 8. [Al3P4O16][CH3(CH2)2NH3]3; 9. [Al3P4O16H] [C3N2H5]2; 10. [Al3P4O16][(CH2)4(CHNH3)2]1.5; 11. [AlP2O8H2][Et3NH].

successfully synthesize this predicted layered aluminophosphate and determine its structure via single-crystal X-ray diffraction analysis. (5) The combination of arylamines and s-butyl alcohol solvent favors the formation of anionic aluminophosphate frameworks containing high-coordination Al atoms. For example, 2,2,6,6-tetramethylpiperidine, triethylamine, propylamine, 1,2-cyclohexanediamine, s-butylamine, imidazole, 4-methylpyridine, and pyridine have been used as the template, and s-butyl alcohol was used as the solvent in the synthesis of aluminophosphates (Table 3.11). The templates directing the high-coordination Al atoms in the structures were imidazole, 4-methylpyridine, and pyridine. Using the same solvent, when 2-aminopyridine was used as organic additive, we successfully synthesized two anionic aluminophosphate frameworks, containing only 6-coordinated Al atoms (AlPO-CJ4) and 5-coordinated Al atoms (AlPO-CJ5), respectively. AlPO-CJ4 and AlPO-CJ5 are the first open-framework aluminophosphates of which the primary Al building unit is solely made up of 6- or 5-coordinated Al atoms with all six or five oxygen vertices being shared by adjacent P atoms, respectively. (6) Among those templates directing the layered aluminophosphates with an Al/P ratio of 3/4, tertiary amines of large size and weak H-bonding ability always led to the introduction of protonated water or ammonium into the resulting structure. These extra introduced positively charged ions can enhance the nonbonded interaction between the inorganic host and guest species and can also balance the negative charge of the inorganic layers. For example, when 1,2-dimethylimidazole and tetramethylethylenediamine were used as templates, ammonium or protonated water was introduced into the final structures. Matching of the Charge Density of the Inorganic Framework and Organic Templates Based on a large number of reported results, chemists working in this field believe that organic amine cations play a templating role during the crystallization process of

152

Chemistry of Zeolites and Related Porous Materials

inorganic frameworks due to the similarity of the size and shape between the organic amine and the channels or cages holding them. One remarkable example is zeolite ZSM-5. The original template for zeolite ZSM-5 is a tetrapropylammonium cation located in the intersection of two channels painting in two different directions. The tetrapropylammonium cation’s four propyl chains point into four branches of two-dimensional interconnected channels. Many studies have suggested that during the hydrothermal synthesis of zeolites such as ZSM-5, the inorganic species such as silicate and aluminate condensed around the organic cations, creating a crystalline phase with pores filled with organic template molecules. However, the real crystallization process of zeolites in the hydrothermal or solvothermal synthesis is not as simple as that. For example, many organic amines other than tetrapropylammonium cation can direct the structure of zeolite ZSM-5. In the presence of specific inorganic cations, zeolite ZSM-5 can be synthesized from a purely inorganic system. In addition, another characteristic of a hydrothermal or solvothermal synthesis system is that changing the crystallization conditions or parent gel composition can result in the crystallization of several different structures from the same synthetic system. In some cases, the correlation between the size or shape of the organic templates and that of the channel or cage holding them is not very obvious. For example, more than 85 organic amines can direct the formation of microporous aluminophosphate AlPO4-5 (AFI), while microporous aluminophosphate AlPO4-11 (AEL) can be templated by more than 20 organic amines. These amines have totally different sizes and shapes. In comparison to the fact that many organic amines can also direct the same structure, one organic amine can template many different structures in some cases. For example, dipropylamine (Pr2NH) can template the structures of AlPO4-8 (AET), AlPO4-11 (AEL), AlPO4-31 (ATO), AlPO4-39 (ATN), AlPO4-41 (AFO), MgAlPO-46 (AFS), and CoAlPO-50 (AFY). Moreover, the guest molecules encapsulated in the metal phosphates with large-pore-size channels such as VPI-5, JDF-20, AlPO4-8, cloverite, ULM-5, and ULM-16 are water or small amines instead of the expected large ones. In these cases, the organic amines may not play a templating role. Structure-directing agent may be a better term for them. During the synthesis, the complicated synergistic interactions among the precursors such as H-bonding, van der Waals, and Coulombic interactions are unavoidable in the building of 1-, 2-, or 3-dimensional ordered structures from isolated precursor molecules. In spite of the different names of these three interactions, they are basically the same in nature: each is an electromagnetic interaction, one of the four basic interactions. The intensity of the interaction depends on the charge and the distance between the two charged species. The charge distribution on the inorganic species and the organic additives, with equal quantities of negative and positive charges, will have an important influence on the interaction intensity. The properties of the organic species such as volume, shape, protonation ability, and the number of the hydrogen atoms that could be protonated will determine which of these three interactions (H-bonding, van der Waals, and Coulombic) will guide the formation of the final structure. In fact, Xu and coworkers found a clear linear correlation between the charge density of the organic additives and that of the inorganic open framework of the aluminophosphate family, which implies that the synergistic interaction between the organic and inorganic species may play an important role in the early stages of the formation of the inorganic structure.[75] This linear correlation may also imply an as-yet-undiscovered, more basic interaction that controls the formation of the final structure during the synthesis.

Synthetic Chemistry of Microporous Compounds (I)-Fundamentals and Synthetic Routes

153

˚ 3) of the In their studies,[75] Xu and colleagues defined the charge density (e 103A organic template as the ratio of total positive charge to the total volume of the charged species encapsulated in the anionic framework (for a unit cell), while that of the framework was defined as the ratio of the total negative charge of the framework in an unit cell to the volume of the unit cell. The correlation between the charge/volume ratio of the organic template and the charge density of the inorganic framework is plotted in Figure 3.21 according to the Al/P ratio, dimension, structure type of the aluminophosphates, and the type of the organic template. The correlation coefficient for the plot in Figure 3.21: (a) is 0.993, which indicates an excellent linear relationship. In these organic amines, triethylamine gives the smallest charge/volume ratio. Accordingly, the anionic inorganic framework of [AlP2O8H2] directed by this amine gives the smallest negative charge density in these four compounds. Compounds [AlP2O8H][H3NCH2CH2NH3] and [AlP2O8][NH4] [H3NCH2CH2NH3], which are directed by same organic additive of ethylenediamine, have an edge- and a corner-shared four-membered ring chain, respectively. Interestingly, the anionic framework of the former compound has one P OH bond in the pendant  PO4H side group, while that of the latter compound has two P  O bonds for each P-atom, which results in an increased negative charge density of the framework. The introduction of ammonium ion objectively increased the positive charge density of the organic species. Simulation studies on the substitution of the ammonium ion with positively charged ethylenediamine to balance the negative charge of the anionic framework indicated that the energy of the system will significantly increase. Therefore, the introduction of the ammonium ion not only increased the positive charge density of the organic species, but also decreased the energy of the compound. Compared with fully charged 1,3-propanediamine, fully charged ethylenediamine has a higher charge/volume ratio. Therefore, it can be expected that the anionic framework directed by ethylenediamine has a higher negative charge density than that directed by 1,3-propanediamine even though they have identical topological structures. Similarly, as shown in Figure 3.21: (b), (c), (d), and (e), there exists an excellent linear correlation between the charge/volume of the charged species encapsulated in the framework and the charge density of the anionic framework for the anionic aluminophosphates with Al/P ratio of 1/2 directed by arylamines, the anionic aluminophosphates with Al/P ratio of 2/3 directed by cyclic amines, the layered anionic aluminophosphates with Al/P ratio of 3/4 and 4,6,8-net topological sheet structure directed by chain-shape alkylamines, and the layered anionic aluminophosphates with Al/P ratio of 3/4 and 4,6,12-net topological sheet structure directed by chain-shape alkylamines as well. The Structural Construction Regularity of Aluminophosphates and Prediction of Open-framework Aluminophosphate Structures with Specified Al/P Stoichiometry It was reported in a previous section that a large number of new microporous aluminophosphate materials have been successfully synthesized under hydrothermal or solvothermal conditions after the discovery of aluminophosphate AlPO4-n (n denotes a specific structure type) molecular sieves in the early 1980s. Unlike the neutral framework of AlPO4-n, which is built from strict alternation of AlO4 and PO4 tetrahedra via cornersharing vertex oxygen atoms with an Al/P ratio of exclusively unity, a variety of

154

Chemistry of Zeolites and Related Porous Materials

Figure 3.21 The correlation between the charge/volume ratio of the organic species and the charge density of the inorganic framework of the anionic aluminophosphate chains with Al/P ratio of 1/2 (a), of the anionic aluminophosphates with Al/P ratio of 1/2 directed by arylamines (b), of the anionic aluminophosphates with Al/P ratio of 2/3 directed by cyclic amines (c), of the layered anionic aluminophosphates with Al/P ratio of 3/4 and 4,6,8-net topological sheet structure directed by chain-shape alkylamines (d), and of the layered anionic aluminophosphates with Al/P ratio of 3/4 and 4,6,12-net topological sheet structure directed by chain-shape alkylamines (e)

Synthetic Chemistry of Microporous Compounds (I)-Fundamentals and Synthetic Routes

Figure 3.21

(Continued )

155

156

Chemistry of Zeolites and Related Porous Materials

Figure 3.21

(Continued )

aluminophosphates with chains, layers, and open frameworks and an Al/P ratio of less than unity have anionic frameworks, which are constructed from the alternation of Al polyhedra (AlO4, AlO5, and AlO6) and P tetrahedra (PO4b, PO3bOt, PO2bO2t, and PObO3t where b represents bridging oxygens and t represents terminal oxygens). On investigation of the structural features of all aluminophosphates, Xu and coworkers found one structural construction regularity.[75] The connectivity between Al atoms and P atoms via O atoms can be represented by Equation (3.2): X X Mki  iðAlki Oib Þ ¼ Nlj  jðPlj OjbÞ ð3:2Þ k

l

where i(j) is the number of bridging oxygen connected with the Al (P) atom and Mk ðNl Þ is the number of Al (P) atoms which share i(j) bridging oxygens with an adjacent P (Al) atom Mk , Nl ¼ 1; 2; 3; . . ., and Mk =Nl is the Al/P ratio. i is 3, 4, 5 or 6, which means that Al atoms share i oxygen atoms with adjacent P atoms. j is 1, 2, 3 or 4, which means that P atoms share j oxygen atoms with adjacent Al atoms with the remaining ð4  jÞ  O and/or P oxygen atoms as terminal P OH groups. k and l mean the number of types  of Al and P atoms possessing the different bridging conditions. According to this equation, Xu and coworkers developed a method to generate hypothetical open-framework aluminophosphate structures with specified Al/P stoichiometry using the automated assembly of SBUs.[76] For each specified Al/P stoichiometry, all the possible combinations of Al and P atoms with different coordination states could be calculated according to Equation (3.2). The Al and P atoms of different coordination states, together with the clusters constructed by them, could be selected as the building

Synthetic Chemistry of Microporous Compounds (I)-Fundamentals and Synthetic Routes

157

units for the structure generation. The detailed procedures of this method can be described as follows: (1) Determine what kinds of building units need to be included in the simulation and their numbers in the unit cell. (2) Generate the required number of building units in a specified unit cell randomly. (3) Optimize the ‘energy’ of the system using our predefined force field. After this step, the building units are assembled together by the attractions between O_al and O_p atoms, that is, ‘sticky pairs’. [O_al (O_p) denotes a terminal O-atom connecting with only one Al (p)-atom.] It should be noted that, in this step, the linkages between the building units are nonbonding interactions between O_al and O_p atoms. (4) ‘Glue’ each ‘sticky’ atom pair into one atom, that is, combine each pair of O_al and O_p atoms into one O_b atom. Any O_al or O_p atom which has not found its ‘sticky’ opposite is turned into an O_t atom. In this step, the building units are linked together through Al O P bonds, and the hypothetical frameworks are built up.

3.2.3

Crystallization of Zeolites under Microwave Irradiation

Microwaves are a form of electromagnetic energy. There are two specific mechanisms of interaction between materials and microwaves: (1) dipole interactions and (2) ionic conduction. Both mechanisms require effective coupling between components of the target material and the rapidly oscillating electrical field of the microwaves. Dipole interactions occur with polar molecules. The polar ends of a molecule tend to align themselves and oscillate in step with the oscillating electrical field of the microwaves. Collisions and friction between the moving molecules result in heating. Generally speaking, the more polar a molecule, the more effectively it will couple with (and be influenced by) the microwave field. Ionic conduction is only minimally different from dipole interactions. Obviously, ions in solution do not have a dipole moment. They are charged species that are distributed and can couple with the oscillating electrical field of the microwaves. The effectiveness or rate of microwave heating of an ionic solution is a function of the concentration of ions in solution. The first papers on the use of microwaves for synthesis reactions appeared in the open, peer-reviewed literature. Since that time, over a thousand articles have been published, numerous conferences have focused on the advance of microwave techniques, and the use of microwave processing is now a ‘hot’ topic for combinatorial and parallel strategies. One of them is the synthesis and modification of zeolites under microwave irradiation. In principle, the microwave dielectric heating effect, microwave ionic conduction loss, and the local superheating effect are three main factors in the acceleration of chemical reactions. The coupling ability of the microwave and molecules depends on the properties of the molecules, which can be used to control the properties of materials and the selectivity of the resulting reaction. In other words, the reactants or complex in the transient state that can determine the reaction rate or nature of the reaction intermediate can selectively adsorb microwave energy, which can greatly increase the reaction rate. Besides their heating effect, microwaves can change the steric configuration of molecules or activate the molecules as well. So far, the interpretation of the nonheating effect of the microwaves is incomplete in both theory and experiments.

158

Chemistry of Zeolites and Related Porous Materials

Sometimes the interpretations conflict with each other. The mechanism for the acceleration effect of microwaves on the reactions and molecular sieve crystallization needs to be studied in a more in-depth way. The crystallization of microporous compounds under microwave irradiation was developed in the 1970s, and was characterized by mild conditions, low energy consumption, rapid reaction, small particles, and uniform particle-size distribution. For example, under microwave irradiation and normal pressure, zeolite NaA could crystallize with a high crystallinity in a short time, even less than 1 minute. Thus, this new synthesis approach can quickly and continuously produce molecular sieves with low energy consumption. The microwave synthesis of zeolite NaA and microporous AlPO4-5 and the ion exchange reaction of zeolite Ce-b will be discussed in detail later in this section. The successful synthesis of microporous FeAPO4-5, CoAPO-5, CoAPO-44, VPI-5,[77] zeolites Na-X, Na-Y, ZSM-5, and TS-1,[78] the preparation of molecular sieve membranes,[79] the dispersion of salts or oxides into the channels of molecular sieves, and the modification or functionalization of the channel or structure of the molecular sieves under microwave irradiation have been reported as well.[80] Microwave Synthesis of Zeolite NaA In this study, the microwave oven used a specific, fixed frequency of 2450 MHz (2.45 GHz) with a 100% power rating of 650 W. The synthesis was performed for 5–20 minutes with a 10–50% power level. Systematic studies on the microwave synthesis of zeolite A revealed that: (1) High-quality small zeolite A crystals (
0.3 mm) can crystallize from the parent mixture with a gel composition of (1.5–5.0) Na2O: 1.0 Al2O3: (0.5–1.7) SiO2: (40–120) H2O. When the ratio of H2O/Al2O3 is greater than 150, amorphous by-product starts to appear. When the ratio of Na2O/Al2O3 is greater than 8.0, hydroxysodalite is the only product. When the ratio of SiO2/Al2O3 reaches 2.0, no NaA crystals are formed. (2) Higher microwave power will shorten the reaction time. Overall, using the 20% microwave power level and a 15–20 minute reaction time can result in the formation of NaA crystals with high crystallinity. If the microwave power is greater than 50%, hydroxysodalite impurity will exist in the product. (3) Aging and stirring play an important role in the successful synthesis of zeolite NaA. The effect of aging of the parent mixture on the microwave synthesis of zeolite NaA was studied by P.M. Slangen et al. in 1997.[81] The gel composition of the studied system was 1.5 Na2O: 1.0 Al2O3: 1.0 SiO2: 96.5 H2O. The gels, aged at ambient temperature for various times, were cooked under microwave irradiation for 5 min at 100  C and the results are summarized in Table 3.12. The results summarized in Table 3.12 clearly suggest that ‘aging’ time has a big influence on the crystallization of the final product. To investigate more deeply the role of aging in the crystallization of the zeolite, the author mixed reaction gel (10 wt%) aged for 20 h and unaged parent gel and cooked the mixture for 5 min at 100  C under microwave irradiation. The resulting product was fully crystallized pure zeolite NaA, which strongly suggested that an appropriate number of nuclei had existed in the parent gel aged for 20 h and that short-time microwave irradiation greatly accelerated the crystallization process by nonspontaneous nucleation. This phenomenon frequently happened in the synthesis of other zeolites under microwave irradiation, showing that microwave heating is different from conventional heating. Regarding the microwave irradiation mechanism in the

Synthetic Chemistry of Microporous Compounds (I)-Fundamentals and Synthetic Routes

159

Table 3.12 Effect of aging time on the crystallization of the product under microwave irradiation Aging time (t/min) 5 60 120 180 240 1200

Product 100% amorphous <10% HS þ >90% amorphous <10% HS þ >90% amorphous >80% NaA þ <10% HS þ <1% amorphous >90% NaA (0.4–2 mm) þ <10 HS 100% NaA 0.1–0.4 mm

synthesis of zeolites, Whittington proposed an ionic conduction model, i.e., the rotation of the dipole of water molecules and Na+ existing in the zeolite led to the ionic conduction.[82] This model was used to explain the following results[83]: calcination of zeolite NaA at 800  C resulted in collapse of the framework and finally led to the formation of the dense phase of nepheline. However, if zeolite NaA was irradiated in a microwave oven operated at 300 W (2.45 GHz) for several hours, the structure of NaA would be transferred to carnegeite (another microporous structure containing 6- and 8membered rings) after identical treatment via rearrangement of the lattice atoms of NaA. It is believed that Na+ played an important ionic-conduction role in this process. Microwave Synthesis of AlPO4-5 Normally, AlPO4-5 molecular sieve was hydrothermally synthesized from the parent mixture that contains phosphorus source (phosphoric acid), aluminum source (aluminum oxide), organic template (tetraethylammonium hydroxide or triethylamine), and water. The pH of the parent mixture was adjusted with hydrochloric acid or ammonia water. In the microwave synthesis, the well mixed parent mixture was loaded into a Teflon autoclave and heated in a microwave oven for 7–25 min with 10–40% of full power. The XRD pattern of the as-synthesized AlPO4-5 perfectly agreed with those reported in the literature. Scanning electron microscope (SEM) analysis indicated that the size of the crystals synthesized by microwave heating is smaller (50 nm–3 mm) than those synthesized by conventional hydrothermal heating, which is usually greater than 5 mm. The crystallization process of AlPO4-5 under microwave irradiation with 10–40% of full power only took 7–25 min, which is much shorter than conventional hydrothermal synthesis (at least 5 h). Experiments further indicated that the composition of the parent gel for the formation of crystalline AlPO4-5 in the microwave synthesis is broader than that in the conventional hydrothermal synthesis, which may reduce the consumption of expensive amines. Recently, S. Yamanaka and coworkers reported the microwave synthesis of microporous AlPO4-H1, H2 (AHT), and H3 (APC) without the use of an organic template. Numerous microporous crystallines with various structure types have been successfully synthesized by using microwave irradiation in the last decade. Compared with conventional hydrothermal synthesis, microwave synthesis has many advantages, such as narrow particle-size distribution, controllable morphology, broader parent mixture composition, and short crystallization time. It can be expected that microwave synthesis will play an important role in the rapid, energy-saving, and continuous production of microporous compounds.

160

Chemistry of Zeolites and Related Porous Materials

Figure 3.22 Microwave-heated, continuous-flow, high-pressure tube reactor comprising reaction mixture inlet (1), pressure jacket (2), isolation jacket (3), reactor tube coil in the microwave-heated cavity (4), thermocouple (5), cooling jacket (6), and reaction mixture outlet (7). Reproduced with permission from [84]. Copyright (1998) Elsevier

Based on the characteristics of microwave synthesis, I. Braun and coworkers designed a microwave-heated high-pressure tube reactor (Figure 3.22) in which high-quality microporous aluminophosphate AlPO4-5 could be continuously synthesized.[84] The well mixed parent mixture (1.0 Al2O3: 1.0 P2O5: 1.5 Pr3N: 150 H2O) was pumped through the reactor under high pressure. The product from the outlet, containing a fraction of crystalline material, was added to the storage vessel under stirring and thus recirculated through the reactor. Optimum process parameters turned out to be: (1) a flow rate of 900 mL/h (residence time: 8 min) and (2) a temperature range of 180 to 190  C. The synthesis gel was fully crystallized (crystal sizes: 1–10 mm) after several cycles. Later, S.E. Park and coworkers reported the rapid and mass production of zeolite ZSM-5 and NaY using continuous microwave equipment, which provided a good example of the large-scale production of microporous materials for industrial use.[85] Ion-exchange of Zeolites under Microwave Irradiation Because of its unique heating mechanism, microwave irradiation has a special effect on the ion-exchange behavior of zeolite molecular sieves. Compared with traditional ionexchange of rare-earth ions, microwave-irradiation-conducted ion-exchange has many

Synthetic Chemistry of Microporous Compounds (I)-Fundamentals and Synthetic Routes

161

advantages. In a typical exchange experiment, mixed solutions containing zeolite Beta (BEA) solid and 0.05 mol/L rare earth ions solution (Ce3+, Eu3+, and Sm3+) with an l=s ratio of 25 (by weight) was heated by a microwave oven at 20% power (full power: 650 W, 2.45 GHz). The products, i.e., Ce-Beta, Eu-Beta, and Sm-Beta, were filtered off, washed with deionized water, and dried. The Ce-Beta produced from microwave-irradiation-driven ion-exchange gave an excitation and emission spectra of Ce3+, which is very similar to that of Ce-Beta produced from traditional ion-exchange methods. However, the emission spectrum consists by at least three spectral bands, with the strongest peak at 400 nm. Under microwave heating, the movement rate of water molecules and rare earth ions is much higher than that under traditional heating, resulting in rare earth ions entering into the c direction channel, which is normally not accessible to rare earth ions in traditional ion-exchange methods. Occupation of positions, in the c direction of the channel by rare earth ions will lead to significant change in the emission spectra. This is one advantage of the microwave-driven ion-exchange method. In a traditional ion-exchange experiment, the final degree of ion-exchange is not affected by the ion concentration of the solution. However, in microwave-driven ionexchange, the concentration of the rare-earth ion solution has a significant effect on the extent of ion-exchange. Higher rare earth ion concentrations of the solution will lead to a higher degree of ion-exchange, which results in a stronger intensity of the emission spectrum. This is another advantage of the microwave driven ion exchange method. Compared with the traditional ion-exchange method, the exchange time in microwavedriven ion-exchange with fixed rare-earth ion concentration and microwave power level will affect the luminescence intensity of the exchanged product. The start-quenching time of Ce3+ luminescence concentration in microwave-driven ion-exchange is 8 min, as compared with 8 h in traditional ion exchange. Therefore, the exchange rate of the ions under microwave heating is much higher than that in the traditional ion-exchange process. This is the third advantage of the microwave-driven ion-exchange method. In a word, ion-exchange with the assistance of microwaves is feasible, convenient, and fast. It can reach a higher exchange degree than can traditional methods and make the inaccessible ions in traditional methods exchangeable. This method is especially appropriate for the laboratory preparation of ion-exchanged zeolite molecular sieves. The microwave technique is very successful in the synthesis of microporous crystals, modification of the properties of zeolites, secondary synthesis of microporous materials, and the preparation of ultra-fine particles and films, and has attracted the wide interest of chemists in the field of molecular sieves. 3.2.4

Hydrothermal Synthesis Approach in the Presence of Fluoride Source

Hydrothermal synthesis of microporous compounds in the presence of fluoride source refers to the hydrothermal or solvothermal crystallization of aluminosilicate zeolites or microporous aluminophosphate such as AlPO4-n series in the presence of a fluoride source. The successful introduction of fluoride ion into the hydrothermal or solvothermal synthesis of microporous materials paves the way for the introduction of other complex-ion or chelation agent’s to the hydrothermal crystallization of microporous compounds.

162

Chemistry of Zeolites and Related Porous Materials

Synthesis of Zeolites The most common mineralizing agent is the hydroxide ion, i.e. OH. The mineralizing agent can dissolve silica or other source of T-atoms and provides the reversible T O T hydrolysis, which is important for the formation of crystalline structures. In addition to OH, F can also be used as a mineralizing agent. The F was first introduced into the zeolite crystallization process as an alternative mineralizer of OH by E.M. Flanigen in the synthesis of zeolites under hydrothermal conditions. Flanigen’s studies indicated that the prue-silica materials could be prepared in F-containing syntheses. Subsequently, J.L. Guth and H. Kessler systematically studied this field and developed this synthetic approach.[86] The introduction of fluoride source allows the crystallization of zeolite molecular sieves under nearly neutral or acidic conditions. Compared with the hydrogel formed in hydroxide solutions, the resulting mixture is a slurry with a viscosity that depends on the type of SDA and the final water content. Numerous microporous compounds with known or new structures, such as MFI, FER, MTT, MIN, GIS, BEA, UTD-1, ITQ-3, ITQ-4, and TON, have been successfully synthesized from an F synthesis system. Recently, several all-silica and high-silica crystalline zeolites with low framework density, including CHA, BEA, ITE, and IFR, were obtained in a modified fluoride media route by Camblor et al.[87] In their studies, they found that the H2O/SiO2 ratios they used to produce low-framework-density zeolites are normally lower than those typically used in the hydrothermal synthesis of zeolites in F or OH medium, and sometimes approach the values found in the reagent rather than the solvent level. They concluded that the H2O/SiO2 ratio may have a strong influence on the phase selectivity of the crystallization, which can be used as a useful strategy to discover new silica zeolites, specially low density ones. Very recently, Corma et al. explored the use of fluoride anions in synthesis with much-lower-than-usual amounts of water, which resulted in a very concentrated hydrogel. This approach led to the discovery of a number of all-silica materials with low framework density and high microporosity.[88, 89] This had rarely been seen in silicate synthesis chemistry. The authors believed that the fluoride provides an important structure-determining factor under these concentrated conditions. Heteroatoms (B, Al, Fe, Ga, and Ti) may be incorporated into the framework of highsilica and all-silica materials in the presence of fluoride as well, giving rise to active acid catalysts. Usually, transition metal ions will hydrolyse to form hydroxide or oxide precipitates in a high-pH solution. Therefore, there is a limitation to the content of transition metals in heteroatom-substituted zeolites. However, this limitation can be significantly increased by using fluoride during the synthesis because fluoride can coordinate to the transition metal atoms to form stable complex, which will help transition metal atoms incorporate into the framework of zeolites. In hydroxide medium, according to Zones et al., the C/N ratio of the SDA is a critical parameter in determining the ability of the organic additive to produce high-silica zeolites.[90,91] Those SDAs with a C/N ratio between 11 and 15 work well in the synthesis of very high or pure silica zeolites. In the synthesis of pure silica zeolites in hydroxide medium, the charge on the SDAs need to be counterbalanced by Si O connectivity defects, which results in the formation of highly defective materials. However, in fluoride medium, charge balance of the SDAs is generally achieved by occluded fluoride, and the concentration of Si O or Si OH groups is typically very low. Thus, nearly perfect zeolite single crystals and those with few defects could be obtained from the fluoride-ion

Synthetic Chemistry of Microporous Compounds (I)-Fundamentals and Synthetic Routes

163

synthesis system. Owing to charge balance of the SDA and to the high pH of the synthetic system, silica materials prepared in OH medium using SDA cations usually have a large concentration of Q3 defect sites. By contrast, silica materials prepared in fluoride media have a small concentration of Q3 sites because the charge balance is generally achieved with occluded F, and the low synthesis pH of the synthetic system favors a more complete condensation of the silica. Usually, the sources of fluoride are NH4F, NH4HF2, or HF. Sometimes, a framework element compact fluoride such as (NH4)SiF6, AlF3H2O, or NH4BF4 could be used as a fluoride source as well. The synthesis of heteroatom-substituted silicalite-I and ZSM-5 are described below as two typical examples: Example 1: the synthesis of Silicalite-I Gel composition: 1.0 SiO2: 0.08 TPABr: 0.04 NH4F: 20 H2O; Crystallization temperature: 200  C; Crystallization time: 15 days; Fluoride source: NH4F; Product: perfect single crystals (95  80 mm), analysed composition: Si96O192 F4(TPA)4. Increasing the amount of NH4F, for example, NH4F/SiO2 ¼ 1, can decrease the crystallization time to as short as 2 days. Example 2: the synthesis of heteroatom-substituted ZSM-5 Pang et al., chemists at Jilin University, systematically studied the crystallization of heteroatom-substituted ZSM-5 [denoted M-ZSM-5, where M ¼ Ti(IV), B, Ga, Fe(III), Ni, Mn, Co, Zn, and Be(II)] in the presence of fluoride ions.[92] In their studies, an aerosil was used as the silicon source, TPABr was used as the template, NH4F was the fluoride source, and NH4TiF6, Ti(O2)F2, H3BO3, Ga(NO3)3, Fe2(SO4)3, or other soluble salts of Ni2+, Mn2+, Co2+, Zn2+, or Be2+ were used as heteroatom sources. The final composition of the parent mixture was 1 SiO2: (0
X) MpOq: (0.2
10) NH4F: (0.1
0.8) TPABr: (30
300) H2O, where MpOq is transition metal oxide [X ¼ 2 for M(IV) and M(II), and X ¼ 1 for M(III)]. The final gel with a pH range of 6
6.5 was usually heated at 170–190  C for 1–14 days. Because most elements can react with fluoride to form complex ions, soluble fluoride 2 complex ions such as MF3 6 (M ¼ Fe, Ga), MF6 (M ¼ Ti, Zr) could be formed in the presence of fluoride ions in a weakly acidic medium. During the hydrothermal process, these fluoride complex ions will react with water molecules to form oxyfluorinated 2 complex ions such as FeOF3 and TiOF2 to 4 4 , which can further react with SiOF4 finally form transition-metal-atom-substituted ZSM-5 (denoted M-ZSM-5). The introduction of fluoride source into the synthesis system overcomes the drawback that transition metal ions will form hydroxides in basic media, resulting in a phase separation in the product. The use of a fluoride source can help transition metal atoms enter the framework of ZSM-5. Compared with the synthesis in basic media, the crystallization of M-ZSM-5 in weakly acidic media in the presence of fluoride source has many advantages, such as the formation of large perfect or single crystals (10 – 80 mm) and one-step preparation of H-ZSM-5 by calcination of the as-synthesized product, because no strong base such as

164

Chemistry of Zeolites and Related Porous Materials

NaOH was involved in the synthesis. After systematic studies, W.Q. Pang and S.L. Qiu developed a general approach to growing large single crystals of zeolites and related microporous materials from fluorine ion synthesis systems. Later, J.L. Guth and W.Q. Pang expanded the fluoride source hydrothermal synthesis approach to the synthesis of microporous aluminophosphates and other metal phosphates. Synthesis of Microporous Aluminophosphates and other Metal Phosphates in the Presence of Fluoride Source The crystallization of microporous aluminophosphates and other metal phosphates in the presence of fluoride ion is similar to that of zeolites, i.e., F acts as mineralizer and coordination agent to affect the crystallization process and rate. Xu systematically studied the crystallization of CHA-type microporous SAPO-34 and CoAPO-34 in the presence of F and found that the induction period was significantly reduced to 1/3 of the original time due to the rapid formation of a small number of nuclei and a decreased crystal grow rate resulting in large perfect single crystals.[93] In the crystallization of microporous AlPO4-n and related heteroatom-containing derivatives, the introduction of F produces two new characteristics compared with the normal zeolite-synthesis system: 1) the formation of new structures and 2) the entrance of F into the framework. First, a large number of new structures were synthesized in the presence of a fluoride source. One remarkable example is the crystallization of the extra-large microporous gallophosphate cloverite. Cloverite is a microporous gallophosphate compound with 20-membered-ring channel and a composition of C7 H14 Nþ Þ24 j8 [F24Ga96P96O372(OH)24]8, where quinuclidine is C7H13N. The pore structure of cloverite is h100i20 4:0  13:2***jh100i8 3:8  3:8***. Cloverite crystals could be prepared by heating a gel formed by mixing phosphoric acid (85 wt% in water), gallium sulfate, hydrofluoric acid (40 wt%), and quinuclidine (Q) in the proportions Ga2O5: P2O5: HF: 80 H2O: 6Q at 150  C for 24 h. Besides cloverite, a large number of aluminophosphates and gallophosphates with new structures were synthesized in the presence of F. Secondly, F was incorporated into the framework of metal phosphates during the synthesis. In this case, F played the roles of mineralizer, structure director, and composition element of the framework. For example, F can enter a double-four-membered ring (denoted by D4R) or to join two Al atoms. The following Table 3.13 summarizes the location of F in selected molecular sieves. 3.2.5

Special Synthesis Approaches and Recent Progress

Secondary Synthesis of Microporous Materials Nearly all syntheses of zeolites and microporous aluminophosphates have limitations to gel composition and other parameters. For example, some zeolites with special compositions such as high-silica Y zeolite and low-silica ZSM-5 cannot be directly synthesized. A secondary framework modification is necessary for their preparation. For instance, dealuminization, isomorphous substitution of extraneous silicon for aluminum, and removal of the sodium process in Y zeolite are necessary to prepare ultra-stable zeolite Y (USY); isomorphous replacement of framework atoms of boron with aluminum in a presynthesized silicon–boron structure is often used to prepare some specific aluminosilicate zeolites that cannot be directly synthesized, such as Al-SSZ-24 (AFI) and Al-CIT-1. Secondary synthesis (post-treatment) will be discussed in detail in Chapter 6.

Synthetic Chemistry of Microporous Compounds (I)-Fundamentals and Synthetic Routes

165

Table 3.13 Location of F in selected molecular sieves[94] Type of IZA microporous code compound LTA

CLO CHA GIS GIS

AlPO AlPO CoAPO SAPO GaPO4 GaPO4 GaPO4 AlPO AlPO GaPO4

Composition

[AlPO4]968 RF264 H2O

[GaPO4]3RF |Qþ 24j8 [F24: Ga96P96O372(OH)24]8 [Al6P6O24F][C4H10NO] [Al2P2O8F][NH2Me2] ULM-16[Ga4(PO4)4F2] 1.5 C6H14N0.5 H2O0.5 H3O

Template1 TMAOH, diethanolamine Kryptofix 222 TMAOH, diethanolamine TMAOH, diethanolamine n-Pr2NH K+, 18-crown-6 quinuclidine (Q) morpholine Me2NH cyclohexylamine

Location of F center of D4R

center of D4R bridge in 4MR atom, bridge in 4MR atom, joined to Ga atoms

Crystal Transition Because most of the microporous open structures are metastable thermodynamically, many zeolites and microporous aluminophosphates can transfer to more stable structures under heating, which can be used to prepare some specific zeolite structures. For example, boron-substituted zeolite b was used to prepare boron-substituted microporous SSZ-24 via crystal transition. More examples come from the crystal transition of Li-Losod (LOS) to Li-cancrinite (CAN) in 0.1 mol/L NaOH solution at 353 K, and Na-EAB (EAB) to sodalite (SOD) at elevated temperature. However, K-EAB is very stable and cannot be converted into sodalite at elevated temperature (K+ stabilized the EAB structure). Completion of this transition only needs to break 1/12 of the T O T bonds and form some new bonds. Water molecules act as catalysts in this process. More solid-crystal transition examples come from aluminophosphate systems. For example, phase transitions of AlPO-21 to AlPO-25, VPI-5 to AlPO-8, and JDF-20 to AlPO-5 have been observed. Nonzeolitic structures can be transferred to some specific zeolitic structures under certain conditions as well. The first example is the heating of solid kanemite (NaH[Si2O5]3 H2O), a layered hydrated sodium silicate, at high temperature to form ZSM-5.[95] The structure of kanemite contains boat-form 6-membered-ring Q3 silicate layers, which are separated by octahedral [NaO6] via sharing corners and edges with both layers. Typically, tetrapropylammonium hydroxide was embedded into kanemite and dried. The resulting product was mixed with a pre-dried silicate gel, ground into fine powder, and pressed to tablets (1.0
1.5 mm in thickness). The tablets were first heated at 573 K in a sealed system for 69 h, followed by continuous heating in an open system at the same temperature for 2 more hours before calcination at 773 K to remove the organic species. The second example is the heating of hydrated layered silicate (inorganic layers were separated by organic templates) at 823 K to form pure ferrierite (FER). Powder X-ray diffraction analysis indicated that this hydrated layered silicate already contained the 2-dimensional sheet of FER.[96] MCM-22 can be formed in a similar way.[97] The third

166

Chemistry of Zeolites and Related Porous Materials

example comes from the formation of microporous AlPO4-5 by heating two-dimensional layered structures at high temperature.[98] Dry Gel Conversion In 1990, Xu et al. first reported the transformation of a dry aluminosilicate gel to crystalline MFI by contact with vapors of water and volatile amines, which was named dry gel conversion (DGC).[99] Since then, this method has been extensively studied and a large number of microporous materials with new compositions and structures were prepared. Generally, DGC can be divided into ‘vapor-phase transport (VPT)’ and ‘steamassisted conversion (SAC)’ according to the volatility of the SDAs. For volatile SDAs such as ethylenediamine, a mixture of water and SDA was poured into the bottom of the autoclave and then a dry gel, which does not contain any SDAs, was placed over the liquid surface. Water and SDAs were vaporized at elevated temperature (150
200  C), reached the dry gel, and initiated the crystallization, which was called VPT. Less volatile SDAs such as tetrapropylammonium hydroxide were usually involved in the dry gel. Only water steam is supplied during the reaction, which was called SAC. By using VPT synthesis, various types of zeolites such as MFI, FER, MOR can be prepared in the presence of vapors of water, ethylenediamine, and triethylamine at 180–200  C.[99–101] By using SAC synthesis, aluminosilicate and pure-silica BEA zeolite with a wide range of SiO2/Al2O3 ratios from 30 to infinity can be prepared,[102] which is in sharp contrast to the conventionally synthesized BEA zeolite with TEAOH as an SDA. Significantly different to the conventional hydrothermal synthesis, the SAC synthesis using TEAOH easily gives even pure silica BEA with a very fast crystallization rate. Generally, it is difficult to fully crystallize a parent gel even after 1 week in a conventional hydrothermal synthesis. In addition, it has been found that high-silica BEA prepared by the SAC method has much higher thermal stability than that synthesized by conventional hydrothermal synthesis.[103] Besides zeolite BEA, hexagonal faujasite (EMT) and cubic faujasite (FAU) can be prepared by using SAC synthesis methods and 1,4,7,10,13,16-hexaoxacyclooctadecane (18-crown-6) and 1,4,7,10,13-pentaoxacyclopentadecane (15-crown-5), respectively, as SDA,[104] Interestingly, field-emission SEM analyses of these EMT crystals indicated that the formation mechanism of EMT crystals in hydrothermal synthesis and in SAC synthesis may be different. The EMT crystals hydrothermally synthesized consisted of thin hexagonal plates and have obvious facets randomly aligned at the edges of plates, with a thickness of
10–20 nm. Such observation is in good agreement with a layer-bylayer growth mechanism.[105] However, EMT crystals prepared by the SAC method have very different morphological features. Stripes with steps less than 10 nm thick were observed on the surface of EMT crystals. In the SAC method, EMT crystals may grow by a spiral growth mechanism.[104] In addition to pure-silica zeolites, titaniumsilicates such as TS-1 and TS-2, Ti-BEA, borosilicates, zincosilicates, and microporous aluminophosphates can be synthesized by using the DGC (SAC) method as well.[106] Zeolite membranes can be prepared by using the DGC method[107,108] using ethylenediamine and triethylamine as SDAs. Generally, the porous support was immersed in the starting hydrogel. After being taken out from the hydrogel, the resulting thin gel layer formed on the support surface was dried and crystallized under SAC or VPT conditions.

Synthetic Chemistry of Microporous Compounds (I)-Fundamentals and Synthetic Routes

167

High-temperature Rapid Crystallization Synthesis Examples of high-temperature rapid crystallization synthesis are very rare. One example is the synthesis of AlPO4-5 by direct heating of aluminophosphate gel containing triethylamine template and triethylene glycol solvent to high temperature (e.g., 600  C, be careful!). The solid product obtained after the decomposition of organic species is AlPO4-5.[109] The whole process only took a couple of minutes. The product prepared by this method did not contain the organic template. The crystallization mechanism of this process is not clear as yet. Synthesizing Zeolites in Outer Space In order to synthesize larger, high purity, and perfect zeolite single crystals, scientists tried to carry out synthesis experiments in outer space, which offers a microgravity environment.[110] The crystallization rate of zeolites in a microgravity environment is slower than that under regular gravity due to a decreased mass-transport rate caused by a decreased convection rate under microgravity conditions. Therefore, the complete crystallization of zeolite structure in outer space will take a longer time than that on earth. In addition, the crystalline product synthesized in outer space will not aggregate together and deposit at the bottom of the reactor. Synthesizing Zeolites in Ionic Liquids Computational enumeration predicts that several thousand zeolite structures are possible.[111] However, only about 179 zeolite framework types have been discovered so far.[32] Therefore, the improvement associated with synthesizing new porous materials together with new methods of preparing them continue to be of great importance. Normally these materials are prepared hydrothermally or solvothermally with water or organic chemicals as the solvent in a sealed autoclave under autogenous pressure at an elevated temperature. The reaction mixture usually includes an organic template or SDA as well as solvent. In contrast to conventional preparation methods of the porous materials, Morris and coworkers recently reported a novel method for the synthesis of zeolitic materials, which was named ionothermal synthesis.[112] In ionothermal synthesis, porous materials were synthesized in an ionic liquid or eutectic mixture, commonly defined as salts that are fluid at near-ambient temperatures and which consist of predominantly ionic species. Because of the vanishingly low vapour pressure of ionic liquids or eutectic mixtures, the synthesis took place at ambient pressure, which eliminated the safety concerns associated with high autogenous pressure generated during the hydrothermal or solvothermal synthesis process. The ionic liquids acted as both solvent and template in the synthesis. By using this novel synthesis method, Morris and coworkers successfully synthesized a couple of aluminophosphate zeolite analogues (SIZ-1-SIZ-5 and AlPO-CJ2) both in and without the presence of fluoride. Their studies indicated further that synthesis of the framework relied on the solvent’s being predominantly ionic because they found that sufficient quantities of molecular water disrupt the reactions by preventing the formation of the porous aluminophosphates. According to the fact that the ionic liquid solubilized the starting materials almost completely at the reaction temperatures, it was concluded that the synthesis mechanism is a solution-mediated mechanism rather than a solid-to-solid transformation. Subsequently, Xu et al. studied the microwave-enhanced ionothermal synthesis of aluminophosphate molecular sieves.[113] By using the ionic liquid 1-ethyl-3-

168

Chemistry of Zeolites and Related Porous Materials

methylimidazolium bromide and in the presence of fluoride, they successfully synthesized aluminophosphate molecular sieve AlPO4-11 and silicoaluminophosphate molecular sieve SAPO-11. Their results indicated that the presence of fluoride was a key factor in the formation of AEL-type aluminophosphate molecular sieves and that fluoride played an important role in ionothermal synthesis and affected both the yield of the crystalline product and its crystallinity. Very recently, Tian and coworkers studied the structure-directing role of amines in the ionothermal synthesis of AFI-type molecular sieves in 1-butyl-3-methylimidazolium bormide ([bmim]Br) ionic liquid.[114] Their results indicated that the original crystallization process can be altered by the addition of amine. The 1H NMR spectra of [bmim]Br and amine liquid mixtures revealed that the presence of amine caused a downfield shift of the signal corresponding to the proton of the imidazolium ring, which was considered to be strong evidence for the formation of hydrogen bonds between the ionic liquid and the added amine. The use of ionic liquids and eutectic mixtures as solvent and template may open up many new possibilities in the preparation of porous materials. A New Synthesis Method for Zeolites To avoid the high-temperature calcination usually used to remove organic SDA within the as-synthesized zeolites, Davis and coworkers developed a new methodology for the synthesis of zeolite and molecular sieves that utilized SDA that can be cleaved within zeolite pore spaces for easy removal of these fragements via ion-exchange.[115] The schematic diagram of this new methodology is shown in Figure 3.23. The key feature of this new methodology lies in the fact that the organic SDA can be cleaved into smaller pieces by the reaction condition changes such as pH and be easily recombined into the original molecule, i.e., this organic molecule is cleaved inside the zeolite pores after zeolite synthesis, the pieces could be easily removed via ion-exchange, and then the fragments are recombined into the original SDA for further zeolite synthesis. By using this new methodology and a dimethylated ammonium derivative of 1,4dioxa-8-azaspiro[4,5]decane as the SDA, they successfully synthesized ZSM-5. The organic molecule is cleaved inside the zeolite pores of the as-synthesized ZSM-5, and the fragments are easily removed via ion-exchange. This method provides an alternative to the removal of the organic SDA occluded in the framework of porous materials, which can overcome some drawbacks of high-temperature calcination such as the release of NOx resulting from the combustion of organic SDAs (most SDAs are quaternary ammonium cations or amines). Recycling of the SDAs can significantly decrease the overall zeolite-synthesis cost. 3.2.6

Application of Combinatorial Synthesis Approach and Technology in the Preparation of Microporous Compounds

Combinatorial chemistry is one of the important achievements in science in recent years.[116] In the combinatorial synthesis approach, the traditional synthesis rules are replaced by reliable reactions and simple separation methods, allowing the preparation of a large number of products at the same time under the same reaction conditions in a specific reactor, together with rapid analysis and high-throughput screening processes.

Synthetic Chemistry of Microporous Compounds (I)-Fundamentals and Synthetic Routes

169

Figure 3.23 Schematic representation of the new methodology for synthesis of zeolites and molecular sieves. Reproduced with permission from [115]. Copyright (2004) Elsevier

Conventional synthesis : R þ S ! RS

ðone productÞ

Combinatorial synthesis : ½Ri i¼l;n þ ½Sj j¼l;m ! ½Ri Sj i¼l;n;j¼l;m ðn  m productsÞ

ð3:3Þ ð3:4Þ

Another example of progress in combinatorial chemistry is the application of the combinatorial approach in the hydrothermal synthesis of zeolites and microporous compounds. In 1998, Akporiaye firstly invented a multi-autoclave hydrothermal synthesis reactor (Figure 3.24), which contains many independent mini-autoclaves with a volume of 50 mL
0.5 mL. Chemists at the State Key Laboratory of Inorganic Synthesis and Preparative Chemistry located at Jilin University improved this apparatus by adding an autosampler (Figure 3.25). Studies on the system Na2O–Al2O3–SiO2–H2O indicated

Figure 3.24 Multi-autoclave hydrothermal synthesis reactors. Reproduced with permission from [118]. Copyright (1998) Wiley-VCH

170

Chemistry of Zeolites and Related Porous Materials

Figure 3.25 Multi-autoclave hydrothermal synthesis reactor equipped with autosampler (State Key Laboratory of Inorganic Synthesis and Preparative Chemistry, Jilin University)

that the efficiency of hydrothermal synthesis was increased by two order of magnitude due to the use of a combinatorial approach. Using a home-made multi-autoclave hydrothermal synthesis reactor, J. Klein et al. studied the synthesis of zeolite TS-1 in 1998.[117] Analysis of the products was directly carried out on the library substrates. Later, T. Bein and co-workers studied the crystallization of aluminophosphates from the system 1.0 Al2O31.0 P2O5x [Co(Cp)2]OHy Pr2NH190 H2O. Recently, Yu and coworkers were the first to apply combinatorial synthesis to the system 1.0 Zn(OAc)2–x H3PO4–y N,N 0 -dimethylpiperazine–H2O and discovered three new phases of microporous zinc phosphates. These studies will be described in the following paragraphs. Example 1: Combinatorial synthesis studies on the system Na2O–Al2O3–SiO2– H2O[118] In 1998, Akporiaye first reported the crystallization field diagram (Figure 3.27) of the system of Na2O–Al2O3–SiO2–H2O (90
98 mol% H2O) obtained by using a combinatorial approach. The apparatus they used is shown in Figure 3.24. The sample volume in each micro-autoclave is about 0.5 mL and the crystallization was performed at 100  C. They compared their diagram to that of a sample obtained from the same system under the same crystallization conditions but using conventional synthesis and published in 1974 as shown in Figure 3.26 (D.W. Breck, Ref. [8]). The main parts of these two crystallization field diagrams are similar. However, a little difference in the location of the crystallization field for each phase could be found. Sodalite phase appeared in the Na2O-rich area in the field diagram obtained by using the combinatorial approach. Because most of the above-mentioned crystalline phases are metastable, a little change in the crystallization field diagram is reasonable due to the slightly changed crystallization conditions. In view of the high synthetic efficiency of the combinatorial approach, Akporiaye and coworkers expanded their studies to the system R2O–Na2O– Al2O3–SiO2–H2O (R ¼ [TMA2O, Cs2O, Li2O ¼ 0.09 mol%], Na2O ¼ 0
1.0 mol%) and obtained some interesting results. Example 2: Combinatorial synthesis studies on the system 1.0 Al2O31.0 P2O5 x [Co(Cp)2]OHy Pr2NH190 H2O[119]

Synthetic Chemistry of Microporous Compounds (I)-Fundamentals and Synthetic Routes

171

Figure 3.26 Crystallization field diagram of Na2O–Al2O3 –SiO2 –H2O (90
98 mol% H2O) at 100  C GME: gmelinite; FAU: faujasite; A: zeolite A; CHA: chabazite. Reproduced with permission from [8]. Copyright (1974) John Wiley & Sons, Inc.

Using a combinatorial synthesis approach, T. Bein and coworkers systematically studied the crystallization of AlPO4 in the presence of two templates of [Co(Cp)2]OH and Pr2NH at 160  C for 24 h and the results are summarized in Figure 3.28. Example 3 Combinatorial synthesis studies on the system ZnO–P2O5–N,N 0 dimethylpiperazine–H2O[120] In 2002, J. Yu and coworkers applied the combinatorial approach to this system by using an improved 64-autoclave (800 mL for each micro-autoclave) hydrothermal

Figure 3.27 Crystallization field diagram of Na2O–Al2O3–SiO2–H2O (90
98 mol% H2O) at 100  C obtained by using combinatorial approach. GME: gmelinite; FAU: faujasite; A: zeolite A; SOD: sodalite. Reproduced with permission from [118]. Copyright (1998) Wiley-VCH

172

Chemistry of Zeolites and Related Porous Materials

Figure 3.28 Crystallization phase diagram of system 1.0 Al2O3  1.0 P2O5  x [Co(Cp)2]OH  y Pr2NH  190 H2O in the presence of two templates at 160  C.  AFI; $ AEL; & noncrystalline; AFI/AST; AFI/tridymite; ~ tridymite. Reproduced with permission from [119]. Copyright (1999) Wiley-VCH

synthesis reactor. The crystallization was carried out at 180  C for 60 h. Two new microporous ZnPO structures were discovered from this system. One is [Zn6P5O20 (H2O)]0.5 C6H16N2C5H14N23 H2O with a 16-membered-ring channel system, and the other is [Zn5P4O20(H2O)]C4H14N2 with a 10-membered-ring channel system. A combinatorial synthesis approach provides a new method for the systematic hydrothermal synthesis of microporous compounds and new molecular sieve structures in a highly efficient way. In combinatorial studies of the above-described systems, all systematic characterizations of the product were performed by a computer-controlled xyz stage GADDS microdiffractor. Further characterization of the structure and properties of the product was carried out at other facilities.

3.3 Typical Synthetic Procedures for some Important Molecular Sieves So far, a number of molecular sieves have been widely used in catalysis and other fields of science and advanced technologies. The methods, techniques, and procedures involved in the synthesis of these molecular sieves are very typical and representative. Therefore, in this section, we will discuss the synthetic procedures for these important molecular sieves. The procedures described here are mainly selected from Ref. [121]. 3.3.1

Linde Type A (LTA)

Linde-A (LTA) molecular sieve is characterized by a 3-dimensional system of channels parallel to all crystallographically equivalent axes of the cubic structure ð< 100 >Þ, i.e., ˚  4:1 A ˚ ***). The pore along x, y and z with circular 8-ring apertures (< 100 > 8 4:1 A ˚ diameter is defined by an 8-membered oxygen ring and is small at 4.1 A, which leads into ˚ . The cavity is surrounded by eight a larger cavity of minimum free diameter 11.4 A sodalite cages (truncated octahedra) connected by their square faces in a cubic structure. The crystallographic and chemical data for the framework of LTA are cubic, Pm 3m

Synthetic Chemistry of Microporous Compounds (I)-Fundamentals and Synthetic Routes

173

˚ and for the type material are jNaþ ðH2 OÞ j ½Al12 Si12 O48  , cubic, (# 221), a ¼ 11:919 A 12 27 8 8 ˚ (Relationship to unit cell of Framework Type:  Fm3c (# 226), a ¼ 24:61 A a0 ¼ b0 ¼ c0 ¼ 2a). Kþ- and Ca2þ-exchanged Linde-A molecular sieves are named 3A and 5A, respectively. Owing to its unique channel system and high ion-exchange capacity, Linde-A is one of the most widely used molecular sieves in the production of detergents, the drying and cleaning of gases, separation of oxygen and nitrogen from air, etc. Linde-A molecular sieves could crystallize from a wide range of gel compositions at 60–110  C. The formation of LTA structure is not sensitive to source materials of silicon and aluminum. For example, fumed SiO2, sodium silicate, Ludox, or silica gels could be used as silicon source; pure aluminum powder, aluminum wire, or sodium aluminate could be used as aluminum source for a successful synthesis. The laboratory synthetic procedure described here was developed by R.W. Thompson and M.J. Huber.[122] Deionized water, sodium hydroxide (NaOH), sodium aluminate (Na2O  Al2O33 H2O), and sodium metasilicate (Na2SiO35 H2O) were used to prepare the reaction gel with a composition of 3.165 Na2O: Al2O3: 1.926 SiO2: 128 H2O. The detailed procedure for the preparation of 10 g of dry Linde-A molecular sieve is described below: (1) Mix 80 mL of deionized water and 0.723 g of NaOH, stir the mixture gently for 10– 20 min until NaOH is completely dissolved. Divide the solution into two equal volumes in autoclavable polypropylene bottles with a capacity of 100 or 150 mL; (2) Add 8.258 g of sodium aluminate to one bottle, cap the bottle, and stir the mixture for 10–20 min until clear; (3) Add 15.48 g of sodium metasilicate to the other bottle, cap the bottle, and stir the mixture for 10–20 min until clear; (4) Pour silicate solution (3) into aluminate solution (2) quickly and a thick gel should form immediately. Cap the bottle tightly and stir the mixture until homogeneous. The capped autoclavable polypropylene bottle with the resulting gel was heated at 99 1  C for 3–4 h and cooled to ambient temperature. The product was filtered off, and washed with deionized water until the filtrate pH was below 9. The solid product on filter paper was transferred to a watch glass and dried at 80–110  C overnight. Approximately 10.4 g of dry Linde-A with a size of 2–3 mm and a cubic shape could be obtained. The XRD pattern gave characteristic strong reflections of Linde-A at d ¼ 4:107, 3.714, 3.293, ˚ . Elemental analysis of the solid product showed a composition of and 2.987 A Na2OAl2O32 SiO2. 3.3.2

Faujasite (FAU)

Faujasite (FAU) is characterized by a 3-dimensional pore structure with pores running perpendicular to each other in the x, y, and z planes similar to LTA, with circular 12-ring ˚  7:4 A ˚ ***). The pore diameter is large at 7.4 A ˚ since the apertures (< 100 > 12 7:4 A aperture is defined by a 12-membered oxygen ring, and leads into a larger cavity of diameter ˚ . The cavity is surrounded by ten sodalite cages connected on their hexagonal faces. 12 A ˚ The crystallographic data for the framework of FAU are cubic, Pm3m (# 227), a ¼ 24:3 A. Linde type X and Y zeolites are two typical microporous materials with the FAU framework. So far, both low-silica type X and high-silica type Y molecular sieves are

174

Chemistry of Zeolites and Related Porous Materials

most widely used catalysts for oil refining. Moreover, low-silica type X is also a very important adsorbent for gas separation and cleaning. Low-Silica Type X (LSX) Molecular Sieve: 0.77 Na2O  0.23 K2O  Al2O3  2.04 SiO2 The Si/Al ratio for X-type molecular sieve normally falls in the range 1–1.5. The laboratory synthetic procedure described here was developed by Ku¨hl.[123] Distilled water, sodium aluminate (45.6% Al2O3, 29.65% Na2O), sodium hydroxide (reagent grade, usually
97% NaOH), potassium hydroxide (reagent grade, usually
86% KOH), and sodium silicate solution (28.7% SiO2, 8.9% Na2O) were used to prepare the reaction gel with a composition of 5.5 Na2O: 1.65 K2O: Al2O3: 2.2 SiO2: 122 H2O. The detailed procedure for the preparation of 29 g of LSX molecular sieve dry product is described below: (1) Add 22.37 g of sodium aluminate to 30 g of distilled water in an autoclavable polypropylene bottle, and stir the solution until sodium aluminate is completely dissolved; (2) Add 21.53 g of potassium hydroxide and 31.09 g of sodium hydroxide to 70 g of distilled water in another autoclavable polypropylene bottle, and stir the solution until KOH and NaOH are completely dissolved; (3) Mix solutions (1) and (2) thoroughly; (4) Add 46.0 g of sodium silicate solution to 71.8 g of distilled water in the third autoclavable polypropylene bottle and blend them well; (5) Mix solutions (3) and (4) thoroughly. The resulting solution was sealed in an autoclavable polypropylene or Teflon bottle and incubated at 70  C for 3 h followed by crystallization at 93–100  C for 2 h. The product solution was diluted with the addition of distilled water, filtered, washed with 0.01 mol/L NaOH solution, and dried at ambient temperature or 110–125  C. The solid product was multi-faceted spherulites of 2–6 mm diameter with 111 faces exposed. XRD characterization confirmed the FAU structure of the product, and elemental analysis suggested a composition of 0.77 Na2O  0.23 K2O  Al2O3  2.04 SiO2, which gave a Si/ Al ratio of 1.02. To further decrease the Si/Al ratio of LSX-type molecular sieve, Norby et al.[124] mixed Na-LTA zeolites and LiCl solution (10 g Na-LTA zeolites and 10 g LiCl dissolved in 150 mL water) and crystallized this solution in a Teflon-lined stainless steel autoclave at 200–250  C for 72 h. The resulting product was pure low-silica X-type molecular sieve (Li1.02Na0.004AlSiO4 . 1.1 H2O) with an Si/Al ratio of 1. Linde-Y Type Molecular Sieve: Na56[Al56Si136O384]  250 H2O Unlike X type molecular sieve, the Si/Al ratio of Y type molecular sieve normally falls in the range 1.5–3. The laboratory synthetic method introduced here was developed by D.M. Ginter, A.T. Bell, and C.J. Radke (Ref. [125]). Deionized water, sodium aluminate (Na/Al ¼ 1.27, 6.1% H2O), sodium hydroxide pellets, and sodium silicate solution (28.7% SiO2, 8.9% Na2O) were used to prepare the reaction gel with a composition of 4.62 Na2O: Al2O3 : 10 SiO2 : 180 H2O. The detailed procedure for the preparation of 32 g of dry product is described below: (I) The preparation of a colloidal seed with a composition of (5% of Al): 10.67 Na2O: Al2O3 : 10 SiO2 : 180 H2O

Synthetic Chemistry of Microporous Compounds (I)-Fundamentals and Synthetic Routes

175

(1) Mix 4.07 g of sodium hydroxide, 2.09 g of sodium aluminate, and 19.95 g of water in a 50 mL plastic bottle, and stir the mixture until the solid is completely dissolved; (2) Add 22.72 g of sodium silicate solution to the solution in (1), stir moderately for at least 10 min, cap the bottle, and let the colloid age at ambient temperature for 1 day without stirring; (II) The preparation of feedstock gel with a composition of (95% of Al): 4.30 Na2O : Al2O3 : 10 SiO2 : 180 H2O (3) Mix 0.14 g of sodium hydroxide, 13.09 g of sodium aluminate, and 130.97 g of deionized water in a 500 mL plastic beaker, and stir the mixture until the solid is completely dissolved; (4) Add 142.43 g of sodium silicate solution into a beaker, stir vigorously until the gel appears somewhat smooth, then cover the beaker with a watch glass; (III) The preparation of overall gel with a composition of 4.62 Na2O: Al2O3 : 10 SiO2 : 180 H2O (5) Slowly add the prepared colloidal seed (2) to the feedstock gel (4) with vigorous stirring and keep stirring for 20 min after the complete addition of seed gel; Finally, the overall gel was transferred into a 300 mL autoclavable polypropylene bottle (sealed) and aged at ambient temperature for 1 day. After that, the sealed bottle was put into a 100  C oven for further crystallization until the gel separated into a solid and a clear supernatant liquid indicating a complete crystallization (Note: this process normally takes about 5–7 hours. Longer heating times, i.e., more than 7 h, will lead to the formation of competing phases of GIS, etc.). The solid product could be recovered by centrifugation or filtration. The wet product was washed with distilled water until the pH of centrifugatate or filtrate was below 9, and dried at 110  C. Approximately 32 g of dry NaY with a particle size of <1 mm and octahedral shape could be obtained. Powder XRD pattern of the solid product showed three characteristic strong reflections of the FAU ˚ and a0 ¼ 24:72 A. ˚ Elemental analysis suggested a structure at d ¼ 14.28, 8.75 and 7.46 A composition of NaAlO22.43 SiO2. Normally, it is difficult to get high silica type Y zeolite with an Si/Al ratio >3 by using a routine synthetic protocol. So far, the only way to synthesize Y zeolite with Si/Al > 3 was developed by Chatelain, et al.[126] They successfully prepared Y zeolite with an Si/Al ratio as high as 3.8 or higher by treating a gel with the composition 2.1 Na2O : 10 SiO2 : Al2O3 : 0.5 (15-crown-5): 100 H2O at 110  C for 8 days. 3.3.3

Mordenite (MOR)

Mordenite is characterized by sinusoidal channels (with limiting 8-ring windows) parallel to [010] that intersect with 1-dimensional 8- and somewhat elliptical 12-ring channels parallel to [001], which could be described by a shorthand notation of [001] 12 6:5  7:0 $ f½0108 3:4  4:8 $ ½001 8 2:6  5:7g*. The unit cell is orthorhombic ˚ with Cmcm symmetry. ða ¼ 18:3; b ¼ 20:5; c ¼ 7:5 AÞ Mordenite ðjNaþ8 ðH2 OÞ24 j ½Al8 Si40 O96 Þ is a very important catalysis, adsorption, and separation material, widely used in petroleum refining and the fine-chemicals industry.

176

Chemistry of Zeolites and Related Porous Materials

The laboratory synthetic procedure described here was developed by Kim and Ahn.[127] Sodium hydroxide, sodium aluminate (32.6% Na2O, 35.7% Al2O3), silica powder (91.8% SiO2, 8.2% H2O, Note: sodium silicate can also be used as a silica source, but crystallization rates are lower), and distilled water were used to prepare the reaction gel with a composition of 6 Na2O : Al2O3 : 30 SiO2 : 780 H2O. The detailed procedure for the preparation of 56 g of anhydrous product is described below: (1) Add 19.0 g of NaOH to 40 g of water, and stir the solution until clear; (2) Add 14.3 g of sodium aluminate to solution (1), and stir until the solid is completely dissolved; (3) Add 645 g of H2O to solution (2); (4) Add 98.2 g of silica powder, and stir the mixture for 30 min. The final mixture was loaded into a Teflon-lined stainless steel autoclave and heated at 170  C for 24 h. The solid product was filtered off, washed with distilled water until the pH of filtrate was below 10, and dried at 110  C. Approximately 56 g of anhydrous mordenite with a size of
5 mm and irregular spherical to prismatic shape could be obtained. Powder XRD pattern of the solid product showed five characteristic strong ˚ . Elemental reflections of the MOR structure at d ¼ 3:45, 3.97, 9.02, 3.27 and 3.21 A analysis suggested a formula of Na2O : Al2O3 : 17.2 SiO2. 3.3.4 ZSM-5 (MFI) ZSM-5 is a highly porous material and throughout its structure it has an intersecting twodimensional pore structure. ZSM-5 has two types of pores, both formed by 10-membered oxygen rings. The first type of these pores is straight and elliptical in cross section; the second type of pores intersects the straight pores at right angles, in a zig-zag pattern and is circular in cross section, which could be described by a shorthand notation of {[100] 10 5:1  5:5 $ ½010 10 5:3  5:6}***. The unit cell of MFI structure is orthorhombic ˚ with Pnma symmetry. (a ¼ 20:1, b ¼ 19:7, c ¼ 13:1 A) The aluminosilicate molecular sieve (jNaþn ðH2 OÞ16 j ½Aln Si96n O192 , n < 27) with MFI structure could have a very high silica-to-alumina ratio. The all-silica analogue of zeolite ZSM-5, silicalite-1, can form as well. The substitution of an aluminum ion (charge 3+) for a silicon ion (charge 4+) requires the additional presence of a positive charge (i.e., Hþ) to balance the negative charge of the framework. This additional proton gives the zeolite a high level of acidity (HZSM-5), which causes its activity. The acidic catalytical activity of HZSM-5 strongly depends on the Al component in the framework. ZSM-5-type molecular sieve is one of the most important molecular sieve catalysis materials in the petroleum industry. Like widely used LTA, FAU, and MOR molecular sieves, ZSM-5-related molecular sieves have been industrially produced for long time. In the recent decade, more than 100 patents and papers related to the synthetic method and process of ZSM-5-type molecular sieve were published each year. The laboratory synthetic procedure introduced here was developed by H. Lechert and R. Kleinwort (Ref. [121], p. 198). Sodium hydroxide, tetrapropylammonium hydroxide (20% solution), silicic acid (Merck, technical grade, SiO2 : 0.5 H2O), sodium aluminate (Roth, Al2O3 : 1.24 Na2O : 0.57 H2O), and distilled water were used to prepare the reaction gel

Synthetic Chemistry of Microporous Compounds (I)-Fundamentals and Synthetic Routes

177

with a composition of 3.25 Na2O : Al2O3 : 30 SiO2 : 958 H2O. The detailed procedure is described below: (I) Preparation of the colloidal seed (1) Mix 710.3 g of water, 13.8 g of sodium hydroxide, and 117.0 g of TPAOH solution, and stir the mixture until the solid is completely dissolved; (2) Add 158.9 g of silicic acid to solution (1) in portions under stirring; shake the resulting mixture for 1 h at ambient temperature followed by aging of the mixture at 100  C for 16 h; (II) The synthesis of ZSM-5 (3) Mix 867.8 g of water, 8.8 g of sodium hydroxide, and 10.3 g of sodium aluminate, and stir the mixture until the solid is completely dissolved; (4) Add 113.1 g of silicic acid to solution (3) in portions under stirring. Shake the resulting mixture vigorously for 1 h at ambient temperature; (5) Add 50 g of seeding gel prepared in mixture (2) to solution (4), and shake for 1 h. The final mixture was loaded into a 50 mL Teflon-lined stainless steel autoclave and heated at 180  C for 40 h without agitation. The solid product was filtered off, washed with distilled water, and dried at 105  C. The product was uniform small crystals with a size of
6 mm. Powder XRD pattern characterization of the solid product showed a fully crystalline MFI structure. Elemental analysis gave an Si/Al ratio of 12–13.5, which is tunable by adjusting the ratio of source materials in the precursor gel. The absence of Al species in the present synthetic system leads to the formation of the high-quality all-silica ZSM-5-type molecular sieve termed silicalite-I. Moreover, high-quality [B]-ZSM-5, [Fe]-ZSM-5, or [Ti]-ZSM-5-type molecular sieves could be obtained by the addition of B2O3, Fe2(SO4)3, or Ti(OC2H5)4 to the synthetic system of silicalite-I and crystallization at 200, 180, or 175  C for several days, respectively. 3.3.5

Zeolite Beta (BEA)

Zeolite beta ðjNaþ7 j½Al7 Si57 O128 Þ has a three-dimensional network of 12-ring pores, which could be presented by a short notation of < 100 > 12 6:6  6:7

$ ½001 12 ˚ with 5:6  5:6*. The unit cell of BEA structure is tetragonal ða ¼ 12:661; c ¼ 26:406 AÞ P4122 symmetry. Zeolite beta has been widely used in the petroleum refining and fine chemical industries due to its high thermal stability and strong acidity. Like LTA-, FAU-, MOR-, and MFI-type molecular sieves, zeolite beta molecular sieve has been industrially produced for a long time. The laboratory synthetic procedure described here was developed by M.A. Camblor and J. Pe´rez-Pariente.[128] Tetraethylammonium hydroxide (Alfa 40 wt% TEAOH), sodium chloride, potassium chloride, silica, sodium hydroxide, sodium aluminate, and deionized water were used to prepare the reaction gel with a composition of 1.97 Na2O : 1.00 K2O : 12.5 (TEA)2O : Al2O3 : 50 SiO2 : 750 H2O : 2.9 HCl. The detailed procedure for the preparation of 20 g of anhydrous product is described below: (1) Mix 89.6 g of TEAOH (40%), 0.53 g of sodium chloride, 1.44 g of potassium chloride, and 59.4 g of water, and stir the mixture until the solid is completely dissolved; (2) Add 29.54 g of silica to solution (1), and stir until homogenized;

178

Chemistry of Zeolites and Related Porous Materials

(3) Mix 0.33 g of sodium hydroxide, 1.79 g of sodium aluminate, and 20.0 g of water, and stir the mixture until the solid is completely dissolved; (4) Mix solutions (2) and (3), and stir for 10 min until a thick gel forms. The thick gel was loaded into a 60 mL stainless steel autoclave with Teflon lining and heated at 135 1  C for 15–20 h. After crystallization, the hot autoclave was quenched in cold water. The solid product was separated on a high-speed centrifuge (10 000 rpm), washed with deionized water until the pH of the centrifugate was less than 9, and dried at 77  C overnight. Uniform round-shaped small crystals with a narrow particle size distribution of 0.1–0.3 mm were obtained. Powder XRD pattern characterization showed a single phase of zeolite beta. Elemental analysis suggested a formula of Na0.92K0.62 (TEA)7.6[Al4.53Si59.47O128] per unit cell (the excess of cations, assumed to be occluded TEAOH or TEAþ, compensates for the SiO-structure defects). 3.3.6

Linde Type L (LTL)

Linde type L zeolite is characterized by a 1-dimensional system of channels parallel to [001] or c with circular 12-ring apertures, which could be presented by a short notation of ˚ [001] 12 7:1  7:1*. The unit cell of LTL structure is hexagonal (a ¼ 18:40, c ¼ 7:52 A) with P6=mmm symmetry. The laboratory synthetic method introduced here was developed by J.P. Verduijn.[129] Potassium hydroxide (pellets, 86.8% KOH), alumina [92.6% Al(OH)3], silica sol (Ludox HS-40, 40% SiO2), magnesium nitrate [Mg(NO3)2  6 H2O], and deionized water were used to prepare the reaction gel with a composition of 2.35 K2O : Al2O3 : 10 SiO2 : 160 H2O: trace MgO. The detailed procedure for the preparation of 59 g of anhydrous product is described below: (1) Heat a mixture of 50.00 g of water, 30.39 g of potassium hydroxide, and 15.82 g of alumina to boiling until the solution is clear. Cool the solution to room temperature and correct water loss due to boiling; (2) Mix 150.24 g of silica sol, 99.0 g of water, and 14.5 g of Mg(NO3)2 until the solution is homogeneous; (3) Add 25 g of (rinse) water to the mixture of (1) and (2), and mix until a gel forms (
3 min). Before the gel became fully stiff, it was transferred to a 300 mL stainless steel autoclave and held at 175  C for 48 h without agitation to bring about crystallization. After that, the autoclave was allowed to cool to ambient temperature. The solid product was recovered by filteration, washed 5 times with 650 mL of deionized water (the pH of the last filtrate should be
10), and dried at 150  C for 16 h. Approximately 59 g of anhydrous product with crystals of a cylindrical shape and a particle size of 0.2–0.4 mm diameter and 0.4–0.7 mm length were obtained. Powder XRD pattern characterization showed a single phase of LTL-type molecular sieve. Elemental analysis suggested a composition of 6.2 SiO2/Al2O3, 1.0 K2O/Al2O3. 3.3.7

AlPO4-5 (AFI)

AlPO4-5 type molecular sieve is characterized by a 1-dimensional system of channels parallel to [001] or c with circular 12-ring apertures, which could be presented by a short

Synthetic Chemistry of Microporous Compounds (I)-Fundamentals and Synthetic Routes

179

notation of [001] 12 7:3  7:3*. The unit cell of AFI structure is hexagonal (a ¼ 13:8, ˚ with P6cc symmetry. c ¼ 8:6 A) Different from the above discussed aluminosilicate molecular sieves, microporous aluminophosphate and related materials were normally synthesized under slight acidic or neutral conditions in the presence of organic additives such as organic amine or quaternary ammonium salt. It is commonly accepted that these organic additives normally play three roles in the crystallization of inorganic frameworks of aluminophosphates: 1) templating role; 2) SDA role; and 3) filling role. So far, it is still a big challenge to clarify the nature of these three roles and the correlation among them even though many molecular sieve chemists in this field are working on this topic. For example, as many as 85 different amines can lead to the formation of the framework of AlPO4-5 (AFI), but the essential correlation among these amines has not been elucidated so far. Therefore, a general explanation for the role that the organic amines played in the formation of AlPO4-5 framework is still not available. This issue will be further discussed in Chapter 5. Many procedures for the synthesis of high quality AlPO4-5 in the laboratory have been developed by different molecular sieve research groups worldwide. The procedure described here was developed by Girnus et al.[130] Orthophosphoric acid (85 wt% H3PO4), triethylamine, aluminum triisopropylate Al(OiC3H7)3, hydrofluoric acid (40 wt% HF in water), and deionized water were used to prepare the reaction gel with a composition of Al2O3 : 1.3 P2O5 : 1.6 TEA : 1.3 HF : 425 H2O : 6 C3H7OH (Note: other Al sources such as pseudoboehmite and Al hydroxide also give good large crystals). The detailed procedure for the preparation of 3 g of anhydrous product is described below: (1) Mix 7 g of water and 3.84 g of phosphoric acid; (2) Add 2.07 g of triethylamine dropwise to solution (1) and mix; (3) Add 5.23 g of aluminum triisopropylate in small amounts at 0  C with intense stirring, then stir the mixture at room temperature for 2 h; (4) Mix 0.83 g of hydrofluoric acid and 89.2 g of water; (5) Mix solutions (3) and (4) and stir for 2 h. The resulting gel was loaded into a 150 mL Teflon-lined stainless steel autoclave and crystallized at 180  C for 6 h without agitation (Note: for microwave synthesis: hold for 15 minutes at 180  C with a heating rate of 4  C/s). The solid product was filtered off, washed four times with 100 mL of deionized water, dried, and calcined at 600  C in air until the product was colorless (white). The single crystals of AlPO4-5 had the appearance of hexagonal columns with a size of up to 50 mm in length. Powder XRD pattern of the solid product showed six characteristic strong reflections of AFI structure at ˚ . Elemental analysis suggested a composid ¼ 11:90, 5.93, 4.48, 4.24, 3.96, and 3.42 A tion of 42.9 wt% P2O5, 30.5 wt% Al2O3 (P/Al ¼ 1.00). 3.3.8

AlPO4-11 (AEL)

AlPO4-11-type molecular sieve is characterized by a 1-dimensional system of channels parallel to [001] or c with elliptic 10-ring apertures, which could be presented by a short notation of [001] 10 4:0  6:5*. The crystallographic and chemical data for the framework ˚ b ¼ 18:729 A, ˚ and c ¼ 13:392 A, ˚ of AEL are orthorhombic, Imma (# 74), a ¼ 8:312 A, and for type material are [Al20P20O80], orthorhombic, Ibm2 (# 46), a ¼ 13:534,

180

Chemistry of Zeolites and Related Porous Materials

˚ (relationship to unit cell of framework type: a0 ¼ c, b0 ¼ b, b ¼ 18:482, c ¼ 8:370 A 0 c ¼ a). The laboratory synthetic procedure introduced here was developed by Szostak, et al.[131] Aluminum hydroxide (Aldrich, 50–57.5% Al2O3), phosphoric acid (85% H3PO4), di-n-propylamine (DPA), hydrofluoric acid (48% HF), and deionized water were used to prepare the reaction gel with a composition of 1.0 Al2O3: 1.25 P2O5: 2.37 DPA: 1.80 HF: 156 H2O. The detailed procedure for the preparation of 4 g of dry product is described below: (1) Add 7.8 g of aluminum hydroxide to 20.0 g of water, and stir until homogenous; (2) Add 14.4 g of phosphoric acid dropwise to solution (1) and stir until effervescence is complete; (3) Add 100 g of water to solution (2), and stir thoroughly; (4) Add 12.0 g of DPA dropwise to solution (3), and stir well; (5) Add 10.0 g of water and 3.75 g of hydrofluoric acid to solution (4), and stir the mixture for 2 h. The final pH value should be 6.0. The resulting gel was loaded into a Teflon-lined stainless steel autoclave and crystallized at 145  C for 18 h without agitation. The hot autoclave was quenched in cold water and the solid product was filtered off immediately, washed, and dried at ambient temperature overnight. The yield was about 70% (bow-tie crystals constructed of long needles). Powder XRD pattern characterization showed a single phase of AEL-type molecular sieve. Elemental analysis suggested a composition of Al2O31.1 P2O5. 3.3.9

SAPO-31

SAPO-31 molecular sieve is widely used in alkylatioin, amination, and isomerization. A synthetic recipe and procedure for SAPO-31 could vary with the variation of Si/Al ratio of the inorganic framework. A routine synthetic method commonly used is based on a procedure developed in a US patent with dipropylamine as organic template.[132] However, the product always contains impurity of AlPO4-11 and AlPO4-41. The laboratory synthetic procedure introduced here was developed by Kikhtyanin et al.[133] Phosphoric acid (85% H3PO4), aluminum triisopropylate Al(OiC3H7)3, fumed silica (Cab-O-Sil M5), di-n-pentylamine, hydrochloric acid (HCl), and deionized water were used to prepare the reaction gel with a composition of 1.0 Al2O3 : 1.0 P2O5 : (0–1.0) SiO2 : 1.2 R : 60 H2O. The detailed procedure is described below: (1) Mix H3PO4 and H2O, and stir for a while; (2) Add aluminum triisopropylate to solution (1); (3) Add pre-weighed Cab-O-Sil to solution (2) with stirring, and then stir the solution for a while; (4) Add di-n-pentylamine to solution (3) with stirring, and then stir the solution for 30 min; (5) Slowly add HCl solution (5–10 min) to solution (4) with stirring, and then stir the solution for 15–30 min. The resulting gel was loaded into a Teflon-lined stainless steel autoclave and crystallized at 175  C for 10 h with agitation of 75 RPM, followed by further crystallization at 175  C for 4.5 days without agitation. The solid product was filtered off, washed with

Synthetic Chemistry of Microporous Compounds (I)-Fundamentals and Synthetic Routes

181

deionized water, dried, and calcined at 600  C for 2–5 hours in sufficient air. Powder XRD pattern characterization showed a single phase of SAPO-31-type molecular sieve. Adsorption analysis indicated that the calcined product has a micro-volume as high as 82 mL/g, which is much higher than that of the di-n-propylamine-templated product. 3.3.10

SAPO-34 (CHA)

SAPO-34 {mR[Al17P12Si7O12] (R ¼ morpholine)} has a chabazite (CHA) structure and 3-dimensional interconnected channels with circular 8-ring apertures. Many important industrially applicable reactions could be catalyzed by SAPO-34. For example, SAPO-34 has a high selectivity for ethylene and propene in MTO process. The laboratory synthetic procedure described here was developed by Prakash and Unnikrishnan.[134] Phosphoric acid (85% H3PO4), pseudo boehmite (70% Al2O3), fumed silica (Degussa Aerosil-200 99þ% SiO2), morpholine (Aldrich, 99% C4H9O), and distilled water were used to prepare the reaction gel with a composition of Al2O3 : 1.06 P2O5 : 1.08 SiO2 : 2.09 R : 66 H2O (Note: H2O includes water from phosphoric acid, pseudoboehmite, and added water). The detailed procedure is described below: (1) (2) (3) (4) (5) (6)

Mix 18.0 g of H2O and 15.37 g of H3PO4; Add 9.20 g of pseudoboehmite to solution (1) slowly (2 h) with stirring; Add 10 g of water to solution (2), and stir the solution thoroughly for 7 h; Mix 4.09 g of fumed silica, 11.62 g of morpholine, and 15 g of water thoroughly; Add solution (4) dropwise to solution (3) with stirring; Add 24 g of water to mixture (5) and stir thoroughly for 7 h. The pH of the gel should be between 6.4 and 7.5.

The resulting gel was loaded into a 150 mL Teflon-lined stainless steel autoclave and incubated at 38  C for 24 h without agitation followed by crystallization at 200  C for 24 h without agitation. After the reaction, the hot autoclave was cooled and the mother liquid was decanted. The solid product was diluted with distilled water, filtered off, washed 3–4times with distilled water, and dried at 100  C for 6 h. The yield was about 98% based on alumina. Although SAPO-34 can crystallize without aging, the crystallinity of the product could be improved by aging. The amount of SiO2 and organic template can vary over a range without affecting the phase purity. With a small amount of silica (SiO2/Al2O3  0.3) or template (R/Al2O3  1.5), however, dense-phase AlPO4-crystobalite co-crystallizes with SAPO-34. The as-synthesized SAPO-34 has a cubic-like rhombohedral morphology with a particle size of 5–20 mm. Elemental analysis of the solid product suggested a composition of 1.0 Al2O3 : 0.68 P2O5 : 0.87 SiO2 : 0.59 R : 1.07 H2O. Powder XRD characterization confirmed the CHA structure of the solid product. 3.3.11

TS-1 (Ti-ZSM-5)

TS-1-type molecular sieve is a widely used and very important oxidation catalyst, and industrial production of this molecular sieve started a long time ago. The laboratory synthetic procedure introduced here was originally developed by M. Taramasso, G. Perego, and B. Notari in 1998 and improved many times later (Ref. [121], p. 207). Tetraethyl orthosilicate [Si(OC2H5)4], tetraethyl orthotitanate [Ti(OC2H5)4], tetra-npropylammonium hydroxide (TPAOH, 40% solution in water), and distilled water

182

Chemistry of Zeolites and Related Porous Materials

were used to prepare the reaction gel with a composition of TiO2 : 70 SiO2 : 1980 H2O : 30 TPAOH. The detailed procedure is described below: (1) (2) (3) (4)

Mix 163.3 g of tetraethyl orthosilicate and 2.56 g of tetraethyl orthotitanate at 35  C; Add 170 g of TPAOH to mixture (1) slowly at 0  C (to prevent hydrolysis); Heat mixture (2) at 80  C to evaporate the ethanol; Add water to mixture (3) to restore initial volume; the final pH of the solution should be 12.2.

The resulting gel was loaded into a 500 mL autoclave and crystallized at 175  C for 2 days with a stirring speed of 120 rpm. The solid product was recovered with centrifugation, washed 3 times with distilled water, dried at 120  C, and calcined at 550  C for 3 h. Approximately 43 g of dry TS-1 with a particle size of 0.3 mm and cubic shaped crystals could be obtained. Powder XRD pattern characterization showed a single phase of TS-1 molecular sieve with an orthorhombic MFI structure. Elemental analysis indicated a 1.37 mol% of Ti loading and an SiO2/TiO2 ratio of 72 (by AAS). In 1998, using source materials and a gel composition similar to Taramasso’s, Ugnina et al. reported a rapid microwave synthesis (20–30 min) of high-crystallinity TS-1 with a narrow particle-size distribution (0.33–0.42 mm).[135] Various characterization results confirmed that the microwave-synthesized TS-1 had a higher Ti content than that synthesized from routine methods and that all Ti atoms were incorporated into the framework as TiO4 tetrahedra (no extra-framework TiO2 phase existed).

References [1] D.M. Bibby and M.P. Dale, Synthesis of Silica-Sodalite from Non-Aqueous Systems. Nature (London), 1985, 317, 157–158. [2] R.E. Morris and S.J. Weigel, The Synthesis of Molecular Sieves from Non-Aqueous Solvents. Chem. Soc. Rev., 1997, 26, 309–317. [3] R.M. Barrer. Hydrothermal Chemistry of Zeolites, Academic Press, New York, 1982. [4] A. Rabenau, The Role of Hydrothermal Synthesis in Preparative Chemistry. Angew. Chem., Int. Ed. Engl., 1985, 24, 1026–1040. [5] R.R. Xu and W.Q. Pang. Inorganic Synthesis and Preparative Chemistry, Higher Education Press, Beijing, 2001. [6] J.C. Jansen, Synthesis of Zeolites, ed. H. VanBekkum and E.M. Flanigen, Elsevier, 2001, 187). [7] E.M. Flanigen, B.M. Lok, R.L. Patton, and S.T. Wilson, Aluminophosphate MolecularSieves and the Periodic-Table. Pure Appl. Chem., 1986, 58, 1351–1358. [8] D.W. Breck. Zeolites Molecular Sieves, John Wiley & Sons Inc., New York, 1974. [9] K. Suzuki, Y. Kiyozumi, S. Shin, K. Fujisawa, H. Watanabe, K. Saito, and K. Noguchi, Zeolite Synthesis in the System Pyrrolidine-Na2O–Al2O3–SiO2–H2O. Zeolites, 1986, 6, 290–298. [10] H. Kacirek and H. Lechert, Investigations on Growth of Zeolite Type NaY. J. Phys. Chem., 1975, 79, 1589–1593. [11] H. Kacirek and H. Lechert, Rates of Crystallization and a Model for Growth of Nay Zeolites. J. Phys. Chem., 1976, 80, 1291–1296. [12] D. Domine and J. Quobex, Molecular Sieves 78, Society of Chemistry Industry, London, 1968.

Synthetic Chemistry of Microporous Compounds (I)-Fundamentals and Synthetic Routes

183

[13] W. Meise and F.E. Swochow, Kinetic Studies on Formation of Zeolite-A. Adv. Chem. Ser., 1973, 121, 169–178. [14] C.L. Angell and W.H. Flank, Molecular Sieve II. ACS Symp. Ser., 1977, 40, 194–201. [15] S.P. Zhdanov, Molecular Sieves 70, Society of Chemical Industry, London, 1968. [16] S.P. Zhdanov and N.N. Samuelevich, Proceedings of 5th International Conference on Zeolites, ed. L.V.C. Rees, Heyden, London, 1980, 75. [17] J. Ciric, Kinetics of Zeolite A Crystallization. J. Colloid Interface Sci., 1968, 28, 315–324. [18] E.F. Freund, Mechanism of Crystallization of Zeolite X. J. Cryst. Growth, 1976, 34, 11–23. [19] R.M. Barrer and P.J. Denny, Hydrothermal Chemistry of Silicates. 9. Nitrogenous Aluminosilicates. J. Chem. Soc., 1961, 971–982. [20] C. Baerloch and W.M. Meier, Synthesis and Crystal Structure of Tetramethylammonium Sodalite. Helv. Chim. Acta, 1969, 52, 1853–1860. [21] C. Baerloch and W.M. Meier, Synthesis and Crystal Structure of Tetramethylammonium Gismondine. Helv. Chim. Acta, 1970, 53, 1285–1293. [22] R.M. Barrer, Cation Templates, in Hydrothermal Chemistry of Zeolites, Academic Press, London, 1982, Ch. 4, section 5.5, 166–170. [23] E.M. Flanigen, Review and New Perspectives in Zeolite Crystallization. Adv. Chem. Ser. 1973, 121, 119–139. [24] E.M. Flanigen, B.M. Lok, and R.L. Patton, Proceedings of 7th International Zeolites Conference, ed. Y. Murakami, A. Iijima, and J. Ward, Aluminophosphate Molecular Sieves and the Periodic Table, Elsevier, Kodansha, Tokyo, 1986. [25] S.T. Wilson, B.M. Lok, C.A. Messina, T.R. Cannan, and E.M. Flanigen, Aluminophosphate Molecular-Sieves – a New Class of Microporous Crystalline Inorganic Solids. J. Am. Chem. Soc., 1982, 104, 1146–1147. [26] F.S. Xiao, S.L. Qiu, W.Q. Pang, and R.R. Xu, New Developments in Microporous Materials. Adv. Mater., 1999, 11, 1091–1099. [27] J.S. Chen, W.Q. Pang, and R.R. Xu, Mixed-bonded Open-framework Aluminophosphates and Related Layered Materials. Top. Catal., 1999, 9, 93–103. [28] J.H. Yu, R.R. Xu, and J.Y. Li, Structural Diversity of a Family of Aluminophosphates with Al/P Ratio of Non-unity. Solid State Sci., 2000, 2, 181–192. [29] R.R. Xu, Q.H. Huo, and W.Q. Pang, The Crystallization of Aluminophosphates Microporous Compounds in Alcoholic System: Proceedings of 9th International Zeolites Conference, ed. R. Ballmoos, J.B. VonHiggins, and M.M.J. Treacy, Butterworth-Heinemann, Toronto, 1992, 271–278. [30] K.X. Wang, J.H. Yu, Y. Song, and R.R. Xu, Assembly of One-dimensional AlP2O83- Chains into Three-dimensional MAlP2O8C2N2H9 Frameworks through Transition Metal Cations (M ¼ Ni2þ, Co2þ and Fe2þ). Dalton Trans., 2003, 99–103. [31] J.M. Bennett, W.J. Dytrych, J.J. Pluth, J.W. Richardson, and J.V. Smith, Structural Features of Aluminophosphate Materials with Ai/P ¼ 1. Zeolites, 1986, 6, 349–361. [32] C. Baerlocher, W.M. Meier, and D.H. Olson. Atlas of Zeolite Framework Types, 5th Edn., Elsevier, Amsterdam, 2001. [33] W.F. Yan, J.H. Yu, Z. Shi, and R.R. Xu, A New fluoroaluminophosphate Chain with an Al/P Ratio of Unity. Inorg. Chem., 2001, 40, 379–383. [34] K.O. Kongshaug, H. Fjellvag, and K.P. Lillerud, Layered Aluminophosphates I. Crystal Structure of Two Novel Layered Aluminophosphates Solved ab initio from Powder Diffraction Data. Microporous Mesoporous Mater., 1999, 32, 17–28. [35] K.O. Kongshaug, H. Fjellvag, and K.P. Lillerud, Layered Aluminophosphates II. Crystal Structure and Thermal Behaviour of the Layered Aluminophosphate UiO-15 and its High Temperature Variants. J. Mater. Chem., 1999, 9, 1591–1598.

184

Chemistry of Zeolites and Related Porous Materials

[36] S.F. Cheng, J.N. Tzeng, and B.Y. Hsu, Synthesis and characterization of a Novel Layered Aluminophosphate of Kanemite-like Structure. Chem. Mater., 1997, 9, 1788–1796. [37] H.M. Yuan, J.S. Chen, Z. Shi, W. Chen, Y. Wang, P. Zhang, J.H. Yu, and R.R. Xu, Synthesis and Structural Characterisation of a New Layered Aluminophosphate [C3H12N2][Al2P2O8(OH)2]H2O. J. Chem. Soc., Dalton Trans., 2000, 1981–1984. [38] K. Maeda, A. Tuel, and C. Baerlocher, Synthesis and Characterization of a New Layered Aluminophosphate templated with 1,3-Diaminopropane: [H3N(CH2)3NH3]0.5[AlPO4(OH) (OH2)]H2O. J. Chem. Soc., Dalton Trans., 2000, 2457–2462. [39] W.F. Yan, J.H. Yu, Z. Shi, and R.R. Xu, A Novel Open-framework Aluminophosphate [AlP2O6(OH)2][H3O] containing Propeller-like Chiral Motifs. Chem. Commun., 2000, 1431–1432. [40] J.H. Yu and I.D. Williams, Two Unusual Layer Aluminophosphates templated by Imidazolium Ions; [N2C3H5][AlP2O8H22 H2O] and 2[N2C3H5][Al3P4O16H]. J. Solid State Chem., 1998, 136, 141–144. [41] K.R. Morgan, G.J. Gainsford, and N.B. Milestone, A New Type of Layered Aluminium Phosphate [NH4]3[Co(NH3)6]3[Al2(PO4)4]2 assembled about a Cobalt(III) Hexammine Complex. Chem. Commun., 1997, 61–62. [42] M.A. Leech, A.R. Cowley, K. Prout, and A.M. Chippindale, Ambient-temperature Synthesis of New Layered AlPOs and GaPOs in Silica Gels. Chem. Mater., 1998, 10, 451–456. [43] R.H. Jones, J.M. Thomas, R.R. Xu, Q.S. Huo, Y. Xu, A.K. Cheetham, and D. Bieber, Synthesis and Structure of a One-Dimensionally Extended Aluminum Phosphate – Et3NHþ(H2AlP2O8). J. Chem. Soc., Chem. Commun., 1990, 1170–1172. [44] I.D. Williams, J.H. Yu, Q.M. Gao, J.S. Chen, and R.R. Xu, New Chain Architecture for a One-dimensional Aluminophosphate, [H3NCH2CH2NH3][AlP2O8H]. Chem. Commun., 1997, 1273–1274. [45] A.A. Ayi, A. Choudhury, and S. Natarajan, Linear-chain AlPOs obtained by the Reaction of Amine Phosphates with Al3þ ions. J. Solid State Chem., 2001, 156, 185–193. [46] B. Wei, G.S. Zhu, J.H. Yu, S.L. Qiu, F.S. Xiao, and O. Terasaki, Solvothermal Synthesis and Characterization of a New 3-D Open Framework Aluminophosphate [Al2P3O12][C4N3H16]. Chem. Mater., 1999, 11, 3417–3419. [47] J. Yu, K. Sugiyama, K. Hiraga, N. Togashi, O. Terasaki, Y. Tanaka, S. Nakata, S. Qiu, and R. Xu, Synthesis and Characterization of a New Two-dimensional Aluminophosphate Layer and Structural Diversity in Anionic Aluminophosphates with Al2P3O3 12 Stoichiometry. Chem. Mater., 1998, 10, 3636–3642. [48] A.M. Chippindale, A.V. Powell, L.M. Bull, R.H. Jones, A.K. Cheetham, J.M. Thomas, and R.R. Xu, Synthesis and Characterization of 2 Layered Aluminophosphates, (T)2HAl2P3O12 (T ¼ 2-BUNH3þ) and (T)H2Al2P3O12 (T ¼ PYHþ). J. Solid State Chem., 1992, 96, 199–210. [49] S. Oliver, A. Kuperman, A. Lough, and G.A. Ozin, Aluminophosphate Chain-to-layer Transformation. Chem. Mater., 1996, 8, 2391–2398. [50] S. Oliver, A. Kuperman, A. Lough, and G.A. Ozin, The Synthesis and Structure of Two Novel Layered Aluminophosphates containing Interlamellar Cyclohexylammonium. Chem. Commun., 1996, 1761–1762. [51] P. Lightfoot, Z.A.D. Lethbridge, R.E. Morris, D.S. Wragg, and P.A. Wright, Synthesis and Structure of an Unusual New Layered Aluminophosphate containing Oxalate Groups. [NH3CH2CH2NH3]2.5[Al4H(HPO4)4(H2PO4)2(C2O4)4]. J. Solid State Chem., 1999, 143, 74–76. [52] A.M. Chippindale and R.I. Walton, [C9H20N][Al2(HPO4)2(PO4)]: An Aluminium Phosphate with a New Layer Topology. J. Solid State Chem., 1999, 145, 731–738. [53] D.J. Williams, J.S. Kruger, A.F. McLeroy, A.P. Wilkinson, and J.C. Hanson, Iridium(III) Amine Complexes as High-stability Structure-directing Agents for the Synthesis of Metal Phosphates. Chem. Mater., 1999, 11, 2241–2249.

Synthetic Chemistry of Microporous Compounds (I)-Fundamentals and Synthetic Routes

185

[54] Y.H. Xu, B.G. Zhang, X.F. Chen, S.H. Liu, C.Y. Duan, and X.Z. You, An Open Framework Aluminophosphate with Unique 12-Membered Ring Channels: Al9(PO4)12(C24H91N16)  17 H2O. J. Solid State Chem., 1999, 145, 220–226. [55] R.H. Jones, J.M. Thomas, R.R. Xu, Q.S. Huo, A.K. Cheetham, and A.V. Powell, Synthesis and Structure of a Novel Aluminum Phosphate Anion (Al3P4O16)3. J. Chem. Soc., Chem. Commun., 1991, 1266–1268. [56] Q.M. Gao, B.Z. Li, J.S. Chen, S.G. Li, R.R. Xu, I. Williams, J.Q. Zheng, and D. Barber, Nonaqueous Synthesis and Characterization of a New 2-Dimensional Layered Aluminophosphate [Al3P4O16]33[CH3CH2NH3]þ. J. Solid State Chem., 1997, 129, 37–44. [57] J.M. Thomas, R.H. Jones, R.R. Xu, J.S. Chen, A.M. Chippindale, S. Natarajan, and A.K. 2þ Cheetham, A Novel Porous Sheet Aluminophosphate Al3P4O3 16 1.5[NH3(CH2)4NH3] . J. Chem. Soc., Chem. Commun., 1992, 929–931. [58] A.M. Chippindale, A.R. Cowley, Q.S. Huo, R.H. Jones, A.D. Law, J.M. Thomas, and R.R. Xu, Synthesis and Structure of a New Layered Aluminium Phosphate: [BuNH3]3[Al3P4O16]. J. Chem. Soc., Dalton Trans., 1997, 2639–2643. [59] I.D. Williams, Q.M. Gao, J.S. Chen, L.Y. Ngai, Z.Y. Lin, and R.R. Xu, Organo-template Control of Inorganic Structures: A Low-symmetry Two-dimensional Sheet Aluminophosphate 3[NH3CHMeCH2NH3][Al6P8O32]H2O. Chem. Commun., 1996, 1781–1782. [60] J. Yu, J. Li, K. Sugiyama, N. Togashi, O. Terasaki, K. Hiraga, B. Zhou, S. Qiu, and R. Xu, Formation of a New Layered Aluminophosphate [Al3P4O16][C5N2H9]2[NH4]. Chem. Mater., 1999, 11, 1727–1732. [61] R.H. Jones, A.M. Chippindale, S. Natarajan, and J.M. Thomas, A Reactive Template in the Synthesis of a Novel Layered Aluminum Phosphate (Al3P4O16)3[NH3(CH2)5NH3]2þ (C5H10NH2)þ. J. Chem. Soc., Chem. Commun., 1994, 565–566. [62] K. Morgan, G. Gainsford, N. and Milestone, A Novel Layered Aluminum Phosphate [Co(en)3Al3P4O163 H2O] Assembled About a Chiral Metal-Complex. J. Chem. Soc., Chem. Commun., 1995, 425–426. [63] P.A. Barrett and R.H. Jones, The Pore-Size Modification of a Layered Aluminophosphate [Al3P4O16C9H24N3] by Rational Selection of the Intercalated Template Cation. J. Chem. Soc., Chem. Commun., 1995, 1979–1981. [64] S. Oliver, A. Kuperman, A. Lough, and G.A. Ozin, Synthesis and Crystal Structures of Two Novel Anionic Aluminophosphates: A One-dimensional Chain, UT-7 ([Al3P5O20H]5[C73 þ H13NHþ 3 ]5, and a Layer Containing Two Cyclic Amines, UT-8 ([Al3P4O16] [C4H7NH3 ]2þ [C5H10NH2 ]. Inorg. Chem., 1996, 35, 6373–6380. [65] J. Yu, K. Sugiyama, S. Zheng, S. Qiu, J. Chen, R. Xu, Y. Sakamoto, O. Terasaki, K. Hiraga, M. Light, M.B. Hursthouse, and J.M. Thomas, Al16P20O80H44 C6H18N2: A New Microporous Aluminophosphate containing Intersecting 12- and 8-Membered Ring Channels. Chem. Mater., 1998, 10, 1208–1211. [66] L. Vidal, V. Gramlich, J. Patarin, and Z. Gabelica, Synthesis and Structure of Mu-4, the New Layered Aluminophosphate [(C2H5)2NH2]4[Al8P10O40H2][H2O]2.5. Eur. J. Solid State Inorg. Chem., 1998, 35, 545–563. [67] Q.H. Huo, R.R. Xu, S.G. Li, Z.G. Ma, J.M. Thomas, R.H. Jones, and A.M. Chippindale, Synthesis and Characterization of a Novel Extra Large Ring of Aluminophosphate JDF-20. J. Chem. Soc., Chem. Commun., 1992, 875–876. [68] W.F. Yan, J.H. Yu, Z. Shi, P. Miao, K.X. Wang, and Y. Wang, R.R. Xu, An Anionic Framework Aluminophosphate [(CH2)6N4H3H2O] [Al11P12O48] and Computer Simulation of the Template Positions. Microporous Mesoporous Mater., 2001, 50, 151–158. [69] K.X. Wang, J.H. Yu, Z. Shi, P. Miao, W.F. Yan, and R.R. Xu, Synthesis and Characterization of a New Three-dimensional Aluminophosphate [Al11P12O48][C4H12N2][C4H11N2] with an Al/P Ratio of 11 : 12. J. Chem. Soc., Dalton Trans., 2001, 1809–1812.

186

Chemistry of Zeolites and Related Porous Materials

[70] W.F. Yan, J.H. Yu, R.R. Xu, G.S. Zhu, F.S. Xiao, Y. Han, K. Sugiyama, and O. Terasaki, [Al12P13O52]3[(CH2)6N4H3]3þ: An Anionic Aluminophosphate Molecular Sieve with Bronsted Acidity. Chem. Mater., 2000, 12, 2517–2519. [71] P.Y. Feng, X.H. Bu, and G.D. Stucky, Control of Structural Ordering in Crystalline Lamellar Aluminophosphates with Periodicity from 51 to 62 Angstrom. Inorg. Chem., 2000, 39, 2–3. [72] Q.M. Gao, J.S. Chen, S.G. Li, R.R. Xu, J.M. Thomas, M. Light, and M.B. Hursthouse, Synthesis and Structure of a Chain Aluminophosphate filled with [NHþ 4 ] and [H3NCH2CH2NH3]2þ Cations. J. Solid State Chem., 1996, 127, 145–150. [73] C. Reichardt. Solvents and Solvent Effects in Organic Chemistry, W. 2nd Edn., VCH, Weinheim, 1988. [74] S. Oliver, A. Kuperman, A. Lough, G.A. Ozin, J.M. Garces, M.M. Olken, and P. Rudolf, Zeolites and Related Microporous Materials: State of the Art 1994, Elsevier Science Publishers B V, Amsterdam, 1994, 219–225. [75] W.F. Yan, Syntheses, Structures, and Design of Aluminophosphates with Anionic Frameworks. PhD Thesis, Jilin University, China (2002). [76] Y. Li, J.H. Yu, J.X. Jiang, Z.P. Wang, J.N. Zhang, and R.R. Xu, Prediction of Openframework Aluminophosphate Structures using the Automated Assembly of Secondary Building Units Method with Lowenstein’s Constraints. Chem. Mater., 2005, 17, 6086– 6093. [77] K. Kunii, K. Narahara, and S. Yamanaka, Template-free Synthesis of AlPO4-H1,-H2, and-H3 by Microwave Heating. Microporous Mesoporous Mater., 2002, 52, 159–167. [78] X.P. Meng, W. Xu, and S. Tang, Hydrothermal Synthesis of Zeolite NaA by Microwave Radiation. Chin. Chem. Lett., 1992, 3, 69–70. [79] Y. Han, H. Ma, S.L. Qiu, and F.S. Xiao, Preparation of Zeolite A Membranes by Microwave Heating. Microporous Mesoporous Mater., 1999, 30, 321–326. [80] F.S. Xiao, S. Zheng, J.M. Sun, R.B. Yu, S.L. Qiu, and R.R. Xu, Dispersion of Inorganic Salts into Zeolites and their Pore Modification. J. Catal., 1998, 176, 474–487. [81] P.M. Slangen, J.C. Jansen, and H. vanBekkum, The Effect of Ageing on the Microwave Synthesis of Zeolite NaA. Microporous Mater., 1997, 9, 259–265. [82] B.I. Whittington and N.B. Milestone, The Microwave Heating of Zeolites. Zeolites, 1992, 12, 815–818. [83] Z. Pilter, S. Szabo, M. Hasznos-Nezdei, and E. Pallai-Varsanyi, X-Ray Diffraction Study of the Effect of Microwave Treatment of Zeolite Na-A. Microporous Mesoporous Mater., 2000, 40, 257–262. [84] I. Braun, G. Schulz-Ekloff, D. Wohrle, and W. Lautenschlager, Synthesis of AlPO4-5 in a Microwave-heated, Continuous-flow, High-pressure Tube Reactor. Microporous Mesoporous Mater., 1998, 23, 79–81. [85] D.S. Kim, S.M. Kim, J.S. Chang, and S.E. Park, Rapid and Mass Production of Porous Mateials Using a Continuous Microwave Equipment. Stud. Surf. Sci. Catal., 2001, 135, 333–340. [86] H. Kessler, J. Patarin, C. Schott-Darie, and J.L. Guth, Synthesis of High-Silica Zeolites and Phosphate-based Materials in the Presence of Fluoride. Stud. Surf. Sci. Catal., 1994, 85, 75–113. [87] M.A. Camblor, L.A. Villaescusa, and M.J. Diaz-Cabanas, Synthesis of All-silica and Highsilica Molecular Sieves in Fluoride Media. Top. Catal., 1999, 9, 59–76. [88] A. Corma, M. Diaz-Cabanas, J. Martinez-Triguero, F. Rey, and J. Rius, A Large-cavity Zeolite with Wide Pore Windows and Potential as an Oil Refining Catalyst. Nature (London) 2002, 418, 514–517. [89] G. Sastre, J.A. Vidal-Moya, T. Blasco, J. Rius, J.L. Jorda, M.T. Navarro, F. Rey, and A. Corma, Preferential Location of Ge Atoms in Polymorph C of Beta Zeolite (ITQ-17) and

Synthetic Chemistry of Microporous Compounds (I)-Fundamentals and Synthetic Routes

[90]

[91]

[92] [93]

[94] [95]

[96]

[97]

[98]

[99] [100] [101]

[102] [103] [104]

[105] [106] [107] [108]

187

their Structure-directing Effect: A Computational, XRD, and NMR Spectroscopic Study. Angew. Chem., Int. Ed., 2002, 41, 4722–4726. S.I. Zones, Y. Nakagawa, G.S. Lee, C.Y. Chen, and L.T. Yuen, Searching for New High Silica Zeolites through a Synergy of Organic Templates and Novel Inorganic Conditions. Microporous Mesoporous Mater., 1998, 21, 199–211. Y. Nakagawa, G.S. Lee, T.V. Harris, L.T. Yuen, and S.I. Zones, Guest/host Relationships in Zeolite Synthesis: Ring-substituted Piperidines and the Remarkable Adamantane Mimicry by 1-Azonio Spiro[5.5]undecanes. Microporous Mesoporous Mater., 1998, 22, 69–85. W.Q. Pang, S.L. Qiu, and F.Q. Zhou, Advances in the Synthesis of Heteroatom-substituted Zeolites. J. Jilin Univ. (Sci. Ed.) 1992, Special Issue (Chemistry), 78–84. Y. Xu, P.J. Maddox, and J.W. Couves, The Synthesis of SAPO-34 and CoSAPO-34 from a Triethylamine Hydrofluoric-Acid Water-System. J. Chem. Soc., Faraday Trans., 1990, 86, 425–429. H.V. Bekkum, E.M. Flanigen, P.A. Jacobs, and J.C. Jansen, Introduction to Zeolite Science and Practice. Stud. Surf. Sci. Catal. 2001, 137, 247–254. I. Kiricsi, S. Shimizu, Y. Kiyozumi, M. Toba, S. Niwa, and F. Mizukami, Catalytic Activity of a Zeolite Disc Synthesized through Solid-state Reactions. Microporous Mesoporous Mater., 1998, 21, 453–459. L. Schreyeck, P. Caullet, J.C. Mougenel, J.L. Guth, and B. Marler, A Layered Microporous Aluminosilicate Precursor of Fer-Type Zeolite. J. Chem. Soc., Chem. Commun., 1995, 2187–2188. Y.J. He, G.S. Nivarthy, F. Eder, K. Seshan, and J.A. Lercher, Synthesis, Characterization and Catalytic Activity of the Pillared Molecular Sieve MCM-36. Microporous Mesoporous Mater., 1998, 25, 207–224. T. Yanagisawa, T. Shimizu, K. Kuroda, and C. Kato, The Preparation of Alkyltrimethylammonium-Kanemite Complexes and Their Conversion to Microporous Materials. Bull. Chem. Soc. Jpn., 1990, 63, 988–992. W.Y. Xu, J.X. Dong, J.P. Li, J.Q. Li, and F. Wu, A Novel Method for the Preparation of Zeolite ZSM-5. J. Chem. Soc., Chem. Commun., 1990, 755–756. M.H. Kim, H.X. Li, and M.E. Davis, Synthesis of Zeolites by Water–Organic Vapor-phase Transport. Microporous Mater., 1993, 1, 191–200. M. Matsukata, N. Nishiyama, and N. Yeyama, Synthesis of Zeolites under Vapor Atmosphere: Effect of Synthetic Conditions on Zeolite Structure. Microporous Mater., 1993, 1, 219–222. P. Rao and M. Matsukata, Dry-gel Conversion Technique for Synthesis of Zeolite BEA. Chem. Commun., 1996, 1441–1442. P. Rao, C. Leon, K. Ueyama, and M. Matsukata, Synthesis of BEA by Dry Gel Conversion and its Characterization. Microporous Mesoporous Mater., 1998, 21, 305–313. M. Matsukata, K. Kizu, M. Ogura, and E. Kikuchi, Synthesis of EMT Zeolite by a Steamassisted Crystallization Method using Crown Ether as a Structure-directing Agent. Cryst. Growth Des., 2001, 1, 509–516. C.B. Khouw and M.E. Davis, Catalytic Activity of Titanium Silicates Synthesized in the Presence of Alkali-Metal and Alkaline-Earth Ions. J. Catal., 1995, 151, 77–86. M. Matsukata, M. Ogura, T. Osaki, P. Rao, M. Nomura, and E. Kikuchi, Conversion of Dry Gel to Microporous Crystals in Gas Phase. Top. Catal., 1999, 9, 77–92. M. Matsukata, N. Nishiyama, and K. Ueyama, Zeolitic Membrane Synthesized on a Porous Alumina Support. J. Chem. Soc., Chem. Commun., 1994, 339–340. N. Nishiyama, K. Ueyama, and M. Matsukata, Synthesis of Defect-free Zeolite-alumina Composite Membranes by a Vapor-phase Transport Method. Microporous Mater., 1996, 7, 299–308.

188

Chemistry of Zeolites and Related Porous Materials

[109] Q.S. Huo and R.R. Xu, A New Route for the Synthesis of Molecular-Sieves – Crystallization of AlPO4-5 at High Temperature. J. Chem. Soc., Chem. Commun., 1992, 168–169. [110] E.N. Coker, J.C. Jansen, J.A. Martens, P.A. Jacobs, F. DiRenzo, F. Fajula, and A. Sacco, The Synthesis of Zeolites under Micro-gravity Conditions: A Review. Microporous Mesoporous Mater., 1998, 23, 119–136. [111] O.D. Friedrichs, A.W.M. Dress, D.H. Huson, J. Klinowski, and A.L. Mackay, Systematic Enumeration of Crystalline Networks. Nature (London) 1999, 400, 644–647. [112] E.R. Cooper, C.D. Andrews, P.S. Wheatley, P.B. Webb, P. Wormald, and R.E. Morris, Ionic Liquids and Eutectic Mixtures as Solvent and Template in Synthesis of Zeolite Analogues. Nature (London) 2004, 430, 1012–1016. [113] Y.P. Xu, Z.J. Tian, S.J. Wang, Y. Hu, L. Wang, B.C. Wang, Y.C. Ma, L. Hou, J.Y. Yu, and L.W. Lin, Microwave-enhanced Ionothermal Synthesis of Aluminophosphate Molecular Sieves. Angew. Chem., Int. Ed., 2006, 45, 3965–3970. [114] L. Wang, Y.P. Xu, Y. Wei, J.C. Duan, A.B. Chen, B.C. Wang, H.J. Ma, Z.J. Tian, and L.W. Lin, Structure-directing Role of Amines in the Ionothermal Synthesis. J. Am. Chem. Soc., 2006, 128, 7432–7433. [115] H. Lee, S.I. Zones, and M.E. Davis, Recent Advances in the Science and Technology of Zeolites and Related Materials, Pts A – C, Elsevier Science BV, Amsterdam, 2004, 102–109. [116] B. Jandeleit, D.J. Schaefer, T.S. Powers, H.W. Turner, and W.H. Weinberg, Combinatorial Materials Science and Catalysis. Angew. Chem., Int. Ed., 1999, 38, 2495–2532. [117] J. Klein, C.W. Lehmann, H.W. Schmidt, and W.F. Maier, Combinatorial Material Libraries on the Microgram Scale with an Example of Hydrothermal Synthesis. Angew. Chem., Int. Ed., 1998, 37, 3369–3372. [118] D.E. Akporiaye, I.M. Dahl, A. Karlsson, and R. Wendelbo, Combinatorial Approach to the Hydrothermal Synthesis of Zeolites. Angew. Chem., Int. Ed., 1998, 37, 609–611. [119] K. Choi, D. Gardner, N. Hilbrandt, and T. Bein, Combinatorial Methods for the Synthesis of Aluminophosphate Molecular Sieves. Angew. Chem., Int. Ed., 1999, 38, 2891–2894. [120] Y. Song, J.H. Yu, G.H. Li, Y. Li, Y. Wang, and R.R. Xu, Combinatorial Approach for the Hydrothermal Syntheses of Open-framework Zinc Phosphates. Chem. Commun., 2002, 1720–1721. [121] Verified Synthesis of Zeolite Materials, ed. H. Robson, 2nd Edn, Elsevier Synthesis Commission of the International Zeolite Association, Amsterdam, 2001. [122] R.W. Thompson and M.J. Huber, Analysis of the Growth of Molecular-Sieve Zeolite Naa in a Batch Precipitation System. J. Cryst. Growth, 1982, 56, 711–722. [123] G.H. Kuhl, Crystallization of Low-Silica Faujasite (SiO2/Al2O3-Approximately-2.0). Zeolites, 1987, 7, 451–457. [124] P. Norby, A.N. Christensen, and I.G.K. Andersen, Hydrothermal Preparation of Zeolite Li-A(BW), LiAlSiO4H2O, and Structure Determination from Powder Diffraction Data by Direct Methods. Acta Chem. Scand. Ser., A, 1986, 40, 500–506. [125] D.M. Ginter, A.T. Bell, and C.J. Radke, Synthesis of Microporous Materials, Vol. I, Molecular Sieves, ed. M.L. Occelli and H.E. Robson, Van Nostrand Reinhold, New York, 1992, p. 6. [126] T. Chatelain, J. Patarin, M. Soulard, J.L. Guth, and P. Schulz, Synthesis and Characterization of High-Silica EMT and FAU Zeolites Prepared in the Presence of Crown-Ethers with Either Ethylene-Glycol or 1,3,5-Trioxane. Zeolites, 1995, 15, 90–96. [127] G.J. Kim and W.S. Ahn, Direct Synthesis and Characterization of High-SiO2-Content Mordenites. Zeolites, 1991, 11, 745–750. [128] M.A. Camblor and J. Pe´rez-pariente, Crystallization of Zeolite Beta – Effect of Na and K-Ions. Zeolites, 1991, 11, 202–210. [129] J.P. Verduijn, US Patent 5,242,675 (1993).

Synthetic Chemistry of Microporous Compounds (I)-Fundamentals and Synthetic Routes

189

[130] I. Girnus, K. Jancke, R. Vetter, J. Richter-Mendau, and J. Caro, Large AlPO4-5 Crystals by Microwave-Heating. Zeolites, 1995, 15, 33–39. [131] R. Szostak, B. Duncan, R. Aiello, A. Nastro, K. Vinje, and K.P. Lillerud, Synthesis of Microporous Materials, ed. M.L. Occelli and H.E. Robson, Van Nostrand Reinhold, New York, 1992, 240–247. [132] B.M. Lok, C.A. Messina, R.L. Patton, R.T. Gajek, T.R. Cannan, and E.M. Flanigen, US Patent 4,440,871 (1984). [133] O.V. Kikhtyanin, R.F. Vogel, C.L. Kibby, T.V. Harris, K.G. Ione, and D.J. O’Rear, Proceedings of the 12th International Zeolite Conference, ed. M.M.J. Treacy, B.K. Marcus, M.E. Bisher, and J.B. Higgins, MRS, Baltimore, 1998, 1743. [134] A.M. Prakash and S. Unnikrishnan, Synthesis of SAPO-34 – High-Silicon Incorporation in the Presence of Morpholine as Template. J. Chem. Soc., Faraday Trans., 1994, 90, 2291– 2296. [135] M.A. Ugnina, D.P. Serrano, R. Song, and T. Castillo, Proceedings of the 12th Interntional Zeolite Conference, ed. M.M.J. Treacy, B.K. Marcus, M.E. Bishor, and J.B. Higgins, MRS, Warrendale, PA, 1998, 1917–1924.

4 Synthetic Chemistry of Microporous Compounds (II) – Special Compositions, Structures, and Morphologies In the previous chapter, the basic synthetic routes and approaches, and the primary synthetic rules, in the synthetic chemistry of microporous crystals were discussed. As examples, the syntheses of some important molecular sieves and microporous compounds have been described in detail. This should allow readers to have a relatively good understanding of the synthetic chemistry of molecular sieves and microporous compounds. To continue the discussion on synthetic chemistry, this chapter will focus on microporous compounds with special types, structures, and aggregation morphologies, and discuss their synthesis more deeply and more widely on the basis of the previous chapter. The so-called ‘special type’ refers mainly to the compounds that are different from zeolites and microporous aluminophosphates, such as M(III)X(V)O4-type transition metal phosphates, aluminoborates, microporous oxides, sulfides, selenides, halides, etc. Microporous compounds with special structures refer to molecular sieves and microporous crystals, whose structures are different from the general (4,2)-connected threedimensional (3-D) open frameworks. Here, four types of special structures will be emphatically described, including the extra-large-microporous structure, the intersecting channel structure, the pillared layered microporous structure, and the chiral structure. Finally, the synthetic chemistry of several types of microporous compounds and related materials with special aggregation morphologies will be described, which includes single crystals and perfect crystals, nanocrystals and ultrafine particles, zeolite membranes, and microporous compounds with special crystal morphologies and

Chemistry of Zeolites and Related Porous Material – Synthesis and Structure Ruren Xu, Wenqin Pang, Jihong Yu, Qisheng Huo and Jiesheng Chen # 2007 John Wiley & Sons, (Asia) Pte Ltd

192

Chemistry of Zeolites and Related Porous Materials

multi-level porous channels. These descriptions and discussions will not only help the readers to further understand the synthetic chemistry of microporous compounds, but also update them on the recent development of this subject and help them to better understand some leading-edge studies and development directions of the synthesis chemistry of microporous compounds.

4.1 Synthetic Chemistry of Microporous Compounds with Special Compositions and Structures 4.1.1

M(III)X(V)O4-type Microporous Compounds

Apart from a large variety of AlPO4-n compounds, M(III)X(V)O4-type microporous crystals have been prepared by hydrothermal reactions among the other elements of the third (III) and fifth (V) main groups in the periodic table, such as Al, Ga, and In, and P and As.[1] The first microporous gallophosphate with open-framework structure was synthesized by Parise in 1985.[2] The following year, R. Xu and S. Feng of Jilin University systemically synthesized twelve open-framework structures of gallophosphates, denoted as GaPO4-Cn (C stands for China, n ¼ 1
12), by using hydrothermal and solvothermal methods.[3,4] Their hydrothermal syntheses were performed in the reaction system of ð0:5
3Þ R : 1 Ga2O3 : ð1
1:5Þ P2O5 : ð25
100Þ H2O, where R represents various structure-directing agents (SDAs). Typically, GaOOH, H3PO4, and water were mixed with stirring at 80  C, followed by the addition of various SDAs. The reaction mixture was heated at 150–190  C for 72
144 h. The as-synthesized products usually possess neutral frameworks with a P/Ga ratio of 1, and their empirical compositions can be expressed as xRGa2 O3 1:0  0:1 P2 O5 yH2 O. Details of the synthesis conditions are listed in Table 4.1. The successful synthesis of series GaPO4-Cn provides the foundation for further exploration and study of microporous gallophosphates. In the last decade, a large variety of novel GaPO4s have been prepared by using similar hydrothermal and solvothermal synthetic routes in the presence of different SDAs, and particularly in the F -ioncontaining systems. A large number of crystal products suitable for structural analysis have been produced, which include cloverite, GaPO4-M1, GaPO4-M2, ULM-n, MIL-n, etc. Different from AlPO4-n, the Ga atom in the structures of GaPO4s is usually five- or Table 4.1 ð0:5
3Þ R : 1 Ga2O3 : ð1
1:5ÞP2O5 : ð25
100ÞH2O at 150–190  C Code GaPO4-C1 GaPO4-C2 GaPO4-C3 GaPO4-C4a GaPO4-C5a GaPO4-C6 a

R(SDA)

X

Y

TMAOH HDA Et3N EAN PriNH2 HDA

0.32 0.40 0.24 0.58 0.36 0.62

0 0 0.30 0 0.22 0.36

Code GaPO4-C7b GaPO4-C8 GaPO4-C9 GaPO4-C10 GaPO4-C11 GaPO4-C12c

R(SDA)

X

Y

PrNH2 DPA HDA CHA DMA

b 0.82 0.26 0.64 0.88

0.34 0.34 0.66 0

GaPO4-C4 and GaPO4-C5 are isostructural with AlPO4-21 and AlPO4-15, respectively. GaPO4-C7 has the composition GaPO4  0.5NH3  1.5 H2O  0.08 PrOH. c GaPO4-C12 is obtained by the calcination of GaPO4-C4 at 500  C for 2 h, and is isostructural with AlPO4-25. b

Synthetic Chemistry of Microporous Compounds (II)

193

Table 4.2 ð0:6
3Þ R  1Al2O3  ð1:0
1:5Þ As2O5  ð25
100Þ H2Oa at 200  C for 4–10 days AlAsO4-n

R(SDA)

m

n

AlAsO4-n

R(SDA)

m

n

AlAsO4-1 AlAsO4-2 AlAsO4-3 AlAsO4-4 AlAsO4-5 AlAsO4-6

EAN TMAOH DAP DAP EDA EDA

0.98 0.60 0.56 0.48 0.80 0.80

0 0 0.54 0.82 1.52 1.20

AlAsO4-7 b AlAsO4-8 AlAsO4-9 AlAsO4-10c AlAsO4-11c AlAsO4-12c

PriNH2 EtNH2 BuNH2 CHA HDA

0.50 0.52 0.70 0.60 0.62

1.80 1.34 1.00 1.50 2.00

a

Al(OC3H7)3, H4As2O7 and various amines act as the reaction materials. R is DMA, but AlAsO4-7 has the composition AlAsO4  0.30 NH3  2.0 H2O. c AlAsO4-10, AlAsO4-11, and AlAsO4-12 have layered structures. b

six-coordinated with O or F atoms, and even with N atoms in some cases. The frameworks of microporous GaPO4s exhibit rich structural architectures due to the variety of basic building units. In 1989, series of open-framework aluminoarsenates AlAsO4-n ðn ¼ 1
12Þ½1 and galloarsenates GaAsO4-n ðn ¼ 1
12Þ½1 were explored by Xu and Chen of Jilin University. Their synthesis conditions and empirical compositions of mR  Al2 O3  1:0  0:1 As2 O5  nH2 O and pR  Ga2 O3  1:0  0:1 As2 O5  qH2 O are listed in Tables 4.2 and 4.3, respectively. Typical structures of AlAsO4-1,[5] AlAsO4-2,[6] and GaAsO4-2[7] are shown in Figure 4.1. Series of M(III)X(V)O4-type compounds, such as GaPO4-n, AlAsO4-n, GaAsO4-n, and InPO4-n, which will be introduced later, are prepared in neutral and weakly acid solutions, and perfect single-crystal products are favored to be obtained. Thus, it is helpful to determine their structures and understand their formation rules. Compared with AlPO4-n, the structures of AlAsO4-n and GaAsO4-n have the following features:[1] (1) Different from AlPO4-n, the six-coordinated Al and Ga atoms connected with O atoms (or F atoms in some cases) are dominant in the frameworks of AlAsO4-n and GaAsO4-n, respectively. In general, the occluded SDA molecules interact with inorganic frameworks through H-bonds, leading to the difficult removal of SDAs or detemplating. These compounds have poor thermal stabilities, making it difficult to form molecular sieves from them through calcinations.

Table 4.3 Ga2O3 As2O5 R H2O systema at 200  C for 5–15 days GaAsO4-n

R(SDA)

p

GaAsO4-1 GaAsO4-2b GaAsO4-3 GaAsO4-4 GaAsO4-5 GaAsO4-6

TMAOH DMA DAP EDA EAN EAN

a

q

GaAsO4-n

R(SDA)

p

q

0.38

0

0.58 1.06 1.42 0.46

0.46 1.78 1.64 2.46

GaAsO4-7 GaAsO4-8 GaAsO4-9c GaAsO4-10c GaAsO4-11c GaAsO4-12c

EtNH2 PrNH2 PriNH2 BuNH2 CHA HDA

0.74 0.42 1.44 1.82 1.46 0.94

1.02 2.34 1.98 1.92 1.98 1.36

GaOOH, H4As2O7 and various amines act as reaction materials, and HF is used in some cases. HF is used during the synthesis of GaAsO4-2 with the composition GaAsO4  0.59 DMA  0.32 HF  0.30 H2O. GaAsO4-9, GaAsO4-10, GaAsO4-11, and GaAsO4-12 have layered structures.

b c

194

Chemistry of Zeolites and Related Porous Materials

Figure 4.1 Structure of (a) AlAsO4-1 (Ref. [5]), (b) AlAsO4-2 (Ref. [6]) and (c) GaAsO4-2 (Ref. [7])

(2) Different from AlPO4-n, AlAsO4-n and GaAsO4-n are usually prepared by using small organic amines as SDAs. Based on our experience, 3-D open frameworks AlAsO4-n and GaAsO4-n are difficult to synthesize when the organic amine has more than four C atoms. If large organic amines, such as BuNH2, 1,2-cyclohexane, and 1,6-hexanediamine, act as SDAs in the above reaction systems, the products often have layered structures. Since 1994, W. Pang,[8] Y. Xu[9,10] and coworkers have explored the family of indium phosphates with open-framework structures. The structural feature of this type of compounds is that the inorganic framework is constructed by the InO6 and PO4 units, and occluded SDA molecules interact with framework host through H-bonds or coordination bonds. Their thermal stabilities are poor compared with others. To the best of our knowledge, there has been no report of an InPO4-n molecular sieve prepared by the removal of organic SDAs until now. There are two features in the synthesis of these compounds. One is that a large proportion of these compounds are prepared under solvothermal conditions, and the other is that they are more easily obtained in reaction systems containing F ions. For examples, InPO4-Cn ðn ¼ 1
12Þ synthesized in the R(SDA)–In2O3 –P2O5 –HF-alcohol system,[8] InPO4-1[10] with variscite-type structure, In5(PO4)4F3(H2O)2  en3[11] with 10-ring channels, and [In3(PO4)6(HPO4)2F16][11] with 14-ring channels were all prepared by hydrothermal/solvothermal reactions under the influence of F ions. 4.1.2

Microporous Transition Metal Phosphates

Since the incorporation of transition metals into the frameworks of zeolites or microporous aluminophosphates to form heteroatom-containing molecular sieves with important application values, the synthesis, structure, and characterization of microporous transition metal phosphates have been extensively studied in the last decade. In particular, because transition metal cations possess redox and coordination features, they are a kind of catalytic material with useful applications, and promise potential

Synthetic Chemistry of Microporous Compounds (II)

195

applications in advanced technology and the life sciences. Therefore, it is important to study the synthesis–structure–function relationship of microporous materials with transition metal atoms as the framework compositions. Up to now, microporous transition metal phosphates with framework components covering almost all the elements of the 3d period except for Cu, and some elements of the 4d period, such as Mo and Zr, have been prepared hydrothermally/solvothermally in the presence of various kinds of organic amines, and even in the presence of inorganic ions. In this section, microporous zinc phosphates will be described as the example. Since the work of Gier and Stucky in the early 1990s,[12] a large number of openframework zinc phosphate crystals have been synthesized and characterized, which greatly promotes the growth of the zinc phosphate family. Most of these compounds have novel topological structures except that only very few are of structures analogous to known molecular sieves. Open-framework zinc phosphates without occluded organic SDAs have been prepared, which are known as MZnPO4H2O (M ¼ Li, Na, K) with Zn/ P ¼ 1, and M[Zn4(PO4)3]xH2O (M ¼ H, Na, Rb, etc.) with Zn/P ¼ 4/3. On the other hand, more zinc phosphates containing organic SDAs have been reported, and their Zn/P ratios are 1/1, 2/3, 3/4, etc. Structural analysis shows that P-centered tetrahedra (PO4, HPO4, H2PO4) and Zn-centered tetrahedra (ZnO4) are the basic building units for the structural construction of zinc phosphates. Moreover, ZnO6 octahedra, ZnO5 square pyramids, and ZnO3(H2O)2 trigonal bipyramids as basic building units also appear in the structures of zinc phosphates. Recent studies indicate that N-atoms of organic amines can coordinate with Zn-atoms to form more building units, such as ZnO3N, ZnO2N2, ZnO3N2, etc. These building units connect with PO4 tetrahedra to form some interesting open-framework zinc phosphates. In addition, Zn2O2 dimer, Zn2O3 trimer, ZnO4 tetrahedral cluster, and Zn7O6 cluster also occur in the structures of zinc phosphates. Of particular interest is that the existence of three-bridging O atoms in some zinc phosphates results in the formation of 3-rings, Zn O Zn bonds, and infinite -Zn O Zn- chains. This is a unique feature in the structure chemistry of zinc phosphates, and has not been found in the structures of aluminophosphates. Furthermore, various rings including 4-, 5-, 6-, 8-, 10-, 12-, 16-, 18-, 20-, and 24-rings present in the zinc phosphates lead to the channel structural diversity and complexity of zinc phosphates. Low-dimensional zinc phosphates have been frequently reported although most zinc phosphates are of 3-D structures, which include 1-D chain compounds [C4N2H10][Zn(HPO4)2] composed of corner-sharing 4-rings, [C6N4H22]0.5[Zn(HPO4)2] and [C3N2H12][Zn(HPO4)2] composed of edge-sharing 4-rings; a 2-D layer structure with 1-D tube channels and wrinkled-layer zinc phosphate with ladder channels. Recently, 0-D zinc phosphate monomers [C6N2H18][Zn(HPO4)(H2PO4)2] and [C6N4H21][Zn(HPO4)2(H2PO4)] comprising 4-rings have also been prepared. Because of the diversity of basic building units in the 3-D open frameworks and the existence of a large variety of low-dimensional zinc phosphates, large numbers of open-framework zinc phosphates with special channel structures have been reported. Notable examples are chiral zinc phosphate with intersecting helical channels reported by C.R.N. Rao,[13] NaZn PO4H2O[14] with a chiral tetrahedral framework, [{Zn2(HPO4)4}{Co(dien)2}]H3O[15] with multi-directional intersecting helical channels by using a racemic mixture of a chiral [Co(dien)2]Cl3 complex as the SDA, zinc phosphate with 20-ring channels by using 1, 6-HDA as the SDA,[16] ND-1 with 24-ring channels by using 1,2-diaminocyclohexane

196

Chemistry of Zeolites and Related Porous Materials

(DACH) as the SDA, zinc phosphate with 24-ring channels by using 1,3,5-benzenetricarboxylic acid and ethylenediamine as SDAs,[17] and zinc phosphate with 16-ring channels prepared by a combinatorial approach by Yu and coworkers in 2002.[18] Large numbers of experiments indicate that the formation of open-framework zinc phosphates intensely depends on the type of organic amine, the concentration of H3PO4, and the molar ratio of the batch composition in the reaction system. It means that the pH value of the reaction system and pKa value of the organic amine are crucial factors affecting the formation of framework structures because the protonated organic amine usually stabilizes the inorganic framework through H-bonds. Various organic amines, including monoamine, diamine, and multi-amines (DETA, TETA, TEPA), are all used as SDAs to prepare zinc phosphates, but diamines and triamines are more favorable for directing the formation of 3-D open-framework structures than monoamines. Despite this, it is still difficult to control and conclude the relationship between the synthesis conditions and structural features of zinc phosphates. As with microporous aluminophosphates, one organic SDA can lead to different framework structures in the same reaction system by varying other reaction conditions, such as the molar ratio of the gel composition, pH value, and so on. This character is particularly obvious in the synthesis of zinc phosphates. One notable example is that five different zinc phosphates have been synthesized by Choudhury in the presence of triethylenetetramine as the SDA, including 1-D ladder, 2-D layer, and three 3-D openframework structures.[19] It is found that the structure of the product depends sensitively on the relative concentrations of the amine and phosphoric acid, with high acid concentrations yielding lower-dimensional structures, and comparable concentrations of the amine and the acid yielding three-dimensional open-framework structures. When the amine concentration is far in excess of that of the acid, the amine molecule binds covalently with the zinc atom. On the other hand, zinc phosphate crystals with the same framework structure can also be prepared in the presence of various SDAs under different crystallization conditions. For example, layer compounds (C6H17N3)[Zn3(HPO4) (PO4)2]H2O templated by 1-(2-aminoethyl)piperazine and (C10N4H28)[Zn6(HPO4)2 (PO4)4]2H2O templated by 1,4-bis(3-aminopropyl)piperazine have the same layer structure. Synthetic Approach of Zinc Phosphates and Exploration of New Synthetic Route The typical synthesis of microporous zinc phosphates is that a zinc source [ZnCl2, Zn(NO3)2, ZnO, etc] is first dissolved in water, and then H3PO4 and organic amines are added dropwise while the mixture is stirred. The formed homogeneous reaction gel is crystallized in the hydrothermal system for some days. The reaction temperatures vary in the range of room temperature to 180  C. On the other hand, a few zinc phosphate crystals have been prepared under solvothermal conditions, such as the synthesis of extralarge 20-ring [H3N(CH2)6NH3][Zn4(PO4)2(HPO4)2]3H2O reported by Harrison in which ethylene glycol was used as the solvent.[16] Recently, Rao[20] proposed a new synthetic route for zinc phosphates in which H3PO4 and organic amines first reacted together to form amine phosphates, and then these amine phosphates further reacted with Zn2þ cations. By means of this method, it is possible to generate not only some zinc phosphates which can be prepared by normal hydrothermal method (but the temperature is lower), but also some zinc phosphates with novel structures which had not been

Synthetic Chemistry of Microporous Compounds (II)

197

prepared before. The most obvious advantage of this method is that the reaction is performed at room temperature. This avoids the use of the traditional hydrothermal approach which is carried out in a sealed vessel, like a black box, making it difficult to obtain any information on the reaction process occuring in it. The synthetic route involving the reaction of amine phosphates with metal salts is helpful for us to understand the formation process of various zinc phosphates. Synthesis of Zinc Phosphates and Discussion of Dimension Buildup Mechanism Recently, Rao and coworkers reported the synthesis of two 0-D 4-ring zinc phosphate monomers, [C6N2H18][Zn(HPO4)(H2PO4)2] and [C6N4H21][Zn(HPO4)2(H2PO4)], and successfully transformed these two monomers into 1-D ladder, 2-D layer, and 3-D open-framework structures.[21] They also achieved the transformations of 1-D ladder structures [C6N4H22]0.5[Zn(HPO4)2] and [C3N2H12][Zn(HPO4)2] into 2-D and 3-D complex structures, and the transformation of 2-D layer into 3-D structure.[22] In 2001, Rao et al. published an article in which the basic building units for the construction of open-framework zinc phosphates and the transformations among zinc phosphates with various structures under suitable conditions were discussed in detail.[23] The transformation reactions are normally achieved by the reactions of monomer or low-dimensional zinc phosphates with suitable media, such as water, zinc salt, H3PO4, or other organic amines. The study shows that 0-D 4-ring monomers can assemble to form 1-D, 2-D, and 3-D complex structures under suitable conditions. The 1-D ladder structure is the most reactive, rather than the 1-D linear-chain, because there are many pendant PO4 groups. It can transform into 2-D and 3-D structures under different conditions, and the formed complex structures all feature the 1-D ladder-chain. Interestingly, the transformations  O groups and the loss of normally come along with the deprotonizing process of P  -HPO4 groups. Thus, the 0-D 4-ring monomer and 1-D chain structures of zinc phosphates are the basic building units for the construction of open-framework structures. Acting as the precursors, these building units can self-assemble to form 3-D complex structures of zinc phosphates. This is the reason why 3-D structures of zinc phosphates are the most commonly observed, and lower-dimensional architectures such as 1-D chains and 4-ring monomers are somewhat rare. More recently, open-framework metal phosphates (metal ¼ Ti, Mo, V, Fe, Co, etc.) with redox features have been continually reported. Cheetham summarized them in 1999.[11] 4.1.3

Microporous Aluminoborates

Although B belongs to the III main group in the periodic table, it is different from other elements, like Al, in the same group. Up to now, there have been rare reports of openframework silicates and phosphates prepared by using B as the primary component. Even in the heteroatom-containing frameworks of aluminophosphates and aluminosilicates containing B atoms, the B content is normally less than 5%. Pang and coworkers at Jilin University first synthesized BAlPO4-5 crystals, in which the B content was only about 1%.[24] In 1996, Sevov reported the single example of mircoporous borate (C2H10N2)[CoB2P3O12(OH)] by using B as the framework element, and its structure was built up from BO4, PO4, and CoO6 units.[25]

198

Chemistry of Zeolites and Related Porous Materials

Strikingly, after a long time of exploration, Xu, Yu, Feng, et al. at Jilin University found that two kinds of framework structures could be constructed by the connections of BO3, BO4, and AlO6 polyhedra in the B Al O system. One is the anionic framework of aluminoborates, such as the natural mineral hilgandite Ca2B5O9ClH2O. In 1989, Feng and coworkers added Et3N, TMAOH, and TEAOH as SDAs in the system Na2O CaO B2O3 Al2O3 H2O at 150–200  C to obtain several open-frame˚ ) built up from BO3, BO4, and AlO6 work aluminoborates (pore size 3.6–6.0 A [26,27] units. The other is the B Al O positive framework first successfully prepared by Yu et al. in 1992,[28] which is different from the anionic framework of molecular sieves and the neutral framework of aluminophosphates. For example, the structure of BAC-1 with an experimental composition of B0.15Al9.85O9(OH)10Cl2 is composed of BO4 and AlO6 primary building units. It was prepared by using H3BO3, AlCl3, and CaO as the reaction materials in the B2O3 Al2O3 CaO H2O system at 200  C. Highly pure BAC-1 was crystallized in a reaction mixture of 0.75 B2O3 : 3.0 Al2O3 : 2.5 CaO : 200 H2O (H2O ¼ 95 mol, pH ¼ 3.5) at 200  C. Further investigation of the crystallization field of such a synthetic system and the influence of the temperature on crystallized products led to the formation of two other open-framework boron– aluminium oxo chlorides, BAC-8 (1.1 B2O31.0 Al2O30.70 HCl5.0 H2O) and BAC10[29] (0.5 B2O31.0 Al2O30.40 HCl3.0 H2O). Their structures are both composed of BO3, BO4, and AlO6 as basic building units, giving rise to microporous cationic B–Al frameworks. Microporous BAC-10 was synthesized from a reaction mixture similar to that for BAC-1, the only difference being the reaction temperature. Four reactant compositions chosen in the crystallization field of BAC-1 were heated at 200  C and 160  C for 72 h, resulting in the production of BAC-1 and BAC-10, respectively (see Table 4.4). Another microporous BAC-3[30] (0.5 B2O31.0 Al2O30.4 HCl3.0 H2O) was synthesized from a reaction mixture of 3.25 B2O3 : 1.0 Al2O3 : 0.75 Na2O : 100 H2O at 180  C for 9 days. A suitable pH value for the crystallization of BAC-3 was 1.8–3.0, similar to those for other BAC-n. In the synthesis of BAC-3, CaO and ammonia can replace Na2O to adjust the pH value of the reaction system because the final product does not contain Naþ , Ca2þ , etc. cations, which affect the pH value of the reaction mixture. Yu et al. studied the crystallization field, structural features and adsorption capability of BAC-3.[30] Although a detailed structural analysis result had not been obtained, it was found that, similarly to other BAC-n, BAC-3 can adsorb H2O molecules with the I type of adsorption isotherm. Table 4.4 The effect of the temperature on the product Initial reactant molar composition

pH

Time (t/h)

1.5 B2O32.25 Al2O32.5 CaO200 H2O

3.5 3.5 3.5 3.5 3.5 3.5 3.5 3.5

72 72 72 72 72 72 72 72

1.0 B2O31.5 Al2O35.0 CaO200 H2O 2.5 B2O32.0 Al2O31.0 CaO200 H2O 3.0 B2O31.5 Al2O31.0 CaO200 H2O

Temperature ( C) 160 200 160 200 160 200 160 200

Product BAC-10 BAC-1 BAC-10 BAC-1 BAC-10 BAC-1 BAC-10 BAC-1

Synthetic Chemistry of Microporous Compounds (II)

199

Figure 4.2 H2O adsorption isotherms of BAC-n. Reproduced from [30]. Copyright (1996) Elsevier

Further study showed that adsorption isotherms of BAC-n by using H2O, hexane, and cyclohexane as ‘plug gauge’ molecules were all of the I type. To illustrate these results, H2O adsorption isotherms of BAC-5, -3, and -10 compared with those of NaX, AlPO4-17, and silicalite-1 are shown in Figure 4.2. The characterization results of microporous boron–aluminum basic chlorides BAC-n are listed in Table 4.5. Another structural feature of these compounds is that Cl ions can be exchanged, which provides the conditions for both modification and application of BAC-n family. 4.1.4

Microporous Sulfides, Chlorides, and Nitrides

The known classes of open-framework materials, such as zeolites, phosphates, and oxides, are primarily dominated by oxo frameworks. In 1989, however, Bedard and Wilson at of UOP[31] extended this kind of material to include metal sulfide compounds R–M0 MS-n (R indicates organic SDA; M0 is a 3d or 4d metal; M is Ge, Sn, Sb, or In). Their frameworks are constructed by the corner-sharing or bridging of connected MSn clusters

Table 4.5 The compositions, structures, and collapse temperatures of BAC-n Mol composition BAC-n

B2O3

Al2O3

HCl

H2O

Primary building units

BAC-1 BAC-8 BAC-3 BAC-5 BAC-10

0.015 1.1 0.2 1.4 0.5

1.00 1.00 1.00 1.00 1.00

0.42 0.70 0.60 0.70 0.40

0.80 5.0 5.7 5.0 3.0

BO4, AlO6 BO3, BO4, AlO6 BO3, BO4, AlO6 BO3, BO4, AlO6 BO3, BO4, AlO6

Pore size (A˚)

Collapse temperature ( C)

3–3.6 3–3.6 3.6–4.3

450 450 325 325 350

200

Chemistry of Zeolites and Related Porous Materials

Figure 4.3

The structure of the [Ge4S10] cluster. Reproduced from [31].

(such as Ge4S10 and Sn3S4) and M0 Sm clusters (such as MnS4) (see Figures 4.3, 4.4). These compounds are usually prepared under hydrothermal conditions from a reaction mixture of metal sulfides and amine or quaternary alkylammonium species heated at 25–180  C for several days. In some preparations, the organic template is used in the carbonate ðCO3 2 Þ or hydrosulfide ðHS Þ form as a mineralizer. Later, Yaghi et al.[32] proposed an alternative route for the synthesis of these sulfides, in which crystals were obtained from the diffusion of an aqueous solution of the 3d metal into a solution of R–Ge4S10 complex at room temperature. In 1997, Martin and Greenwood[33] successfully prepared a new class of microporous metal chlorides, named CZX-n, under solvothermal conditions (160  C) in benzene solution. The framework of CZX-1 is isostructural with the aluminosilicate sodalite (SOD), while CZX-2

Figure 4.4 The connections of the [Ge4S10] cluster. Reproduced from [31]. Copyright (1989) Elsevier

Synthetic Chemistry of Microporous Compounds (II)

201

and CZX-3 have new framework topologies. All these three structures are based on corner-sharing ZnCl4 and CuCl4 tetrahedra, and the organic molecules are trapped in the micropores. This work provides a new direction for the synthesis of nonoxo framework structures. The preparation of microporous nitrides, such as Zn7[P12N24]Cl2 and Ba2Nd7Si11N23, has also been achieved by means of a high-temperature (700
800  C) solid-state reaction. Following the above introduction of four special types of microporous compounds, the synthesis of four classes of microporous compounds with special structures and functions will be discussed below in detail. 4.1.5

Extra-large Microporous Compounds

The synthesis of extra-large microporous (more than 12-rings) molecular sieves and microporous compounds has been an important target and of particular interest in the field of the synthesis of porous materials. After several decades of effort, the first aluminophosphate molecular sieve VPI-5 with 18-ring channels was successfully synthesized by Davis et al. in 1988.[34] In the following years, a few extra-large microporous molecular sieves and phosphates have been synthesized by using various organic SDAs. Some typical examples are listed in Table 4.6. Except for high-silica molecular sieves UTD-1 (DON) and CIT-5 (CFI) with 14-ring channels and germanates FDU-4, ASU-16 and FJ-1 with 24-ring channels, others listed in Table 4.6 are almost extra-large microporous phosphates including synthesized compounds and natural cacoxenite. The structural features of UTD-1 and CIT-5 are different from those of extra-large microporous phosphates. All secondary building units (SBUs) in their structures, like zeolites, are TO4 tetrahedra, while those in extra-large microporous phosphates are a mixture of TO4, TO5, and TO6 polyhedra with rich  O groups. Thus, as-synthesized extra-large microporous phosterminal P OH and P  phates normally occlude organic molecules which strongly interact with inorganic frameworks through nonbonding interactions, and which are difficult to remove to form extra-large microporous phosphate-based molecular sieves with hollow channels. In the following sections, the synthetic routes and conditions for several important extralarge-pore compounds will be described. Synthesis of UTD-1 (DON) In 1996, the high-silica zeolite UTD-1 (DON) with 14-ring channels was first successfully synthesized by Balkus et al.[35,36] The structure of UTD-1 with a formula of (Cp*2Co)2(OH)2[Si64O128] has 1-D straight channels running along the [010] direction, ˚ . Up to now, it still possesses the and its elliptical 14-ring pore has dimensions 7:5  10 A largest pore opening among the known zeolite-type molecular sieves. UTD-1 was hydrothermally synthesized in the silicate system by using bis(pentamethylcyclopentadienyl)cobalt(III) hydroxide Cp*2Co(III)OH[35] or bis(tetramethylcyclopentadienyl) cobalt(III) hydroxide Cp0 2 Co(III)OH[36] as the SDA. We use the latter as an example to describe its synthesis. Cp0 2 Co(III)OH was prepared by the reaction of tetramethylcyclopentadienyl (Aldrich) as ligands with Co(III) cations. By using Cp0 2 Co(III)OH as the SDA, UTD-1 crystals were synthesized from a reaction mixture of 1.0 SiO2 : 0.125 Cp0 2 CoOH : 0.1 NaOH : 60 H2O, stirred and aged at room temperature for 1 h, and then heated at 175  C for 2 d in

202

Chemistry of Zeolites and Related Porous Materials

Table 4.6 Some typical extra-large-pore compounds Typical compounds

Ring number

Cacoxenite

Year of SDA and synthesis discovery system

Framework composition

Channel size

1983

natural mineral

Al, Fe, P

˚ ,1-D 14.2 A channel

AlPO4-8(AET)

14

1982

dipropylamine

Al, P

˚ ,1-D 7:9  8:7 A channel

VPI-5(VFI)

18

1988

organic amine not needed

Al, P

˚ , 1-D 12.7 A channel

Cloverite (CLO)

20

1991

cycloamine, F system

Ga, P

˚ , 3-D 13 A channel

JDF–20

20

1992

triethylamine, nonaqueous system

Al, P

˚, 14:5  6:2 A 1-D channel

ULM-5

16

1994

1,6-diammoniohexane, F system

Ga, P

˚, 12:2  8:3 A 2-D channel

UTD-1 (DON)

14

1996

[(Cp*)2Co]OH

Si, Al

˚, 7:5  10 A 1-D channel

ULM-16

16

1996

cyclopentylamine, F system

Ga, P

10:5  11 A˚, 1-D channel

CIT-5 (CFI)

14

1997

N-methyl-() -sparteinium

Si

˚, 7:2  7:5 A 1-D channel

ND-1

24

1999

1,2-diaminocyclohexane

Zn, P

8.6 A˚, 1-D channel

VSB-1

24

1999

tris(2-aminoethyl) amine (TREN), F system

Ni, P

8.8 A˚, 1-D channel

FDU-4

24

2001

TREN

Ge, O

˚ , 3-D 12.65 A channel

NTHU-1

24

2001

diethylenetriamine (DETA)

Ga, P

˚ , 1-D
11 A channel

ASU-16

24

2001

1,4-diaminobutane, F system

Ge, O

˚, 8:5  15 A 1-D channel

VSB-5

24

2001

1,3-diaminopropane

Ni, P

˚ , 1-D 10.2 A channel

FJ-1

24

2005

ethylenediamine (en) or 1,2diaminopropane (enMe)

Ge, O

˚, 8:3  13:6 A 3-D channel

ZnHPO-CJ1

24

2006

butylamine

Zn, P

˚ , 2-D
11 A channel

Synthetic Chemistry of Microporous Compounds (II)

203

a Teflon-lined stainless steel autoclave. The formed yellow crystals were filtered off, washed with distilled water, and then dried at 90  C for 2 h. Organic templates Cp0 2 Coþ are usually occluded in the as-synthesized UTD-1. The typical method to remove these organic species from pores is that the samples are heated at 500  C to decompose Cp0 2 Coþ to cobalt oxide, followed by treatment with HCl to remove cobalt oxide, and then are washed by water and dried in air. The obtained UTD-1 molecular sieve has good thermal stability, and its crystal lattices remain stable upon being heated to 1000  C. The synthesis of Al-containing high-silicate UTD-1 is very difficult, which needs a two-step synthesis in general. Synthesis of CIT-5 (CFI) CIT-5 is another high-silicate molecular sieve with extra-large 14-ring channels. It was synthesized in 1997 by Davis at the California Institute of Technology by using the organic molecule N(16)-methylsparteinium (MeSPAOH) as the SDA.[37,38] CIT-5 was synthesized from a reaction mixture of 1.0 SiO2 : 0.02 Al2O3 : 0.1 MOH : 0.2 MeSPAOH : 40 H2O sealed in a Teflon-lined stainless steel autoclave and heated at 175  C for 12 d. In the absence of aluminium, pure-silica CIT-5 was prepared in 5 days. The organic template MeSPAOH was prepared in a patent.[39] Heteroatom-containing highsilica CIT-5 could be prepared from a reaction mixture of 1.0 SiO2 : 0.1 MOH : 0.2 MeSPAOH : xW : 40 H2O at 50
175  C, in which W indicates the type of heteroatomcontaining compound used, such as Ga(NO3)3, H3BO3, etc. In general, x ¼ 0
0:02, MOH indicates LiOH or a mixture of LiOH and NaOH (KOH). According to the report, CIT-5 could also be prepared in the absence of Liþ cations if alkali metal cations (Kþ or Naþ ) were present in very low concentration, [MOH]/SiO2 < 0.05 and [OH ]/SiO2 ¼ 0.3. After calcination of as-synthesized CIT-5 in air, the microporous molecular sieve CIT-5 with extra-large 14-ring channels was formed, which still had high hydrothermal and thermal stability upon being heated to 900  C.

Synthesis of VPI-5 (VFI) VPI-5 is the first extra-large microporous molecular sieve synthesized by Davis et al. in 1988.[34] Its synthesis opened the door for the exploration and development of extra-large microporous compounds in the following years. The structure of VPI-5 with a formula of [Al18P18O72]  42H2O contains 18-ring channels along the [001] direction with a pore ˚ . Two common synthetic methods for VPI-5 involve the use of diameter of 12:7  12:7 A dipropylamine (DPA) and tetrabutylammonium hydroxide (TBAOH) acting as the SDA, respectively. For the synthesis of VPI-5 by using DPA as the SDA, the reaction was performed in a gel mixture of 1.00 Al2O3 : 1.00 P2O5 : 1.00 DPA : 40 H2O in which pseudo-boehmite, H3PO4, DPA, and water were used as the reaction materials. Typically, pseudo-boehmite was dissolved in water, and then H3PO4 was added while the mixture was stirred. After aging of the mixture for 2 h, DPA was added and the mixture further stirred for 2 h. The formed gel was sealed in a Teflon-lined stainless steel autoclave and heated at 142  C for 4 h. The reaction product was washed with distilled water by

204

Chemistry of Zeolites and Related Porous Materials

Figure 4.5 Structural conversion of VPI-5 into AlPO4-8. Reproduced from [41]. Copyright (1999) Springer-Verlag

decantion three times, and then filtered off and dried at 50  C to obtain spherical crystals with a size of
100 mm. When TBAOH instead of DPA was used as the SDA, VPI-5 was synthesized from a reaction mixture with gel composition 1.00 Al2O3 : 1.00 P2O5 : 1.12 TBAOH : 50 H2O at 150  C for 20 h. The reaction process is similar to the above description, and only the crystallization conditions are slightly different. The as-synthesized product was needleshape crystals with size of
10 mm. However, the crystallized product VPI-5 (compositions of Al2O3 : 0.9 P2O5 : 0.04 DPA and Al2O3 : P2O5 : 0.006 TBAþ ) contains only very few organic SDA molecules, which rarely occur in the extra-large microporous phosphates and molecular sieves. Moreover, the as-made product of VPI-5 is very sensitive to the humidity in the air, and hydrated VPI-5 has low thermal stability. Anderson[40] reported that when VPI-5 was heated in air at 100  C for 1.2 h (from 20 to 100  C, 1 h), 19% of VPI-5 samples transformed into AlPO4-8; when heated at 130  C for 3 h (from 20 to 130  C, 1 h), 48% of VPI-5 samples transformed into AlPO4-8 (Figure 4.5). AlPO4-8 is another extra-large microporous molecular sieve with 1-D 14ring channels. However, studies show that there usually exist a large number of faults in the crystal lattices of AlPO4-8 transformed from VPI-5 by heating, which block the 14-ring windows of the largest channels and reduce the free volume. Thus, the study of increasing thermal stability of VPI-5 and its sensitivity to humidity is an important task in its synthetic chemistry. Synthesis of Cloverite (GaPO4) Cloverite was successfully synthesized by Estermann et al. in 1991, and was the second extra-large microporous compound at that time.[42] Cloverite is a microporous gallophosphate with the formula [Ga96P96O372(OH)24](QF)24(H2O)n (Q is quinuclidine) (F is fluorinion). Its channels are defined by a 20-ring window with the channel structure h100i20 4:0  13:2 jh100i8 3:8  3:8 . The synthesis of cloverite involves the use of Ga2(SO4)3xH2O, H3PO4, HF, and quinuclidine (Q) as the reaction materials to form a gel with molar proportions Ga2O3 : P2O5 : HF : 80 H2O : 6Q. Typically, Ga2(SO4)3xH2O, H3PO4, and water were mixed and stirred until completes dissolution, and then HF and quinuclidine were added to the above mixture. A homogeneous reaction gel

Synthetic Chemistry of Microporous Compounds (II)

205

ðpH ¼ 4
4:5Þ was formed after stirring, and was then sealed in a PTFE-lined stainless steel autoclave heated at 150  C for 24 h. The product was washed with water and dried at 60  C to obtain small cubic crystals (
1 mm) with the composition Ga0.48P0.52O2 Q0.14F0.13. Similarly to VPI-5, cloverite is also very sensitive to the humidity in the air. Particularly, when the temperature increases, the moisture will destroy the crystallinity of cloverite. In order to synthesize cloverite with higher thermal ability, Huo and Xu[43] tried to use ethylene glycol as the solvent and piperidine as the SDA to prepare cloverite with higher crystallinity from a reaction mixture of 1.0 Ga2O3 : 1.8 P2O5 : 4.3 piperidine : 1.7 HF : 44 EG at 140 or 180  C for 18 or 10 d. Its crystallinity was unchanged when heated at 400  C, and transformed into tridymite GaPO4 phase when heated to 700  C. If cloverite was exposed in the air after being heated at 400  C, the moisture would also destroy its crystallinity. As with VPI-5, studying the increasing thermal and hydrothermal stability of cloverite is also a vital task in its synthetic chemistry. Synthesis of JDF-20 Following the discoveries of extra-large microporous aluminophosphate VPI-5 (18-rings) and gallophosphate cloverite (20-rings) by Davis and Estermann in 1998 and 1991, respectively, in 1992, Xu and Huo at Jilin University successfully prepared aluminophosphate JDF-20 with extra-large 20-ring channels, and this promoted the study of extralarge microporous aluminophosphate to a new stage. The structure of JDF-20 is composed of large 20-ring elliptical channels intersected with small 10-MR and 8-MR channels. Up to now, JDF-20 with a formula of [Al5P6O24H]22[Et3NH]þ 2H2O has the largest known pore opening and pore size in the microporous aluminophosphates. Of particular interest is that the P/Al ratio in its framework is 6/5, not the typical 1/1. This is because terminal P OH groups exist in the framework, resulting in the interrupted structure of JDF-20. Four P OH groups in one 20-ring window close interact with four triethylamine (TEA) molecules through H-bonds, as seen in Figure 4.6. JDF-20 was synthesized by using TEA as the SDA in an alcoholic solvent with weak polarity, such as diethylene glycol, triethylene glycol, tetraethylene glycol, or 1,4-butanediol, whereas AlPO4-5 would be formed if a solvent with high polarity (ethylene glycol or ethanol) were used under the same reaction condition and composition.

Figure 4.6

Structure of JDF-20

206

Chemistry of Zeolites and Related Porous Materials

The synthesis of JDF-20 involves the use of Al(OPri)3, H3PO4, triethylamine (TEA), and triethylene glycol (tEG) as the reaction materials to prepare a gel mixture of composition 1.0 Al2O3 : 1.8 P2O5 : 5.9 Et3N : 14 tEG. Typically, Al(Opri)3 and tEG solvent were mixed, and then TEA was added dropwise to the above stirred mixture, and the mixture was further stirred for 10 min. Finally, H3PO4 was added to the stirred mixture. After aging for 2 h, the formed homogeneous gel was sealed in a Teflon-lined stainless steel autoclave and heated at 180  C for 10 d. The as-synthesized product was washed with water and dried at room temperature to obtain regular crystals with sheet-like morphology. JDF-20 crystals could also be prepared by using tetraethylene glycol, diethylene glycol, or 1,4-butanediol as the solvent instead of tEG under the similar conditions.[44] JDF-20 is very sensitive to the humidity in the air, like VPI-5 and cloverite. When heated above 300  C, JDF-20 transforms into AlPO4-5 after removal of organic Et3N molecules.[45] Synthesis of Gallophosphates ULM-5 and ULM-16 In 1994, ULM-5 was first successfully synthesized by Loiseau et al. in an F -containing system.[46] ULM-5 is a fluorinated gallophosphate containing 16-ring channels, which has the formula Ga16(PO4)14(HPO4)2(OH)2F7  4 [H3N(CH2)6NH3]  6 H2O, and its ˚ $ [100] 8. The synthesis of channel structure is described as [100] 16 12:2  8:34 A ULM-5 involves the use of Ga2O3, H3PO4, HF, and H2N(CH2)6NH2 (DAH) as the reaction materials to form a gel mixture of 1 Ga2O3 : 1 P2O5 : 2 HF : 1 DAH : 80 H2O, which was heated at 180  C for 24 h. The location of organic SDAs in the 16-ring channel of ULM-5 is shown in Figure 4.7. The structure of ULM-5 is not stable to calcination, and collapses after the removal of organic molecules upon being heated to
330  C. ULM-16 is another gallophosphate with extra-large 16-ring channels, synthesized by Fe´rey and colleagues in an F -ion-containing system in 2002.[47] ULM-16 has the formula Ga4(PO4)4F1.33(OH)0.67  1.5 NC5H12  0.5 H3O  0.5 H2O, which also contains F ions, similar to ULM-5. It was synthesized in a reaction system similar to that for ULM-5 by using cyclopentylamine (CPA) as the SDA from a reaction mixture of 1 GaOOH : 1 H3PO4 : 0.5 HF : 0.6 CPA : 40 H2O at 180  C for 3 d. Figure 4.8 shows the location of organic SDA molecules in the 16-ring channel of ULM-16. When ULM-16 is heated to 250
400  C, organic CPA molecules and some of the F ions are removed, which results in collapse of the structure.

Figure 4.7

Structure of ULM-5

Synthetic Chemistry of Microporous Compounds (II)

207

Figure 4.8 Structure of ULM-16

Based on the successful synthesis of ULM-5 and ULM-16, Fe´rey and coworkers prepared MIL-31[48] with extra-large 18-ring channels (Figure 4.9) by using larger alkyldiamines as the SDA, such as H3N(CH2)9NH3 (DAN) and H3N(CH2)10NH3 (DAD). The structure of MIL-31 collapses after the removal of organic amines upon heating above 300  C. Synthesis of Phosphates NTHU-1 and ND-1 In 2001, Lii and coworkers successfully prepared gallophosphate NTHU-1 with 24-ring channels, which had the formula [Ga2(DETA)(PO4)2]2H2O (DETA ¼ diethylenetria˚ ) among reported mine).[49] Up to now, it possesses the largest pore channels (
11 A gallophosphates, as seen in Figure 4.10. NTHU-1 was synthesized in the mixed solvents of water and ethylene glycol from a reaction mixture of 1 Ga2O3 : 6 P2O5 : 5.1 DETA : 333 H2O : 110 EG heated at 180  C for 3 d. When the crystal product of NTHU-1 is heated to 300  C, the structure collapses with removal of the organic SDAs, and finally converts into dense-phase GaPO4 upon being heated to 1150  C. ND-1 is zinc phosphate with extra-large 24-ring channels reported by Yang and Serov in 1999, which has the formula Zn3(PO4)2(PO3OH)(H2DACH)2 H2O (DACH ¼ 1,2diaminocyclohexane).[50] It is the second synthesized phosphate with extra-large 24-ring channels (Figure 4.11). ND-1 was prepared from a reaction mixture of H3PO4, Zn(Ac)2,

Figure 4.9 Structure of MIL-31

208

Chemistry of Zeolites and Related Porous Materials

Figure 4.10 Elsevier

Structure of NTHU-1. Reprinted with permission from [36]. Copyright (1997)

DACH, and H2O with molar proportions 1.0 P2O5 : 1.0 ZnO : 2.5 DACH : 117 H2O at 180
210  C for 2 d. Like other extra-large microporous phosphates described above, the thermal stability of ND-1 is poor. Upon heating to 350  C, its structure collapses with the loss of organic amines, and converts into Zn2P2O7 above 700  C. Synthesis of Nickel Phosphates VSB-1 and VSB-5 VSB-1[51] and VSB-5[52] are two extra-large microporous nickel phosphates with 24-rings reported by Guillou et al. in 1999 and 2001, respectively. VSB-1 with the formula Ni18(HPO4)14(OH)3F9(H3O,NH4)412H2O has 24-ring channels along the [001] ˚ . It was prepared in the F system direction, and its free diameter is approximately 9.0 A from a hydrothermal reaction mixture of nickel(II) chloride hexahydrate, phosphoric acid, tris(2-aminoethyl)amine (TREN), pyridine, HF, and water in the molar proportions 5 : 5 : 2 : 9 : 12 : 200 at 180  C for 6 d. However, it was found that neither TREN nor pyridine was incorporated into the resulting structure, and the same product could be obtained from an aqueous solution in the presence of ammonium or potassium fluoride with Ni/NH4þ or Ni/Kþ ¼ 1=3. However, VSB-1 counld not be synthesized in the absence of fluoride ions. VSB-5 with the formula Ni20(HPO4)8(PO4)4(OH)1212H2O also has one-dimensional ˚, 24-ring channels along the [001] direction, and its free diameter is approximately 10.2 A slightly larger than that of VSB-1. VSB-5 could be synthesized from a hydrothermal

Figure 4.11 Chemistry

Structure of ND-1. Reprinted from [37]. Copyright (1997) Royal Society of

Synthetic Chemistry of Microporous Compounds (II)

209

reaction with pH values of 7:3
11:0 in a wide range of Ni/P ratios and concentrations. Although VSB-5 has not been prepared in the absence of an organic amine, the particular amine does not seem to be important. For example, a range of diamines from 1,2ethylenediamine through to 1,8-octanediamine could be used as the SDA. In a typical synthesis, NiCl26H2O (6.0 g) was dissolved in water (60 mL), followed by the addition of H3PO4 (6.2 g, 85%) and 1,3-diaminopropane (9.8 g). The final solution with a pH of 9.0 was sealed in a 120 mL Teflon-lined Parr autoclave heated at 180  C for 5 d to form pure phase of VSB-5. Different from other phosphates, VSB-1 and VSB-5 have good thermal stability, and their frameworks could be sustained in the air up to approximately 823 K and 723 K, respectively. This exhibits typical zeolitic properties, and provides their potential applications in ion-exchange, catalysis, etc. Synthesis of Germanates FDU-4, ASU-16, and FJ-1 Besides extra-large microporous phosphates and silicates, in 2001, Zhou et al. reported the first zeolite-like germanate [Ge9O17(OH)4][N(CH2CH2NH3)3]2/3[HCON(CH3)2]1/6 (H2O)11/3 (FDU-4) with 24-ring channels.[53] Its open-framework structure has a 3-D ˚ ) in FDU-4 is intersecting channel system. Each 24-ring channel (about 12.65 A ˚ surrounded by six 12-ring (about 9.52 A) channels; both channel types are connected by alternating 8-ring pores. The organic amine molecules are located within the 12-ring channels and some solvent species such as DMF and water molecules are disordered and located in the 24-ring channels (Figure 4.12). FDU-4 was prepared by using the organic multiamine tris(2-aminoethyl)amine (TREN) as an SDA and a mixture of N,N-dimethylformamide (DMF) and water as the solvent. A reaction mixture containing GeO2, TREN, water, and DMF in the molar proportions 3.95 : 16.1 : 278 : 110 was heated at 180  C for

Figure 4.12 Structure of FDU-4

210

Chemistry of Zeolites and Related Porous Materials

Figure 4.13

Structure of ASU-16

7 d to produce colorless needle-like crystals of FDU-4. The structure of FDU-4 collapses upon heating to 400  C for 3 h, suggesting an unstable framework for FDU-4. ASU-16, formulated as Ge14O29F4[H2DAB]3[DAB]0.516H2O, is another germanate with extra-large 24-ring channels, synthesized by Ple´vert et al. in 2001.[54] Its structure features 24-ring channels along the [100] direction, and 8-, 10-, and 12-rings perpendicular to the channel direction. Only two TREN molecules were clearly located by X-ray analysis, which were found in the 10- and 12-ring windows (Figure 4.13). Remaining species are disordered in the channels. ASU-16 was synthesized under hydrothermal conditions from a mixture of GeO2, water, 1,4-diaminobutane (DAB), pyridine, and hydrofluoric acid (48 wt %) with typical molar proportions of 1.0 : 70 : 12 : 40 : 2 heated at 160  C for 4 d. The resulting product was washed, filtered off, and then dried at room temperature. As with FDU-4, ASU-16 has bad thermal stability, its framework collapsing upon being heated above 150  C. In 2005, the novel extra-large microporous germanate FJ-1 with 24-rings was successfully prepared by Lin et al. in the presence of metal complex [NiCl2(L)3] [L ¼ ethylenediamine (en), 1,2-diaminopropane (enMe)] as SDA.[55] The open framework of FJ-1 formulated as Ni@Ge14O24(OH)32Ni(L)3 (L ¼ en/enMe) has unique Ge Ni Ge linkages and chiral [Ni@Ge14O24(OH)3] cluster motifs, which may open up possibilities for the synthesis of novel frameworks with T M T linkages. [Ni(L)3]2þ ˚ ) (Figure 4.14), which intertemplates occluded in the 24-ring channels (8:3  13:6 A sected with two 12-ring channels running along the a and b axes, respectively. FJ-1 was synthesized from the solvothermal reaction of [NiCl2(L)3]2H2O and GeO2 in mixed

Synthetic Chemistry of Microporous Compounds (II)

Figure 4.14

211

Structure of FJ-1

solvents of ethylene glycol and L (L ¼ en/enMe) at 170  C for 7 d in a Teflon-lined steel autoclave. Otherwise, FJ-1 was also obtained if complex [Ni(L)3Cl2]2H2O was replaced by NiCl22H2O and en/enMe. Synthesis of Zinc Phosphite ZnHPO-CJ1 More recently, Liang et al. at Jilin University reported the first metal phosphite jðC4 H12 NÞ2 j½Zn3 ðHPO3 Þ4  (ZnHPO-CJ1) with extra-large 24-ring channels.[56] The ˚ ) and 8-ring structure of ZnHPO-CJ1 possesses parallel 24-ring (11:0  11:0 A ˚ (3:4  3:4 A) channels extending along the [001] direction. Eight protonated CH3 ðCH2 Þ3 NH3 þ cations reside in each 24-ring window, and inteact with the framework through H-bonds (Figure 4.15). The 8-ring channels are hollow but contain no encapsulated molecules. ZnHPO-CJ1 was prepared by using butylamine as the SDA via hydrothermal reaction of a mixture of 1.0 Zn(OAc)2H2O : 4.5 butylamine : 3.0 H3PO3 : 488 H2O. Typically, Zn(OAc)2H2O was first dissolved in water, and then butylamine and H3PO3 was added while the mixture was stirred. After further stirring, the formed homogeneous reaction gel ðpH
6Þ was sealed in a Teflon-lined stainless steel autoclave and heated at 180  C for 3 d. The colorless rod-shaped single crystals were separated by sonication, washed with distilled water, and then dried in air. Like known metal phosphates with extra-large microporous channel structures, the structure of ZnHPO-CJ1 is unstable upon heating due to the existence of P H groups and strong H-bonding guest–host interactions. When heated at 280  C for 3 h, its structure collapses after the removel of the occluded organic molecules. From the above descriptions of many extra-large microporous phosphates, the syntheses and structural features of such phosphates can be further understood and some common rules can be started as follows: (1) As opposed to the extra-large microporous high-silica compounds UTD-1 and CIT-5, which are templated by bulky organic cations with low charge density, many metal

212

Chemistry of Zeolites and Related Porous Materials

Figure 4.15

Structure of ZnHPO-CJ1. Reproduced from [56]. Copyright (2006) Wiley-VCH

phosphates with extra-large pores have been prepared in the presence of relatively small organic amines which yield ammonium ions with high charge density upon protonation. In such structures, several ammonium ions are included in one extralarge pore. This exemplifies the idea of a supramolecular templating process being used for the preparation of mesoporous materials. (2) The organic SDA molecules favor the formation of H-bonds with the terminal P OH  O groups of the primary and secondary building units, or H-bonds with the and P  bridging oxygen atoms between the atoms with small electronegativity and the P-atom, such as Ga O P linkages, or even coordination bonds with Ga-atoms. Further polymerization tends to result in channel structures with large pores. (3) On the other hand, due to the strong and complex H-bonding interactions between the organic guests and inorganic frameworks in such extra-large-pore phosphates, it is very difficult to form extra-large microporous molecular sieves with uniform structures by calcination to remove the organic SDAs. Recently, many researchers on molecular sieves have been trying to employ other routes to remove the SDA molecules under mild conditions, and to keep the integrity of channel structures. This is an interesting scientific issue, and will be further described in Chapter 6. 4.1.6

Zeolite-like Molecular Sieves with Intersecting (or Interconnected) Channels

In terms of the demands of catalytic reactions, the synthesis and exploration of molecular sieves with intersecting channels, in particular with interconnected 10-ring (middle pore) and 12-ring (large pore) channels, are of great significance in the aspects of molecular diffusion and shape-selective catalytic reactions. Up to now, such structural molecular sieves include only nature zeolite boggsite (BOG) with an intersecting 10-ring and

Synthetic Chemistry of Microporous Compounds (II)

213

12-ring tridirectional channel structure. The structure of Nu-87 contains interconnected 10-ring and 12-ring channels, but only the 10-ring windows open outwards. Although the structures of MCM-22 and its analogs SSZ-25, ERB-1, and PSH-3 possess 10-ring and 12-ring channels, they are isolated by a large cavity, and the guest molecules cannot diffuse between the different channels. In addition, SSZ-26, SSZ-33, and CIT-1 (½00112, ˚ $ ½01012, 7:0  5:9 A; ˚ $ ½01010, 5:1  4:5 A) ˚ have intersecting 10- and 6:4  7:0 A 12-ring channels, but, regrettably, all of them are faulted intergrowth materials, and their channels are all composed of three polymorphs A, B, and C. In the past few years, Corma’s research group [57] first successfully synthesized the ˚ 10-ring Ge-containing zeolite ITQ-22 with interconnected 8-ring ð4:52  3:32 AÞ, ˚ ˚ ð5:86  4:98 AÞ and 12-ring ð6:66  6:66 AÞ pores by combining the structure-directing effect of the organic guest 1,5-bis(methylpyrrolidinio)pentane and the framework isomorphic substitution of germanium for silicon. A [4458612] cage is the basic building unit (BBU) [see Figure 2.43(a)], which is connected with double 4-rings (D4R) to form the interconnected 8-, 10-, and 12-ring channels of ITQ-22 [see Figure 2.43(c)]. In 2003, Corma and coworkers[58] reported the synthesis of zeolite Al-ITQ-24 by using hexamethonium dihydroxide [R(OH)2] as an SDA in the reaction mixture of 5.0 SiO2 : 1.0 GeO2 : 0.15 Al2O3 : 1.5 R(OH)2 : 30 H2O at 150  C for 15 d. Its structure presents tri˚ running directional pore systems, which include a 12-ring straight channel ð7:7  5:6 AÞ ˚ running along the perpendicular to the ab plane, a 12-ring sinusoidal channel ð7:7  6:2 AÞ ˚ intersecting perpendicularly both 12-ring a axis, and a 10-ring channel ð5:7  4:8 AÞ channel systems (see Figure 2.49). Ti-containing ITQ-24 can also be prepared by the same synthetic route, and its structure resembles that of ITQ-24. In the structure of ITQ-24, the 12-ring sinusoidal channel parallel to the ab plane is composed of the D4R cage, which is similar to the structure of polymorph C in SSZ-33/SSZ-26/CIT-1. In fact, the synthesis of zeolite materials with 10- and 12-ring pores, as well as the D4R cage, by Corma and coworkers in the presence of a large SDA and by using Ge and Si as starting reaction materials to polymerize with Al is based on the understanding of the structural features of polymorphs and the utilization of Ge to form an open-channel structures.[58] That is, compared with an Si O bond, a Ge O bond length is longer and more easily bend to small T O T angle, which favors the formation of a small D4R cage.[59] By using the same idea, Corma et al.[60] synthesized polymorph C of Ge-containing zeolite BEA in 2001 that had a tridirectional pore system with intersecting 12-ring channels, and the D4R cage acted as the BBU. The synthesis of this pure polymorph is important in the synthetic chemistry of microporous compounds, and it opens up a new research direction. Therefore, the important status of zeolite BEA within the molecular sieves will be particularly described. One reason for its importance is that the structure of BEA contains an intersecting 12-ring 3-D channel system. However, as with SSZ-33/SSZ-26/CIT-5, BEA is also formed by an intergrowth of polymorphs A, B, and C, which in general results in the existence of defaults. Notably, the hypothetical structure of polymorph C has a threedimensional pore topology, in which all three 12-ring channels are linear (P42/mmc), while in the case of polymorphs A (P4122) and B one of the channels is sinusoidal. The reason is that polymorph C contains double 4-ring (D4R) cages acting as BBUs (see Figure 2.38), while the other two polymorphs do not contain such D4R cages. Based on previous experience in the synthesis of structures with D4R cages, Corma et al. carried out a series of syntheses by using different organic SDAs (Figure 4.16), and

214

Chemistry of Zeolites and Related Porous Materials

Figure 4.16 Various SDAs used in the synthesis of polymorph C of Ge-containing zeolite BEA. Reproduced with permission from [60]. Copyright (2001) Wiley-VCH

Ge and Si as the starting reaction materials in the gel compositions ð1  xÞ SiO2 : xGeO2 : (0.5
0.25) SDAOH : 0.5 HF : wH2O at 135
175  C for 15
120 h. The crystallized products are listed in Table 4.7. Polymorph C of Ge-containing zeolite BEA with Si/Ge ratios of 1.8 : 1, 5.0 : 1, and 11.6 : 1 is prepared from the following three reaction systems at 150  C for 15 h, respectively. In the absence of Ge, general zeolite BEA or pure silica ZSM-12 is obtained. The three crystallization systems adopted by Corma are: (1) 0.666 SiO2 : 0.333 GeO2 : 0.5 BDOH : 0.5 HF : 8 H2O (2) 0.833 SiO2 : 0.166 GeO2 : 0.5 BDOH : 0.5 HF : 8 H2O (3) 0.937 SiO2 : 0.062 GeO2 : 0.5 BDOH : 0.5 HF : 8 H2O In conclusion, some interesting structures with intersecting or interconnected channel systems have been presented in this section, which have important applications in

Table 4.7 Synthesis conditions and zeolite structures obtained by using different SDAs with and without germanium in the crystallization gel SDA BQþ BQþ m-XydQ2þ M4BQ2þ M4BQ2þ CyHMPþ MCyHMPþ TEAþ

H2O[w] 8 8
24 8 7.5
15 15 7.25 7.25 7.5

Si/Ge 1
10 0.5
30 5 2
10 2
20 5 5 2

T[ C]

Time (t/h)

Ge zeolite

150 135
175 150 175 175 135
175 135
175 140

15
120 15
120 16
96 24
96 24
96 15
96 15
96 96

Polymorph polymorph polymorph polymorph polymorph polymorph polymorph polymorph

Si zeolite C C C C C C C C

ITQ-4 ITQ-4 Beta Beta, ZSM-12 ZSM-12 ZSM-12 Beta Beta

Synthetic Chemistry of Microporous Compounds (II)

215

shape-selective catalysis. It is a long-term goal to explore new synthetic routes for aluminosilicate zeolites with similar channel systems. Such efforts will enhance the study and exploration of extra-large microporous compounds and molecular sieves. 4.1.7

Pillared Layered Microporous Materials

Another approach to the preparation of extra-large microporous or mixed-channel structures is based on dimension expansion from low-dimensional compounds to three-dimensional frameworks. A feasible method is the pillaring of inorganic layers or sheets with certain compositions and structures by pillars with specific compositions and lengths to form pillared layered microporous or mesoporous materials. Owing to the variability of layer or sheet structures (including layered porous structures), and the lengths and structures of pillars, novel structural materials with mixed-channel systems can be prepared through control of the reaction conditions. The layer sources commonly used are synthetic and natural mineral layered silicates, layered aluminosilicates and other heteroatom-substituted layered compounds, layered aluminophosphates, layered titanates, layered manganates, layered niobates, and related sheets, etc. The low polymers of silicon–oxygen, aluminum– oxygen, titanium–oxygen and chromium–oxygen compounds, and some organic compounds or low organic polymers, are common pillar materials. In the following, we use layered silicates as an example and TEOS or [Al13O4(OH)24(H2O)12]7þ as a pillar agent to discuss the synthesis of pillared layered materials. Cheng and coworkers at Taiwan University[61] have discussed the silica- and aluminapillared derivatives of four layered silicates with different charge density and thickness, as well as related synthetic problems. Here, this system will be presented as a typical example. In general, its synthesis includes three steps. Preparation of Layered Silicates Four layered silicates with different thicknesses and structures were used, i.e., ˚ ), Na2Si8O17xH2O (6.5 A ˚ ), Na2Si14O2911H2O (9.6 A ˚ ), and NaHSi2O53H2O (4.5 A ˚ ). The last three layered compounds were hydrothermally K2Si20O4110H2O (15.1 A prepared in the NaOH (or KOH)–SiO2–H2O system at 100  C for several days. NaHSi2O53H2O was prepared by infusing water glass (sodium silicate) solution into methanol. The formed precipitate was filtered off, dried at 100  C, then heated in the muffle at 700  C for 5
6 h, and followed by rehydration. Preparation of Hexylamine-expanded Silicates The hexylamine-intercalation method was usually used: four layered compounds were treated by the exchange of acid (1M-HNO3 with stirring for several hours at 70  C) to form Hþ -type product, which was dried in air, and then suspended in the excess of hexylamine solution at room temperature and the mixture was stirred for 28 d. Hexylamine was removed by filtration and the hexylamine-expanded silicates were used immediately for the pillaring reaction in order to prevent hexylamine evaporation. Preparation of Silica- and Alumina-pillared Silicates (1) Silica-pillared silicates were prepared by stirring hexylamine-expanded silicates with an excess of TEOS at room temperature for 1
3 days, and then filtered, washed with ethanol, and heated at 360  C for 2 h.

216

Chemistry of Zeolites and Related Porous Materials

Table 4.8 Free interlayer spaces of pillared silicates (A˚) Hexylamine-expanded NaHSi2O53H2O Na2Si8O17xH2O Na2Si14O2911H2O K2Si20O4110H2O

21 23 23 26

Silica-pillared 15 10 10 9

Alumina-pillared 5.1 1.6 4.4

(2) Alumina-pillared silicates were prepared by ion-exchanging the interlayer hexylammonium ions with an excess of 0.086M-[Al13O4(OH)24(H2O)12]7þ liquid solution with stirring at 50  C for 1 d, and then filtered, washed, and heated at 300  C. After the XRD characterization, the measured free interlayer spaces are listed in Table 4.8. The surface areas of silica- and alumina-pillared silicates usually distribute in the range 300
500 m2 g1 and 180
280 m2 g1 , respectively, and their adsorption isothermals are consistent with a microporous character. As shown in Table 4.8, the free interlayer spaces of alumina-pillared silicates are smaller than those of silica-pillared ˚ ). silicates, and also smaller than the length of the [Al13O4(OH)24(H2O)12]7þ cation (8.4 A This suggests that decomposition of the Al13 7þ pillar may occur in the process of intercalation-exchange, which results in pillaring of the silicates by more low polymers of aluminum hydroxide. As with silica- and alumina-pillared silicates, attention should be focused on several problems in the synthesis of various pillared microporous materials. The first one is the preparation of the layer. In general, the synthesized clays, such as synthesized SMM,[62] sapornite,[63] and their heteroatom substitutions of Fe, Ni, Co, Ga, etc.,[63] are all prepared at
300  C by hydrothermal methods, while various layered aluminophosphates with rich structural architectures are synthesized at
200  C in either hydrothermal or solvothermal systems.[64] The synthesis and structural characterization of layered compounds with high purity are the foundation for the preparation of pillared materials. Next is the problem of swelling or preswelling of layered compounds. This is an important procedure to enable the pillars (chemical species such as molecule, ion, complex, etc.) to enter the interlayer spaces. Some layered compounds favor swelling and peeling off, therefore only pillars need to be added to the liquid solution of layered compounds. However, the addition conditions must be carefully studied to facilitate the pillars’ entering into the interlayer spaces and to achieve ‘pillaring’. On the other hand, some layered compounds, such as layered silicates and numerous transition metal oxide layers including titanates, manganates, and niobates, are not easily swelled in water. A preswelling step should be undertaken. For example, the long-chain amines are intercalated into the interlayers to expand the interlayer spaces, making possible the pillars’ entrance. Furthermore, the intercalation of amines creates an organophilic environment between layers, which facilitates the entering of TEOS into the interlayer spaces. Based on this, the pillars are intercalated or exchanged into the interlayer regions. After that, the problem of ‘pillaring’ becomes more complex. The complexity can be briefly explained by using the example of the ‘pillaring’ of TEOS in the silicate layers. It includes TEOS, the hydrolysis of Si(OC2H5)4, i.e., SiðOC2 H5 Þ4 þ nH2 O ! SiðOC2 H5 Þ4n ðOHÞn þ nC2 H5 OH, and condensations among

Synthetic Chemistry of Microporous Compounds (II)

217

Figure 4.17 Stepwise ion-exchange process. Reproduced with permission from [69]. Copyright (2003) Science in China Press

the hydrolysis products of SiðOC2 H5 Þ4n ðOHÞn . The formation and distribution of low polymers of silicon–oxygen produced in these two reactions are different owing to the change in reaction conditions. According to the differences of interlayer structures and spaces of hexylamine-expanded silicates, some low polymers of silicon–oxygen selectively enter the interspaces, and rationally interact with the Si OH groups on the surfaces of silicates. After calcination at a certain temperature, bonds are formed between the pillars and layers to connect them together. Not only the condensation of related groups between the layers and pillars, but also the bonding or H-bonding interactions, are all closely related to the structures of layers and the compositions of pillars. The selection of matched pairs and the conditions of ‘pillaring’ and ‘formation of pores’ based on the demand of channel functions and the structural characters of layers and pillars are of particular importance. Some preliminary rules can be concluded from the following examples. Apart from the study of silica- and alumina-pillared silicates by Cheng,[61] Valverde[65] studied the titanium-pillared clays using hydrolysed titanium ethoxide Ti(OC2H5)4 as a pillaring agent. Alberti et al.[66] investigated the organic phosphonate-pillared a-Zr(HPO4)2H2O. Cheng and colleagues also studied the pillaring of layered manganese oxide[67] and layered titanates[68] using [Al13B4(OH)24(H2O)12]7þ as the pillaring agent. Recently, Hou and coworkers[69] at Nanjing University presented a comprehensive summary of the synthetic chemistry of pillared layered transition metal oxides, dividing the synthetic routes into two modes on the basis of previous work. (1) Stepwise ion-exchange process (Figure 4.17) (2) Delamination procedure (Figure 4.18)

Figure 4.18 Press

Delamination procedure. Reproduced [69]. Copyright (2003) Science in China

218

4.1.8

Chemistry of Zeolites and Related Porous Materials

Microporous Chiral Catalytic Materials

The individual chemical species with chiral catalytic properties, such as complex, organometallic compounds, organic ligands or molecules, anchored or grafted into the channels of microporous and mesoporous materials, and some microporous compounds possessing chiral channels or their pore structures composed of the chiral motifs, all promise further development and potential application in microporous chiral (asymmetric) catalysis and separations. It is an important frontier direction in the zeolite catalytic field at present. Therefore, the synthesis and assembly of chiral microporous compounds and materials are of particular interest for researchers engaged in porous materials. This is a research field in rapid development. Assembly of Chiral Catalytic Centers in the Pores (Microporous and Mesoporous) of Molecular Sieves Example 1: Chiral rhodium complexes anchored on modified USY zeolites.[70] For the asymmetric catalytic hydrogenation reaction of N-acylphenylaniline, the product not only has high translation yield and ee% value (>95%), but also has a long reaction life. The assembly process of rhodium complex catalyser is as follows: First step: L-Proline as the chelate agent to prepare Rhþ complex.

Second step: The preparation of extra-stable Y-type zeolite USY with more surface Si OH groups. NH4Y zeolite was calcined at 1000  C for 2 h, followed by the surface treatment of dealuminum and hydroxidation by citric acid at 130  C to obtain USY (pore ˚ ) with surface Si size>15 A OH groups. Third step: Rhþ complex anchored on USY.

Synthetic Chemistry of Microporous Compounds (II)

219

Example 2: Sharpless catalysts grafted onto mesoporous MCM-41[71] Katsuki and Sharpless have found a highly efficient catalytic material for the epoxidation of allylic alcohols which consists of titanium tetraisopropoxide and a chiral dialkyl tartrate.[72] However, this homogeneous catalytic system has a disadvantage: it is necessary to separate the product and catalyst. In order to solve this problem, Li and coworkers at the Dalian Institute of Chemical Physics tried to graft tartaric acid derivatives into the channels of mesoporous MCM-41,[71] which shows high efficiency in the application of heterogeneous asymmetric epoxidation of allyl alcohol. The chiral tartaric acid derivatives are grafted onto inorganic supports in several steps, as outlined in Figure 4.19. Apart from the above described chiral porous assembly materials, zeolites with chiral structural features, i.e., helical channel structure, the channels of microporous materials are composed of chiral motifs, have been continuously studied by researchers in an attempt to explore microporous chiral catalytic materials. Synthesis of Microporous Compounds with Chiral Channels or Chiral Structural Features In theory, there are 66 chiral space groups among 230 space groups, such as P41, P43, P61, P62, P31, P32, P65, P64, etc. The framework structures crystallized in these space groups do not contain any symmetric element, and this results in the chiral structures. However, the occurrence of chiral frameworks is rare, and an optically pure chiral zeolite material has never been found. Among the large variety of zeolites and related compounds, only a few are known to have chiral channels or chiral structural features. Therefore, it is difficult to summarize the rules for the synthesis of such compounds. Herein, we will present a rough overview of the synthesis of chiral microporous structures or microporous compounds with chiral structural features based on their framework compositions. (1) The synthesis of chiral microporous silicates In the case of silicates, only a few chiral silicates are known, such as the natural mineral goosecreekite (GOO) with the formula CaAl2Si6O165H2O,[73] b Zeolite (BEA),[74] beryllium silicate OSB-1 (OSO),[75] and zincosilicate NaZnSiO3OH.[76] b Zeolite is a notable example of a chiral zeolite proposed by Newsam in 1988.[74(a)] It is an intergrowth of three different but structurally related polymorphs A, B, and C, in which polymorph A has chiral pores.[74] Zeolite BEA with various Al/Si ratios can be synthesized by using many different SDAs in the hydrothermal systems, which usually have different compositions of polymorphs A, B, and C, as well as different defaults. Davis has reported the synthesis of zeolite BEA with enriched polymorph A by using a chiral organic SDA, which has been confirmed by the ee% value in the enantioselective catalysis for trans-stilbene oxide.[74(b)] However, there is still a lack of powerful evidence supporting this result, and further results related to this study have not been reported. In 2001, OSB-1[75] with the formula K6[Be3Si6O18]H2O (zeotype OSO) was successfully prepared by Cheetham and coworkers. It is the first molecular sieve framework constructed by 3-rings only, whose 14-ring channels are chiral and formed by a double-helix chain out of 3-rings (Figure 4.20). OSB-1 was synthesized from a reaction mixture of K6BeO4, K2Si2O5, KOH, and H2O in the molar proportions 1 BeO : 18 K2O : 2.6 SiO2 : 180 H2O in a stainless steel autoclave at 175  C for 4 d.

220

Chemistry of Zeolites and Related Porous Materials

Figure 4.19 Sharpless catalytic system grafted onto MCM-41. Reproduced with permission from [71]. Copyright (2002) Wiley-VCH

Synthetic Chemistry of Microporous Compounds (II)

Figure 4.20

221

14-ring channels of OSB-1 viewed along [010] (up) and [001] (down) directions

NaZnSiO3OH was synthesized by Healey et al. in 1999.[76] Its structure, constructed from ZnO4 and SiO3OH tetrahedra, crystallizes in the noncentrosymmetric space group P212121, producing a chiral open framework. Sodium ions are housed in the 8-ring channels. In the synthesis of NaZnSiO3OH, Na2[ZnSiO4] as a precursor was first synthesized hydrothermally in Teflon-lined steel autoclaves from zinc oxide and sodium metasilicate and then reacted with the sodium hydroxide solution at 180  C for 1 d. Then Na2[ZnSiO4] was further treated in the same autoclaves with dilute sodium hydroxide solution (2 M) at 225  C for 5 d. The obtained products are impure, and the crystals of predominant phase NaZnSiO3OH can be easily separated and washed with distilled water to remove any of the second phase. (2) The synthesis of chiral microporous germanates A typical example of chiral germanates is zeolite-like UCSB-7 with 3-D cross-linked helical pores (Figure 4.21) reported by Stucky and coworkers in 1998.[77] UCSB-7 with various compositions (zinc arsenate, beryllium arsenate, and gallium germanate) was prepared in the aqueous system by using inorganic cations (Kþ , Naþ ) or organic amines (en) as SDAs at low temperature (20–100  C). For example, (NaGaGeO4)8xH2O (x
7) was synthesized from a reaction mixture of Na5GaO4, Na4GeO4, H2O, and HNO3 heated at 100  C for 7 h. The framework structure of UCSB-7 is not stable to water loss in the range 250–400  C depending on chemical composition. Besides this, more recently, several chiral germanates or borogermanates have also been prepared, which include Ge9O19(OH)2(N2C2H10)2(N2C2H8)0.5H2O (ICMM-6),[78] KBGe2O6 (FJ-9),[79] and K2[Ge(B4O9)]2 H2O.[80] A chiral germanium zeotype of ICMM-6 was

222

Chemistry of Zeolites and Related Porous Materials

Figure 4.21 Helical ribbons are cross-linked in three dimensions to give 12-ring pores. Reproduced with permission from [77]. Copyright (1998) Nature Publishing

synthesized by Medina et al. hydrothermally in a stainless steel autoclave from a reaction mixture containing GeO2, H2O, DABCO, and 1-butanol (molar proportions of reactants 1 : 100 : 2.5 : 10) at 170  C for 5 d.[78] Its structure, built up from connections of GeO4, GeO5, and GeO6 units, possesses three kinds of interconnected helical 8-, 11-, and 11-ring channels. It is stable in both water and organic solvents, and is an active chiral catalyst for both the Michael addition reaction and the acetalization of aldehydes. KBGe2O6 (FJ-9)[79] and K2[Ge(B4O9)]2 H2O[80] were prepared by Yang and coworkers in 2003 and 2004, respectively. FJ-9 is the first chiral zeotype borogermanate with 7-ring channels. Its structure is chiral (space group P212121), containing helices composed of GeO4 tetrahedra, which arrange around 21 screw axes. FJ-9 was synthesized from a reaction mixture of GeO2, K2B4O7, DABCO, H2O, and EG in the molar proportions 1 : 0.5 : 2 : 33 : 16.6 sealed in a Teflon-lined steel autoclave, and heated at 170  C for 11 d. K2[Ge(B4O9)]2 H2O is a noncentrosymmetric potassium templated borogermanate crystallizing in the monoclinic space group Cc. Its structure possesses a unique 3-D open framework composed of alternating linkages of B4O9 clusters and GeO4 units, which contains two pairs of interweaving double-helical channels with 10-rings. In each one, the right- and left-handed helices couple together. K2[Ge(B4O9)]2 H2O was synthesized by the solvothermal reaction of K2B4O5(OH)42 H2O, GeO2, pyridine, and H2O in the presence of diethylenetriamine (DETA) under basic conditions. The reaction mixture was sealed in a Teflon-lined stainless steel autoclave and heated at 170  C for 7 d under autogenous pressure to obtain colorless crystals of K2[Ge(B4O9)]2 H2O. Of particular interest is that the syntheses of ICMM-6, FJ-9, and K2[Ge(B4O9)]2 H2O are all carried out in mixed solvents and in the presence of organic amines. The difference is that DABCO breaks down into ethylenediamine molecules residing in the as-synthesized ICMM-6, while Kþ ions rather than organic amines are occluded in the final products FJ-9 and K2[Ge(B4O9)]2 H2O. (3) The synthesis of chiral open-framework uranyl molybdates More recently, several chiral open-framework uranyl molybdates have been successfully synthesized by Krivovichev et al., and this greatly enriches the family of

Synthetic Chemistry of Microporous Compounds (II)

223

chiral microporous compounds. These compounds include (NH4)4[(UO2)5(MoO4)7] (H2O) 5,[81] [(C2H5 )2NH2 ]2[(UO2)4(MoO4)5(H2O)](H2O),[82] (UO2 )0.82[C8H20N]0.36 [(UO2)6(MoO4)7(H2O)2](H2O)n,[83] [C6H14N2][(UO2)6(MoO4)7(H2O)2](H2O)m,[83] and [C6H16N]2[(UO2)6(MoO4)7(H2O)2](H2O)2.[84] One common structural feature of these compouds is that they all crystallize in chiral space groups, such as P21, C2221, P212121, and P6522, and their structures all consist of UO7 pentagonal bipyramids that share equatorial corners with MoO4 tetrahedra, giving rise to multi-dimensional channel structures. In these compounds, the U O Mo angles vary in a large range of 120 to 180 , which may be a factor favoring the formation of chiral open-framework structures. The synthesis of (NH4)4[(UO2)5(MoO4)7](H2O)5 was carried out in a 23 mL Teflon-lined steel autoclave from a reaction mixture of (NH4)6Mo7O24, (UO2)(CH3CO2)22H2O, and ultrapure water at 180  C for 4 d. The others are synthesized in the presence of organic amines (R), such as octylamine, 1,4-diazabicyclo[2.2.2]octane (DABCO), triethylamine, and diethylamine, and were obtained from a reaction mixture of UO2(CH3COO)22H2O, MoO3, organic amine, HCl, and H2O in a Teflon-lined Parr bomb heated at 180–220  C for 24–65 h. (4) The synthesis of chiral phosphates Compared with chiral silicates and germanates, relatively more metal phosphates with chiral channels or chiral structural features have been synthesized, such as [(CH3)NH2]K4[V10O10(H2O)2(OH)4(PO4)7]4H2O,[85] chiral cobalt phosphates with zeolite ABW and related frameworks,[86] several metal phosphates or metal borophosphates with zeolite CZP structures,[87] [CN3H6][Sn4P3O12],[88] [NH3(CH2)2NH2(CH2)2NH3] [Zn(PO4)(HPO4)],[13] [Zn0.5(H2PO4)0.5]H2O(fJ-13)[89] and other aluminophosphates, gallium phosphates, and indium phosphates. The presence of helical channels is one of their structural features. However, except for NaZnPO4H2O and related compounds with CZP zeotype[87] and MCoPO4 (M ¼ Na, K, NH4, etc.) with ABW zeotype,[86] other chiral phosphates all have low thermal stability, making it difficult to prepare zeolites with chiral channel structures. Hence, we use these two compounds as examples to discuss their syntheses. NaZnPO4H2O is the first zeolite structure with CZP topology, prepared by Rajic et al. in 1995.[87] After that, several metal phosphates and metal borophosphates with CZP structures have been prepared by using inorganic cations Naþ , Kþ , NH4 þ and organic amines DETA and TETA as an SDA. The structure of NaZnPO4H2O built up from ZnO4 and PO4 tetrahedra possesses 12-ring channels composed of double helical ribbons of edge-sharing 4-rings. It is noted that two enantiomeric phases of NaZnPO4H2O can be prepared by two different methods. NaZnPO4H2O of space group P6122 was prepared hydrothermally from a reaction mixture of water, ZnO, H3PO4, and NaOH heated at 70  C for 2 d. The formed crystals were recovered from the mother liquor by vacuum filtration, and washed with acetone. NaZnPO4H2O of space group P6522 was prepared from a reaction mixture of NaOH, Mg(OH)2, H3PO4, and water heated at 100  C for 24 h, then the product was isolated by filtration, and washed with water. In 1997, five chiral cobalt phosphates were prepared by Feng et al.[86] Two of them (NH4CoPO4-ABW, RbCoPO4) has zeolite ABW topology containing 8-ring channels, and the others (NaCoPO4, KCoPO4, NH4CoPO4-HEX) exhibit a hybrid hexagonal structure of tridymite and an ABW framework which contains 6-ring channels. All

224

Chemistry of Zeolites and Related Porous Materials

these structures crystallize in chiral space group, such as P61, P63, and P21. The transformation of these two kinds of structures is affected by the size of extra-framework cations, such as Naþ , Kþ , NH4þ , or Rbþ. NaCoPO4, KCoPO4, and RbCoPO4 were prepared in the CoCO3xH2O–H3PO4–H2O–NaOH/KOH/RbOH systems heated at 150 or 180  C for 4 or 8 d. NH4CoPO4-ABW and NH4CoPO4-HEX were prepared in the CoCO3xH2O–H3PO4–ethylene glycol–NH4OH system with a pH value of 7.5 (adjusted by 29.9% NH4OH) and 9.0 (adjusted by 1,4-diaminobutane), respectively, heated at 180  C for 12 d. Two aspects are of particular importance for the synthetic chemistry of these microporous phosphates. One is the exploration of the synthesis of microporous phosphates with chiral channels, and the study of the synthetic rules in detail, particularly our understanding of the relationships between the SDAs and the formation of chiral channels. The other is the investigation of the preparative chemistry of chiral channel zeolites. However, up to now, no valuable synthetic rules have been found. Therefore, in this synthetic field, the main goal is to gather more experimental results and to intensify the search for new chiral structures. On the other hand, in recent years some synthetic researchers have tried to use chiral organic SDAs in the synthesis, and several layered aluminophosphate compounds have been successfully synthesized by the use of chiral metal complexes as SDAs. Interestingly, experimental results indicate that the chirality can be transferred from the chiral metal complexes to the inorganic layers. Thus, the use of chiral complexes as SDAs provides a feasible approach to the rational synthesis of inorganic frameworks with chiral structural features. Two zinc phosphates with novel frameworks have been prepared by using a racemic mixture of the chiral metal complex Co(en)3Cl3 as SDA under hydrothermal conditions, and the transformation of chirality from chiral metal complex to inorganic framework has been studied.[90] This work provides a good example of how recognition phenomena between a chiral metal complex guest and an inorganic host framework can lead to the crystallization of structures that retain the chiral character of the guest. This issue will be further discussed in Chapter 7. In conclusion, the synthesis of chiral microporous compounds or microporous compounds with chiral channels or chiral features is a great challenge. Although there is great variation in the probability of formation of chiral structures or helices, or some potential synthetic rules have not been known until now, some preliminary rules can be concluded from the above discussions, and there may be helpful in the exploration of more chiral microporous compounds. (1) The synthesis of chiral silicates usually involves the use of alkali metal or alkali-earth metal cations, such as Ca2þ and Kþ , and the reactions are carried out at higher temperature. Moreover, the structures of chiral silicates all consist of TO4 tetrahedra. (2) Compared with chiral silicates, Ge-atoms in the structures of chiral germanates generally adopt more complex coordination modes, such GeO4, GeO5, and GeO6. Organic amines and inorganic cations are all used in the synthesis of these compounds. (3) Most chiral phosphates are prepared in the presence of organic amines or metal complexes with or without chiral features. By using chiral metal complexes as the SDAs, low-dimensional structures or interrupted open-framework structures tend to

Synthetic Chemistry of Microporous Compounds (II)

225

form, and the reactions are carried out at a lower reaction temperature of 110–150  C in order to avoid racemization of the chiral SDAs. (4) In the structures of many chiral microporous compounds, helical chains or ribbons following certain screw axis can be found. Notably, most helices arranging around the 21 screw axis is a single helical chain of -O T O-, and helices arranged around the 61 screw axis are usual helical ribbons comprising 4-rings or more complex structures. (5) The framework elements with considerable flexibility of the O T O and T O T angles favor the formation of helices. Coordination and Condensation of Chiral Building Blocks – the Formation of Openframework with Optical Activity One feasible method for the exploration of chiral open-framework compounds is the use of chiral chemical units as primary building blocks by coordinating with metal or other assembly methods to form 2-D layer or 3-D open-framework structures with optical activity. A notable example is the enantiomerically pure zinc phosphonate based on a mixed phosphonic acid–phosphine oxide chiral building block reported by Bujoli and coworkers in 2001.[91] The reaction procedures are shown as follows. (1) The synthesis of chiral ligands

(2) Optical resolution of chiral ligands and their condensation with Zn2þ The optical resolution of H2O3PCH2P(O)(R)(C6H5) racemates (1: R ¼ CH3; 2: R ¼ C2H5) was performed by using quinine or quinidine as resolution agent to give (R)-2 in 84% yield and (S)-2 in 76% yield (based on one of the enantiomers), respectively.[91] Then the condensation reaction was carried out on a mixture of Zn(Ac)2 (0.3 mmol), (R)-2 [or (S)-2](0.2 mmol) and 20 mL of NaOH (1 M) in PTFElined autoclaves at 110  C for 3 d. The product is enantiomerically pure white crystals of a-Zn-(R)-2 in 87% yield [or a-Zn-(S)-2 in 78% yield] with a composition of (R)- or (S)-aZn[O3PCH2P(O)(C2H5)(C6H5)]H2O. Their structures are shown in Figure 4.22. It is noted that the control of the condensation conditions is very important. In the slightly acid medium, the functional groups (CO2H, NH2) on the phosphonic acid do not coordinate with Zn2þ, but instead crystallize to give a 2-D framework (R)-bZn[O3PCH2 P(O)CH3(C6H5)]H2O, and its structure is shown in Figure 4.23. This is the first reported enantiomerically pure zinc phosphonate with optical activity formed by the coordination of chiral building units, laying the foundation for further development of this method.

226

Chemistry of Zeolites and Related Porous Materials

 CH3 and C2H5). Figure 4.22 Layer structure of ðRÞ-a-Zn[O3PCH2P(O)R(C6H5)].H2O(R  Reproduced with permission from [91]. Copyright (2001) Royal Society of Chemistry

4.2 Synthetic Chemistry of Microporous Compounds with Special Morphologies 4.2.1

Single Crystals and Perfect Crystals

Large single crystals are indispensable for structural analysis, studies of crystal growth mechanism, adsorption and diffusion, the determination of optical and electrical

Figure 4.23 ðRÞ-b-Zn[O3PCH2P(O)(CH3)(C6H5)]H2O. Reproduced with permission from [91]. Copyright (2001) Royal Society of Chemistry

Synthetic Chemistry of Microporous Compounds (II)

227

properties, and applications in advanced functional materials. In general, ultra-fine powders and crystallite congeries are formed in the hydrothermal systems because most microporous crystals are metastable phases. Because the crystallization mechanism is rather complicated and unclear, and many factors affect the crystallization kinetics, there has been no routine method to obtain large single crystals or perfect crystals up to now. However, a number of manageable factors and conditions can be controlled to promote the formation of large-size crystals or perfect crystals. Pang, Qiu, et al. at Jilin University have carried out systematic studies in this respect, and have gained some important results.[92,93] Herein, their work together with others’ will be described in detail. It is necessary to strictly control the various factors affecting the crystallization process in the synthesis of large single crystals. In general, the crystallization process in the hydrothermal or solvothermal system includes the following steps: (1) achievement of supersaturation of reactive species; (2) nucleation; (3) crystal growth. First, the nucleation and the crystal growth (including the growth rate and the ultimate crystal size) strongly depend on supersaturation of the solution. However, the supersaturation is not an independent variable in most syntheses. Supersaturation of specifically reactive species in the solution is controlled by the composition, structure, and solubility of amorphism gel precursor, and other reaction conditions. Second, nucleation is of crucial importance in the whole crystallization process. No matter whether we consider homogeneous nucleation or heterogeneous nucleation, a few nuclei will provide enough reactive species in the reaction system for crystal growth until the crystals reach maximum dimensions. Synthetic Route under the Influence of Nucleation Suppressors In 1993, Morris[94] reported that triethanolamine (TEA) tended to chelate Al, which could suppress the nucleation rate and favor the formation of large and uniform crystals of LTA and FAU (X) in the Na2O–SiO2–Al2O3–TEA–H2O system. This system has also been investigated by Qiu et al.[92] and their results are listed in Table 4.9. Scanning electron micrographs of as-synthesized LTA and FAU (X) are shown in Figure 4.24. These results indicate that uniform and pure large single crystals of LTA and FAU can be produced after the addition of TEA to the reaction system, and the crystal sizes are normally larger than 50 mm. This result cannot be obtained in the Table 4.9 Formation of large single crystals of LTA and FAU in the presence of TEA Gel molar composition No. 1 2 3 4 5 6 7 a

SiO2 0.84a 1.12a 1.12a 1.12b 1.12b 1.87a 1.87b

Al2O3 Na2O 1.0 1.0 1.0 1.0 1.0 1.0 1.0

1.94 2.84 2.5 2.5 2.5 4.17 4.17

SiO2 powder. Si(OEt)4, crystallization temperature: 85  C.

b

Crystallization

Crystal size

TEA

H2O Time (days) Product

(mm)

10 3.17 5.8 5.8 10 10 10

194 276 276 276 276 460 460

40 50 60 80 50 þ 50 80 150

8 6 10 10 12 21 30

LTA LTA LTA LTA LTA þ FAU (X-type) FAU (X-type) FAU (X-type)

228

Chemistry of Zeolites and Related Porous Materials

Figure 4.24 Large single crystals of LTA (left) and FAU (X) (right). Reproduced from [92]. Copyright (1998) Elsevier

absence of TEA. Further investigation of reaction conditions shows that using less reactive Si(OEt)4 as the silicon source instead of SiO2 powder will decrease the rates of condensation and nucleation, and favor the formation of larger-size single crystals of FAU (X) (
150 mm). Other nucleation-suppressor-containing systems studied by Qiu and coworkers involve the use of pyrocatechol as an additive in the systems 7.0 SiO2 : 2.5 NaOH : 65 EG : (0.5
3.0) R at 180  C for 9
20 d[95] and SiO2 : 0.2 TPABr : 0.5 NaOH : 30 H2O : (0.2
1.0) R at 180  C for 7 d,[96] in which uniform and larger-size crystals of pure silica sodalite (60
70 mm) and various size-controlled crystals of pure silica silicalite-I ð26  24  5 mm to 165 30  30 mmÞ, respectively, are prepared. The reason is that pyrocatechol tends to chelate Si, which reduces the polymerization reactions among silicate species, and further decreases the rates of nucleation (in the induction period) and crystallization to play the role of nucleation suppressor. This result is clearly observed in the crystallization curve shown in Figure 4.25. Synthetic Route under Organic Solvothermal Conditions The successful synthesis of a zeolite, typically sodalite, in ethylene glycol starts from the initial work of Bibby and Dale in 1985.[97] Since 1986, Xu and coworkers at Jilin University have expanded this synthetic route to the fields of microporous aluminophosphates and transition metal phosphates with various structures, and found another advantage of the organic solvothermal synthetic route; that is, a great quantity of large single crystals and perfect crystals could be prepared in this system. An example is the research group’s work in which around one hundred large single crystals with 2-D layered and 3-D open-framework structures have been prepared. Together with other researchers’ work, organic solvothermal synthesis has become one of the most important synthetic routes for the preparation of large single crystals and perfect crystals. Notable examples (including zeolites) are the single-crystal syntheses of sodalite (EG),[92] Si-ZSM-39 (BuOH),[98] Si-ZSM-48, and silicalite-1[99] under alcohol thermal conditions;

Synthetic Chemistry of Microporous Compounds (II)

229

Figure 4.25 Crystallization curves for Si-SOD zeolites from batch compositions with molar proportions 7.0 SiO2 : 2.5 NaOH : y R : 65 EG at 180  C. Curves a) y ¼ 0; b) y ¼ 1.0; c) y ¼ 2:0; d) y ¼ 3:5. Reproduced from [95]. Copyright (1999) Elsevier

and the synthesis of giant crystals of ZSM-35 by Kuperman in 1993 using the pyridine and alkylamine system.[100] Notably, perfect crystals of cancrinite (CAN) were prepared by Xu et al in the same year by using 1,3-butanediol as the solvent. Different from assynthesized CAN prepared in the aqueous system, there were no barriers and defaults in the 12-ring channels in respect of adsorption abilities for many kinds of absorbents.[101] The advantage of this synthetic route has been powerfully displayed in the syntheses of single crystals of aluminophosphates in alcohol thermal systems. During the last decade, numerous new aluminophosphate compounds with 2-D layer and 3-D open-framework structures have been explored based on the syntheses and structural analyses of large-size crystals. Some examples include the first aluminophosphate molecular sieve AlPO-CJB1 [Al12P13O52][(CH2)6N4H3][102] with Bro¨nsted acidity, the first open-framework aluminophosphate AlPO-CJ4 [AlP2O6(OH)2][H3O][103] (Al/P ¼ 1/2) containing propeller-like chiral motifs, the anionic framework aluminophosphate AlPO-CJB2 [Al11P12O48] [(CH2)6N4H3H2O][104] built up from the strict alternation of AlO4/AlO6 and PO4 units, aluminophosphate APO-HDA with intersecting 12-ring and 18-ring channels, and numerous other 3-D anionic framework and 2-D layer structures. The Al/P ratios of these compounds include 1/1, 1/2, 2/3, 3/4, 4/5, 5/6, 11/12, 12/13, 13/18, etc. This greatly enriches the open-framework aluminophosphate family, showing their structural and compositional diversity.[64] Recently, the organic solvothermal synthetic technique has been expanded to the synthesis of single crystals of other microporous phosphates, such as gallium, indium, and a large variety of transition metal phosphates. Many research results have been reported, in particular its contribution to the exploration of new compounds and new phases. In the following, several typical examples will be presented. Yu et al. at Jilin University prepared Co-GaPO4-LTA[105] with a crystal size of
100 mm in the ethylene glycol system, and studied the incorporation of CoII in the framework of GaPO4-LTA based on single-crystal structure analysis. Chippindale et al. synthesized a 3-D anionic

230

Chemistry of Zeolites and Related Porous Materials

framework gallophosphate jMe2 NHðCH2 Þ2 NHMe2 j½Ga4 P5 O20 H with 16-ring channels in the butanol solvothermal system, and the crystal size was up to 0:85  0:45  0:2 mm3 .[106] In 2001, Lii and coworkers[49] prepared single crystals of [Ga2(DETA)(PO4)2]2H2O with extra-large 24-ring pores under mixed ethylene glycol and water conditions. In 1997, Pang and Du at Jilin University systematically studied the synthesis and formation of InPO4-n single crystals with 2-D and 3-D frameworks in the alcohol and amine thermal systems,[107] such as InPO4-C11 and -C12. In recent years, the syntheses of single crystals of microporous transition metal phosphates by using this method have been continually reported. Notable examples include jH3 NCH2 CH2 NH3 j0:5 ½FePO4  containing a fourcoordinated Fe center prepared in ethylene glycol system,[108] the first zeolite structure CoPO-GIS, jH3 NCH2 CH2 NH3 j0:5 ½CoPO4  synthesized in ethylene glycol system,[109] single crystals of a novel titanium phosphate synthesized under butan-1-ol solvothermal conditions,[110] single crystals ð0:5  0:05  0:05 mmÞ of microporous zinc phosphate jH3 NðCH2 Þ6 NH3 j ½Zn4 ðPO4 Þ2 ðHPO4 Þ2 3H2 O½16 with extra-large 20-ring channels synthesized by Rodgers and Harrison in the ethylene glycol system in 2000, etc. The alcohol or amine solvothermal synthetic route favors the formation of large single crystals and perfect crystals. One reason is that the organic solvent has lower polarity (smaller permittivity) and larger viscidity compared with water. The organic solvents, such as alcohol or amine, usually interact with the reaction materials and organic templates, i.e., H-bonding or coordination interactions. Moreover, their abilities to accelerate the decomposition and rearrangement of T O T bonds are lower than those of water. All these properties of organic solvents do not favor the dissolution and diffusion of reactants, aq. their aggregation and crystallization. The overall result is a decrease of nucleation rate, and a decrease of crystal growth rate to a certain degree, which results in few nuclei being formed in the reaction system, as well as a slow crystallization process, favoring the formation of large single crystals and perfect crystals. The above discussion indicates that there are more factors affecting the nucleation and growth of crystals in the organic solvothermal systems compared with water. At the same time, it is noted that a small amount of water is brought into the organic solvothermal systems from the solvents and reactants, such as H3PO4 (85%), which may affect the crystallization. Therefore, the conclusion of some rules will provide useful directions for the controlled synthesis of crystal products with various morphologies, crystal sizes, and crystal distributions. Synthetic Route in the Presence of F Ions As one of the synthetic routes for zeolites and microporous crystals, the F -containing hydrothermal approach will be described in Section 3.2.4. Based on this, we will further introduce the advantages and features of this synthetic route for the preparation of large single crystals and perfect crystals in this section. The hydrothermal system involving certain mounts of F ions as the mineralizing agent allows the crystallization of zeolites under nearly neutral conditions, and avoids the disadvantage of strongly alkaline systems in which crystals with defaults are always formed. The reason is that the positive charges of organic templates are usually balanced by the negative charges coming from the defaults in the framework, while in the F system the positive charges of organic templates are balanced by the F ions to reduce the possibility of formation of defaults. In addition, F has a tendency to form complexes with silicon, aluminum, phosphorus,

Synthetic Chemistry of Microporous Compounds (II)

231

boron, etc, in the reaction materials. These complexes slowly hydrolyse to release less fluorinated silicon, aluminum, phosphorus, or boron species, which gradually supply the nutrients for the growth of large single crystals. Owing to this feature, crystallization in the F -containing reaction system has become an important approach for the preparation of large single crystals and perfect crystals at the present time. This synthetic approach was first proposed by the Mulhouse group in France, and developed by Guth and Kessler at the Universite´ de Haute Alsace. Qiu and Pang at Jilin University have done a lot of work on the growth of large single crystals, in particular their outstanding work on the syntheses of large single crystals of zeolites and heteroatom-substituted zeolites. Their representative work is the first syntheses of B-ZSM-5 single crystals ð120  30  30 mmÞ (Figure 4.26) and Ti-ZSM-5 single crystals ð150  45  45 mmÞ in the 1 SiO2 : 0.2 B2O3 : 0.25 TPABr : 0.5 NH4F :70 H2O system (pH ¼ 6.8, T ¼ 180  C for 9 d) and 1 SiO2 : 0.2 TiO2 : 0.25 TPABr :1 NH4F : 70 H2O system (pH ¼ 6.5, T ¼ 170  C for 10 d), respectively, reported in 1989[111]. The crystallization curves in the strongly alkaline system and F system were studied for comparison. After that, single crystals of ZSM-5 containing Ni2þ, Fe2þ, Fe3þ, and Ga3þ were synthesized from similar reaction systems.[112] In 1992, Pang and coworkers employed this synthetic route in the synthesis of perfect single crystals of ZSM-5 (Figure 4.27) by using seven different organic amines as the SDAs.[113] Perfect single crystals of MAPO-5 were also prepared in the MOX–Al2O3–P2O5–R–NH4F–H2O system in the presence of various organic amines as the SDAs (see Table 4.10[112] and Figure 4.28[24]). Large single crystals of numerous heteroatom-containing zeolites have been synthesized by applying this route. This is because many heteroatoms tend to form

Figure 4.26 Scanning electron micrograph of single crystals of B-ZSM-5. Reproduced from [111]. Copyright (1989) Elsevier

232

Chemistry of Zeolites and Related Porous Materials

Figure 4.27 Scanning electron micrograph of single crystals of ZSM-5. Reproduced from [113]. Copyright (1992) Elsevier

fluorinated complexes, which slowly release the reactive species through hydrolysis to gradually supply the nutrients for the growth of large single crystals. This is a unique feature of this synthetic route. In the past decade, Fe´rey’s research group has synthesized a series of microporous gallophosphates in F -containing hydrothermal systems. Several gallophosphates with special structural features have been explored, such as j2N2C6H18j[Rb2Ga9(PO4)8(HPO4)(OH)F6]7 H2O (MIL-50)[114]with 18-ring channels, and MIL-31, MIL-46, and ULM-5, and ULM-16[47] with 16-ring channels. Interestingly, some scientists have introduced F ions into the organic solvothermal system and obtained surprising results Table 4.10 The formation of MAlPO4-5 in the presence of TPAOH, Pr3N, tropine, and Et3N MAlPO4-5

SDA

Crystal size ( mm)

AlPO4-5 AlPO4-5 SAPO4-5 LiAlPO4-5 BAlPO4-5 ZnAlPO4-5 MnAlPO4-5 CoAlPO4-5 TiAlPO4-5

TPAOH Pr3N Tropine Tropine Tropine Et3N Et3N Pr3N Pr3N

740  120 310  85 110  50 320  60 260  50 130  25 120  30 120  30 350  50

Synthetic Chemistry of Microporous Compounds (II)

233

Figure 4.28 SEMs of single crystals of SAPO-5 (a), BAPO-5 (b), and LiAPO-5 (c). Reproduced with permission from [24]. Copyright (1991) Elsevier

sometimes. In 1993, Kuperman reported in Nature[100] giant crystals of ZSM-35, with sizes up to the centimeter scale, obtained in the pyridine and alkylamine system containing F ions. ‘Two Silica Sources’ Synthetic Route in the Hydrothermal Crystallizations In the first report of zeolite synthesis under hydrothermal conditions by Pang and coworkers in 1995,[115] the crystallization time would be prolonged and the perfect single-crystal product would be obtained if two silica sources, sodium silicate solution and SiO2 powder (Aerosil), were used. Their first successful example was the synthesis of MOR in the Al2O3–SiO2–Na2O–NaCl–H2O system at 150  C by the use of two different silicon sources (sodium silicate and Aerosil). The results are listed in Table 4.11 and shown in Figure 4.29. By applying the ‘two silica sources’ synthetic approach, perfect crystals of MFI and BEA were produced in the presence of TEA and butan-1-amine as SDA, respectively.[116] Table 4.11 Crystallization conditions and corresponding products in the Al2O3–SiO2–Na2O– NaCl–H2O system Mol gel composition No. Al2O3

SiO2

Na2O

A

1

a

60 þ 15

B

1

C

b

Mc

550

150

15

60a þ 15b 15 4KCl

550

150

15

1

75a þ 0b

15 4NaCl

550

150

5

D

1

0a þ 75b

15 4NaCl

550

150

5

E

1

60a þ 50b 15 4NaCl

550

145

25

a

Aerosil. Sodium silicate.

b

15 4NaCl

Time H2O Temp. ( C) (days)

Reactant Crystal status size (mm) clear 185  125 solution clear 85  50 solution clear 83 solution clear 21 solution gel 110  55

Product MOR MOR MOR MOR MOR

234

Chemistry of Zeolites and Related Porous Materials

Figure 4.29 Scanning electron micrographs of MOR single crystals prepared using two silica sources. Reproduced with permission from [92]. Copyright (1998) Elsevier

The good effect of this route may be due to the different reactivities coming from different silicon sources, which reduce the rates of nucleation and crystallization. Hydrothermal Synthetic Route in the Clear Homogeneous System At the end of the 1970s and in the early days of the 1980s, Ueda and Koizumi systematically studied the crystallization of ANA, SOD, zeolite B, MOR, and FAU (Y), which were synthesized directly from clear homogeneous systems of Na2O–Al2O3–SiO2–H2O.[117] In 1986, Pang et al. investigated the crystallization of zeolite LTA in the clear homogeneous system 10 Na2O : (0.1
0.5) Al2O3 : (1.0
9.0) SiO2 : (100
300) H2O at 60
100  C.[118] It was found that the as-synthesized product crystallized from clear homogeneous solution usually had perfect crystal morphology. In 1989, Qiu and coworkers applied this method to the syntheses of AlPO4-5 and many heteroatom-containing MAPO-5[119] in the 1.0 P2O5 : 0.77 Al2O3 : (0.03
0.10) M2O3(M0 O) : (1.8
2.2) Et3N : (0.1
1.0) HF : (40
100) H2O (M ¼ B, Fe; M0 ¼ Co, Ni) system to produce perfect crystal products. The clear homogeneous system can be prepared by using different methods. It is formed under very excessive NaOH conditions for aluminosilicate zeolites, while for AlPO4-5 and heteroatomsubstituted MAPO-5 it is formed under conditions of a slight excess of SDA and the addition of HF. One important reason for the favorable formation of perfect crystals in the clear homogeneous solution is that the large viscosity of such systems does not favor reactant diffusion or homogeneous nucleation. On the other hand, sol decomposition may be more difficult than fast polymerization to form the amorphous gel, which is not suitable for the formation of a nucleus with regular structures. The difficulties of nucleation and the slow crystallization rate result in the formation of perfect crystals. In addition to the five synthetic routes mentioned above, it is worth mentioning that prefect crystals usually tend to form in the synthesis of microporous M(III)X(V)O4-type crystals in slightly acidic solution. This has been described in the previous section. Synthesis of Giant Zeolite Crystals by a BMD (Bulk Material Dissolution) Technique In 2001, Shimizu et al.[120] developed a ‘BMD’ (bulk materials dissolution) technique for the synthesis of giant zeolite crystals based on others’ work. A piece of bulk material,

Synthetic Chemistry of Microporous Compounds (II)

235

Figure 4.30 Schematic illustration of the experimental setup for BMD crystallization. Reproduced with permission from [120]. Copyright (2001) Elsevier

such as glass tube, ceramic tube, or ceramic boat, was used as the Si (or Si and Al) source, which was placed in a PTFE sleeve inside an autoclave. Liquid materials containing SDA, HF, and NaOH were used to fill the sleeve, and the crystallization was performed under hydrothermal conditions. The experimental setup is illustrated in Figure 4.30. In the reaction process, the material sources were slowly dissolved to provide reactive species continually and adequately. The giant zeolite crystals were obtained after a long period of crystallization (10
45 days) by varying the reaction composition. The BMD crystallization conditions and results are listed in Table 4.12. This is an interesting technique for the synthesis of giant zeolite crystals. However, one problem still unsolved up to now is that of crystal breakage in the calcination to remove the SDAs. This problem must be solved before its application. 4.2.2

Nanocrystals and Ultrafine Particles

The crystal sizes of molecular sieves used as catalytic and adsorption materials are usually in the range 0.5
10 mm. Larger crystals favor the shape-selectivity of molecular sieves due to their uniform channels and large inner surfaces, whereas the employment of ultra-fine powders or even nano-sized crystals of molecular sieves as catalytic materials will favor the heat and mass transfer, and the reactive centers are exposed due to the increase of external surfaces, enhancing catalytic efficiency. On the other hand, the deactivation phenomena caused by the channels filled with deposited species will decrease because more pore windows are exposed outside. Moreover, molecular sieve nanocrystals exhibit potential applications in sensing membranes, low-k films, and new type of electronic materials. The study of Camblor et al.[121] shows that the physicochemical properties and structure of a crystalline material vary as the crystal size decreases, particularly after it decreases to the nanocrystalline range. Thus, the synthesis of nanocrystals and ultra-fine particles is an important aspect in the discussion of the synthetic chemistry of molecular sieves and microporous materials with special aggregation morphologies.

236

Chemistry of Zeolites and Related Porous Materials

Table 4.12 Synthesis of zeolite crystals using BMD technique Zeolite Bulk source framework (mmol) MFI

MFI

MFI

ANA JBW and CAN CAN

SOD

SOD

SiO2 tube SiO2 (25.2)

Crystallization Liquid phase Maximun (mm) crystal size (mmol) Temperature ( C) Time (t/d)

TPAOH (8.9) HF (9.7) H2O (870) SiO2 tube TPAOH (8.9) SiO2 (18.4) HF (14.6) H2O (885) SiO2 tube TPAOH SiO2 (680) (88.4) HF (137.9) H2O (3616) Ceramic boat NaOH (20.3) SiO2 (12.1) H2O (847) Al2O3 (3.0) Ceramic boat NaOH (51.2) SiO2 (12.6) H2O (683) Al2O3 (9.6) Ceramic boat NaOH (99.0) SiO2 (11.4) H2O (832) Al2O3 (2.8) Ceramic boat NaOH (51.0) SiO2 (11.0) H2O (833) Al2O3 (2.7) Ceramic boat NaOH (98.7) SiO2 (21.7) H2O (697) Al2O3 (11.7)

200

25

2000  1000  1000

200

46

3200  2800  2600

200

34

3900  2600  2000

200

31

3000  2800  2500

200

7

JBW 6740  320  1000 CAN 300  5  5

200

13

100  20  20

100

19

60  60  60

200

12

120  120  120

In general, molecular sieve nanocrystals and ultra-fine powders(<200 nm)are prepared by the following methods: (1) Controlled crystallization of sol; (2) Hydrothermal crystallization under controlled conditions; (3) Controlled crystallization in the micro-reactor spaces (such as the mesoporous spaces of carbon black and micro-emulsion). Controlled Crystallization of Sol Colloidal sols containing homogeneous particles of nano-scale size can be formed under special conditions after the reaction materials are mixed. The zeolite and some microporous aluminophosphate nanocrystals (size in the range 50
100 nm) tend to be prepared through the controlled crystallization at low temperature in the clear sol reaction system. For example, Long and coworkers[122] synthesized colloidal TBA-silicalite-2 crystals with sizes of 60
90 nm from the homogeneous sol formed in the reaction mixture of 0.35 TBAOH : 1.0 TEOS : 12 H2O at 114  C for 7 d. Van Griekan et al.[123]

Synthetic Chemistry of Microporous Compounds (II)

237

synthesized ZSM-5 nanocrystals with sizes of 10
100 nm through hydrothermal crystallization of clear supersaturated homogeneous mixture 1 Al2O3 : 60 SiO2 : 21.4 TPAOH : 650 H2O at 170  C, and studied the crystallization process of nanocrystals. Many characterization results indicate that the formation mechanism of ZSM-5 nanocrystals involves three steps. First, amorphous particles with sizes of 8
10nm are gradually formed within 18 h after the reactants are mixed. Second, amorphous particles aggregate into larger units (secondary particles), which are gradually ordered, and nuclei appear. Finally, the zeolitization of these secondary particles yields ZSM-5 nanocrystals. By utilizing controlled crystallization in the clear and homogeneous sol system, many kinds of molecular sieve nanocrystals have been successfully prepared, including TPAsilicalite-1,[124] TBA-silicalite-2,[122] ZSM-5,[123] ZSM-12, SOD, LTA, FAU, BEA, LTL, TS-1, and AlPO4-5.[122] The synthesis of molecular sieve nanocrystals from the initial reaction sol can only be achieved by strictly controlling the crystallization conditions. Many crystallization conditions, such as SiO2/Al2O3, OH /SiO2 and H2O/SiO2 ratios in the reaction mixture, the types of reaction materials and crystallization temperature, all influence the sizes and distributions of nanocrystals. Therefore, further study of the effect of numerous crystallization factors on the crystal size not only favors the controlled synthesis of molecular sieve nanocrystals, but also provides the direction for the synthesis of molecular sieve ultra-particles with larger sizes. Here, we use the syntheses of ZSM-5[123] together with other molecular sieves as examples to briefly discuss this point. Several factors affect the crystal sizes and yields of ZSM-5 molecular sieve (sizes ranging from tens to hundreds of nm) synthesized from the batch with composition 1 Al2O3 : 60 SiO2 : 21.4 TPAOH : 650 H2O at 80  C, such as the aging time before crystallization (16
66 h) and the type of aluminum source [Al2(SO4)318 H2O, Al(NO3)39 H2O, Al(iPrO)3]. The crystal sizes become slightly large (increasing by 20%) when the aging time is short and inorganic aluminum salts are used as the reaction source. The alkalinity of the reaction system has an intense influence on the average crystal size ðDC Þ. If the amount of TPAOH increases from 21.4 mmol to 41.5 mmol, DC increases from 70 to 170 nm. However, the influence of alkalinity on the crystal sizes exhibits different rules in some reaction systems, such as the syntheses of FAU and TPA-silicalite-1 nanocrystals. It is noted that the amount of water in the initial reactants also affects the crystallized products. In the above crystallization system, when the amount of water increases from 650 ! 1300 ! 3000 mmol, and the pH value is kept at 14 by using other alkali, DC of ZSM-5 increases from 70 ! 300 ! 1000 nm, and crystallized yield decreases from 75:1 ! 65:6 ! 22:8%. Interestingly, this phenomenon can also be observed in the crystallization system of (Na,TPA)-ZSM-5.[125] Regarding the influence of crystallization time on the as-synthesized product, Persson et al. have particularly investigated it in the crystallization of ZSM-5 from the reaction system xTPABr : yTPAOH : 0.1 Na2O : 25 SiO2 : 480 H2O : 100 EtOH.[125] It is found that crystal size increases in linearity with the increase of crystallization time in the initial period of crystallization, and then remains almost unchanged after a certain time (see Table 4.13). The relationship between the DC and crystallization time is shown in Figure 4.31. The above rules are consistent with the assumptions about the crystallization process of nanocrystals proposed by Van Grieken et al.[123] These rules provide useful guidance for the best conditions in the synthesis of nanozeolites and ultra-fine particles.

238

Chemistry of Zeolites and Related Porous Materials

Table 4.13 The synthesis of ZSM-5 in the xTPABr : y TPAOH : 0.1 Na2O : 25 SiO2 : 480 H2O : 100 EtOH system No

x

y

DC (nm)

Yield (%)

S1 S2 S3 S4 S5

0.0 3.0 0.0 4.5 0.0

9.0 6.0 5.0 4.5 3.0

95 110 155 188 307

64 81 82 86 88

Hydrothermal Crystallization under Controlled Conditions High-quality molecular sieve nanocrystals can be prepared by hydrothermal crystallization in the initial reaction gel system. The key point is the strict control of the reaction conditions. In 2003, Yan and colleagues[126] investigated the influence of [TMAþ ] concentration and anion [Br ] and [OH ] concentrations on the formation and yield of zeolite Y nanocrystals in the 1.00 Al2O3 : 4.35 SiO2 : 2.4 (TMA)2O(2 OH ) : (1
1.2) (TMA)2O(2 Br ) : 0.048 Na2O : 249.0 H2O system at 100  C. When TMABr was added to the reaction system as a second SDA, the nanocrystal size decreased (generally stable under 40 nm), and a yield improvement of zeolite Y could be observed due to the increase of nucleation rate. Furthermore, the effect of the TMABr/TMAOH ratio in the reaction system on the nanocrystal sizes and yield of as-made product was also studied, and, based on this, zeolite Y nanocrystals with controllable particle sizes of 32
120 nm were synthesized. Ultra-fine particles of molecular sieves (0.2
0.5 mm) can also be synthesized by hydrothermal crystallization in a suitable reaction gel system. The key to this synthesis is again the study of the crystallization process and the control of crystallization conditions.

Figure 4.31 Relationship between DC and crystallization time. Reproduced with permission from [125]. Copyright (1995) Elsevier

Synthetic Chemistry of Microporous Compounds (II)

239

In general, the product sizes and distributions are determined by the rates of crystal nucleation and growth that are closely related to the supersaturation of reactive species in the crystallization process. Hence, enhancing the supersaturation of reactive species for the nucleation rate increase is a vital direction for the synthesis of ultra-fine particles of molecular sieves. Another important aspect for the preparation of nanocrystals and ultrafine particles following this route is the isolation of as-made product from the mother liquid. At present, high speed centrifugation (>20 000 rpm) is commonly adopted to separate nanocrystals from the mother liquid. The separated nanocrystals are dispersed in water and washed by ultrasonication to remove the residual mother liquid, and then centrifuged again and finally dried. The drying methods involve low-temperature drying, freeze drying, and drying under supercritical or subcritical conditions, which avoid aggregation and keep the integrity of the molecular sieve nanoparticles. However, generally speaking, molecular sieve nanocrystals and ultra-fine particles prepared by this method usually have low yields due to so many treatment processes. The Controlled Crystallization in the Micro-reactor This is where the synthesis of nano-sized molecular sieves is carried out in the template matrix within confined spaces. This is an ideal synthetic route if the space size and uniformity favor the crystallization, and the as-synthesized product is easily isolated from the templates. Mesoporous molecular sieves with uniform mesopore structures can be adopted as the template, such as MCM-41. In 2000, Schmidt et al.[127] first proposed such a route to synthesize ZSM-5 nanocrystals. The synthesis procedure consisted of the impregnation of mesoporous carbon black with reaction solution, followed by treatment with steam at 150  C, and the combustion of carbon black. Compared with other methods, the advantage of this one is that the nano-sized product is easily isolated and the yield is relatively higher. However, it also has some drawbacks. First, there is a high requirement for the preparation of carbon black as the template matrix, i.e., the mesopore sizes in carbon black must be uniform. Second, the crystallization must be performed in the mesopores, not on the extra surfaces of the carbon black. Third, a large amount of carbon black will be consumed (about four-times that of the nanozeolite product). All of these factors affect the further development of this route to some degree. In 2002, Chiang and coworkers[124] developed a new scheme for the confined synthesis of TPA-silicalite nanocrystals. The surfactant cetyltinmethylammonium bromide (CTAMeBr) (in ethanol solution) was added to the single- and double-heated TPAsilicalite precursor sols (SHPS and DHPS), and the mixture was flocculated at a certain pH value to collect the nano-size silicate species in the precursors, and then dried. The dried precursor/surfactant hybrid was pressed into pellets and then steamed in a stainless steel autoclave at 110
150  C for 7
36 h. Finally, the product was calcined to remove the surfactants and TPA. The particle sizes of silicalite-1 produced in this method are about 30 nm. The study indicates that the nanoparticles collected by surfactants already exhibit the structural features of MFI. They crystallize entirely to form silicalite-1 nanocrystals after steam treatment at 110
150  C. This new solid-phase approach provides a way to synthesize MFI nanocrystals without the problem of separation and collecting nanocrystals from suspension, and it also avoids the large consumption and cost of special mesoporous templates used in the confined-synthesis methods.

240

Chemistry of Zeolites and Related Porous Materials

In 2003, Yan and coworkers[128] successfully explored a new route of using thermoreversible polymer hydrogels (Methyl-Cellulose, MC) to synthesize zeolite NaA and NaX nanocrystals in the system Na2O–Al2O3–SiO2–H2O–MC. This synthetic route exhibits several advantages. First, water-soluble MC is gelated at 50  C to form a hydrogel with three-dimensional structure. The liquids entrapped in the frameworks of hydrogels can potentially serve as micro-reactors or nano-reactors for the crystallization of zeolite NaA (crystallinity: 95%, average size: 98 nm) and NaX (crystallinity: 90%, average size: 70 nm) nanocrystals at 80  C. Second, the as-made product is separated from MC by washing due to the water solubility of MC, which is a vital virtue of this synthetic route. It avoids the aggregation of nanocrystals caused by the high-temperature treatment for the removal of templates or micro-reactors including carbon black, mesoporous materials, surfactants, etc. Furthermore, this route synthesizes zeolite NaA and NaX nanocrystals under template-free conditions, avoiding the over-consumption and cost of templates. These virtues have attracted increasing attention to this route’s development. In conclusion, there is a lot of room for the further development of this nanocrystal synthetic route in confined spaces, but there are also many aspects that need improvement and innovation. Preparation of Nanozeolite Catalytic Materials There are many difficulties involved in the application of pure nanozeolites as catalytic materials, such as the aggregation of nanozeolites, their low thermal and hydrothermal stabilities, and the difficulties in regeneration, filtration, and recycling. Hence, how to develop nanozeolites into new catalytic materials with application values in industry is of particular interest. Here, we use b-type (BEA) nanozeolite catalytic material as an example to introduce this topic. At present, two methods are adopted to deal with the above problems: (1) In situ-crystallization method: a hydrothermal crystallization method is used to make nanozeolites grow directly in the channel and on the extra surface of carriers. However, the large zeolite particles tend to form on the extra surface of carriers in this method, which destroys the continuity of carriers, and prevents the contact of reactant molecules and reactive centers. (2) Slurry coating method: zeolite and carrier slurry are mixed automatically to make the zeolite disperse in the slurry, and the mixture is then calcined. However, some carriers block the zeolite pore openings, decreasing catalyst efficiency. The sol–gel method controls the microstructures of materials in a small range by lowtemperature chemical techniques, enabling the uniformity of structures to be achieved on a sub-micrometer, nanometer, or even molecule scale. The catalysts prepared by the sol– gel method have high purity, good uniformity, controlled pores, and other virtues. Particularly, compounds with large surfaces can be prepared at low temperatures. Wang[129] prepared a b-Al2O3 composite by employing a two-step sol–gel method. Al2O3 sol was first prepared under acid conditions, followed by the addition of zeolite slurry to make the carrier sol form a gel immediately. In this way, zeolites were formed and filled into the matrix channels of carriers, forming the b-Al2O3 composite nanocatalytic materials with relatively high stability.

Synthetic Chemistry of Microporous Compounds (II)

241

The micropore structure of the sol–gel composite could be controlled by varying the pH value of the sol. As the pH value increases, the pore volume of the composite decreases. This is the result of zeolite aggregations caused by the highly alkaline sol. From this example we can conclude that the two-step sol–gel method has the following features: (1) the addition of alkali solution of zeolite synthesis before gel formation enables the highly dispersed zeolite nanoparticles to assemble in the framework of the carrier, which restrains the aggregation of nanozeolites; (2) the sol–gel composites are filtered easily, and zeolites are not easily lost in the washing and drying processes; (3) series composite catalysts with different zeolite contents can be prepared by varying the assembly amount of zeolite slurry (upper limit amout of zeolite is 84%); (4) the channel in the carrier of the sol–gel composite is like a micro-reactor, which reduces the diffusion resistance of the contact between gas reactant molecules and zeolite reactive centers, enhancing catalyst efficiency. Therefore, the two-step sol–gel method can promote the applications of nanozeolites in industry. 4.2.3

The Preparation of Zeolite Membranes and Coatings

Zeolites have become increasingly important due to their applications in catalysis, adsorption and separation, as well as in advanced materials for photoelectronics and sensors. In the past decade, the synthesis and preparation of zeolite membranes and their complex materials have been of particular interest and have made great progress. Generally, zeolite membranes can be classified into two types: zeolitic films on stable supports, and self-supporting zeolitic crystalline membranes. In this section, the former will be mainly described because zeolite membranes in applications usually involve the growth of zeolites on the supports and substrates. The supports and substrates can be roughly classified as porous supports including porous Al2O3, porous ceramic and stainless steel, etc., and glaze substrates including silicon wafer, quartz plate, glass, and LiTO3 single-crystal plate, etc. Up to now, many zeolites have been used as materials to prepare membranes with various applications. For instance, silicalite-1, silicalite-2, ZSM-5, TS-1, LTA, LTL, and X- and Y-type membranes have been applied in gas permeation; ZSM-5, LTA, and silicalite-1 membranes in pervaporation; LTA, FAU, and silicalite-1 membranes in sensor materials; silicalite-1 membranes in microcalorimetry; and TS-1, MOR, GME, ZSM-35, and AlPO4-5 membranes in optics materials. In terms of the demands on membrane capability, the ideal structural features should be thin (generally <1 mm), dense (without secondary channels among particles), uniform, and ordered in arrangement. Thus, the preparation of continuous membranes, the decrease of the thickness and increase of the uniformity and order of membranes, as well as the enhancement of the orientations of crystals in the membrane, are all focal points in the preparation of membranes. Techniques involved in the preparation of zeolitic films on stable supports can be divided into four categories: (1) pre-treatment of the supports, in which the most important is chemical treatment (such as the siloxation of surface-to-anchor zeolite crystallites), (2) synthesis of zeolite film precursors, (3) growth of films on the support surfaces, and (4) elimination of small defects, which can be achieved by chemical vapor deposition (CVD) of silica sources (by reaction with silicon alkoxide or other silylation agents) and other techniques. Processes (2) and (3) are crucial in the preparation of films,

242

Chemistry of Zeolites and Related Porous Materials

which will be discussed in detail. A well known procedure for preparing a continuous and highly oriented film with sub-micrometer thickness involves the formation of crystallite particles of nanometer size on the pre-treated surface of the support, followed by secondary growth of these particles to form continuous films under suitable crystallization conditions. This is generally called a two-step method. However, different processes and conditions will be adopted according to the crystallization features of zeolites and the demand of the functions of zeolite membranes. Example 1: Synthesis of LTA and MFI zeolite membranes for gas separation[130,131] In 2001, Lin and coworkers at the Dalian Institute of Chemical Physics synthesized LTA membranes on a porous a-Al2O3 support from clear solution.[130] A porous a-Al2O3 disk (30 mm diameter, 3 mm thickness, 0.1–0.3 mm pore radius, about 50% porosity) was used as the support. After the surface treatment, the support was impregnated in a 0.5 wt% NaA zeolite suspension for 30 min, and then dried and microwave irradiated to produce NaA zeolite seeds with a uniform size of 1 mm on the surface. The seeded support was placed vertically with a Teflon holder in a stainless steel autoclave, and a clear and homogeneous solution with molar proportions 5 SiO2 : Al2O3 : 50 Na2O : 1000 H2O was added. Highquality LTA membranes were obtained after two crystallizations carried out at 90  C for 2 h each. The permselectivity of LTA membranes for H2/n-C4H10 and O2/N2 was 19.1 and 1.8, respectively. When the crystallization time was increased, the membrane thickness increased, and the uniformity and orientation of the membranes decreased. The formation process of LTA membranes is shown in Figure 4.32. The preparation of MFI membranes for high-quality gas separation by Noack et al. in 2001 also involved two steps.[131] The first step was the seeding of colloidal crystals on the surface of the support, and the second step was the synthesis of the membranes under hydrothermal conditions. However, the seeding method of microcrystals on the support is different from that in the above example. An a-Al2O3 disk was also used as the support; however, after a cleaning treatment of the surface, the support was treated with diallyldimethylammonium chloride to provide a positive charge to the surface of the alumina support. Then the support was put into the clear, homogeneous TPA-silicalite-1 nanocrystal solution, and the seed crystals carrying negative charges in aqueous solution were attracted to the surface of the support by electrostatic interaction under controlled conditions. The seeded support was washed with dilute aq. NH3 and then placed vertically with a Teflon holder in a stainless steel autoclave to crystallize in a reaction

Figure 4.32 The formation process of LTA membranes. Reproduced with permission from [130]. Copyright (2001) Elsevier

Synthetic Chemistry of Microporous Compounds (II)

243

mixture of 3 TPAOH : 25 SiO2 : 1500 H2O : 100 EtOH for several hours. The assynthesized MFI membranes were washed in turn with 0.1 M NH3, water and acetone, and then calcined at 600  C for 20 min. In the next example (Example 2), a different approach for the preparation of zeolite films will be presented. This method was first proposed by Iler in 1979,[132] and involved the assembly of charged nano-sized particles layer by layer to form multilayer films based on electrostatic interaction, hence called the layer-by-layer (LBL) self-assembly technique. Example 2: Applications of LBL self-assembly technique in the preparation of zeolite films.[133] In this example, negatively charged polystyrene beads with a size of 4–8 mm were applied as supports to prepare LTA, FAU, BEA, and MFI hollow spheres by the LBL technique. The main procedure is as follows. First, polystyrene beads were treated with the cationic polymer Redifloc 4150 (Akzo Nobel, Sweden) as a charge-reversing agent. Then the particles in the nanozeolite systems were aggregated layer by layer to form membranes. Finally, calcination was carried out at 550  C in air to remove beads to leave hollow spheres without supports. The key problems are the synthesis of colloidal zeolite suspensions and the self-assembly of zeolite nanocrystals on the charged beads, which will be discussed one by one. (1) Synthesis of colloidal zeolite suspensions. The synthesis conditions of LTA, FAU, BEA, and MFI nanocrystals are listed in Table 4.14. (2) LBL self-assembly of zeolite nanocrystals on the surface of polystyrene beads. First, the as-synthesized nanozeolites were separated from the mother liquid by highspeed centrifugation, and then redispersed in water (2
3 wt%), and the pH value was adjusted to
9.5 by 0.1M aq. NH3 to form zeolite suspensions. Second, hollow nanozeolite spheres were prepared by the LBL technique according to the procedure presented in Figure 4.33. Zeolite membranes can also be prepared by the LBL self-assembly technique by using other inert supports, such as gold and carbon, instead of polystyrene beads. This technique for film preparation has several advantages. First, it is the first example of Table 4.14 Synthesis conditions and ultimate size of LTA, FAU, BEA, and MFI nanocrystals Code LTA FAU BEA MFI

Molar composition of the starting solution 0.3Na2O : 5 SiO2 : 0.6 Al2O3 : 9 TMA2O : 400 H2O 0.08 Na2O : 5 SiO2 : 1.15 Al2O3 : 2.7 TMA2O : 285 H2O 0.35 Na2O : 2.5 SiO2 : 0.5 Al2O3 : 4.5 TMA2O : 295 H2O 4.5 TPA2O : 2.5 SiO2 : 480 Al2O3 : 100 EtOH

T ( C)

Time (t/h)

80

4

150

100

76

50

100

196

40

60

300

50

Crystal size (nm)

244

Chemistry of Zeolites and Related Porous Materials

Figure 4.33 Procedure of the LBL technique used for preparation of hollow zeolite spheres. Reproduced with permission from [133]. Copyright (2001) Elsevier

the preparation of highly uniform zeolite films, especially unsupported films, under mild conditions by using supports with various morphologies, sizes, and 3-D surfaces. Second, different from other synthetic routes to zeolite membranes under hydrothermal conditions, there are almost no special demands for the thermal and chemical stability of supports. Finally, this method provides the foundation for the techniques of preparing various zeolite blend membranes. Example 3: Preparation of b-axis-oriented MFI zeolite thin films. In recent years, great success has been achieved in the preparation of oriented and continuous zeolite A and MFI monolayer films on ceramic substrates by using seededgrowth methods[134–136]. However, the MFI film prepared by this method is usually c-axis oriented. Because the straight channels of MFI zeolite run along the b direction, it is desirable to prepare b-oriented MFI films. In 2001, Yan and coworkers[134,135] made excellent contributions to this field. He found the relationship between batch composition and crystal orientation based on a systematic study of various crystallization conditions in the system TPAOH–NaOH–TEOS–H2O–NaCl at 165  C[134] (Figure 4.34). Highly oriented MFI films were synthesized by using polished and cleaned stainless steel substrates that were placed horizontally at the bottom of the autoclave to enable crystallization in the system 0.32 TPAOH : TEOS : 165 H2O at 140
175  C for 1
7 h. Synthesis conditions including the crystallization temperature and time, surface roughness, position and chemical nature of the support, and aging time were investigated

Synthetic Chemistry of Microporous Compounds (II)

245

Figure 4.34 Crystal orientation–synthesis composition map of MFI film. Reproduced with permission from [134]. Copyright (2001) American Chemical Society

carefully. Yan and coworkers found the relative conditions for formation of MFI zeolite seeds at 165  C in the form of b-oriented 0.35 mm disks tightly deposited on the surface of the substrate. This provides the key techniques for the synthesis of b-oriented MFI films. Based on this, Yan and coworkers prepared oriented and continuous MFI film with a thickness of less than 0.4 mm, and proposed the formation mechanism for this kind of oriented thin film (Figure 4.35). Example 4: Preparation of a-axis-oriented MFI zeolite films. The seeded approach to zeolite film synthesis was utilized by Tsapatsis and coworkers in preparing a-oriented silicalite-1 membranes on a-alumina disks.[137] Previously, the group had used similar approaches in synthesizing both c- and b-oriented membranes. a-Oriented siliceous MFI seed crystals were formed in the presence of trimer-TPA as the SDA along with KOH, triethyl orthosilicate (TEOS), and water. The substrate was prepared with a (3-chloropropyl)trimethoxysilane-functionalized silica layer and was then coated with the seeds. Secondary growth was performed, in Teflon-lined autoclaves, in a synthesis sol with molar composition 60 SiO2 : 9 TPAOH : 9500 H2O : 240 EtOH at 90  C for 96 h. Figure 4.36 shows SEM images of the films. The group is currently investigating the membranes for gas permeation applications. Example 5: Spin-on zeolite films. Thin zeolite membranes can also be prepared through a spin-coating process of a nanoparticle suspension. Yan and coworkers synthesized silicalite-1 and silicalite-2 in a nanoparticle suspension using a two-stage hydrothermal process.[138,139] First, the precursor

246

Chemistry of Zeolites and Related Porous Materials

Figure 4.35 Possible formation mechanism for the b-oriented MFI thin film. Reproduced with permission from [135]. Copyright (2001) Elsevier

compounds were aged in a polypropylene bottle at molar compositions of 1 TPAOH : 2.8 SiO2 : 22.4 EtOH : 40 H2O and 0.3 TBAOH : 1 SiO2 : 4 EtOH : 10 H2O for MFI and MEL, respectively, where TEOS was the silica source and TBAOH stands for tetrabutylammonium hydroxide. The first stage was a low-temperature synthesis that allowed for nucleation, and the temperature was then ramped up to a higher second stage where both nucleation and crystal growth occurred. After the synthesis, the resulting suspension was then deposited onto a silicon wafer and spun at 3300 rpm for about 20 s. The resulting film, which was a composite of amorphous silica and nanocrystalline zeolites, was about 400 nm thick. Figure 4.37 shows SEM images of the MFI films. The thin films were intended for low-dielectric-constant applications, specifically for next-generation microprocessors, and

Synthetic Chemistry of Microporous Compounds (II)

247

Figure 4.36 SEM images of a) zeolite MFI crystals obtained with trimer-TPA as SDA, b) magnified crystal with indication of b and c axes, and c) chemically bonded monolayer of MFI seeds after seed deposition with most of the MFI crystals lying flat on the support and a few misoriented seeds indicated by arrows. Top-view SEM images of MFI films after secondary growth are shown in d) and e), with lower and higher magnification, respectively, and a crosssectional view is shown in f). Scale bars are 1 mm. Reproduced with permission from [137]. Copyright (2006) Wiley-VCH

the spin-on process is the ideal method for the preparation of these films because it is compatible with the existing semiconductor infrastructure. Example 6: Synthesis of patterned zeolite films. Patterned zeolite films have been investigated for possible optical, magnetic, and electronic applications. Zeolite nanocrystals have been used as building blocks for patterned films.[140] Silicalite-1 nanocrystals were synthesized hydrothermally and the resulting crystals, after redispersion in ethanol, were deposited onto a smooth surface, such as a silicon wafer. A polydimethylsilane (PDMS) stamp was applied with adequate

248

Chemistry of Zeolites and Related Porous Materials

Figure 4.37 SEM images of pure-silica-zeolite spin-on films. Top views of films at second-stage synthesis times of a) 8 h, b) 10 h, c) 69 h. d) Cross-sectional view of a film with a second-stage synthesis time of 69 h. Reproduced with permission from [138]. Copyright (2004) Wiley-VCH

pressure yielding a micro-patterned zeolite film. Other 3-D structures and transparent self-standing membranes can also be synthesized by using silicalite-1 nanocrystals as building blocks through gel-casting and high-pressure compression techniques.[141,142] Nonregular surface-patterned films were also prepared with similar nanocrystals through convection-assisted self-assembly of silicalite-1 nanocrystals on a silicon wafer.[143] In this case, the films probably formed during solvent (ethanol) evaporation due to Benard–Marangoni convection flow patterns. Examples of these films are shown in Figure 4.38.

4.2.4 Synthesis of Microporous Material with Special Aggregation Morphology in the Presence of Templates As mentioned in the previous section, hollow zeolite spheres of LTA, FAU, BEA, MFI can be prepared in the presence of polystyrene beads as ‘templates’ by using an LBL selfassembly technique. Recently, several research groups have tried to adopt similar methods to synthesize zeolite-template composites on the surfaces of templates with various shapes and sizes, properties, and structures through self-assembly or in situcrystallization approaches. Subsequent removal of the templates forms zeolite materials with analogical skeletons of the templates. Up to now, the reported templates include microspheres, carbon fibers, polyurethane foams, and microbe structures,[144,145] as well

Synthetic Chemistry of Microporous Compounds (II)

249

Figure 4.38 Patterned zeolite films prepared by using a) soft-lithography. Reproduced with permission from [140]. Copyright (2000) American Chemical Society; b) convection-assisted self-assembly. Reproduced with permission from [143]. Copyright (2001) American Chemical Society); c) gel-casting. Reproduced with permission from [141]. Copyright (2001) Royal Society of Chemistry

as natural woods used by Tang and coworkers at Fudan University to prepare porous zeolite materials with cellular structures.[146] In the following, paragraphs their work will be described as examples to illustrate the evolution of this method. Two kinds of tissue, cedar with relatively uniform-sized pores and bamboo with nonuniform-sized pores, were adopted as the templates. The typical synthetic procedure was that the cut wood slices ð1:0 cm  0:5 cm  0:5 cmÞwere first soaked in a solution of cationic polyelectrolyte, poly(diallyldimethylammonium chloride) (PDDA, Mr < 200 000), for 2 h. After being washed with distilled water, the wood slices were dipped into a suspension of silicalite-1 nanocrystals (80 mm, 1 wt%) for 12 h, followed by washing with water at pH 10 to remove the excess of seeds. Finally, the wood slices were placed in a reaction mixture (10 mL) of 3 TPAOH : 25 SiO2 : 1500 H2O : 100 EtOH (TPAOH: tetrapropylammonium hydroxide; tetraethyl orthosilicate as the Si source) and crystallized at 110  C for 24 h. The as-synthesized zeolite/template composite was calcined at 600  C in air to obtain zeolite materials with biomimetic morphology. XRD analysis shows that the product is the MFI phase with high purity. The zeolite materials obtained after removal of the templates inherit the initial wood cellular structure well. Their structures are composed of bundles of hollow fibers, and the size and shape of the hollow fibers are very similar to those of the wood cells. At high magnifications, the growth of zeolite twin crystals on the fiber walls can be clearly observed on SEM images. Initial study of the formation mechanism of these materials shows that adjacent tracheids of woods are linked by middle lamellae, and zeolite crystals predominantly grow on the inner cell walls under electrostatic interactions, resulting in continuous crystal seed films. After the secondary growth, the crystal seed films are gradually thickened into a dense zeolite film to form the zeolite/wood composite. Upon removal of the wood tissue by calcination, the porous zeolite materials with wood-like

250

Chemistry of Zeolites and Related Porous Materials

Figure 4.39 SEM images of (a) as-synthesized silicalite-1 microspheres and (b) calcined samples prepared by the use of MSA-1 resin. Reproduced with permission from [147]. Copyright (2000) Elsevier

structure are produced. A nitrogen adsorption–desorption experiment indicates that, besides the microporosity of the zeolite, the additional adsorption step at P=P0 ¼ 0:35 is clearly indicative of the mesoporous zeolitic tissue. The mesopores may be the intercrystalline voids between the individual zeolite particles. The total surface area of 315 m2 g1 and micropore surface area of 233 m2 g1 are close to those values in the reported nanozeolites. These experimental results indicate that hierarchical pore zeolite materials can be prepared through a secondary growth method by using wood tissue as the template. Because the woods used are easily obtained, and the product overcomes the limitation of the single micropore of zeolites on mass transfer, there are potential applications in the adsorption and catalysis fields. It also provides a precedence for the further exploration of this kind of material. In 2000, Tosheva et al.[147] employed anion-exchange resins (macroporous basic styrene–divinylbezene) from Dowex as shape-directing macrotemplates to successfully prepare uniform silicate-1 microspheres. Typically, a strongly basic resin (MSA-1) with a bead size distribution in the range 0.3
1.2 mm, an ion-exchange capacity of 1.0 meq/mL, and pore size of 20
100 nm was used as the template, which was placed in a reaction system of composition 9 TPAOH : 25 SiO2 : 480 H2O : 100 EtOH at 100  C. Crystallization was performed in the pores of MSA-1 resin to produce resin/silicalite-1 microspheres. The product was filtered off, washed, dried, and calcined at 600  C for 5 h to remove the template. Uniform Silicalite-1 microspheres were then obtained (Figure 4.39). Figure 4.40 schematically illustrates the procedure for preparing silicalite-1 microspheres. In 2003, Yates and coworkers[148] explored a new method to prepare microporous AlPO4-5 fibers by using micro-emulsion droplets as the templates. The surfactant cetylpyridinium chloride with the cosurfactant butanol in the ratio 1/2 forms a stable micro-emulsion in the

Figure 4.40 The procedure for preparing silicalite-1 microspheres using anion-exchange resins as macrotemplates. Reproduced with permission from [147]. Copyright (2000) Elsevier

Synthetic Chemistry of Microporous Compounds (II)

251

Figure 4.41 Formed micro-emulsion phase diagram in the cetylpyridinium chloride–butanol– toluene–H2O system. Reproduced with permission from [148]. Copyright (2003) American Chemical Society

AlPO4-5 synthesis gel [H2O : Al(OPri)3 : H3PO4 : (C2H5)3N : HF ¼ 50 : 0.8 : 1.0 : 0.6 : 0.5] at room temperature, and the phase diagrams (Figure 4.41) were obtained to guide the synthesis. Crystallization was achieved at 180  C for 6 h by choosing various microemulsion compositions. The products obtained from compositions A, C, and D were AlPO45 crystals with a fibrous morphology (Figure 4.42). As a comparison, Figure 4.42(E) shows the product obtained from the control experiment without the micro-emulsion. It should be noted that similar products could also be prepared by microwave heating at 180  C for 17 min (Figure 4.43). These results show that AlPO4-5 fibers tend to form by using a microemulsion as the template. In conclusion, the employment of templates provides a feasible strategy to prepare microporous compounds with special morphologies. However, the formation mechanism of these compounds is still not very clear, and more work on this subject should be carried out. 4.2.5

Applications of Zeolite Membranes and Films

Zeolites are traditionally used in catalysis/purification and separation applications in the petrochemical industry but are rapidly finding new uses. This section discusses membranes for low-dielectric-constant, corrosion-resistant, hydrophilic and antimicrobial, and pervaporation applications. Example 1: Low-dielectric-constant films. Low-dielectric-constant (low-k) materials are needed to reduce signal delays, crosstalk noise, and power consumption in the next generation of faster and more powerful

252

Chemistry of Zeolites and Related Porous Materials

Figure 4.42 Products crystallized at 180  C for 6 h from microemulsion compositions (A and C) and traditional hydrothermal conditions without micro-emulsion (E). Reproduced with permission from [148]. Copyright (2003) American Chemical Society

microprocessors. The sub-90 nm technology is expected to be accomplished only with the use of a dielectric material with a k value lower than 2.4, but a manufacturable solution of such a low-k material is still unknown.[149] Porous silicas, including sol–gel silica and surfactant-templated mesoporous silica, have been considered as potential candidates for low-k materials and have been shown to be effective in lowering k values by taking advantage of the low dielectric constant of air ðk ¼ 1Þ. However, the trade-off between k value and mechanical-strength raises concerns. Amorphous porous silicas exhibit mechanical-strength values (through the elastic modulus) below 6 GPa at k values below 2.0.[150] The semiconductor industry generally considers 6 GPa to be the minimum threshold for survivability during chemical-mechanical processes.[151] Based on their previously reported in-situ MFI low-k films,[152] Yan and coworkers explored the possibility of extending the k value to below 2 by different methods without significantly lowering the mechanical strength. They successfully prepared pure-silica zeolite (PSZ) MFI films by in-situ, seeded growth, and spin-on methods. b-Oriented in-situ films yielded k values of 2.7 and elastic modulus ðEÞ values of 30–40 GPa.[152] Spin-on MFI films that incorporated g-cyclodextrin

Synthetic Chemistry of Microporous Compounds (II)

253

Figure 4.43 Products by microwave heating at 180  C for 17 min from micro-emulsion compositions (A and C) and traditional hydrothermal conditions without microemulsion (E). Reproduced with permission from [148]. Copyright (2003) American Chemical Society

as a porogen had a lower k value of 1.8 and an E of 14.3 GPa.[153] Using nanoparticle suspensions with high crystallinity, the k value of the spin-on MFI membranes was recorded as 1.6 and the mechanical strength was about 12 GPa.[138] By changing the zeolite framework, Yan and colleagues also prepared highly crystalline PSZ MEL films with an ultra-low k value of 1.5–1.7 and a reduced modulus of 10.1–17.6 GPa.[139] Small MEL nanoparticles (
50 nm) were obtained in high yield (>50%) by using a two-stage synthesis method. Figure 4.44 shows that zeolite films are far superior to amorphous porous silicas in terms of mechanical strength at similar k values. The plot also contains calculated values for different zeolite framework types, as well as experimental mechanical and dielectric values of single zeolite crystals.[150,154] Example 2: Corrosion-resistant coatings. Zeolite coatings have been investigated as corrosion-resistant coatings for aluminum alloys in aerospace applications. The currently used chromic acid anodization and chromate conversion coatings are effective but release hexavalent chromium, which is

254

Chemistry of Zeolites and Related Porous Materials

Figure 4.44 Elastic modulus versus dielectric constant for amorphous porous silicas and pure-silica zeolites. The open circles and solid data-fitting line correspond to experimental data taken from Xu et al. Reproduced with permission from [150]. Copyright (2006) Wiley-VCH

a proven carcinogen. Several types of zeolite offer excellent corrosion resistance. First, high-silica zeolites are thermally stable up to 1000  C and pure-silica zeolites are chemically stable in all mineral acids except hydrofluoric acid. Secondly, although zeolites are porous, many synthetic zeolites require an organic SDA to correctly obtain the desired framework. After the synthesis, the SDA is still entrapped in the pores and renders the zeolite membrane gas-tight.[155] Yan and coworkers showed, through DC polarization and AC impedance tests, that high-silica ZSM-5 coatings are highly resistant to corrosion by strong acids, bases, and pitting aggressive media.[155] Aluminum alloy 2024-T3 substrates were placed in a synthesis solution of molar composition 0.16 TPAOH : 0.64 NaOH : 1 TEOS: 92 H2O : 0.0018 Al and zeolite coatings were synthesized hydrothermally in Teflonlined autoclaves at 175  C. SEM images of the coatings are shown in Figure 4.45. The coatings were then exposed to 0.5M H2SO4, 0.5M NaCl/HCl (pH ¼ 1), or 0.5M NaCl þ 0.26 g/L CuCl2 þ HAc (pH ¼ 3) solutions for 2–24 h and they subsequently outperformed both anodization coatings and chromate conversion coatings in DC polarization tests. AC impedance tests on the zeolite coatings also showed better results than did anodization and chromate conversion coatings. Furthermore, the corrosionresistance properties were demonstrated to hold using a single zeolite synthesis recipe on aluminum alloy substrates of significantly different chemical compositions (e.g., AA2024-T3, AA-5052-H32, AA-6061-T4, and AA-7075-T6).[156] The thermal and mechanical properties of the coatings were also demonstrated by the same group. Corrosionresistant coatings of other framework types, including BEA and MTW, also showed good corrosion resistance in similar experiments.[157, 158] Example 3: Hydrophilic and antimicrobial coatings. Hydrophilic and antimicrobial zeolite coatings have also been shown to be effective for gravity-independent water-separation applications on manned spacecraft. Condensing

Synthetic Chemistry of Microporous Compounds (II)

255

Figure 4.45 SEM images of an as-synthesized zeolite coating on Al-2024-T3 a) top view b) cross-sectional view. Reproduced with permission from [155]. Copyright (2001) Electrochemical Society Inc.

heat exchangers are used for controlling the cabin humidity and temperature, and, as the air cools within the condenser, water droplets form. Because of the microgravity environment, the small droplets are carried in the air stream, causing the cabin to become foggy. Covering the fins of the condenser with a hydrophilic coating causes the water to form a film on the surface of the condenser that can then be vacuum sucked away.[159] Yan and coworkers synthesized both ZSM-5 and zeolite A coatings on several aluminum alloy and stainless substrates using an in-situ method.[159] The ZSM-5 coating was synthesized at 175 C for 15 h using a solution of molar composition 16 TPAOH : 0.64 NaOH : 1 TEOS : 92 H2O : 0.0018 Al, and the zeolite A coating was prepared at 65 C for 15 h in a solution of 10 Na2O : 0.2 Al2O3 : 1 SiO2 : 200 H2O. Except for one case, water-contact angles on both coatings averaged less than 10 for all substrates tested, revealing a hydrophilic nature. Furthermore, all coatings received a rating of 5B, the highest possible rating, on ASTM D-3359-02 adhesion tests. To prevent bacterial growth on the films, the sodium ions in the zeolite A pores were exchanged with silver ions. After an initial inoculation with 10[137] colony-forming units (CFUs), the CFU value decreased over a period of 24 h to zero. This result is shown in Figure 4.46. It was also shown that the coatings retained their hydrophilic and antimicrobial properties and were durable under water immersion for eight weeks.[160] The antimicrobial capability is regenerative through further ion-exchange, and silver is nontoxic to humans. Example 4: Zeolite membranes for pervaporation. Pervaporation is a separation technique that has several advantages over other separation methods such as distillation, extraction, and sorption. It is an especially good process for certain azeotropes and mixtures that have components with similar boiling points. Also, pervaporation can be performed at low temperatures and the membranes involved generally do not need to be regenerated in any way.[161] Zeolite membranes have been used for pervaporation processes and offer numerous benefits over other pervaporation materials such as polymers. First, they are chemically

256

Chemistry of Zeolites and Related Porous Materials

Figure 4.46 Surviving colony-forming units on bare stainless steel, zeolite A-coated stainless steel, and silver-ion-exchanged zeolite A over a 24 h incubation period. Reproduced with permission from [159]. Copyright (2005) Wiley-VCH

stable in strong solvents and acids, except hydrofluoric acid, and they can operate at high temperatures. Secondly, as opposed to polymer membranes, zeolites do not swell and are generally mechanically stable. Furthermore, substitution of other T-atoms (e.g., boron, iron, and germanium) into the zeolite framework generally improves the separation performance.[162] Noble and Falconer and colleagues have developed a number of zeolite membranes for separating organic compounds from aqueous solutions. Specifically, they have studied the use of B-ZSM-5 membranes for the separation of butane isomer vapors, methanol, ethanol, 1-propanol, 2-propanol, and acetone.[162,163] The monolith-supported membranes were hydrothermally synthesized in Teflon-lined autoclaves in a synthesis gel of molar composition 1.6 TPAOH : 19.5 SiO2 : 0.2 B(OH)3 : 438 H2O. Figure 4.47 shows the permeances of several organic and inorganic compounds tested over the temperature range 300–475 K. The authors note

Figure 4.47 Permeance values for several organic compounds in water over a temperature range of 300–475 K. The feed pressure was 223 kPA and the trans-membrane pressure drop was 138 kPa.[162] Reproduced with permission from [162]. Copyright (2003) Elsevier

Synthetic Chemistry of Microporous Compounds (II)

257

Table 4.15 Flux and separation factor comparisons for silicalite-1 membranes synthesized with and without a template[165] With template Parameter

Pure

Flux for p-xylenea Flux for o-xylene p=o Separation factor

34 27 1.2

Mixture 15 15 1

Without template Pure

Mixture

28.2 0.4 69

13.7 0.3 40

Unit for flux (102 kg/m2 h).

a

that selective adsorption of the organic compounds was due to the hydrophobicity of the zeolite membranes, as well as diffusion differences. The separation factors for methanol, ethanol, 2-propanol, and 1-propanol were 8.4, 31, 42, and 75, respectively, at 333 K and were all considerably higher than B-ZSM-5 tubular membranes, with the exception of methanol. The group has also written a comprehensive review on pervaporation using zeolite membranes.[161] Pervaporation of xylene isomers is important to the petrochemical industry and is also a widely accepted method for testing zeolite molecular sieving capabilities. Tsapatsis and coworkers synthesized b-oriented silicalite-1 membranes using a two-step seeded method with a p-xylene permeance of 2  107 mol/m2 s Pa and a p- to o-xylene separation factor of 500.[164] The secondary growth step requires use of trimer-TPA as the SDA. Lin and coworkers also studied the separation of xylene isomers by using a silicalite-1 membrane synthesized through a template-free seeded process.[165] Because this method does not require the removal of the organic material, these membranes are free of intercrystalline gaps or mesoscopic defects. Table 4.15 illustrates that, compared with membranes prepared with templates, the template-free membranes perform better, in terms of flux and selectivity.

References [1] R. Xu, J. Chen, and S. Feng, New Families of M(III)X(V)O4-type Microporous Crystals and Inclusion Compounds. Chemistry of Microporous Crystals. Stud. Surf. Sci. Catal., 1991, 60, 63–72. [2] J.B. Parise, Some Gallium Phosphate Frameworks Related to the Aluminum Phosphate Molecular Sieves: X-ray Structural Characterization of {(PriNH3)[Ga4(PO4)4OH]}H2O. J. Chem. Soc., Chem. Commun., 1985, 606–607; J.B. Parise, Preparation and Structural Characterization of Two Metallophosphate Frameworks Clathrating Diprotonated Ethylenediamine: AlPO4-12(en) and GaPO4-12(en). Inorg. Chem., 1985, 24, 4312–4316. [3] S. Feng, Doctoral Thesis, Jilin University, China (1986). [4] S. Feng and R. Xu, Studies on the Syntheses and Structures of a Novel Family of Microporous Gallophosphates. Chem. J. Chin. Univ., 1987, 8, 867–868. [5] G. Yang, L. Li, J. Chen, and R. Xu, Synthesis and Structure of a Novel Aluminoarsenate with an Open Framework. J. Chem. Soc., Chem. Commun., 1989, 810–811. [6] L. Li, L. Wu, J. Chen, and R. Xu, Structure of a Novel Aluminoarsenate with an Occluded (CH3)4Nþ Cation. Acta Crystallogr., Sect. C, 1991, 47, 246–249.

258

Chemistry of Zeolites and Related Porous Materials

[7] J. Chen, L. Li, G. Yang, and R. Xu, Preparation and Structural Characterization of a Novel Galloarsenate Using a Dimethylamine Template. J. Chem. Soc., Chem. Commun., 1989, 1217–1218. [8] H. Du, J. Chen, and W. Pang, Synthesis of Various Indium Phosphates in the Presence of Aminetemplates. Progress in Zeolite and Microporous Materials. Stud. Surf. Sci. Catal., 1997, 105, 397–404. [9] L. Koh, Y. Xu, H. Du, and W. Pang, New Insight into the Study of Indiumphosphate Molecular Sieves. Stud. Surf. Sci. Catal., 1997, 105, 373–380. [10] Y. Xu, L. Koh, and L. An, Synthesis and Characterization of a Novel Microporous Indiumphosphate. Stud. Surf. Sci. Catal., 1994, 84, 2253–2260. [11] A.K. Cheetham, G. Fe´rey, and T. Loiseau, Open-framework Inorganic Materials. Angew. Chem., Int. Ed., 1999, 38, 3268–3292. [12] T.E. Gier and G.D. Stucky, Low-temperature Synthesis of Hydrated Zinco(beryllo)-phosphate and Arsenate Molecular-sieves. Nature (London), 1991, 349, 508–510. [13] S. Neeraj, S. Natarajan, and C.N.R. Rao, A Novel Open-framework Zinc Phosphate with Intersecting Helical Channels. Chem. Commun., 1999, 165–166. [14] W.T.A. Harrison, T. Gier, G. Stucky, R. Broach, and R. Bedard, NaZnPO4 H2O, an Openframework Sodium Zincophosphate with a New Chiral Tetrahedral Framework Topology. Chem. Mater., 1996, 8, 145–151. [15] Y. Wang, J. Yu, M. Guo, and R. Xu, [{Zn2(HPO4)4}{Co(dien)2}] H3O A Zinc Phosphate with Multidirectional Intersecting Helical Channels. Angew. Chem., Int. Ed., 2003, 42, 4089– 4092. [16] J.A. Rodgers and W.T.A. Harrison, H3N(CH2)6NH3 Zn4(PO4)2(HPO4)2 3H2O: a Novel Three-dimensional Zinc Phosphate Framework Containing 5- and 20-rings. J. Mater. Chem., 2000, 10, 2853–2856. [17] J. Zhu, X. Bu, P. Feng, and G. Stucky, An Open-framework Material with Dangling Organic Functional Groups in 24-ring Channels. J. Am. Chem. Soc., 2000, 122, 11563–11564. [18] Y. Song, J. Yu, G. Li, Y. Li, Y. Wang, and R. Xu, Combinatorial Approach for the Hydrothermal Synthesis of Open-framework Zinc Phosphates. Chem. Commun., 2002, 1720–1721. [19] A. Choudhury, S. Natarajan, and C.N.R. Rao, Formation of One-, Two-, and Threedimensional Open-framework Zinc Phosphates in the Presence of a Tetramine. Inorg. Chem., 2000, 39, 4295–4304. [20] C.N.R. Rao, N. Srinivasan, and S. Neeraj, Building Open-framework Metal Phosphates from Amine Phosphates and Monomenic Four-membered Ring Phosphate. J. Solid State Chem., 2000, 152, 302–321. [21] A. Ayi, A. Choudhury, S. Natarajan, S. Neeraj, and C.N.R. Rao, Transformations of Low-dimensional Zinc Phosphates to Complex Open-framework Structures. Part 1: Zerodimensional to One-, Two- and Three-dimensional Structures. J. Mater. Chem., 2001, 11, 1181–1191. [22] A. Choudhury, S. Natarajan, S. Neeraj, and C.N.R. Rao, Transformations of the Lowdimensional Zinc Phosphates to Complex Open-framework Structures. Part 2: One-dimensional Ladder to Two- and Three-dimensional Structures. J. Mater. Chem., 2001, 11, 1537– 1546. [23] C.N.R. Rao, S. Natarajan, A. Choudhury, S. Neeraj, and A. Ayi, Aufbau Principle of Complex Open-framework Structures of Metal Phosphates with Different Dimensionalities. Acc. Chem. Res., 2001, 34, 80–87. [24] S. Qiu, W. Tian, W. Pang, T. Sun, and D. Jiang, Synthesis and Characterization of Single Crystals of SAPO-5, BAPO-5 and LiAPO-5 Molecular Sieves. Zeolites, 1991, 11, 371–375.

Synthetic Chemistry of Microporous Compounds (II)

259

[25] S.C. Sevov, Synthesis and Structure of CoB2P3O12(OH) C2H10N2: The First Metal Borophosphate with an Open Framework Structure. Angew. Chem., Int. Ed. Engl., 1996, 35, 2630– 2632. [26] J. Wang, S. Feng, and R. Xu, Synthesis and Characterization of Zeolitic Microporous Alumino-borates. Stud. Surf. Sci. Catal., 1989, 49, 143–150. [27] J. Wang, S. Feng, and R. Xu, Synthesis and Characterization of a Novel Microporous Alumino-borate. J. Chem. Soc., Chem. Commun., 1989, 265–266. [28] J. Yu, R. Xu, Q. Kan, Y. Xu, and B. Xu, Synthesis and Characterization of a Novel Boroncontaining Aluminum Basic Chloride. J. Mater. Chem., 1993, 3, 77–82. [29] J. Yu, R. Xu, J. Chen, and Y. Yue, On the Crystallization and Nature of the Microporous Boron-aluminum OxO-chloride BAC(10). J. Mater. Chem., 1996, 6, 465–468. [30] J. Yu, R. Xu, Y. Xu, and Y. Yue, Synthesis and Characterization of a Boron-aluminum OxOchloride. J. Solid State Chem., 1996, 122, 200–205. [31] R. Bedard, S. T. Wilson, L. D. Vail, J. M. Bennett, and E. M. Flanigen, The Next Generation: Synthesis, Characterization and Structure of Metal Sulfide-based Microporous Solids. Stud. Surf. Sci. Catal., 1989, 49, 375–387. [32] O.M. Yaghi, Z. Sun, and D. Richardson, Directed Transformation of Molecules to Solids: Synthesis of a Microporous Sulfide from Molecular Germanium Sulfide Cages. J. Am. Chem. Soc., 1994, 116, 807–808. [33] J. Martin and K. Greenwood, Halozeotypes: a New Generation of Zeolite-type Materials. Angew. Chem., Int. Ed. Engl., 1997, 36, 2072–2075. [34] M.E. Davis, C. Saldarriag, C. Montes, J. Garces, and X. Zhao, A Molecular Sieve with Eighteen-membered Rings. Nature (London), 1998, 331, 698–699. [35] K.J. Balkus Jr., A.G. Gabrielov, and S.I. Zones, Synthesis of Porous Materials, ed. M. Occelli and H. Kessler, Marcel Dekker, New York, 1996, 77. [36] K.J. Balkus Jr., M. Biscotto, and A.G. Gabrielov, The Synthesis and Characterization of UTD-1: The First Large Pore Zeolite Based on a 14 Membered Rings System. Stud. Surf. Sci. Catal., 1997, 105A, 415–421. [37] P. Wagner, M. Yoshikawa, M. Lovallo, K. Tsuji, M. Tsapatsis, and M.E. Davis, CIT-5: a High-silica Zeolite with 14-rings Pores. Chem. Commun., 1997, 2179–2180. [38] M. Yoshikawa, P. Wagner, M. Lovallo, K. Tsuji, T. Takewaki, C.-Y. Chen, L.W. Beck, C. Jones, M. Tsapatsis, S.I. Zones, and M.E. Davis, Synthesis, Characterization and Structure Solution of CIT-5, a New High-silica Extra-large-pore Molecular Sieve. J. Phys. Chem. B, 1998, 102, 7139–7147. [39] Y. Nakagawa and S. Zones, US Patent, 5,271,922 (1993). [40] J.R. Anderson, W.R. Jackson, D. Hay, Z. Yang, and E.M. Campi, Optimization of a VPI-5 Synthesis. Zeolites, 1996, 16, 15–21. [41] J. Chen, W. Pang, and R. Xu, Mixed-bonded Open-framework Aluminophosphates and Related Layered Materials. Top. Catal., 1999, 9, 93–103. [42] M. Estermann, L.B. McCusker, Ch.M.A. Baerlocher, A. Merrouche, and H. Kessler, A Synthetic Gallophosphate Molecular Sieve with a 20-tetrahedral-atom Pore Opening. Nature (London), 1991, 352, 320–332. [43] Q. Huo and R. Xu, Synthesis of the Gallophosphate Molecular Sieve Cloverite with 20-ring Channel from an Alcoholic System. J. Chem. Soc., Chem. Commun., 1992, 1391–1392. [44] Q.S. Huo, R.R. Xu, S.G. Li, Z. Ma, J.M. Thomas, R. Jone, and A.M. Chippindale, Synthesis and Characterization of a Novel Extra Large Ring of Aluminophosphate JDF-20. J. Chem. Soc., Chem. Commun., 1992, 875–876. [45] Q.S. Huo, R.R. Xu, S.G. Li, Y. Xu, Z. Ma, Y. Yue, and L. Li, Synthesis, Characterization and Phase Transition of the 20-membered Ring AlPO4: JDF-20. Proceedings of the 9th

260

[46]

[47]

[48]

[49]

[50] [51]

[52]

[53]

[54]

[55]

[56]

[57]

[58]

[59] [60]

[61] [62]

Chemistry of Zeolites and Related Porous Materials International Zeolite Conference, ed. R. Von Ballmoos, J.B. Higgins, and M.M.J. Treacy, Butterworth-Heinemann, Montreal, 1992, 279–286. T. Loiseau and G. Fe´rey, Oxyfluorinated Microporous Compounds: VII. Synthesis and Crystal Structure of ULM-5, a New Fluorinated Gallophosphate Ga16(PO4)14(HPO4)2 (OH)2F7 [H3N(CH2)6NH3]4 6 H2O with 16-membered Rings and Both Bonding and Encapsulated F. J. Solid State Chem., 1994, 111, 403–415. C. Sassoye, J. Marrot, T. Loiseau, and G. Fe´rey, Utilization of Cyclopentylamine as Structure-directing Agent for the Formation of Fluorinated Gallium Phosphates Exhibiting Extra-large Pore Open Frameworks with 16-ring (ULM-16) and 18-ring Channels (MIL-46). Chem. Mater., 2002, 14, 1340–1347. C. Sassoye, T. Loiseau, F. Taulelle, and G. Fe´rey, A New Open-framework Fluorinated Gallium Phosphate with Large 18-ring Channels (MIL-31). Chem. Commun., 2000, 943–944. C.H. Lin, S.L. Wang, and K. Lii, Ga2 [(DETA)(PO4)2] 2H2O (DETA ¼ Diethylenetriamine): A Novel Porous Gallium Phosphate Containing 24-ring Channel. J. Am. Chem. Soc., 2001, 123, 4649–4650. G. Yang and S. Sevov, Zinc Phosphate with Gigantic Pores of 24 Tetrahedra. J. Am. Chem. Soc., 1999, 121, 8389–8390. N. Guillou, Q. Gao, M. Nogue´s, R. Morris, M. Hervieu, G. Fe´rey, and A. Cheetham, Zeolitic and Magnetic Properties of a 24-membered Ring Porous Nickel(II) Phosphate, VSB-1. C. R. Acad. Sci., Ser. IIc: Chem., 1999, 2, 387–392. N. Guillou, Q. Gao, P. Forster, J. Chang, M. Nogue`s, S. Park, G. Fe´rey, and A. Cheetham, Nickel(II) Phosphate VSB-5: A Magnetic Nanoporous Hydrogenation Catalyst with 24-ring Tunnels. Angew. Chem., Int. Ed., 2001, 40, 2831–2834. Y. Zhou, H. Zhu, Z. Chen, M. Chen, Y. Xu, H. Zhang, and D. Zhao, A Large 24-memberedring Germanate Zeolite-type Open-framework Structure with Three-dimensional Intersecting Channels. Angew. Chem., Int. Ed., 2001, 40, 2166–2168. J. Ple´vert, T. Gentz, A. Laine, H. Li, V. Young, O.M. Yaghi, and M. O’Keeffe, A Flexible Germanate Structure Containing 24-ring Channels and with Very Low Framework Density. J. Am. Chem. Soc., 2001, 123, 12706–12707. Z. Lin, J. Zhang, J. Zhao, S. Zheng, C. Pan, G. Wang, and G. Yang, A Germanate Framework Containing 24-ring Channels, Ni-Ge Bonds, and Chiral [Ni@Ge14O24(OH)3] Cluster Motifs Transferred from Chiral Metal Complexes. Angew. Chem., Int. Ed., 2005, 44, 6881–6884. J. Liang, J. Li, J.Yu, P. Chen, Q. Fang, F. Sun, and R. Xu, [(C4NH12)2][Zn3(HPO3)4]: A New Open-framework Zinc Phosphite Containing Extra-large 24-ring Channels. Angew. Chem., Int. Ed., 2006, 45, 2546–2548. A. Corma, F. Rey, S. Valencia, J. Jorda´, and J. Rius, A Zeolite with Interconnected 8-, 10- and 12-ring Pores and its Unique Catalytic Selectivity. Nature Mater. (London), 2003, 2, 493– 497. R. Castanˇeda, A. Corma, V. Fornes, F. Rey, and J. Rius, Synthesis of a New Zeolite Structure ITQ-24 with Intersecting 10- and 12-membered Ring Pores. J. Am. Chem. Soc., 2003, 125, 7820–7821. A. Burton, Zeolite: Porous Architectures. Nature Mater. (London), 2003, 2, 438–440. A. Corma, M. Navarro, F. Rey, J. Rius, and S. Valencia, Pure Polymorph C of Zeolite Beta Synthesized by Using Framework Isomorphous Substitution as a Structure-directing Mechanism. Angew. Chem., Int. Ed., 2001, 40, 2277–2280. S. T. Wong, S. H. Wong, J. Liu, and S. Cheng, Molecular Sieves Frame Pillaring of Layered Silicates. Stud. Surf. Sci. Catal., 1994, 84A, 45–52. W. T. Granquist, and M. Township, US Patent, 3,252,757 (1966).

Synthetic Chemistry of Microporous Compounds (II)

261

[63] Z. Liu, D. Jiang, H. Cai, and E. Min, The Hydroisomerization Nature of Ni-substituted Mica Montmorillonite and its Silica Oligomers Pillared Clay in Hexane Reaction. Chem. J. Chin. Univ., 1991, 12, 397–399. [64] J. Yu, R. Xu, and J. Li, Structural Diversity of a Family of Aluminophosphates with Al/P Ratio of Non-unity. Solid State Sci., 2000, 2, 181–192. [65] J. Valverde, P. Sa´nchez, F. Dorado, C. Molina, and A. Romero, Influence of the Synthesis Conditions on the Preparation of Titanium-pillared Clays Using Hydrolyzed Titanium Ethoxide as the Pillared Agent. Microporous Mesoporous Mater., 2002, 54, 155–165. [66] G. Alberti, U. Costantiono, F. Marmottoni, R. Vivani, and P. Zappelli, Preparation of a Covalently Pillared a-Zirconium Phosphite-diphosphonate with a High Degree of Interlayer Porosity. Microporous Mesoporous Mater., 1998, 21, 297–304. [67] S. Wong and S. Cheng, Synthesis and Characterization of Pillared Buserite. Inorg Chem., 1992, 31, 1165–1172. [68] S. Cheng and T. Wang, Pillaring of Layered Titanates by Polyoxo Cations of Aluminum. Inorg. Chem., 1989, 28, 1283–1289. [69] X. Guo, W. Hou, Q. Yan, and Y. Chen, Pillared Layered Transition Metals Oxides. Chin. Sci. Bull., 2003, 48, 101–110. [70] A. Corma, M. Iglesias, C. Pino, and F. Sa´nchez, New Rhodium Complexes Anchored on Modified USY Zeolites. A Remarkable Effect of the Support on the Enantioselectiveity of Catalytic Hydrogenation of Prochiral Alkenes. J. Chem. Soc., Chem. Commun., 1991, 1253– 1255. [71] S. Xiang, Y. Zhang, Q. Xin, and C. Li, Asymmetric Epoxidation of Allyl Alcohol on Organicinorganic Hybrid Chiral Catalysts Grafted onto the Surface of Silica and in the Mesopores of MCM-41. Angew. Chem., Int. Ed., 2002, 41, 821–824. [72] T. Katsuki and K.B. Sharpless, The First Practical Method for Asymmetric Epoxidation. J. Am. Chem. Soc., 1980, 102, 5974–5975. [73] R. Rouse and D. Peacor, Crystal Structure of the Zeolite Mineral Goosecreekite, CaAl2Si6O16 5H2O. Am. Mineral., 1986, 71, 1494–1501. [74] (a) J. Newsam, M. Treacy, W. Koetsier, and C. de Gruyte, Structural Characterization of Zeolite Beta. Proc. R. Soc. London, Ser. A, 1988, 420, 375–405; (b) M.E. Davis and R.F. Lobo, Zeolite and Molecular Sieve Synthesis. Chem. Mater., 1992, 4, 756–768. [75] K. Kongshaug, H. Fjellva˚g, K. Lillerud, T. Gier, G. Stucky, and A. Cheetham, Personal Communication. [76] A. Healey, M.Weller, and A. Genge, Synthesis and Structure of NaZnSiO3OH, a New Chiral Zincosilicate Framework Material. Inorg. Chem., 1999, 38, 455–458. [77] T. Gier, X. Bu, P. Feng, and G. Stucky, Synthesis and Organization of Zeolite-like Materials with Three-dimensional Helical Pores. Nature (London), 1998, 395, 154–157. [78] M. Medina, M. Iglesias, N. Snejko, E. Gutie´rrez-Puebla, and M. Monge, Chiral Germanium Zeotype with Interconnected 8-, 11-, and 11-Ring Channels Catalytic Properties. Chem. Mater., 2004, 16, 594–599. [79] Z. Lin, J. Zhang, and G. Yang, Synthesis and Structure of KBGe2O6: The First Chiral Zeotype Borogermanate with 7-Ring Channels. Inorg. Chem., 2003, 42, 1797–1799. [80] H. Zhang, J. Zhang, S. Zheng, G. Wang, and G. Yang, K2[Ge(B4O9)] 2H2O: A Unique 3D Alternating Linkage Mode of a B4O9 Cluster and GeO4 Unit in Borogermanate with Two Pairs of Interweaving Double Helical Channels. Inorg. Chem., 2004, 43, 6148– 6150. [81] S. Krivovichev, C. Cahill, and P. Burns, A Novel Open Framework Uranyl Molybdate: Synthesis and Structure of (NH4)4[(UO2)5(MoO4)7](H2O)5. Inorg. Chem., 2003, 42, 2459–2464. [82] S. Krivovichev, C. Cahill, E. Nazarchuk, P. Burns, T. Armbruster, and W. Depmeier, Chiral Open-framework Uranyl Molybdates. 1. Topological Diversity: Synthesis and Crystal

262

[83]

[84]

[85]

[86]

[87] [88]

[89]

[90]

[91]

[92] [93] [94]

[95]

[96]

[97] [98] [99]

Chemistry of Zeolites and Related Porous Materials Structure of [(C2H5)2NH2]2[(UO2)4(MoO4)5(H2O)](H2O). Microporous Mesoporous Mater., 2005, 78, 209–215. S. Krivovichev, P. Burns, T. Armbruster, E. Nazarchuk, and W. Depmeier, Chiral Openframework Uranyl Molybdates. 2. Flexibility of the U:Mo¼6:7 Frameworks: Syntheses and Crystal Structures of (UO2)0.82[C8H20N]0.36[(UO2)6(MoO4)7(H2O)2](H2O)n and [C6H14N2][(UO2)6(MoO4)7(H2O)2](H2O)m. Microporous Mesoporous Mater., 2005, 78, 217–224. S. Krivovichev, T. Armbruster, D. Chernyshov, P. Burns, E. Nazarchuk, and W. Depmeier, Chiral Open-framework Uranyl Molybdates. 3. Synthesis, Structure and the C2221 to P212121 Low-temperature Phase Transition of [C6H16N]2[(UO2)6(MoO4)7(H2O)2](H2O)2. Microporous Mesoporous Mater., 2005, 78, 225–234. V. Sohomonian, Q. Chen, R. Haushalter, J. Zubieta, and C. O’Connor, An Inorganic Double Helix: Hydrothermal Synthesis, Structure and Magnetism of Chiral [(CH3)2NH2]K4[V10O10(H2O)2(OH)4(PO4)7] 4H2O. Science, 1993, 259, 1596–1599. P. Feng, X. Bu, S. Tolbert, and G. Stucky, Syntheses and Characterizations of Chiral Tetrahedral Cobalt Phosphates with Zeolite ABW and Related Frameworks. J. Am. Chem. Soc., 1997, 119, 2497–2504. N. Rajic, N. Logar, and V. Kaucic, A Novel Open Framework Zincophosphate: Synthesis and Characterization. Zeolites, 1995, 15, 672–678. S. Ayyappan, X. Bu, A. Cheetham, and C. Rao, Synthesis and Structural Characterization of a Chiral Open-framework Tin(II) Phosphate, [CN3H6][Sn4P3O12](GUAN-SnPO). Chem. Mater., 1998, 10, 3308–3310. Z. Lin, Y. Yao, J. Zhang, and G. Yang, Synthesis and Structure of a Novel Open-framework Zincophosphate with Intersecting Three-dimensional Helical Channels. J. Chem. Soc., Dalton Trans., 2002, 4527–4528. J. Yu, Y. Wang, Z. Shi, and R. Xu, Hydrothermal Synthesis and Characterization of Two New Zinc Phosphates Assembled about a Chiral Metal Complex: [CoII(en)3]2[Zn6P8O32H8] and [CoIII(en)3][Zn8P6O24Cl] 2H2O. Chem. Mater., 2001, 13, 2972–2978. F. Fredoueil, M. Evain, D. Massiot, M. Bujoli-Doeuff, and B. Bujoli, Enantiomerically Pure Zinc Phosphonates Based on Mixed Phosphonic Acid-phosphine Oxide Chiral Building Blocks. J. Mater. Chem., 2001, 11, 1106–1110. S. Qiu, J. Yu, O. Terasaki, Y. Nozve, W. Pang, and R. Xu, Strategies for the Synthesis of Large Zeolite Single Crystals. Microporous Mesoporous Mater., 1998, 21, 245–251. S. Qiu, W. Pang, and R. Xu, Synthesis of Large Crystals of Molecular Sieves – A Review. Stud. Surf. Sci. Catal., 1997, 105, 301–308. M. Morris, A. Dixon, A. Sacco, and R. Thompson, Investigation on the Relative Effectiveness of some Tertiary Alkanolamines in the Synthesis of Large-crystal Zeolite NaA. Zeolites, 1993, 13, 113–121. C. Shao, X. Li, S. Qiu, and F. Xiao, The Role of Pyrocatechol as a Complex Agent for Silicon in the Synthesis of Large Single Crystals of Silica-sodalite Zeolite. Microporous Mesoporous Mater., 1999, 33, 215–222. C. Shao, X. Li, S. Qiu, F. Xiao, and O.Terasaki, Size-controlled Synthesis of Silicalite-1 Single Crystals in the Presence of Benzene-1,2-diol. Microporous Mesoporous Mater., 2000, 39, 117–123. D.N. Bibby and M.P. Dale, Synthesis of Silica-sodalite from Non-aqueous Systems. Nature (London), 1985, 317, 157–158. Q. Huo, S. Feng, and R. Xu, Synthesis of Pentasil Silica-zeolites from Non-aqueous Systems. Stud. Surf. Sci. Catal., 1989, 49A, 291–298. Q. Huo, R. Xu, and S. Feng, First Synthesis of Pentasil-type Silica Zeolites from Nonaqueous Systems. J. Chem. Soc., Chem. Commun., 1988, 1486–1487.

Synthetic Chemistry of Microporous Compounds (II)

263

[100] A. Kuperman, S. Nadimi, S. Oliver, G. Ozin, J. Garce´s, and M. Olken, Non-aqueous Synthesis of Giant Crystal of Zeolites and Molecular Sieves. Nature (London), 1993, 365, 239–242. [101] C. Liu, S. Li, K. Tu, and R. Xu, Synthesis of Cancrinite in a Butane-1,3-diol Systems. J. Chem. Soc., Chem. Commun., 1993, 1645–1646. [102] W. Yan, J. Yu, R. Xu, G. Zhu, F. Xiao, Y. Han, K. Sugiyama, and O. Terasaki, [Al12P13O52]3 [(CH2)6N4H3]3þ: an Anionic Aluminophosphate Molecular Sieve with Bro¨nsted Acidity. Chem. Mater., 2000, 12, 2517–2519. [103] W. Yan, J. Yu, Z. Shi, and R. Xu, A Novel Open-framework Aluminophosphate [AlP2O6(OH)2][H3O] Containing Propeller-like Chiral Motifs. Chem. Commun., 2000, 1431–1432. [104] W. Yan, J. Yu, Z. Shi, P. Miao, K. Wang, Y. Wang, and R. Xu, An Anionic Framework Aluminophosphate [(CH2)6N4H3 H2O][Al11P12O48] and Computer Simulation of the Template Positions. Microporous Mesoporous Mater., 2001, 50, 151–158. [105] J. Yu, J. Chen, and R. Xu, Formation of Single-crystal Cobalt-substituted Gallophosphate LTA from an Alcoholic System. Microporous Mater., 1996, 5, 333–336. [106] A. Chippindale, R. Walton, and C. Turner, Synthesis and Structure of a Novel Openframework Gallium Phosphate [Me2NH(CH2)2NHMe2]2þ [Ga4P5O20H]2 H2O. J. Chem. Soc., Chem. Commun., 1995, 1261–1262. [107] H. Du, Doctoral Thesis, Jilin University, China (1997). [108] L. Na, Y. Liu, W. Pang, and R. Xu, Solvothermal Synthesis and Characterization of a 3-D Iron Phosphate with Exclusively Four-Coordinated Iron. Chin. J. Inorg. Chem., 2000, 16, 287– 292. [109] H. Yuan, J. Chen, G. Zhu, J. Li, J. Yu, G. Yang, and R. Xu, The First Organo-templated Cobalt Phosphate with a Zeolite Topology. Inorg. Chem., 2000, 39, 1476–1479. [110] Y. Guo, Z. Shi, X. Qu and W. Pang, Studies on Solvothermal Synthesis and Structure of a New Titanium Phosphate Large Single Crystal. Chem. J. Chin. Univ., 2000, 21, 1010–1012. [111] S. Qiu, W. Pang, and S. Yao, Growth of Single Crystals of Borozeosilite and Titanozeosilite. Stud. Surf. Sci. Catal., 1989, 49A, 133–142. [112] R. Xu, Z. Gao, and Y. Xu, Progress in Zeolite Science – A China Perspective, World Scientific, Singapore, 1995. [113] D. Zhao, S. Qiu, and W. Pang, Zeolites Synthesized from Non-alkaline Media, ed. R. Von Ballmoos, J.B. Higgins, and M.M.J. Treacy, Proceedings of the 9th International Zeolite Conference (Montreal, Canada), Butterworth-Heinemann, Montreal, 1992, 337–344. [114] L. Beitone, J. Marrot, T. Loiseau, G. Fe´rey, M. Henry, C. Huguenard, A. Gansmuller, and F. Taulelle, MIL-50, an Open-framework GaPO with a Period Pattern of Small Water Ponds and Dry Rubidium Atoms: A Combined XRD, NMR and Computational Study. J. Am. Chem. Soc., 2003, 125, 1912–1922. [115] Y. Sun, T. Song, S. Qiu, W. Pang, J. Shen, D. Jiang, and Y. Yue, Synthesis of Mordenite Single Crystals Using Two Silica Sources. Zeolites, 1995, 15, 745–753. [116] Y. Sun, T. Song, S. Qiu, and W. Pang, Synthesis of Single Crystal of Zeolite Beta Using Two Silica Sources. Chem. Res. Chin. Univ., 1994, 10, 141–143. [117] S. Ueda, N. Kageyama, and M. Koizumi, Crystallization of Zeolite Y from Solution Phase. Proceedings of the 6th International Zeolite Conference, Butterworths, Guildford, 1984, 905–913. [118] W. Pang, S. Ueda, and M. Koizumi, The Synthesis of Zeolite NaA from Homogeneous Solutions and Studies of Its Properties. Proceedings of the 7th International Zeolite Conference, Kodansha-Elsevier, Tokyo, 1986, 177–184. [119] W. Pang, S. Qiu, and Q. Kan, Synthesis and Characterization of FAPO-5 Crystallized from Clear Homogeneous Solutions. Stud. Surf. Sci. Catal., 1989, 49A, 281–289.

264

Chemistry of Zeolites and Related Porous Materials

[120] S. Shimizu and H. Hamada, Synthesis of Giant Zeolite Crystals by a Bulk Material Dissolution Technique. Microporous Mesoporous Mater., 2001, 48, 39–46. [121] M. Camblor, A. Corma, and S. Valencia, Characterization of Nanocrystalline Zeolite Beta. Microporous Mesoporous Mater., 1998, 25, 59–74. [122] J. Dong, J. Zou, and Y. Long, Synthesis and Characterization of Colloidal TBA – Silicalite-2. Microporous Mesoporous Mater., 2003, 57, 9–19. [123] R. Van Grieken, J. Sotelo, J. Mene´ndez, and J. Melero, Anomalous Crystallization Mechanism in the Synthesis of Nanocrystalline ZSM-5. Microporous Mesoporous Mater., 2000, 39, 135–147. [124] S. Naik, J. Chen, and A.S.T. Chiang, Synthesis of Silicalite Nanocrystals Via the Steaming of Surfactant Protected Precursors. Microporous Mesoporous Mater., 2002, 54, 293–303. [125] A. Persson, B. Schoeman, J. Sterte, and J. Otterstedt, Synthesis of Stable Suspensions of Discrete Colloidal Zeolite (Na, TPA) ZSM-5 Crystals. Zeolites, 1995, 15, 611–619. [126] B.A. Holmberg, H. Wang, J.M. Norbeck, and Y. Yan, Controlling Size and Yield of Zeolite Y Nanocrystals Using Tetramethylammonium Bromide. Microporous Mesoporous Mater., 2003, 59, 13–28. [127] I. Schmidt, C. Madsen, and C. Jacobsen, Confined Space Synthesis. A Novel Route to Nanosized Zeolites. Inorg. Chem., 2000, 39, 2279–2283. [128] H. Wang, B.A. Holmberg, and Y. Yan, Synthesis of Template-free Zeolite Nanocrystals by Using In Situ Thermoreversible Polymer Hydrogels. J. Am. Chem. Soc., 2003, 125, 9928– 9929. [129] Y. Wang, The Preparation of b-type Nanozeolite Catalytic Material, Book of Abstracts. 11th China Catalytic Conference, Zhejiang University Publishing Co., Hangzhou, 2002, 979–981. [130] X. Xu, W. Yang, J. Liu, and L. Lin, Synthesis of NaA Zeolite Membranes from Clear Solution. Microporous Mesoporous Mater., 2001, 43, 299–311. [131] M. Noack, P. Ko¨lsch, R. Scha¨fer, P. Toussaint, I. Sieber and J. Caro, Preparation of MFI Membranes of Enlarged Area with High Reproducibility. Microporous Mesoporous Mater., 2001, 49, 25–37. [132] R.K. Iler, The Chemistry of Silica, John Wiley & Sons, Inc., New York, 1979. [133] V. Valtchev and S. Mintova, Layer-by-layer Preparation of Zeolite Coatings of Nanosized Crystals. Microporous Mesoporous Mater., 2001, 43, 41–49. [134] Z. Wang and Y. Yan, Controlling Crystal Orientation in Zeolite MFI Thin Films by Direct In Situ Crystallization. Chem. Mater., 2001, 13, 1101–1107. [135] Z. Wang and Y. Yan, Oriented Zeolite MFI Monolayer Films on Metal Substrates by In Situ Crystalization. Microporous Mesoporous Mater., 2001, 48, 229–238. [136] S. Lai, T. Louisa, and K. Yeung, Influence of the Synthesis Conditions and Growth Environment on MFI Zeolite Film Orientation. Microporous Mesoporous Mater., 2002, 54, 63–77. [137] J. Choi, S. Ghosh, Z.P. Lai, and M. Tsapatsis, Uniformly a-Oriented MFI Zeolite Films by Secondary Growth. Angew. Chem., Int. Ed., 2006, 45, 1154–1158. [138] Z.J. Li, S.A. Li, H.M. Luo, and Y.S. Yan, Effects of Crystallinity in Spin-on Pure-silicazeolite MFI Low-dielectric-constant Films. Adv. Funct. Mater., 2004, 14, 1019–1024. [139] Z.J. Li, C.M. Lew, S.A. Li, D.I. Medina, and Y.S. Yan, Pure-silica-zeolite MEL Low-k Films from Nanoparticle Suspensions. J. Phys. Chem. B, 2005, 109, 8652–8658. [140] L.M. Huang, Z. Wang, J. Sun, L. Miao, Q. Li, Y. Yan, and D. Zhao, Fabrication of Ordered Porous Structures by Self-assembly of Zeolite Nanocrystals. J. Am. Chem. Soc., 2000, 122, 3530–3531. [141] L. Huang, Z. Wang, H. Wang, J. Sun, Q. Li, D. Zhao, and Y. Yan, Hierarchical Zeolite Structures with Designed Shape by Gel-casting of Colloidal Nanocrystal Suspensions. Chem. Commun., 2001, 1364–1365.

Synthetic Chemistry of Microporous Compounds (II)

265

[142] L Huang, Z. Wang, H. Wang, J. Sun, Q. Li, D. Zhao, and Y. Yan, Hierarchical Porous Structures by Using Zeolite Nanocrystals as Building Blocks. Microporous Mesoporous Mater., 2001, 48, 73–78. [143] H.T.Wang, Z.B.Wang, L.M. Huang, A. Mitra, and Y.S. Yan, Surface Patterned Porous Films by Convection-assisted Dynamic Self-assembly of Zeolite Nanoparticles. Langmuir, 2001, 17, 2572–2574. [144] S. Davis, M. Breulmann, K. Rhodes, B. Zhang, and S. Mann, Template-directed Assembly Using Nanoparticle Building Blocks: A Nanotectonic Approach to Organized Materials. Chem. Mater., 2001, 13, 3218–3226. [145] X. Wang, W. Yang, Y. Tang, Y. Wang, S. Fu, and Z. Gao, Fabrication of Hollow Zeolite Spheres. Chem. Commun., 2000, 2161–2162. [146] A. Dong, Y.Wang, and Y. Tang, Bionic Synthesis of Hierarchical Channel Zeolite Materials. 11th China Catalytic Conference, Zhejiang University Publishing Co., Hangzhou, 2002, 991–992. [147] L. Tosheva, V. Valtchev, and J. Sterte, Silicalite-1 Containing Microspheres Prepared Using Shape-directing Macro-templates. Microporous Mesoporous Mater., 2000, 35–36, 621– 629. [148] J. Lin, J. Dipre, and M. Yates, Microemulsion-directed Synthesis of Molecular Sieve Fibers. Chem. Mater., 2003, 15, 2764–2773. [149] Semiconductors, International Technology Roadmap for Semiconductors–Interconnect, 2005, 1–63. (See, for example, www.itrs.net/Links/2005ITRS /ExecSum2005.pdf). [150] Z. Li, M. C. Johnson, M. Sun, E. T. Ryan, D. J. Earl, W. Maichen, J. I. Martin, S. Li, C. M. Lew, J. Wang, M.W. Deem, M. E. Davis, and Y. Yan, Mechanical and Dielectric Properties of Pure-silica-zeolite Low-k Materials. Angew. Chem., Int. Ed., 2006, 45, 6329–6332. [151] Z.B. Wang, A.P. Mitra, H.T. Wang, L.M. Huang, and Y.S. Yan, Pure Silica Zeolite Films as Low-k Dielectrics by Spin-on of Nanoparticle Suspensions. Adv. Mater., 2001, 13, 1463– 1466. [152] Z.B. Wang, H.T. Wang, A. Mitra, L.M. Huang, and Y.S. Yan, Pure-silica Zeolite Low-k Dielectric Thin Films. Adv. Mater., 2001, 13, 746–749. [153] S.A. Li, Z.J. Li, and Y.S. Yan, Ultra-low-k Pure-silica Zeolite MFI Films using Cyclodextrin as Porogen. Adv. Mater., 2003, 15, 1528–1531. [154] G. Xu, J. He, E. Andideh, J. Bielefeld, and T. Scherban, Cohesive Strength Characterization of Brittle Low-k Films. IEEE International Interconnect Technology Conference, IEEE Standards Office, Pircataway, 2002, 57–59. [155] X.L. Cheng, Z.B. Wang, and Y.S. Yan, Corrosion resistant Zeolite Coatings by In-situ Crystallization. Electrochem. Solid State Lett., 2001, 4, B23–B26. [156] D.E. Beving, A.M.P. McDonnell, W.S. Yang, and Y.S. Yan, Corrosion Resistant High-silicazeolite MFI Coating: One General Solution Formulation for Aluminum Alloy AA-2024-T3, AA-5052-H32, AA-6061-T4, and AA-7075-T6. J. Electrochem. Soc., 2006, 153, B325– B329. [157] A. Mitra, C. W. Kirby, Z. Wang, L. Huang, H. Wang, Y. Huang, and Y. Yan, Synthesis of Pure-silica MTW Powder and Supported Films. Microporous Mesoporous Mater., 2002, 54, 175–186. [158] A. Mitra, Z. Wang, T. Cao, H. Wang, L. Huang, and Y. Yan, Synthesis and Corrosion Resistance of High-silica Zeolite MTW, BEA, and MFI Coatings on Steel and Aluminum. J. Electrochem. Soc., 2002, 149, B472–B478. [159] A.M.P. McDonnell, D. Beving, A.J. Wang, W. Chen, and Y.S. Yan, Hydrophilic and Antimicrobial Zeolite Coatings for Gravity-independent Water Separation. Adv. Funct. Mater., 2005, 15, 336–340. [160] C. O’Neill, D.E. Beving, W. Chen, and Y.S. Yan, Durability of Hydrophilic and Antimicrobial Zeolite Coatings under Water Immersion. AIChE J., 2006, 52, 1157–1161.

266

Chemistry of Zeolites and Related Porous Materials

[161] T.C. Bowen, R.D. Noble, and J.L. Falconer, Fundamentals and Applications of Pervaporation through Zeolite Membranes. J. Membr. Sci., 2004, 245, 1–33. [162] T.C. Bowen, H. Kalipcilar, J.L. Falconer, and R.D. Noble, Pervaporation of Organic/Water Mixtures through B-ZSM-5 Zeolite Membranes on Monolith Supports. J. Membr. Sci., 2003, 215, 235–247. [163] H. Kalipcilar, J.L. Falconer, and R.D. Noble, Preparation of B-ZSM-5 Membranes on a Monolith Support. J. Membr. Sci., 2001, 194, 141–144. [164] Z.P. Lai, G. Bonilla, I. Diaz, G. Nery, K. Sujaoti, M.A. Amat, E. Kokkoli, O. Terasaki, R.W. Thompson, D.G. Vlachos, and M. Tsapatsis, Microstructural Optimization of a Zeolite Membrane for Organic Vapor Separation. Science, 2003, 300, 456–460. [165] W.H. Yuan, Y.S. Lin, and W.S. Yang, Molecular Sieving MFI-type Zeolite Membranes for Pervaporation Separation of Xylene Isomers. J. Am. Chem. Soc., 2004, 126, 4776–4777.

5 Crystallization of Microporous Compounds Chapters 3 and 4 mainly focus on the synthetic chemistry of microporous compounds. In Chapter 3, the synthetic regularities, procedures, and principles of two main classes of microporous compounds, i.e., zeolites and aluminophosphates, are extensively discussed. Over the past decade, microporous compounds with special types, structures, and states of aggregation such as single crystal, perfect crystal, nano- and ultra-fine crystal, molecular sieve membrane, and crystal with special morphology have been prepared and widely used. The synthetic chemistry for these materials is discussed in detail in Chapter 4. Because hydrothermal and solvothermal crystallization synthesis approaches are the main means in the preparation of numerous microporous compounds, it is necessary to further discuss their crystallization mechanism and related issues in an indepth way. This will help readers to fully understand the synthetic chemistry of microporous materials and to develop new synthetic routes and techniques. In this chapter, the following content will be discussed: the polymerized state and the polymerization regularities of the main source materials such as silicate, aluminate, and phosphate in the liquid phase prior to crystallization; the structure of the liquid phase and the gel prior to nucleation; the templating or structure-directing effect (SDE) in the nucleation and crystallization process; regularities of crystal growth; the phase transition of the metastable phase, and so forth. The synthesis and crystallization mechanism of the zeolites will be the focus of discussion because zeolitic materials were the first to be discovered and have been well studied. Owing to the complexity of the pre-described problems and the lack of powerful techniques on in situ characterization of these processes, the crystallization mechanism of microporous compounds has only been partially understood and arguments on the explanation of some results do exist. These issues will be introduced to readers as they are, without any judgement.

Chemistry of Zeolites and Related Porous Material – Synthesis and Structure Ruren Xu, Wenqin Pang, Jihong Yu, Qisheng Huo and Jiesheng Chen # 2007 John Wiley & Sons, (Asia) Pte Ltd

268

Chemistry of Zeolites and Related Porous Materials

5.1 Starting Materials of Zeolite Crystallization 5.1.1

Structures and Preparation Methods for Commonly Used Silicon Sources

Many zeolites were crystallized from strongly basic solution containing silicate and aluminate. In strongly basic solution, silicate exists as polymerized-state polysilicates with a variable distribution of molecular weight. Aluminate reacted with polysilicate to form various zeolite structures via condensation reaction under hydrothermal or solvothermal conditions. Therefore, in the study of the synthetic chemistry of zeolites, one of the key issues is to understand the existing state and reactivity of the polymerized silicate. Studying the polymerization state of the silicate in the solution during the formation of zeolites is a very important and complex issue. After years of study, it has been found that some correlations exist between the polymerization state of the silicate in the solution and the structure type of zeolites, while the polymerization state and its distribution could be affected by many factors such as alkalinity, nature of the cation, organic templates, temperature, and pressure. These parameters could partially determine the condensation rate between polymerized silicate and aluminate, which will finally affect the formation of zeolite structure. The following sections describe, as an example, the Na2O–Al2O3–SiO2–H2O fourcomponent crystallization field diagram. Commonly used silicon sources for the synthesis of zeolites include silica sol, silica gel, amorphous SiO2 powder, soluble silicate solution containing alkali metal (e.g., sodium and potassium, etc.) and organic amines, Si(OCH3)4 (TMOS), and Si(OC2H5)4 (TEOS), etc. A feature of zeolite formation is that the formation of metastable zeolite structure strongly depends on the type (property) and structure of the silicon source. This is illustrated by the crystallization field diagram shown in Figure 5.1, which clearly shows that different zeolite structures could be crystallized under identical crystallization parameters from the parent gel with the same composition but with different silicon sources. For example, the crystallization conditions for the system shown in (a) and (c) of Figure 5.1 are identical except for the silicon source. Silica sol was used in (a), while sodium silicate (water glass) was used in (c). The crystallized product in (a) was zeolite Y, whereas zeolite X and A were crystallized from (c). When the crystallization temperature was increased to 100
C [phase diagrams (b) and (d)], the same phenomenon happened. According to these results, it is necessary to discuss the structure and preparation methods for the silicon sources used in the synthesis of zeolites, which include silica sol, silica gel, amorphous SiO2 powder, and polysilicates in basic solution. Structure of Soluble Silicate in Basic Medium Most syntheses of zeolites were conducted in strongly basic media. The silicate solution containing alkali metal cations such as Naþ or Kþ is one of the main silicon sources in the synthesis of zeolites. The concentration of the basic species in the starting material for the synthesis of zeolites is usually about 0:5  5:5 mol/L. Within this concentration range, aluminate and silicate ions with various polymerization degrees existed in the system. The state and distribution of these ions could be affected by many factors. When the temperature and the concentration of SiO2 were kept constant, the acid–base

Crystallization of Microporous Compounds

269

Figure 5.1 Correlation between crystallization field of zeolites and various silicon sources. The silicon source for (a) and (b) is silica sol: (a) crystallization temperature: 25
C, zeolite Y (FAU); (b) crystallization temperature: 100
C, zeolite Y (FAU), S (SOD), and R (CHA). The silicon source for (c) and (d) is sodium silicate (water glass): (c) crystallization temperature: 25
C, zeolite X (FAU) and A (LTA); (d) crystallization temperature: 100
C, zeolite P (GIS), X (FAU), and A (LTA). Reproduced with permission from [1]. Copyright (1974) John Wiley & Sons, Inc.

equilibrium and condensation between the silanols could be expressed as Equations (5.1) and (5.2):    OH $  O þ Hþ   Si   Si

ð5:1Þ

     OH þ HO Si  O Si    Si  $   Si   þ H2 O

ð5:2Þ

According to Equations (5.1) and (5.2), it could be inferred that the state and structure of the polysilicate anions are mainly determined by the pH of the solution and the concentration of SiO2 and partially by the types of cations, which will be discussed below. 1. The state and structure of silicate ions in sodium salt solution Sodium silicate solution (e.g., water glass) is one of the most commonly used chemicals in the synthesis of zeolites, which will supply the silicon species as well as Naþ and OH. Therefore, it is important to study the polymerization state and distribution of the polysilicate ions in sodium silicate solution under hydrothermal conditions. Studies on sodium silicate solution were started in the 1950s. However, owing to the limitations of techniques at that time, the only information obtained was the weight- and number-average molecular weight with various moduli and SiO2 concentrations. Over the past years, this system has been restudied by using improved techniques such as TMS-GLC, TMS-GPC, and 29Si-NMR.

270

Chemistry of Zeolites and Related Porous Materials

Figure 5.2 Typical 29Si-NMR spectrum of sodium silicate solution with 3 mol% [SiO2] and R ¼ 1.5. Reproduced with permission from [2]. Copyright (1986) Elsevier

In the early 1980s, 29Si-NMR was used to determine the distribution of polysilicate anions in sodium silicate solution for the first time. For example, McCormick et al. studied in detail the state and distribution of polysilicate anions in sodium silicate SiO2 solution with [SiO2] ¼ 1  3 mol% and R ¼ Na ¼ 1  3:[2] A typical 29Si-NMR spec2O trum is shown in Figure 5.2. Based on analyses on the 29Si-NMR spectra of
the sodium silicate solution with  SiO2 various concentrations of SiO2 ([SiO2]) and moduli R Na2 O , McCormick and colleagues obtained detailed information on the state and distribution of the polysilicate anions under the following conditions, summarized in Table 5.1. The results in Table 5.1 indicated that silicate ions with a low degree of polymerization existed in sodium silicate solution with the possible structures shown in the second column of Table 5.1. Besides there low-polymerization silicate ions, highly polymerized silicate ions with more complex structures and higher molecular weights (10 00050 000) existed as well, and could be measured by the TMS-GPC method. Because there are still some technical problems in using GPC for the separation of silicate ions, it is very difficult to determine the distribution of highly polymerized silicate ions. 2. State of silicate ions in potassium silicate solution In 1981, Harris et al. systematically studied the state and distribution of polysilicate anions in potassium silicate solution through analyses of their 29Si-NMR spectra.[3] They found that the state and distribution of polysilicate anions in potassium silicate solution are similar to those in sodium silicate solutions. Figure 5.3 shows the state and distribution abundance (the number in parentheses indicates the relative abundance of the polymerized silicate with the assumption of 100 monomers.) of 18 types of polysilicate anions existing in potassium silicate solution with a concentration of

Crystallization of Microporous Compounds

271

Table 5.1 State and distribution of the polysilicate anions in sodium silicate solutions with different moduli (R) No.

Structure of polysilicate ions

R ¼ 1:0

R ¼ 1:5

R ¼ 2:0

18.0

6.0

5.0

6.0

5.0

2

5.2

1.5

1.2

0.9

1.1

3

2.4

1.0

1.2

0.7

0.7

4

2.8

1.6

0.6

0.3

5

1.0

1.1

0.5

0

0

6

0

0.3

1.2

0.6

n.o.a

7

2.0

1.0

0.9

0.2

8

2.0

2.1

3.3

1.8

9

0.5

0.5

0.5

10

1.0

1.7

1

A

0

R ¼ 2:5

R ¼ 3:0

0

0 1.7

0

0

0

0

11

0

0.8

1.8

0.8

0.6

12

0

0.4

0.3

0

0

13

0

0.4

0.3

0

0

0

0

14

0.3

0.5

0.7

15

0.4

0.6

1.4

16

0.4

17

0

0

0.1

n.o.

n.o.

18

0.2

0.3

0.2

n.o.

n.o.

19

0

0.1

0.2

0.2

0.3

a

n.o. means not observable.

0

0

0.3 0

0.3 0

272

Chemistry of Zeolites and Related Porous Materials

Figure 5.3 The state and distribution abundance of the polysilicate anions that exist in potassium silicate solution (0.63 mol/L, K: Si ¼ 1.5: 1.0)

0.63 mol/L and K: Si ¼ 1.5: 1.0. These data were obtained through analyses of their 29 Si-NMR spectra. Their results indicated the presence of some highly polymerized silicate ions as well. According to the pre-described studies, it could be concluded that the state and distribution of polysilicate anions in both Naþ and Kþ basic solutions are dependent on both [SiO2] concentration and the alkalinity of the solution. Increased pH value and decreased [SiO2] concentration result in an increased concentration of silicate monomer. In addition, Harris’s studies indicated that the polymerization state of the polysilicate anions in the potassium silicate solution is similar to that in the sodium silicate solution with the same silica concentration. The difference lies in their relative distribution abundance. 3. The structure of silicate ions in solutions containing organic bases High-silica zeolites could be crystallized from the parent gel containing organic bases. The questions arising from these results are why high-silica zeolites could be synthesized in the presence of organic bases and how they affect the state of the polysilicate anions. Recently, the tetraalkylammonium silicate aqueous solutions including tetramethylammonium silicate (TMAS), tetraethylammonium silicate (TEAS), tetrapropylammonium silicate (TPAS), and tetrabutylammonium silicate (TBAS) were extensively studied and some interesting results were obtained.

Crystallization of Microporous Compounds

273

Figure 5.4 Distribution of polysilicate anions in saturated TMAxS solution, where x ¼ 0:6  20. Reproduced from [5] with permission of Wiley-VCH

(1) The studies on tetramethylammonium silicate (TMAS) aqueous solution. Hoebbel et al. studied the state of the polysilicate ions in TMAS aqueous solution with 29 TMA Si-NMR and TMS-GC techniques and found that there SiO2 ¼ 0:6  20 by using existed a large quantity of octa-polysilicate anions ([Si8O20]8) in form of a double 48 membered ring.[4,5] When the value of TMA SiO2 is about 1  3, the concentration of [Si8O20] is as high as 60  80% (see Figure 5.4 and Table 5.1 for detailed information), which is totally different from that in the aqueous silicate solutions containing Naþ and Kþ. When the concentration of SiO2 is decreased, the percentage of monosilicate ions is gradually increased, while the concentration of octa-polysilicate anions is gradually decreased. The results are summarized in Table 5.2 and illustrated in Figure 5.4. These results indicated that the polymerization state of the polysilicate anions depends on the concentration of SiO2 as well as on that of TMAþ of the solution. Figure 5.5 shows the correlation between the state of the polysilicate anions and the ratio of TMAþ: Si. When the ratio of TMA: Si is about 1  3, most of the polysilicate anions exist as

Table 5.2 Polymerization state of silicate ions in TMAS solution (data were obtained through TMS and NMR techniquesa) Si8 O20 8

SiO44 TMAS solution (mol/L) 0.1 0.34 0.75 1.0 1.25 2.47 a

TMA20Si TMA4Si TMA1Si TMA1Si TMA1Si TMA.08Si

NMR

TMS

NMR

TMS

50 27 7 5 2 <1

51 29 10 9 5 1

45 66 19 42 50 18

41 53 19 36 48 26

The results are expressed as the weight percentage of SiO2.

274

Chemistry of Zeolites and Related Porous Materials

Figure 5.5 Distribution of the state of polysilicate anions in TMAS solution with various TMA/ SiO2 ratios. Note: Mono means the monomer of the silicate ions; Di means the dimer of the polysilicate anions; D4R means the double 4-membered ring octa-polysilicate. Reproduced from [5] with permission of Wiley-VCH

octa-polymeric silicate in the form of a double 4-membered ring. When the ratio of TMA: Si was decreased to about 0.6, the concentration of octa-polysilicate anions was significantly decreased and the concentration of highly polymerized silicate ions was significantly increased as shown in Figure 5.5. (2) The studies on tetraethylammonium silicate (TEAS) aqueous solution. Hoebbel et al. systematically studied the aqueous solution of tetraethylammonium silicate (TEAS).[6] When the ratio of TEA/SiO2 is about 2:8  1, the main form of the

Crystallization of Microporous Compounds

275

Table 5.3 Distribution of polysilicate anions in various TEAS solutions TEAS solution

Double 3-membered ring

Double 4-membered ring

1.37 1.68 1.98 2.45 3.07 3.95

77 76 84 67 44 20

5 5 3 8 20 26

TEA2.8S TEA1.98S TEA1.55S TEA1.0S TEA0.81S TEA0.62S

Double 5-membered ring

5 11 20

Double 6-membered ring

6 19

Others 18 19 13 20 19 15

polysilicate anions in the solution is hexa-polymeric [Si6O15]6 in the form of a double 3membered ring, besides small amounts of silicates Si1, Si2, Si3, Si4, and Si8. When the TEA/SiO2 ratio is about 0:6  0:8, there are considerable quantities of polysilicate anions in the form of double 3-, 4-, and 5-membered rings, which has been confirmed by experimental results as shown in Table 5.3 obtained by TMS. In addition, Harris and Knight measured the polymeric state of the silicate ions in TEAS solutions by using 29Si-NMR technique and found that the NMR signal is not sensitive for dilute solutions.[7] The correlation between the NMR signals and the concentration of SiO2 can only be obtained from the spectra recorded at high concentration. (3) The studies on tetrabutylammonium silicate (TBAS) aqueous solution. Hoebbel et al. carefully studied the state of polysilicate anions in TBAS solution.[8] Their results indicated that there mainly existed monomeric, dimeric, trimeric, tetrameric, pentameric, and heptameric silicate ions and polysilicate anions in the form of double 3-, 4-, and 5-membered rings in TBAS solution. Compared with the TMAS and TEAS solutions that mainly contain double 3- and 4-membered-ring-type silicate ions under some conditions, the distribution of these species in TBAS solution is quite even. However, the distribution of these species in solid crystalline structures is quite different to that in solution. It was found that there are a large number of double 5-membered rings in the solid crystalline structure obtained from the corresponding solution. Studies indicated that there are some similarities between the properties of TPAS and TBAS solutions. Owing to space limitation, studies on TPAS solution will not addressed here. The above discussion on TMAS, TEAS, and TBAS systems shows that the polymerization state of polysilicate anions in concentrated and dilute tetraalkylammonium silicate (TAAS) solution is significantly different. In general, concentrated solution favors the formation of double ring- and cage-like silicate ions, while the dilute solution favors the formation of noncage-like silicate ions with a relatively low polymeric degree and similar distribution (Table 5.4). For example, Si8 O20 8 in the form of a double 4-membered ring is the main type of species in saturated TMAS solution; polysilicate anions in the form of double 3- (Si6O168), 4-, and 5-membered ring are the main species in saturated TEAS solution; double 5-membered ring-like silicate ions that could only be found in some crystalline structures mainly exist in saturated TBAS solution. Studies on the state and distribution of polysilicate anions under the presence of TAAþ have been conducted over the past decades and new results and concepts continue to

276

Chemistry of Zeolites and Related Porous Materials

Table 5.4 State and distribution of polymerized ions in dilute TMA1Si, TEA1Si, TBA1Si, and Na1Si (0.1 mol/L) solutions Silicate ions

0.1 mol/L TMA1Si

0.1 mol/L TEA1Si

0.1 mol/L TBA1Si

0.1 mol/L Na1Si

SiO44 Si2O76 Si3O108 Si4O1310 Si3O96 Si4O128 Ring-like Highly polymerized

58 14 4 1 3 2 2 16

56 14 7 3 3 4 2 11

59 16 6 1 4 2 2 10

18 14 7 2 3 3 6 17

emerge. For example, Kinrade and colleagues systematically studied tetraalkylammonium silicate solutions by using the 29Si-NMR technique in 1998 and found that the polymerization state of polysilicate anions under the presence of TMAþ strongly depends on the specific conditions (e.g., equilibria or nonequilibria) of the system, which is different from previously reported results.[9] They found that the equilibrium ratio of TMAþ to Si8 O20 8 is 8 : 1 for solutions with a concentration ratio [OH ] : ½Si  1 : 1 and that TMAþ directly associate with Si8 O208 to form a shell of hydrophobic hydration that impedes the hydrolysis of the central polysilicate anions. Kinrade et al. further studied the polymerization kinetics of the polysilicate anions of Si8 O208 and related Kf and showed unequivocally that TMAþ participate directly in the formation and subsequent stabilization of cage-like polysilicate anions such as Si8 O208 .[10] This new information suggests a radically different mechanistic role than ‘templating’ for TMAþ in the synthesis of molecular sieves. In addition, the polymerization state of polysilicate anions in TAAþ system could be affected by the temperature as well. Figure 5.6 shows the decomposition curve of highly polymerized silicate ions and that of low-polymerization

Figure 5.6 Variation of the composition of TMAS solutions with temperature. Reproduced from Stud. Surf. Sci. Catal, 137. Copyright (2001) Elsevier

Crystallization of Microporous Compounds

277

silicate ions with respect to temperature, which have been further confirmed by other experimental results. The Structure and Preparation Methods for Silica Sol 1. Polymerization of silicate The most important property of silicate species is their polymerization state. For a solution containing silicate, low SiO2 concentration can result in the formation of silica sol, while high SiO2 concentration can lead to the formation of silica gel. Many factors can affect the polymerization process of silicate anions. The most important one is the acidity of the solution. The polymerization state of polysilicate anions and the polymerization process of SiO2 strongly depend on the pH value of the solution. Iler systematically studied the polymerization process and the corresponding mechanism of silicate anions in acidic, neutral, slightly basic, and strongly basic solutions.[11] 2. The structure of silica sol[11] The most important property of silicate solution is its self-polymerization, i.e., monomeric silicate ! low-polymerization silicate ! highly polymerized silicate. Silicate solution without the presence of salts can form a silica sol within a pH range of 7  10, while silicate solution with the presence of salts can form a silica gel in the same pH range (e.g., 7  10) as shown in Figure 5.7. Silica sol contains a large amount of well dispersed hydrated SiO2 particles. Owing to its large interface area, silica sol possesses huge free energy. Therefore, silica sol is a thermodynamically unstable system. The colloidal particles can automatically aggregate together with the formation of huge particles. Figure 5.8 schematically illustrates the structure and distatic-charge-layer of colloidal particles of silica sol. The center of the colloidal particle is a colloid nucleus, which consists of thousands of SiO2 molecules. Colloid nucleus is water insoluble (sometimes it

Figure 5.7 Illustration of polymerization process of silicate. Reprinted from Ralph K. Iler: The Colloid Chemistry of Silica and Silicates. Copyright 1995 by Cornell University (Copyright renewed 1983). Used by permission of the publisher, Cornell University Press

278

Chemistry of Zeolites and Related Porous Materials

Figure 5.8 Schematical illustration of the structure and distatic-charge-layer of colloidal particle of alkali silica sol. d: compact layer;
: diffuse layer; jo: total potential;
: electrokinetic potential; - - -: surface Si–O groups; þ: cations such as Naþ;: anions such as OH; m, x, y, z: integral number. Reprinted from Ralph K. Iler: The Colloid Chemistry of Silica and Silicates. Copyright 1995 by Cornell University (Copyright renewed 1983). Used by permission of the publisher, Cornell University Press

may adsorb some water molecules). However, it could selectively adsorb ions from the surrounding aqueous solution to form a compact layer and the counter ions formed a diffuse layer outside the compact layer. Because of the electrostatic attraction between the colloid nucleus and the counter ions layer, movement of the colloid nucleus leads to movement of the compact layer as well as of water molecules. The charged particle which consists of the colloid nucleus and the compact layer is called a colloid. The colloid and the diffuse layer are called a colloidal particle, which is neutral. The electrokinetic potential (
) of the silica sol is defined as the potential difference between the colloidal particle and solution. There are two types of silica sol: basic and acidic. The electrokinetic potential of basic silica sol is negative, while that of acidic silica sol is positive. Because the surface of each individual colloidal particle has the same electric charge, the electrostatic repulsive force of the colloidal particles makes the system stable.

Crystallization of Microporous Compounds

279

A higher electrokinetic potential leads to a higher stability of silica sol. In general, the stability of acidic silica sol is better than that of basic silica sol. 3. The preparation method for silica sol As early as 1915, Schiwerin and coworkers made commercial silica sol by using an electrodialysis method. The production of silica sol has progressed in the past 30 years. It was recently reported that the concentration of SiO2 in silica sol solution has reached as high as 60%.[11] The concentration of [SiO2] in the commercial products Ludox-AS-40 and -HS-40 used in the synthesis of zeolites is about 40%. Many productive methods have been developed for the preparation of silica sol including acidification,[12] electrolysation–electrodialysis,[13] ion-exchange,[14] peptization,[11] and hydrolysis of silicon compounds,[10] which can be grouped into two main types. One is called the aggregation method that contains two steps: the polymerization of silicate ions and the aggregation of these polysilicate anions via condensation reaction between the hydroxy groups of the particles. The other one is called the peptization method, i.e., dispersal of a precipitate of SiO2 to form colloid. The acidification method will be discussed in detail below. Acidification method: Silica sol can be quickly obtained by mixing small amounts of acid and soluble sodium silicate with control of the pH of the solution at 8  9 and the concentration of Naþ < 0:3 mol/L. This procedure has been widely used in industry to prepare silica sol with a SiO2 concentration of 30%. Alexander and Iler first prepared SiO2 sol by using this method.[12] They removed a large quantity of Naþ from sodium silicate solution by using ion-exchange resin and finally obtained a solution with an SiO2/Na2O ratio of 85. This solution was kept at 100
C for 10 min. The resulting solution was mixed with dilute sulfuric acid and the pH of the solution was controlled at about 9. Subsequently, the solution was further vigorously stirred for 8 h. The addition of acid enabled the continuous precipitation of polysilicate anions on the surface of the colloid nucleus, resulting in the growth of the colloidal particle. The final size of the SiO2 colloidal particle was about 37 nm. Several key points should be noted for the above described procedure: 1) the mixing of acid and silicate should be quickly performed, and vigorous stirring should be applied to avoid local aggregation; 2) the pH of the solution should be kept at about 8  9 after the addition of acid; and 3) the concentration of Naþ should be low. In addition, the reaction between sodium silicate and acid was used to prepare SiO2 sol as well. With this method, the acid was used to neutralize the sodium silicate solution containing organic species such as ethanol. Later, it was found that the organic species could be diethylene glycol, oxalic acid, glyoxal, and acetone. Naþ can be removed by many methods, such as ion-exchange. Structure and Preparation of Silica Gel and Amorphous SiO2 1. Silica gel Silica sol, silica gel, and amorphous SiO2 are spherical SiO2 colloidal particles with different aggregation states. As shown in Figure 5.9, silica sol is a dispersed colloid, while silica gel is a continuous solid containing a cross-linked network of colloidal particles in three-dimensional space. The amorphous SiO2 is basically fragments of silica gel.

280

Chemistry of Zeolites and Related Porous Materials

Figure 5.9 Schematic diagram of the structure of (a) silica sol, (b) silica gel, and (c) amorphous SiO2. Reprinted from Ralph K. Iler: The Colloid Chemistry of Silica and Silicates. Copyright 1995 by Cornell University (Copyright renewed 1983). Used by permission of the publisher, Cornell University Press

Silica gel is jelly-like species formed from silicate solution under certain conditions, with properties between those of a solid and those of a liquid. It consists of a cross-linked network of SiO2 colloidal particles in three-dimensional space. If it is an aqueous gel, the pores inside the gel will be filled with water or dilute solution. 2. Gelation of silica sol Silica gel could be prepared via the gelation of silica sol. The process for the formation of water-containing uniform gel from spherical silical colloidal particles is very fast. It is known that there is adhesive force on the surface of spherical silica collodial particles, which could lead to the aggregation of these particles. This process could be described as below. First, chemical bonds form between the SiO2 colloidal particles.  Si OH groups exist on the surface of the spherical particles of silica sol. The collision of these colloidal particles together with the catalysis of OH results in the condensation reaction between the -OH groups on the surface of two different colloidal particles to form a  Si O Si  bond resulting in a joint between two SiO2 colloidal particles. The surface near this joint has an infinitesimal negative camber radius, resulting in a near-to-zero solubility of this area. More and more individual SiO2 colloidal particles precipitate to the area near this joint, leading to the rapid growth of SiO2 colloidal particles. Growth near this area will be terminated when the low solubility dominance of this area disappears. Figure 5.10 schematically shows this process. Owing to different types of connections of the colloidal particles via  Si O Si  bonds, the shape of the aggregated big SiO2 particles could have a chain, rod, or fiber shape. These big SiO2 colloidal particles could further cross-link to form a network structure in three-dimensional space containing a large amount of solution. It is commonly accepted that the SiO2 gel is built from the aggregation of primary SiO2 colloidal particles. Small Angle X-ray Scattering (SAXS) studies on the as-prepared SiO2 gel indicated that it contained spherical primary SiO2 colloidal particles with a size of 3  6 nm. The difference in the size and composition of the primary SiO2 colloidal particles results in the structural complexity of the silica gel. The main information obtained from the structural studies on the SiO2 gel is the size and shape of the primary SiO2 colloidal particles, which depend on the coordination number and arrangement regularity of the primary SiO2 colloidal particles. Therefore, the porosity, pore volume,

Crystallization of Microporous Compounds

281

Figure 5.10 Formation of the bonds between the SiO2 colloidal particles during gelation. Reprinted from Ralph K. Iler: The Colloid Chemistry of Silica and Silicates. Copyright 1995 by Cornell University (Copyright renewed 1983). Used by permission of the publisher, Cornell University Press

and pore diameter derived from the stacking density or coordination number of the primary SiO2 colloidal particle are used to describe the structure of the SiO2 gel. The stacking density (S) is defined as the volume fraction of the primary SiO2 colloidal particles for a given amount of SiO2 gel. Table 5.5 shows the correlation between the Table 5.5 Correlation between the stacking density and coordination number Coordination number (C.N.) 12 6 4 3

Stacking density (S) 0.7405 0.5236 0.123  0.338 0.056  0.185

282

Chemistry of Zeolites and Related Porous Materials

Table 5.6 Correlation between the pore volume and stacking density of SiO2 gel Samples of SiO2 gel 1 2 3 4

Specific surface area/(m2/g) 344 657 248 478

Pore SiO2 density H2 volume/ displacement (mL/g) method 0.870 0.348 0.612 0.575

2.32 2.35 2.32 2.15

Stacking density(S) 0.331 0.522 0.414 0.443

Mean Mean diameter of diameter of colloidal micropore/A˚ particles 92 21 73 41

75 29 81 48

stacking density (S) and coordination number (C.N.). The stacking density or coordination number could be calculated from the pore volume of the primary SiO2 colloidal particles for a given amount of SiO2 gel. The structural properties of SiO2 gel determined from gas adsorption analyses are summarized in Table 5.6. The data in Table 5.6 indicate that the stacking density of the primary SiO2 colloidal particles in SiO2 gel is usually about 0:331  0:522. The most compact arrangement of the primary SiO2 colloidal particles happened in sample 2, which gave a coordination number of 6, while the loosest arrangement happened in sample 1, which gave a stacking density of 0.331 and coordination number of 4. Looser arrangement of the primary SiO2 colloidal particles than that in sample 1 was studied as well. In these cases, the measured coordination number is about 2  3, which leads to a bigger pore volume and higher specific surface area. 3. Preparation of SiO2 gel The preparation of SiO2 gel is not very complicated. Either the reaction of an appropriate amount of acid and silicate or the gelation of SiO2 colloidal particles can result in SiO2 gel. However, the preparation of SiO2 gel with various properties is not that simple. Previous studies indicated that the particle size, porosity, and hardness of the SiO2 gel strongly depended on the gelation rate that was affected by many factors.[11] The practical preparation of SiO2 gel consists of many steps, including the formation of primary SiO2 colloidal particles via the polymerization of silicate, the gelation of the primary SiO2 colloidal particles, aging, modification, drying, and heat treatment of the resultant gel.  The preparation of SiO2 gel with large pore and low specific surface area. Winyall and colleagues reported the preparation of SiO2 gel with large pore and specific surface area lower than 320 m2/g. [15,16] The pore volume of the SiO2 gel they reported is 1.0 mL/g. Specifically, the pH of silicate was adjusted to 9:8  10:4 with acid to prepare the hydrogel of silica. Subsequently, the obtained silica hydrogel was aged for 50 minutes at pH 2  3 followed by further aging for 4 h at pH 8. Finally, the product was washed with water and dried. SiO2 gel with large pore and small specific surface area could be obtained by washing the product with hot water. In this case, the gel should be aged at pH 7. Alternatively, the product could be washed with an organic solvent with small surface tension, which could displace the water inside the gel framework and reduce shrinkage at the drying stage.

Crystallization of Microporous Compounds

283

Table 5.7 Specific surface-area change of the SiO2 gel after calcination at atmospheric pressure Temperature(T/
C) 25 800 1000

Specific surface area(m2/g) 700 500 0

SiO2 gel with large pore could be prepared by dissolving TEOS in ethanol, followed by addition of the calculated amount of water. Hydrolysis of the TEOS forms the poly ethyldrosiloxane (PES). The pore diameter of the resultant SiO2 gel is about 11:6 11.8 nm and the pore volume is about 2.0 mL/g, which could be increased to 3.0 mL/g by the addition of hexane before the initiation of gelation. In addition, the hydrothermal treatment of the initial silica gel at a temperature of 100
C can enlarge the pore without destroying the framework of the product.  The preparation of SiO2 gel with small pore and high specific surface area. The SiO2 gel with small pore could be prepared by using appropriate amounts of acid and silicate. However, in practice, the SiO2 gel with small pore was prepared by heat treatment of SiO2 gel with large pore, which could reduce the pore size as well as the specific surface area (Table 5.7).  The preparation of SiO2 gel with high strength and hardness. In industry, SiO2 gel with high hardness, strength, and small particles size is prepared by using H2SO4, HCl, and SiO2 as source materials at appropriate pH and low temperature. 4. The preparation of amorphous SiO2 Amorphous SiO2 is a fragment of silica gel. The difference between the amorphous SiO2 and silica gel is that the amorphous SiO2 contains only part of the medium (water). Sodium silicate solution could be used to prepare amorphous SiO2. However, the concentration is much lower than that in the preparation of silica gel. For a given silica sol solution with pH 9  10 and Naþ concentration higher than 0.3 mol/L, simple heat treatment of this solution can result in the precipitation of amorphous SiO2. Besides Naþ, Ca2þ or multivalent cations, NH4þ, and F can also lead to the precipitation of amorphous SiO2. In addition, the addition of ammonia in TEOS ethanolic solution can result in the precipitation of amorphous SiO2 as well. The above-described precipitation of amorphous SiO2 refers mainly to the product from aqueous solution. Alternatively, amorphous SiO2 could be prepared via high-temperature hydrolysis of Si sources. The amorphous SiO2 made by this method features low density, high purity, being anhydrous sometimes, and high cost, which is not a suitable source material for the synthesis of zeolites. In industry, amorphous SiO2 is made by the flame (high-temperature) hydrolysis of SiCl4 or SiF4. [17,18] The Organic Chelate of Silicon Our understanding of silicon chelates with organic ligands is still insufficient, and research in this area is sparse.

284

Chemistry of Zeolites and Related Porous Materials

Silicon could form a 4-, 5-, or 6-coordinated inner orbit chelate with organic ligands in neutral or slightly basic media. The chelate of organic silicon could be positively or negatively charged, depending on the structure of the organic ligand. Under an inert environment and mild basic conditions, as well as in the presence of ammonia, pyridine, or alkylamine, silicon could form a 6-coordinated chelate with C6H4(OH)2. The formula is:

The presence of this reaction sometimes could affect the crystallization of zeolites. Silicon could form a 5-coordinated chelate with organic ligands:

The structures of these three chelates have been determined. R in crystal I could be C6H5- or O2NC6H4-, while R in crystal II and III could be CH3- and C6H5-, respectively. Compared with 5-coordinated silicon, 6-coordinated silicon can form more chelates. For example, silicon can form octahedral 6-coordinated chelates with catechol and 2,3naphthalenediol. The stability of these chelates is significantly affected by the pH of the solution. Bauman and coworkers studied the stability of silicon–catechol chelate and obtained a value of 2:1  1011 (pH 8.25) for the stability constant. 5.1.2

Structure of Commonly Used Aluminum Sources

The synthesis of zeolites is usually conducted in slightly basic media via the condensation of silicate and aluminate. Therefore, studies on the structure and state of commonly used aluminum sources in solution are essential for studies on the synthesis of zeolites. Commonly used aluminum sources in the synthesis of zeolites include sodium aluminate (NaAlO2, Na2O: 54%), pseudo-boehmite, home-made aluminum hydroxide, and aluminum isopropoxide, which is the mostly used in the synthesis of microporous aluminophosphates. The structures of the two most important aluminum sources are shown in Figure 5.11 and Figure 5.12, respectively. Studies[20] on the polymerization state of aluminate in solution have been conducted for a long time and it has been found that the pH of the solution directly affects this state in solution. In basic solution, the main form of aluminate is Al(OH)4. When the

Figure 5.11 Structure of pseudo-boehmite. Reprinted from [19]. Copyright (1999) John Wiley & Sons, Inc.

Crystallization of Microporous Compounds

285

Figure 5.12 Structure of aluminum isopropoxide. Reproduced from [19] with permission of John Wiley & Sons, Inc.

alkalinity of the solution is high enough (e.g., the concentration of Na2O is greater than 25%), Al(OH)4 will be transferred to AlO2 or dimer via dehydration via two possible mechanisms described blow in Equations (5.3) to (5.5): I AlðOHÞ4  ! AlOðOHÞ2  þ H2 O

ð5:3Þ

AlOðOHÞ2  ! AlO 2 þ H2 O II

ð5:4Þ 2



O AlðOHÞ3  2AlðOHÞ4 ! ½ðOHÞ3 Al 

þ H2 O

ð5:5Þ




The bond angle of the formed AlO2 is 132 . In sufficiently acidic solutions, the aluminum exists as hydrated Al3þ ion. From pH 2–6, the aluminous cation changes by progressive hydrolysis according to the formula ½Al13 O4 ðOHÞ24þy ðH2 OÞ12y ð7yÞþ , where y  0. These complex cations may co-exist with simpler species such as Al(OH)2þ and Al(OH)2þ, as it is estimated that unit values of concentration ratio of certain pairs of ions occur at the following pH values: Al(OH)2þ/Al3þ ¼ 1 Al(OH)2þ/Al3þ ¼ 1 Al(OH)4/Al3þ ¼ 1

pH ¼ 4.89 pH ¼ 5.28 pH ¼ 5.87

At other pH values, the concentration ratios are as in Table 5.8. One may summarize the various observations as follows. In strongly acid solutions, Al(H2O)63þ is dominant. From pH ¼ 6 upwards, Al3þ becomes wholly insignificant, polymeric ions are no longer important, and Al(OH)4 or AlO2 are dominant. The important Al(OH)4 is tetrahedral in structure and so should favor tectosilicate formation with silicates condensed into a 3-D framework. 5.2

Crystallization Process and Formation Mechanism of Zeolites

Studies on the crystallization process and formation mechanism of zeolites are very important not only because of their theoretical significance but also due to practical

286

Chemistry of Zeolites and Related Porous Materials

Table 5.8 Estimated concentration ratios of ion pairs as functions of pH pH

Al(OH)4/Al3þ

Al(OH)2þ/Al3þ

Al(OH)2þ/Al3þ

1 2 3 4 5 6 7 8 9 10 11 12 13 14

1019.5 1015.5 1011.5 107.5 103.5 100.5 104.5 108.5 1012.5 1016.5 1020.5 1024.5 1028.5 1032.5

103.89 102.89 101.89 100.89 10þ0.11 101.11 102.11 103.11 104.11 105.11 106.11 107.11 108.11 109.11

106.56 104.56 102.56 100.56 101.44 103.44 105.44 107.44 109.44 1011.44 1013.44 1015.44 1017.44 1019.44

values. Even though numerous zeolite structures have been successfully synthesized, it is still necessary to rationally design and synthesize more and more zeolite structures with specific architectures and properties, which requires a fuller understanding of the crystallization process and formation mechanism of zeolites. Extensive studies should be conducted in various areas such as the polymeric state and structure of silicate ions; the condensation reaction between the silicate and aluminate ions; the structure of aluminosilicate ions; the formation, structure, and transition of the sol; the formation and structure of the gel; the nucleation of zeolites and the role of structure-directing agents (SDAs); the growth of zeolite crystal and the transition of metastable phases in zeolite crystallization. So far, there is still no well accepted interpretation of the formation of zeolites. At present, two main mechanisms exist. One is called the solid hydrogel transformation mechanism, and the other is called the solution-mediated transport mechanism. In the first mechanism, it is believed that, under crystallization conditions, the structure of zeolites is obtained by rearrangement of the framework of aluminosilicate hydrogel formed from the condensation of silicate and aluminate ions in the early stage of crystallization, while in the second mechanism, it is believed that, under crystallization conditions, the aluminosilicate hydrogel will be redissolved and the structure of zeolites is formed from the recrystallization of silicate and aluminate ions in the solution. Arguments on these two mechanisms have existed for decades. In the late 1980s, a dualphase transition mechanism was proposed and the view that the formation process of zeolites might follow different mechanisms under different crystallization conditions was also put forward. The proposition of different views on the formation mechanism of zeolites is due to the complexity of the formation of zeolites. The crystallization system of zeolites has solid and liquid components. The solid phase contains amorphous gel and zeolite crystals, while the liquid phase contains silicate ions with various polymeric states, aluminate, and aluminosilicate ions with various structures and polymeric states. The procedure for the synthesis of zeolites is not complicated, but the crystallization mechanism involved is

Crystallization of Microporous Compounds

287

very complicated. Most zeolites are metastable phases, and phase transition could happen for these structures, causing additional difficulty for any study of the formation process and mechanism of zeolites. In addition, experimental methodologies limit the studies and understanding of the crystallization process and mechanism of zeolites as well. For example, we still lack effective methods to measure or determine the structure and state of poly-silicate and-aluminate ions in solutions, the condensation reaction between poly-silicate and-aluminate ions, the structure of hydrogels, the nucleation of zeolites, and the structure-directing role of the organic template. Any breakthrough in these experimental methodologies may greatly improve the studies on the crystallization mechanism of this complicated system. 5.2.1

Solid Hydrogel Transformation Mechanism

Another name for solid hydrogel transformation mechanism is solid-phase mechanism, while solution-mediated transport mechanism is also called liquid-phase mechanism. The main difference in explaining the formation process of zeolites by these two mechanisms lies in whether the liquid component is involved during the crystallization of zeolites. The views of these two mechanisms are opposite to each other and have their own experimental supporting evidence. To date, the liquid-phase mechanism has more experimental support than does the solid-phase mechanism. In 1968, Breck and Flanigen for the first time proposed the solid-phase mechanism based on their studies of the crystallization of aluminosilicate. They found that the formation and transformation of amorphous aluminosilicate hydrogel always happened during the crystallization process of zeolites and that the composition of the hydrogel was similar to that of the resultant zeolites. In a word, in the solid-phase mechanism, it is believed that neither the dissolution of solid gel nor the direct involvement of the liquid phase happened for the nucleation and growth of zeolite crystals during the crystallization process of zeolites. The nucleation and growth of zeolite crystals came from the structural rearrangement of the framework of solid aluminosilicate gel under hydrothermal crystallization conditions. The solid-phase mechanism could be roughly illustrated by Figure 5.13. After the mixing of various source materials, the initial aluminosilicate gel formed due to the condensation reaction between silicate and aluminate ions. Even though the liquid phase could be formed in the void of a hydrogel framework, supporters of the solidphase mechanism believed that the liquid phase did not involve the crystallization reaction of zeolites. Subsequently, depolymerization and structural rearrangement reaction of the solid gel happened under the catalysis of OH, resulting in the formation of the primary structural units needed for the crystallization of zeolites. These primary structural units could form polyhedra via rearrangement around the hydrated cations, which could further polymerize and join to form the zeolite crystal. In the early 1970s, McNicol et al. monitored the entire crystallization process of zeolite A by using molecular spectrum techniques and obtained experimental evidence that supported the solid-phase mechanism.[21] Based on an analysis of the results of Raman and phosphorescence spectra, McNicol claimed that, during the crystallization process of zeolites A, the composition of the

288

Chemistry of Zeolites and Related Porous Materials

Figure 5.13 Illustration of solid-phase mechanism. Used with permission from the Journal of Chemical Education, Vol. 41, No. 12, 1964, p. 683; Copyright 1964, Division of Chemical Education, Inc.

liquid phase did not change and no aluminosilicate ions or secondary structural units were detected, while the amorphous hydrogel was transformed into a zeolite framework with cage-like structural units. Therefore, it was concluded that the crystallization of zeolite A followed the solid-phase mechanism. In the late 1970s, Flanigen found that zeolite X could directly crystallize from pure solid phase. In this study, the synthetic system that had passed the induction period was filtered and the hydrogel was dried at an appropriate temperature. The dried gel was amorphous at this stage and had a molar composition of 1.1 Na2O : Al2O3 : 2.7 SiO2 : 4.6 H2O. After 10 days of treatment of this dry amorphous gel at ambient temperature in air, 2% zeolite X formed, and this amount increased to 20% after 47 days.

Crystallization of Microporous Compounds

289

Strong experimental support for the solid-phase mechanism came from the studies of Xu on the direct crystallization of ZSM-35 and ZSM-5 from a solid phase of hydrogel.[21] In his study, the aluminosilicate hydrogel for the synthesis of ZSM-35 and ZSM-5 was dried and calcined at 550
C prior to crystallization. The calcined product was dispersed in triethylamine (Et3N) that contained the template ethylenediamine (EN) under N2 atmosphere to form a mixture with molar composition (20  45) EN: (1  8) Na2O : (15  50) SiO2 : Al2O3 : (40  420) Et3N. Treatment of this mixture at 160
C resulted in the formation of pure ZSM-35 and ZSM-5. Surprisingly, it was found that the Si2O/ Al2O3 ratio of the solid phase was constant and no silicate or aluminate ions were detected in the liquid phase during the crystallization process, which strongly supported the solid-phase mechanism. Over the past 10 years, more crystallization systems supporting the solid-phase mechanism have been reported. Remarkable examples included Tsapatsis’s study on the crystallization of zeolite L by using High Resolution Transmission Electron Microscope (HRTEM) technique in 1996;[23] Serrano’s study on the crystallization of TS-1 by using a couple of spectral techniques in 1996[24] and on the crystallization of pure silica zeolite beta under the presence of F;[25] and Uguina’s study on the crystallization of TS2 by using multiple techniques.[26] The commonly used dry-gel synthesis of zeolites in recent years (DGC and SAC, see Section 3.2.5, Sub-section ‘Dry Gel Conversion’ in Chapter 3) partially confirms the rationality of the solid-phase mechanism. 5.2.2

Solution-mediated Transport Mechanism

The solution-mediated transport mechanism has been extensively discussed in the literature. In the middle of the 1960s, according to their studies on the crystallization of zeolite A, Kerr and Ciric proposed the solution-mediated transport mechanism. They believed that the nucleation and growth of zeolite crystals happened in solution. The initial gel was partially or completely dissolved in the solution with the formation of active silicate and aluminate ions. These active silicate and aluminate ions could further form the structural units of zeolite crystal. Zhdanov and colleagues for the first time discussed in detail the solution-mediated transport mechanism.[27] They believed that 1) nucleation happened in the solution or at the interface of the solution and solid gel; 2) the further growth of zeolite nuclei consumed the silicate and aluminate ions in solution; 3) the solution supplied the soluble structural units for the growth of zeolite crystal; and 4) the consumption of the liquid component during the crystallization process resulted in the continuous dissolution of solid gel. The liquid-phase mechanism is illustrated in Figure 5.14. After the mixing of source materials, the aluminosilicate hydrogel is formed first, which might contain simple structural units such as 4- and 6-membered rings. A dissolving equilibrium between the solid gel and liquid phase existed at this stage. The solubility product (a parameter) of aluminosilicate ions depended on the structure of the solid gel and the system temperature. When the temperature was increased, a new equilibrium between the solid gel and liquid phase would be established. The increase in the concentration of the polysilicate and aluminate ions in the liquid phase resulted in the formation of nuclei followed by crystal growth. The nucleation and crystal growth consumed the polysilicate and aluminate ions in the liquid phase, leading to the continual dissolution of aluminosilicate gel. Because the solubility of zeolite crystal is much lower than that of amorphous

290

Chemistry of Zeolites and Related Porous Materials

Figure 5.14

Illustration of solution-mediated transport mechanism

gel, the solid aluminosilicate gel was completely dissolved after complete crystallization of zeolite. Zhdanov and colleagues found that the growth rate of zeolite crystal depended on the concentration of polysilicate and aluminate ions in the liquid phase and the composition of the liquid phase was not constant during the crystallization process.[27] This study strongly supported the liquid-phase mechanism. In the 1980s, Xu and Liu studied the crystallization process of zeolite KL and concluded that the formation of zeolite KL followed the liquid-phase mechanism.[28] The experiments they designed were the substitution of Al with Ga for crystallization systems of 7 K2O . Al2O3 . 15 SiO2 . 360 H2O and 7 K2O . Ga2O3 . 15 SiO2 . 360 H2O, which could crystallize the pure zeolite L containing Al and Ga, respectively, under identical conditions. After 13 hours’ crystallization, the liquid phases for both systems were separated and exchanged with each other. The new crystallization systems came to be the solid phase containing Al2O3 and the liquid phase containing Ga2O3, and the solid phase containing Ga2O3 and the liquid phase containing Al2O3. These two new systems continually crystallized. If the formation of zeolite L followed the solid-phase mechanism, the composition of the liquid phase for both new systems should remain constant.

Crystallization of Microporous Compounds

Figure 5.15

291

Concentration change of Ga in liquid phase for the new Al(Ga) system

However, the reality was totally against this assumption. As shown in Figure 5.15, the Ga content in the liquid phase decreased to zero after 6 hours’ crystallization. This result indicated that the Ga2O3 component in the liquid phase was consumed during the crystallization process of zeolite L. For the new system of the solid phase containing Ga2O3 and liquid phase containing Al2O3, the content of Ga2O3 in the liquid phase gradually increased from zero along with progress of the crystallization as shown in Figure 5.16. After a maximum was reached, the content of Ga2O3 was kept constant for a short while followed by a gradual decrease, indicating that the rate of Ga2O3 dissolution was greater than the consumption rate at the initial stage and that the situation was reversed at the later stages after a short equilibrium for both rates. The abovedescribed experiments indicated that the liquid phase participated in the crystallization of

Figure 5.16

Concentration change of Ga in liquid phase for the new Ga(Al) system

292

Chemistry of Zeolites and Related Porous Materials

zeolite and that the consumption of the components in the liquid phase was replenished by the dissolution of the solid phase. In addition, the presence of ultra-small crystals with structure of zeolite L in the liquid phase of the crystallization system was confirmed by electron diffraction analysis, suggesting that the nucleation of zeolite L occurred in the liquid phase. The polymeric state of silicate ions in the liquid phase was analysed by using the trimethylsilylation method. The results indicated that the processes of nucleation and growth of the crystal of zeolite L needed polymerized silicate ions with different polymeric degrees. For example, a large number of polysilicate anions in the form of 6-membered rings were needed at the nucleation stage, while a large number of monomeric or dimeric silicate ions were needed at the growth stage of the zeolite crystal. Moreover, a study of the crystallization mechanism of zeolite A by using the same methodology gave a very similar result.[29] Powerful evidence for the liquid-phase mechanism comes from the direct crystallization of zeolite from clear solution. In the early 1980s, Koizumi and coworkers carried out an extensive study on this topic. They directly synthesized analcime, hydroxysodalite, zeolite B, mordenite, zeolite P, faujasite,[30] erionite, and potassium chabazite from clear solution. Pang et al. directly crystallized zeolite A[31] and FAPO-5[32] from clear solution as well. The study on the direct synthesis of faujasite [30, 33] from clear solution will be elaborated below. Figure 5.17 shows the crystallization field of NaY drawn by Koizumi and coworkers based on 225 experimental points. Zeolite NaY could be directly crystallized from the

Figure 5.17 Crystallization field of zeolite from clear solution. Reproduced with permission from [30]. Copyright (1984) Elsevier

Crystallization of Microporous Compounds

293

Figure 5.18 Direct crystallization of zeolite Y, S, and P from clear solution. Reproduced with permission from [30]. Copyright (1984) Elsevier

clear solution. The features of this field are high alkalinity and Si/Al ratio. The authors emphasized that the use of fresh aqueous colloidal silica sol was extremely important for the reproducibility of their results. When the source materials were mixed, the solid gel immediately formed. After crystallization at 100
C for 2 h, this solid gel was completely dissolved and a clear solution formed. Further crystallization of this clear solution resulted in the formation of zeolite crystals. Undoubtedly, this case is powerful evidence for the correctness of the solution-mediated mechanism. The crystallization kinetics curve for the synthetic system with a composition of 10 Na2O . 0.25 Al2O3 . 25.3 SiO2 . 270 H2O is shown in Figure 5.18. According to the results in Figure 5.18, zeolites NaY, S, and P crystallized from the clear solution after crystallization periods of 12, 140, and 240 h, respectively. The composition of clear solution could be calculated from the yield and composition of the resulting zeolite. The study indicated that the composition of zeolites Y, S, and P was related to the corresponding composition of the clear solution. In addition, several other experiments designed by Koizumi and coworkers showed that the growth of zeolite crystal would not be immediately affected by the change of the composition of the liquid phase. For example, a small amount of liquid was taken from the synthetic system of zeolite Y crystallized for one, two, three, and four days, respectively. The sodium silicate solution was used to tune the composition of this solution to that for the crystallization of zeolite P. The resultant liquid phase was continually crystallized under the original crystallization conditions. The growth of zeolite NaY crystals did not stop immediately but continued for a considerable length of time as shown in Figure 5.19.

294

Chemistry of Zeolites and Related Porous Materials

Figure 5.19 Effect of discontinuous changes of the composition of the clear solution on the crystallization of zeolite Y. Reproduced with permission from [30]. Copyright (1984) Elsevier

In the past 10 years, many characterization techniques have been used in the in situ study of the formation mechanism of zeolite from clear solution. For example, Honssian et al. studied the TPA-silicalite-1 system by using the small-angle X-ray scattering technique;[34] Carlsson et al. performed a modeling study on the crystallization of silicalite-I from a liquid-phase system;[35] Smaihi and colleagues applied in situ 27Al-, 29 Si-, and 13C-NMR techniques to the study of the formation mechanism of zeolite A;[36] Bronic et al. investigated the crystallization mechanism of zeolite A crystallized from clear solution by combining XRD, SEM, Laser Light Scattering (LLS), and NMR techniques;[37] Grizzetti and colleagues studied the nucleation and crystal-formation kinetics of LTA by using the X-ray Powder Diffraction (XRPD) method;[38] and Xiao and coworkers at Jilin University investigated the liquid-phase crystallization mechanism of zeolite X by employing UV Raman spectra.[39] All these studies strongly support the solution-mediated transport mechanism of some zeolites. Several important issues related to the solution-mediated transport mechanism will be discussed below. 5.2.3

Important Issues Related to the Solution-mediated Transport Mechanism

Condensation Reaction between Silicate and Aluminate Ions in Solution Along with the ongoing study of the crystallization mechanism of zeolites, it becomes increasingly necessary to study the polymerization state of aluminosilicate ions in solution and the condensation reaction between silicate and aluminate ions. The polymerization state of individual silicate and aluminate ions has been extensively studied, which has been discussed in Section 5.1.1. However, studies on the

Crystallization of Microporous Compounds

295

polymerization state and reaction of aluminosilicate ions in liquid- and solid-gel phases are far from extensive. The following discussion will focus on the structural states and the formation reactions of aluminosilicate ions under synthetic conditions of zeolites (pH > 11). At this stage, Al species exist as Al(OH)4, while Si species exist as various polysilicate anions. In the early 1980s, Dent Glasser et al. studied aluminosilicate systems with different cations by using 27Al- and 29Si-NMR, light-scattering, and electric-conductance techniques.[40] On the basis of their studies, Dent Glasser and colleagues believed that there are two link-types for the connection of AlO4 tetrahedron and SiO4 tetrahedron.[41] One is a covalent type such as Q1, Q2, Q3, and Q4, which is formed from the polymerization of [Si(OH)2O] and Al(OH)4. The other is a polymerization reaction based on a 5-coordinated intermediate as shown in Equation (5.6) below.

(5:6)

In the spectral area with low Si/Al ratio, besides a strong peak for the species of Al(OH)4, several broadened, weak peaks in the range 72  80 ppm could be found in the 27Al-NMR spectra, which is a typical feature of 5-coordinated intermediates. This structure is unstable in solution, but very stable in the solid phase. The aluminate compounds containing 5-coordinated Al species can be found in nature. One example is andalusite. To interpret the polymerization reaction that occurred in aluminosilicate solution, 5-coordinated aluminate was assumed to be existed. Based on the assumption of the presence of 5-coordinated aluminate, Dent Glasser and colleagues proposed another mechanism (described below) for the polymerization of silicate and aluminate ions in the solution,[41] which was named the 5-coordinated intermediate mechanism or base-delayed mechanism. The reactions will follow the steps described in Equation (5.7):

(5:7)

296

Chemistry of Zeolites and Related Porous Materials

In the early stage of this mechanism, Al(OH)4 reacted with slightly negatively charged silicate monomer to form the 5-coordinated intermediate (Al)* at a very low reaction rate. Subsequently, negatively charged 5-coordinated (Al)*s were adsorbed by cations and polymerized with each other around Mþ, resulting in covalence-type complicated aluminosilicate ions, which reacted further to form gel frameworks. This mechanism mainly occurred in the area with low Si/Al ratio, because a large number of Al(OH)4 could exist under this condition. If the alkalinity and other conditions are changed, the negative charge of the silicate ions could be increased, which is unfavorable for the formation of 5-coordinated intermediate (Al)*. Al(OH)4 could react only with slightly negatively charged silicate monomer. Therefore, increasing the alkalinity of the solution leads to an increase in the gelation time. This mechanism interpreted the predescribed experiments well. In addition, the appropriate volume of the cation of Mþ was necessary for the polymerization of negatively charged 5-coordinated intermediate (Al)*. Among alkali metal ions, Kþ met this requirement perfectly, which resulted in the fastest gelation rate for aluminosilicate systems containing an alkali metal. At the initial stage of the gelation of the potassium-containing system, the main species for Al and Si were Al(OH)4 and SiO4 tetrahedra, respectively, which have relatively low molecular weight. Along with the formation of more and more 5coordinated intermediate (Al)*, a rapid polymerization among (Al)* could happen, leading to the formation of complicated aluminosilicate ions with high molecular weight. Dent Glasser and colleagues found that a correlation exists between the gelation rate and the synthesis of zeolites.[41] They found that a low gelation rate would lead to a higher framework density for the synthesized zeolites. The framework densities for some typical zeolites are summarized in Table 5.9. Structure and Aging of the Aluminosilicate Gel In practice, aging of the primary aluminosilicate hydrogel formed from the mixture of silicate and aluminate ions at ambient temperature is necessary for the synthesis of zeolites, which usually results in shortening of the crystallization process and changes to the particle size and morphology of the crystallized zeolites. Over the past decade, the topic of ‘aging’ has been extensively studied and several regularities have been uncovered. In general, the commonly accepted concept is that aging changes the composition and structure of the primary aluminosilicate gel, which can affect the nucleation and growth of zeolites. In 2003, Okubo and coworkers studied the influence of aging on the crystallization of FAU-type zeolite from the gel system 50 Na2O : 10 SiO2 : 1.0 Al2O3 : 400 H2O at 90
C by employing the 29Si-NMR technique.[42] In their study, the colloidal silica (Ludox HS-40), sodium aluminate (NaAlO2), sodium hydroxide Table 5.9 Framework density of typical zeolites Zeolite X A P HS a

T ¼ Al or Si.

Framework density (T/1000 A˚3)a 12.7 12.9 15.4 17.4

Crystallization of Microporous Compounds

297

Figure 5.20 XRD patterns of the products obtained by (a) 24 h of crystallization without aging; (b) 48 h crystallization with 1 h of aging. Reproduced with permission from [42]. Copyright (2003) American Chemical Society

(NaOH), and water were well mixed. Direct crystallization of the resultant hydrogel led to the formation of zeolites SOD, ANA, and CHA (Figure 5.20). It was found that at least 1 day of aging was required to obtain a pure phase of FAU. Prolonged aging time led to an increase in the amount of aluminosilicate hydrogel during the induction period, which resulted in the appearance of more nucleable aluminosilicate ions. These factors resulted in a shortened crystallization time, a decreased particle size, and a narrowed particle-size distribution (7 days of aging led to a mean particle size of 0.6 mm, while 2 days of aging led to a mean particle size of 3 mm) as shown in Figures 5.21, 5.22, 5.23, and 5.24, respectively. In the following text, we will introduce the results concerning the influence of aging on the composition and structure of the aluminosilicate hydrogel obtained by employment of 29Si-NMR spectroscopy technique by these authors. The data in Figure 5.21 indicated that the structure of the aluminosilicate hydrogel changed along with prolongation of the aging time. The 29Si-NMR spectra of the unaged aluminosilicate gel showed a sharp peak and a shoulder centered at 110 and 100 ppm, respectively. The former sharp peak was attributed to Q4(OAl) from undissolved SiO2, while the latter shoulder was attributed to Q3 of the aluminosilicate that has a side chain. Aging of the initial gel for 2 days resulted in the weakening of the two peaks centered at 100 and 110 ppm and the appearance of one broad peak centered at 84 ppm, suggesting the formation of Q4(4Al) structural units. As shown in Figure 5.22, further prolongation of the aging time to 7 days promoted the formation of more aluminosilicate

298

Chemistry of Zeolites and Related Porous Materials

Figure 5.21 Formation and composition of aluminosilicate species obtained after the aging process at ambient temperature for 0, 2, and 7 days, respectively (29Si-NMR spectra). Reproduced with permission from [42]. Copyright (2003) American Chemical Society

structural units which were needed for the occurrence of nucleation. Two days’ aging of the aluminosilicate gel at ambient temperature resulted in the formation of pure FAUtype zeolite after 6 hours of crystallization, while 7 days’ aging shortened the crystallization time of the pure FAU-type zeolite to 3 hours (Figure 5.23 and Figure 5.24). In addition, compared to the 2-day aging, the 7-day aging of the aluminosilicate gel led to a narrower width of the peaks in the 29Si-NMR spectra and a lower Si/Al ratio of the framework (Table 5.10). Based on their results, Okubo and coworkers proposed a formation mechanism and the promotional effect of aging on the crystallization of zeolite FAU (Figure 5.25).

Figure 5.22 29Si-NMR spectra for the crystallization products obtained after the aging process for 0, 2, and 7 days, respectively. Reproduced with permission from [42]. Copyright (2003) American Chemical Society

Crystallization of Microporous Compounds

299

Figure 5.23 29Si-NMR spectra of the products obtained with 2 days of aging prior to crystallization with different periods. Reproduced with permission from [42]. Copyright (2003) American Chemical Society

Figure 5.24 29Si-NMR spectra of the products obtained with 7 days of aging prior to crystallization with different periods. Reproduced with permission from [42]. Copyright (2003) American Chemical Society

300

Chemistry of Zeolites and Related Porous Materials

Table 5.10 Distribution of Al in the framework of FAU obtained by crystallization after each aging period and the widths of each peak Width (Hz) Q4 (nAl) n Chemical shift/ppm

4 84 93.9 88.1

3 88 Width (Hz) 145 130

Aging time (t/d) 2 93

1 98

0 102

145 120

145 90.1

174 106

2 7

Si/Al

1.46 1.25

Nucleation ‘Nucleation’ is a key step in the crystallization process of microporous compounds. Over the past decade, the nucleation process has been extensively studied. The primary question for these studies is the definition of the term ‘nucleus’. In the field of zeolites and microporous compounds, the unit cell of one specific structure is usually regarded as its ‘nucleus’. In 1998, Pope studied the nucleation of zeolites A and X and suggested that the unit cell of LTA and FAU structure could be regarded as the ‘nucleus’ for zeolite A and X, respectively.[43] A cluster shown in Figure 5.26(a) could be regarded as a ‘nucleus’ of zeolite A and this ‘nucleus’ could grow via the growth points of 24 4-membered rings (^), while a cluster shown in Figure 5.26(b) could be regarded as a ‘nucleus’ of zeolite X and this ‘nucleus’ could grow via the growth points of 6-membered rings ( ). There are several other points of view regarding the definition of the ‘nucleus’ of zeolite. For example, it was suggested that some primary structural units of the framework, such as rings and basic cages, could be defined as the ‘nucleus’ of zeolites and other microporous crystals. It was also proposed that the ‘nucleus’ of zeolite could be defined as particles with critical size. These particles should be stable under crystallization conditions. Compared with the classical theory of nucleation from homogeneous solution, the theory developed by Pope could well explain the significant decrease of the free-energy barrier of nucleation for zeolites and other microporous compounds.[43] This

Figure 5.25 Influence of aging on the crystallization of zeolite FAU structure. Reprinted with permission from [42]. Copyright (2003) American Chemical Society

Crystallization of Microporous Compounds

Figure 5.26 Elsevier

301

‘nucleus’ of LTA and FAU structure. Reprinted from Pope [43]. Copyright (1998)

theory could explain the importance of aging in the crystallization of zeolites as well. In 1998, M. Tsapatsis and coworkers proposed a model for nucleation and crystal growth from the microstructure of aluminosilicate hydrogel.[44] Because aluminosilicate hydrogel is a precursor that contains micro-, meso-, and macro-pore structures, the nucleation of microporous crystals could occur only in the liquid phase located in the void of the gel framework. This model is schematically shown in Figure 5.27 In the following, the most important issue concerning the ‘nucleus’ of zeolites will be discussed, i.e., how the Si and Al species self-assemble together to form the nuclei with specific pore structure under the help of an SDA under crystallization conditions. This process has not been fully understood so far due to the lack of efficient in situ detection

Figure 5.27 Nucleation and growth model in aluminosilicate hydrogel. Reproduced with permission from [44]. Copyright (1998) Elsevier

302

Chemistry of Zeolites and Related Porous Materials

techniques. The reported studies on the discovery of nucleation regularity are mainly based on some specific systems. For example, Burkett and Davis systematically studied the nucleation mechanism for the synthesis of Si-ZSM-5 and Si-ZSM-48 in the presence of SDAs of TPAþ (NPr4þ) and HDA (H2N-CH2-CH2-CH2-CH2-CH2-CH2-NH2) in 1995.[45] In their studies, high-quality Si-ZSM-5 was synthesized from the crystallization of the gel system 0.5 TPA2O : 10 SiO2 : 380 H2O at 110
C for 15 days, while high-quality SiZSM-5 and Si-ZSM-48 could be obtained from the crystallization of the gel system 5 HDA : 2 SiO2 : 10 H2O at 120 and 150
C for 40 days and 10 days, respectively. The structures of Si-ZSM-5 and Si-ZSM-48 are shown in Figure 5.28. The SDAs TPAþ and HDA were located in the intersection of the 2-dimensional channel for ZSM-5 and the straight channel of ZSM-48 (HDA only). Based on these facts, Burkett and Davis proposed a nucleation mechanism for ZSM-5 and ZSM-48 in the presence of TPAþ and HDA. First, they confirmed the presence of the NPr4þ hydrophobic hydration sphere and hydrophobically hydrated domains of soluble silicate species by using XRD, IR,

Figure 5.28 Structures of ZSM-5 and ZSM-48 and the location of structure-directing agents of TPAþ and HDA in ZSM-5 and ZSM-48. Reproduced with permission from [45]. Copyright (1995) American Chemical Society

Crystallization of Microporous Compounds

303

29

Si-NMR, and 1H–29Si-CP-MASNMR (Cross-Polarization-Magic Angle Spinning Nuclear Magnetic Resonance) techniques. In the limited space of the liquid phase among the colloidal particles of the gel, these two species could impact with each other and make overlap of the hydration sphere happen. The release of water molecules from the ordered hydrophobic hydration spheres into the bulk and the subsequent establishment of favorable van der Waals’ contacts between the alkyl chains of TPA and the hydrophobic silica could thus provide the entropic and enthalpic driving forces for the formation of the composite inorganic–organic species as shown in Figure 5.29. Burkett and Davis proposed a similar formation mechanism for Si-ZSM-5 and SiZSM-48 under the influence of the SDA of HDA as shown in Figure 5.30.

Figure 5.29 Nucleation mechanism of ZSM-5. Reproduced with permission from [45]. Copyright (1995) American Chemical Society

304

Chemistry of Zeolites and Related Porous Materials

Figure 5.30 Nucleation mechanism of Si-ZSM-5 and Si-ZSM-48 under the influence of HDA. Reproduced with permission from [45]. Copyright (1995) American Chemical Society

Along with the discovery of numerous microporous aluminophosphates and other metal phosphates, the concept of ‘nucleation’ was modified. Recently, it was proposed by Ozin and coworkers that the nuclei of aluminophosphates could be formed from the selfassembly of some basic secondary building units (SBUs) existing in the liquid phase at the early stage of the crystallization reaction under the assistance of SDAs. Ozin and coworkers believed that the gel system Al2O3–H3PO4–TEG–Et3N should contain a parent chain of [AlP2O8H2] units built from corner-shared 4-membered rings in the liquid phase at the early stage of the crystallization reaction. The stability of this [AlP2O8H2] chain is very sensitive to the water quantity in the gel system. An increase in water content of the system results in the hydrolysis of this chain, with the formation of various new structural units. For example, when the water content in the system is limited, a ladder-like aluminosilicate chain could be formed from the hydrolysis of the cornershared aluminophosphate chain. The framework of JDF-20 could be regarded as the construction of these two kinds of aluminophosphate chain. When water is sufficient in the system, the crankshaft-like chain is the main product of the hydrolysis of the parent chain, which is the precursor needed for the formation of AlPO4-5. When water in the system is at an appropriate level, hydrolysis of the [AlP2O8H2] chain leads to the formation of chains (d) and (e) shown in Figure 5.31. Three-dimensional

Crystallization of Microporous Compounds

305

Figure 5.31 Main AlPO4 species existing in a TEG solution. Reproduced with permission from [46]. Copyright with permission of Wiley-VCH

open-framework aluminophosphates such as JDF-20, AlPO4-C, AlPO4-21, and AlPO4-5 could be regarded as the growth of the nuclei built from the above-described chain-like SBU structure under the influence of the SDA (Figure 5.31). Recently, Fe´rey and coworkers successfully synthesized a series of microporous gallophosphates from the Ga2O3–P2O5–RNH2–HF–H2O system under the influence of various SDAs of type RNH2. [47,48] Structural analyses of these compounds indicated that the framework of these compounds could be regarded as the self-assembly of a hexameric SBU with the help of different RNH2 SDAs (Figure 5.32). 5.2.4

Dual-phase Transition Mechanism

In the 1980s, a dual-phase transition mechanism was proposed, which suggested that both solid- and liquid-phase transition could happen in the same crystallization system. In the late 1990s, the understanding of the formation mechanism of zeolites was greatly improved due to the discovery of new study approaches and detection techniques. For example, Iton et al. studied the crystallization process of ZSM-5 under various conditions by first using small-angle neutron scattering.[49] They found that the formation mechanism of ZSM-5 depended on the silicon source. When low-polymerized silicate anions were used as silicon source, the active site on the surface of the hydrogel would combine with the TPAþ cations to form crystal nuclei. These surface nuclei would promote the rearrangement and crystallization of the framework of the gel, which suggested a solid-phase transition mechanism. However, when the silica sol (Ludox) was used as silicon source, the colloidal particles with a size of 12 nm were gradually dissolved, dispersed, and further polymerized to form primary gel in the presence of TPAþ cations. Subsequently, the primary gel was redissolved and polymerized to form nuclei, which suggested a typical liquid-phase transition mechanism. Iton and colleagues further concluded that the formation of the same zeolite could follow different transition mechanisms under different conditions. In 2000, Van Grieken et al. concluded that both solid- and liquid-phase transition mechanisms existed in the synthesis of nanocrystalline ZSM-5.[50]

306

Chemistry of Zeolites and Related Porous Materials

Figure 5.32 Structures of microporous gallophosphate built from the SBU of with permission from [47]. Copyright (2002) American Chemical Society

. Reproduced

In summary, great progress has been made in understanding the formation mechanism of zeolites and other microporous compounds. However, it is still far from allowing us to draw clear conclusions. To achieve a full understanding of the formation mechanism of zeolites, great advances should be made regarding the following topics: the development of new detection methods and techniques for the crystallization process, especially for the structure determination of the liquid phase in the parent mixture; the polymerization reaction of the components and the state and distribution of the resulting product; the phase transition between the chemical species of reactants, the sol and the gel; the structure of the gel; the role of SDAs in the formation of nuclei in liquid and solid phases; metastable phase and the crystal structure transition; the improvement in in situ detection techniques, and so forth. Recently, Francis and O’Hare reviewed the in situ detection techniques and their applications to the study of the formation mechanism of zeolites:[51] (1) in situ spectroscopic and optical techniques including IR, Raman, light scattering, optical microscopic, and NMR such as 27Al-, 29Si-, 31P-, 1H-, 13C-, and 15N-NMR; (2) in situ diffraction and scattering studies including XRD, neutron diffraction, X-ray and neutron scattering, and Extended X-ray Absorption Fine Structure/X-ray Diffraction (EXAFS/XRD), and so forth.

Crystallization of Microporous Compounds

307

5.3 Structure-directing Effect (SDE) and Templating in the Crystallization Process of Microporous Compounds Organic quaternary ammonium base was first introduced into the synthetic system by Barrer and Denny in 1961 to partially or fully replace inorganic base, and several puresilica or high-Si/Al-ratio zeolite molecular sieves were obtained from the organic quaternary ammonium base system. After extensive studies on the synthetic chemistry of this new system, it was learned that the organic base changed the chemistry of the gel in the synthetic system and partially played a templating role. According to this concept, the organic base used in the synthetic system of zeolites has been called a template or SDA. Subsequently, neutral organic molecules and inorganic ions were used as SDAs as well, resulting in the formation of numerous new structures. The initially proposed SDE of the organic base was that the assembly of TO4 tetrahedra or some specific basic structural units around the SDA formed some specific geometric configuration and structures at the nucleation stage. [52,53] However, studies conducted so far have revealed that the roles of the guest molecules in the synthesis of inorganic microporous compounds are very complicated. Up to now, there is no well accepted theory for the interpretation of the templating and structure-directing roles of organic guest molecules. In this section, various templating theories proposed in the literature and the experimental studies on the structure-directing and templating roles of organic guest molecules will be summarized and analysed. The views of the authors on this topic will be introduced as well. 5.3.1

Roles of Guest Molecules (Ions) in the Creation of Pores

Commonly used templates or SDAs in the synthesis of inorganic microporous compounds are metal cations, organic amines, organic quaternary ammonium base, fluorine ion, metal complex, etc. Their roles in the creation of pores will be individually discussed below. Cations Aluminosilicate zeolite molecular sieves are usually synthesized from basic solutions. The introduction of OH to the solution results in the introduction of corresponding cations as well. Commonly used cations in the synthesis of aluminosilicate zeolites are alkali metal cations such as Liþ, Naþ, and Kþ and alkaline earth metal cations such as Ca2þ and Ba2þ. Different cations can template different structures of zeolites. For example, Naþ can template the structures of zeolite A (LTA), cancrinite (CAN), analcime (ANA), gmelinite (GME), faujasite (FAU), and P (GIS); Kþ together with other cations can direct the structures of zeolite L (LTL), chabazite (CHA), and erionite (ERI); Liþ can template the structure of ABW; and Kþ can direct the structures of ANA, EDI, CHA, LTL, and BPH, but there have been no reports on the synthesis of the structures of FAU and LTA by using Kþ as a template so far. In the early 1960s, Roy and Barrer successfully synthesized the analogues of natural zeolites of heulandite (HEU), yugawaralite (YUG), and phillipsite (PHI) by using Ca2þ, Sr2þ, and Ba2þ as templates. In 1978, Robson pointed out that the alkaline earth metal cations would play an important role in the synthesis of aluminosilicate zeolites, which has been confirmed.[54]

308

Chemistry of Zeolites and Related Porous Materials

Dent Glasser et al. found that metal cations can affect the polymeric state of silicate ions, the polymerization rate and form of silicon and aluminum species, and the properties of the gel.[40, 41] In addition, it has been confirmed by extensive experimental data that there is a tight correlation between the metal cations and small cage-like structural units of the resultant zeolites (see Table 5.11 for details), i.e., the size of the cations or hydrated cations matches that of these small cages. One type of cation could template different structures under different conditions as well. Studies indicate that metal cations will affect the structure of zeolites via their electropositivity, size, and geometric configuration. Flanigen believed that the components of alkali metal cations of the synthetic system would play an important role in two aspects in the synthesis of zeolites:[55] (1) supply of base, i.e., OH and (2) structure directing. Recently, Dutta and colleagues concluded that the hydrated cations can stabilize aluminosilicate ions via static electric and steric interactions.[56] In a word, alkali metal cations or alkaline earth metal cations mainly play the roles of structure direction, framework charge balancing, and pH adjustment in the synthesis of aluminosilicate zeolites. Organic Compounds 1. Organic templates and the corresponding microporous compounds In the early 1960s, organic base such as amines was introduced into the synthetic system of zeolites for the first time, which brought about the breakthrough in the synthesis of zeolite molecular sieves. In 1961, Barrer and Denny reported the synthesis of zeolite A by the addition of TMAþ to the synthetic system.[20] At the same time, Kerr, a scientist at Mobil Petroleum Inc, started adding organic quaternary ammonium cations to the synthetic system of zeolites.[57] Studies indicated that alkylamine (ammonium) could increase the Si/Al ratio of the resulting aluminosilicate molecular sieves. High- or allsilica zeolites such as ZSM-5, ZSM-11, ZSM-12, ZSM-34, ZSM-39, and ZSM-48 could be obtained under the presence of alkylamines. This phenomenon is especially obvious in the synthesis of zeolite sodalite (SOD). The Si/Al ratio of sodalite synthesized in the presence of Naþ is 1. Each SOD cage possesses three negative charges, which are balanced by three free Naþ ions. However, the Si/Al ratio of the same framework crystallized in the presence of TMAþ could reach 5 because each SOD cage can contain only one counter ion of TMAþ.[58]

Table 5.11 Metal cations and the corresponding structural units Cation

Structural unit

Zeolite

Naþ Naþ or TMAþ Naþ or TMAþ Kþ, Ba2þ, Rbþ Naþ Naþ, Kþ, Sr2þ, Ba2þ

a cage SOD cage GME cage CAN cage D4R D6R

LTA, KFI LTA, FAU GME, OFF, MAZ FRI, OFF, LTL LTA FAU, KFI, CHA, GME, ERI/OFF, LTL

Crystallization of Microporous Compounds

309

Compared with rigid spherical inorganic cations, organic molecules or their cations have more complicated shapes and sizes, which means more choices in the synthesis of microporous compounds. So far, most of the microporous compounds have been synthesized under the influence of organic SDAs except for some zeolites which crystallized from metal-cation-containing systems (Table 5.12).[59] Besides organic amines (ammoniums), some other organic compounds such as polymer, alcohol, ketone, and organic sulfur compound could also be used as SDAs.[52] Jansen et al. summarized the typical organic SDAs (Table 5.13).[60] In the early 1980s, a series of aluminophosphate molecular sieves, AlPO4-n, were synthesized in the presence of organic amines.[61] For the first time, no SiO4 tetrahedra were found in the framework of these materials. The strict alternation of PO4 and AlO4 tetrahedra (AlO5 and AlO6 polyhedra existed in some cases) built up a neutral framework of aluminophosphate molecular sieves and resulted in an Al/P ratio of 1/1. Some of them are analogs of corresponding aluminosilicate zeolites, while others have new structures. However, there is no strict correlation between the structure of the organic amines (ammoniums) and that of the resulting microporous aluminophosphates. That is to say, one organic compound could direct toward many structures of microporous aluminophosphate, while one structure of microporous aluminophosphate could also be templated by many SDAs. Lok et al. summarized the organic amines used in the synthesis of AlPO4-n (Table 5.14).[62] Later, a series of anionic open frameworks of aluminophosphate with an Al/P ratio less than 1 including three-dimensional (3-D) interrupted framework, two-dimensional (2-D) layer, and one-dimensional (1-D) chain were solvothermally synthesized in the presence of organic amines, which greatly enriched the structural chemistry of the aluminophosphate family.[63] The negative charge of the framework is usually balanced  by the protonated organic amine. There are terminal P OH groups in the  O or P framework, which form hydrogen bonds with organic amines. The removal of organic amines from the anionic framework of aluminophosphate usually leads to collapse of the framework. In addition, a large number of microporous metal phosphates, arsenates, borates, and titanates have been successfully synthesized by using organic amines as SDAs. 2. Role of organic compounds in the synthesis of microporous compounds Similar to inorganic base, the basic roles of organic base in the synthesis of microporous compounds are to balance charge and adjust pH. However, different from inorganic base, an organic base will play a templating or structure-directing role in the creation of the framework of microporous compounds as well. For example, it was reported that tetramethylammonium (TMAþ) could be used as a template in the synthesis of sodalite ˚ ) well and each SOD cage contained one TMAþ. The size of the SOD cage (6.6 A ˚ ). However, the maximum opening of the SOD cage is a 6matched that of TMAþ (6.9 A membered ring, which will not allow TMAþ come in and go out. Obviously, TMAþ was encapsulated into the SOD cage during the formation of the framework of sodalite (Figure 5.33). Therefore, TMAþ was called an organic template or SDA. Owing to the variety of their sizes and structures, organic guests play different roles under different reaction conditions. Based on extensive analyses of the literature, we

Type of materials

Li-A(BW) ACP-1 AlPO4-18 AlPO4-11 AlPO4-5 AlPO4-8 AlPO4-14 SAPO-40 AlPO4-52 AlAsO-1 AlPO4-16 MAPO-36 AlPO4-33 AlPO4-21 b BePO-H Linde Q CIT-5 SAPO-34 Cloverite DAF-1 DAF-2 Dodecasil-1H EMC-2 Erionite AlPO4-17

Framework

ABW ACO AEI AEL AFI AET AFN AFR AFT APD AST ATS ATT AWO BEA BPH BPH CFI CHA -CLO DFO DFT DOH EMT ERI ERI

SDA Li H2NCH2CH2NH2 Et4Nþ Pr2NH, iPr2NH, Bu2NH Et4Nþ, Pr4Nþ, Pr3N, Et3N, etc. Bu4Nþ, Bu2NH, Pr2NH t-BuNH2, iPrNH2 Pr4Nþ Et4Nþ þ Pr3N HOCH2CH2NH2 quinuclidine Pr3N, Pr4Nþ Me4Nþ Me3N, PrNH2 Et4Nþ Naþ þ Kþ þEt4Nþ Kþ N-methylsparteinium Et4Nþ, PrNH2 quinuclidine H2N(CH2)10NH2 DEA pyrrolidine, MeNH2 18-crown-6 Naþ þ Kþ þ Me4Nþ cyclohexylamine, piperidine

þ

Table 5.12 Typical zeolite structures and corresponding SDAs

EUO FAU FAU FER GME IFR KFI LTA LTA LTL MAZ MEI MEL MFI MTN MTT MTW MWW OFF RHO RUT SBE SBS VFI ZON

Framework EU-1 Faujasite SAPO-37 ZSM-35 Gmelinite ITQ-4 ZK-5 Linde Type A SAPO-42 Linde type L

ZSM-18 ZSM-11 ZSM-5 ZSM-39 ZSM-23 ZSM-12 MCM-22 Offretite Rho Rut-10 UCSB-8Co UCSB-6GaCo VPI-5 ZAPO-M1

Type of materials

Me3N(CH2)6NMe3 Naþ Me4Nþ þ Pr4Nþ H2NCH2CH2NH2, pyrrolidine DABCO N-benzylquinuclidine Triethylenediamine Naþ Me4Nþ þ Naþ,Et4Nþ Kþ Me4Nþ þ Naþ (Me3NþCH2CH2)3CH Bu4Nþ Pr4Nþ, H2N(CH2)6NH2, etc. pyrrolidine, piperidine pyrrolidine Et4Nþ hexamethylenediamine Kþ þ Me4Nþ Naþ þ Csþ Me4Nþ H2N(CH2)6NH2 H2N(CH2)6NH2 Pr2NH MeNþ

SDA

Crystallization of Microporous Compounds

311

Table 5.13 Organic structure-directing agents and their functional groups SDA

Functional group

Amine

SDA

Functional group

Pentaerythritol

n ¼ 4,5 Diamine

amine þ ammonium 3  n  10

n ¼ 2,3 x ¼ 1  3

n ¼ 4,5 Ammonium

ammonium þ alcohol

Diammonium

acetal 3(n(9

Triammonium

amine þ ether

Amine þ ammonium

N O þ ammonium

Alcohol

Cn-OH n¼16

phosphine

Diol

HO-Cn-OH n¼16

diphosphine

Triol

summarize the roles of organic species in the synthesis of microporous compounds as follows: (1) True templating effect. True templating effect means that the organic compounds play a real templating role in the synthesis of microporous compounds. The structure templated by this organic species

312

Chemistry of Zeolites and Related Porous Materials

Table 5.14 Organic amines and corresponding aluminophosphate molecular sieves Organic amine

AlPO4-n

Organic amine

AlPO4-n

nPrNH2 tBuNH2 Cyclohexylamine (nPr)2NH (nBu)2NH Dicyclohexylamine Me3N (nPr)3N TMEOH TPAOH TPAOH DDO Diethylalethanolamine N-formaldehydeDiethanolamine Pyrrolidine 3-Methylpyridine

AlPO4-21 AlPO4-14 AlPO4-5,17 AlPO4-11,31 AlPO4-8,11 AlPO4-5 AlPO4-21 AlPO4-5 AlPO4-21 AlPO4-5 AlPO4-8 AlPO4-22 AlPO4-5 AlPO4-5,7

iPrNH2 nPrNH2 DMBA (iPr)2NH (nPe)2NH N-formaldehydecyclohexylamine Et3N (HOCH2CH2)3N TEAOH TBAOH choline-OH diethylalethanolamine diformaldehydeethanolamine formaldehydeethanolamine

AlPO4-14 AlPO4-17 AlPO4-5 AlPO4-11 AlPO4-8,11 AlPO4-5 AlPO4-5 AlPO4-5 AlPO4-5,18 AlPO4-8 AlPO4-5,7 AlPO4-5 AlPO4-5,21 AlPO4-5,21

AlPO4-21,23 AlPO4-5

2-methylpyridine n-butyldimethylamine

AlPO4-5 AlPO4-5

cannot be synthesized by using any other organic compound as organic additive. In this case, the size of the organic compound well matches that of the void of the microporous compounds, thus limiting the movement of the organic guest. The true templating effect can only happen when the geometric and electron configurations of the organic molecule perfectly match those of the framework of microporous compounds. Therefore, the case is very rare. One remarkable example is the synthesis of ZSM-18 (MEI).[64,65] The geometric configuration of the organic molecule perfectly matched that of the cage in the framework of ZSM-18 (Figure 5.34). Based on the true templating effect that happened in this case, Schmitt and Kennedy successfully synthesized ZSM-18 by using organic species A and B, as shown in Figure 5.35, which have a configuration very similar to that of the triquaternary ammonium species C18H36N33þ.[66]

Figure 5.33

SOD cage and encapsulated TMAþ

Crystallization of Microporous Compounds

Figure 5.34

313

A C18H36Nþ cation located in a MEI cage

(2) Structure-directing effect Most organic amines and quaternary ammonium species play a structure-directing role in the synthesis of microporous compounds, and are thus called structure-directing agents (SDAs) in the literature published in recent years. Compared with metal cations, organic compounds have a stronger SDE. For example, Zones et al. reported the synthesis of high silica zeolite SSZ-13 (CHA) from the synthetic system of zeolite P by adding N,N, N-trimethyldiamantane to the synthetic system.[67,68] However, the SDE of organic compounds varies from case to case and is usually classified into strict SDE and normal SDE. Strict SDE mainly refers to the instance where a zeolite structure can only be directed by one organic compound, which was called one-to-one SDE by Davis.[69] Examples include zeolite SSZ-24[70] and trimethylamine, EMT[71] and Naþ and 18-crown-6, and CIT-5 and sparteine MeSPAOH.[72,73] Normal SDE mainly refers to the case in which organic compounds usually can direct the formation of small structural units, cages, or channels, but there is no one-to-one correlation between the organic molecule and the structure. Normal SDE is very common in the synthesis of microporous compounds. For example, TMAþ usually directs to SOD cage, 4-membered rings, and double 4-membered rings; TEAþ usually directs to double 3-membered rings; while TPAþ and TBAþ usually direct to 5-membered rings and double 5-membered rings, respectively. In some cases, the shape and size of organic molecules perfectly match those of the channel of the framework. For example, TPAþ can direct to zeolite ZSM-5.[74] Structure analysis indicated that TPAþ was located in the

Figure 5.35

Structures of molecule A (left) and B used in the synthesis of ZSM-18

314

Chemistry of Zeolites and Related Porous Materials

intersection of two channels. Four propyl chains stretched into the channels in four directions, suggesting that the guest molecule of TPAþ directed the intersected channel of ZSM-5. In addition, organic guest molecules can direct to various cages. For instance, when the organic diammonium species [(C7H13N)(CH2)n(C7H13N)Br2] (biquinine) was used as SDA in the synthesis of STA-2, the size of the organic cations obviously affected the shape, size, and connection style of the resultant cage or channel as shown in Figure 5.36.[75] The size of the organic cation was determined by the length of the alkyl chain. Microporous compounds of AlPO4-17 (cage connection model: 412626386), STA-2 (412626686), and AlPO4-56 (41243628683) which had different cage structures were successfully synthesized with n ¼ 3, 4, and 5, respectively. These three cages are similar and have the same 4126268 structural unit. The void of these three cages well matched the size of their SDAs (Figure 5.37). Microporous aluminophosphate of AlPO4-5 with onedimensional 12-membered ring channel formed when the size of this organic diammonium species was further increased (n  6). The results in Figure 5.36 indicated that the size of the cages did not increase with ˚  9.64 A ˚ 2) is increasing size of the SDAs. For example, the cage in STA-2 (18.16 A 2 ˚  10.42 A ˚ ). This may be caused by the slightly bigger than that in AlPO4-56 (16.16 A shape change of the organic guest molecule during nucleation due to the interaction of inorganic host and organic guest (chain-shape organic SDA is very flexible). In general, a longer chain for organic molecules means better flexibility and greater possibility of shape changing. Therefore, there is no strict correlation between the length of the chain of organic molecules and the size of the resulting cage or channel. For example,

Figure 5.36 The cage and channel structure directed by [(C7H13N)(CH2)n(C7H13N)Br2]: (a) AlPO4-17 (n ¼ 3); (b) STA-2 (n ¼ 4); (c) AlPO4-56 (n ¼ 5), and (d) AlPO4-5 (n  6)

Crystallization of Microporous Compounds

315

Figure 5.37 The location of the structure-directing agent [(C7H13N)(CH2)4(C7H13N)]2þ in the cage of STA-2

diammonium[76] or diamine[77] species with different lengths can direct to different microporous compounds (Table 5.15). Along with the increase of the chain length of the diammonium or diamine, microporous compounds ZSM-39, ZSM-12, and EU-1/ZSM-23 containing a 6-, 12-, and 10-membered ring, respectively, were synthesized. Ferrierite, ZSM-5, and ZSM-12 containing a 10-membered ring channel were prepared under the influence of diamines with different chain lengths as well. For the normal SDE, the presence of SDA is necessary but not sufficient for the formation of a specific structure. Under different gel compositions and crystallization conditions, the same SDA can direct to many different structures. For example, the framework of CoAPO (AFY) and SAPO-40 (AFR) could only be synthesized in the presence of nPr2NH and Pr4NOH, respectively. However, Pr4NOH can direct to AlPO4-5, MAPO-36, and ZSM-5 as well. AlPO4-20 (SOD) could be synthesized under the influence of TMAþ, while TMAþ is a common SDA in the synthesis of other microporous compounds.

Table 5.15 Zeolites directed by diammonium or diamine species Diammonium [(CH3)3N(CH2)nN(CH3)3]2þ

n

Product

Size of ring or channel/A˚2

3 4 5,6 7,8

ZSM-39 ZSM-12 EU-1 ZSM-23

6MR 12MR(5:5  5:9) 10MR(4:1  5:7) 10MR(4:5  5:2)

Diamine [H2N(CH2)nNH2]

n

Product

Size of ring or channel/A˚2

2–5 5–6 7–10

Ferrierite ZSM-5 ZSM-5 ZSM-11

10MR(4:2  5:4) 10MR(5:3  5:6; 5:1  5:5) 10MR(5:3  5:6; 5:1  5:5) 10MR(5:3  5:4)

316

Chemistry of Zeolites and Related Porous Materials

Different SDAs can direct to different structural units, cages, or channels. Therefore, a mixture of different SDAs may lead to the formation of new structures.[60] In fact, it was reported that mixed TPAOH and TMAOH directed to the framework of SAPO-37 (FAU). NMR analysis indicated that TMAþ and TPAþ were encapsulated in SOD cage and supercage, respectively. Based on these data, it could be inferred that TMAþ directed to the SOD cage and TPAþ directed to the supercage, which are the basic structural units of the framework of SAPO-37. SAPO-LTA could be synthesized under the influence of mixed SDAs as well. In this case, TMAþ, diethanolamine, and F were located in SOD cage; a cage and 8-membered ring channel; and the center of the double 4-membered ring, respectively. Obviously, these three structural units were directed by TMAþ, diethanolamine, and F, respectively. In some cases, the structure can be directed by a mixture of different SDAs. However, the framework of this structure contains only one of the mixed SDAs. For example, the mixture of Pr3N and TEAOH can direct to AlPO4-52, while the molecule of Pr3N was not included in the framework of AlPO4-52 (See Table 5.16 for details). However, the presence of Pr3N is very necessary for the formation of the structure of AlPO4-52. The role that Pr3N played in this case is not yet well understood. However, we could still infer that it played an assistant structure-directing role. The literature reporting the synthesis of microporous compounds in the presence of mixed SDAs is far from abundant. The possible reason is that it is very difficult to control the amount of each individual SDA in the mixture, which usually leads to the formation of impurities. For instance, impurities of SAPO-5 and SAPO-20 were always crystallized from the synthetic system of SAPO-37 in which the mixture of TPAOH and TMAOH was used as SDA. The gel chemistry in the synthesis of microporous compounds is very complicated. Important parameters include the pH of the system, the composition, the solubility of components, aging time of the gel, crystallization temperature and time, the concentration

Table 5.16 Structures directed by a mixture of SDAs Structure SAPO-37 AlPO4-52 SAPO-LTA GaPO-TREN SAPO-40 UiO-6

SDAs TPAOH TMAOH TEAOH Pr3N TMAOH diethanolamine F TREN pyrimidine F TPAOH TMAOH base TEAOH F Base

Location supercage SOD cage large cage SOD cage supercage D4R large channel small channel

Literature [77] [78] [79] [80]

large channel

[81]

large channel

[82]

Crystallization of Microporous Compounds

317

and the polymerization state of the precursors, etc. During the synthesis, the SDE of organic guest molecules (ions) could be significantly affected by these parameters. Organic amines (ammoniums) can only play their structure-directing roles under the appropriate gel chemistry. The fact that one SDA can direct to several different structures indicates that the gel chemistry has a significant influence on the directing ability of organic guest molecules (ions) (see Table 5.17 for details). For example, the SDA of TEAþ can direct to AlPO4-5 (12-membered-ring channel), mordenite (distorted 12membered-ring channel), and ZSM-8 (10-membered ring).[64] The shape of TEAþ in these structures is different in each. (3) Space-filling effect. Any guest molecules or ions located in the inorganic frameworks will play a space-filling role, which could stabilize the resultant structures. During the formation of high-silica zeolites, the surface of the crystal is hydrophobic. Therefore, the organic molecules in the synthetic system can partially enter the cage or channel of the zeolites, which could stabilize the inner hydrophobic surface and increase the thermokinetic stability of the frameworks. There are many examples showing that organic guest molecules play just such a space-filling role during the formation of the resulting structures. The most remarkable one is the synthesis of AlPO4-5.[60] So far, it has been reported that more than Table 5.17 Structures directed by the same SDA SDA Me4Nþ Et4Nþ Pr4Nþ Pr2N Pr3N NH2(CH2)2NH2

(CH3CH2)3N Cyclohexylamine Pyridine

Im C5H9NH2

Products AlPO4-33, erionite, SAPO-42, , offretite, RUB-10, ZAPO-M1 AlPO4-5, -18, b, BePO-H, SAPO-34, SAPO-42 AlPO4-5, MAPO-36, ZSM-5 AlPO4-18, -11, VPI-5 AlPO4-5, MAPO-36 AlPO4-12, A, A1, A2, A3, A4, [Al3P4O16][NH3(CH2)2NH3][OH2(CH2)2OH] [OH(CH2)2OH], [AlP2O8H][H3NCH2CH2NH3], [AlP2O8][H3NCH2CH2NH3]H3O, [AlP2O8][H3NCH2CH2NH3]NH4 AlPO4-5, JDF-20 [Al3P4O16][(CH3CH2)3NH]3 [AlP2O8H2][(CH3CH2)3NH] AlPO4-5, -17 [Al2P3O12H]2[C6H11NH3] [Al2P3O12H]2[C6H11NH3] AlPO4-21, -23 [Al2P3O12H2][PyH] [AlP2O10]2[PyH] [AlP2O8H]2[PyH] [Al3P4O16H]2[N2C3H5] [AlP2O8H2(OH2)2][N2C3H5] [Al2P3O12H]2[C5H9NH3] [Al3P5O20H]5[C5H9NH3]

Dimension of products 3-D 3-D 3-D 3-D 3-D 3-D 2-D 1-D 1-D 3-D 2-D 1-D 3-D 2-D 2-D 3-D 2-D 2-D 1-D 2-D 1-D 2-D 1-D

318

Chemistry of Zeolites and Related Porous Materials

85 organic guest compounds with totally different structures, shapes, and sizes could direct the framework of AlPO4-5. The smallest one is isopropylamine, and the biggest one is hexabutyl-1,6-hexanediamine. The following are some similar cases: more than 49 SDAs could direct the structure of MTW; at least 30 organic guest molecules could direct the structure of CHA, where the smallest amine is isopropylamine that contains 3 C atoms and the biggest amine is N,N,N-trimethyldiamantane that contains 13 C atoms. ZSM-5 and ZSM-48 could be directed by more than 22 and 13 SDAs, respectively.[84] In the synthesis of these microporous compounds, the SDE of organic molecules (ions) is relatively weak. Therefore, it is believed that these organic molecules (ions) mainly play a space-filling role. It should be noticed that the volumes (sizes) of these organic molecules (ions) are quite different, which should have different space-filling effects. However, all of them can direct to the same structure. In addition, oversized organic molecules such as hexabutyl-1,6-hexanediamine, which is much bigger than the diameter of the 12-membered-ring channel of AlPO4-5, will not be encapsulated in the cage or channel of the resulting framework. Therefore, it seems unreasonable to say that they play a space-filling role in the synthesis. In fact, the space-filling effect of the organic molecules (ions) is just used to explain the fact that many organic molecules (ions) significantly different in size and shape can direct to the same structure. Many other unknown interactions or reactions should exist during the synthesis process. For example, the structure of the organic molecules (ions) could be changed via isomerization, fragmentation, cyclization, degradation, or polymerization during the crystallization process of microporous compounds, which could change their structure-directing effect. (4) Charge-balancing effect. As described above, aluminosilicates, heteroatom-substituted zeolite molecular sieves, and open-framework aluminophosphates with nonunity Al/P ratio all have an anionic framework, which needs inorganic cations or organic ammoniums to balance the negative charge of the framework. It was proposed that the charge-density matching between the inorganic anionic framework and the organic cations is an important part of the templating effect of organic molecules (ions). Stucky and coworkers pointed out that the structure of an inorganic framework could reorganize to reach an appropriate charge density, which should be matched to that of the organic molecules (ions).[85] This process could be accomplished by (a) adjusting the curvature and charge of the framework surrounding the template through the creation of expanded or interrupted cages or, (ii) matching the framework charge with a template charge by changing the framework composition appropriately if tetrahedral atoms with different charges (for example, Al3þ and Co2þ) are made available during the assembly. They chose the second method to adjust the charge density of the inorganic framework. The matching of the charge density between the inorganic and organic species resulted in the successful synthesis of UCSB6, -8, and -10, three new metal phosphate frameworks with a multidimensional 12membered-ring system. Wilson and co-workers believed that the volume (size) and charge of the organic base played an important role in the synthesis as well.[86] Another remarkable example that reflected the charge effect of the SDA is the synthesis of AlPO411 under the influence of secondary amines.[87,88] Primary amines with sizes and shapes similar to those of secondary amines cannot direct to the structure of AlPO4-11 even under similar crystallization conditions.

Crystallization of Microporous Compounds

319

(5) Other effects. The organic species also has the following effects: (a) acting as suppressing agent to prohibit the formation of some specific structures. For example, the addition of hexamethyl-1,6-hexamethylenediamine to the synthetic system of ZSM-5 promoted the formation of mordenite instead of ZSM-5; (b) avoiding the introduction of inorganic cations. For example, many microporous compounds could only be synthesized in the presence of organic amine (ammonium); and (c) acting as complexing agent. The complexation of organic species to framework atoms could increase their solubility, making it easier to enter the framework. 3. The SDE in the synthesis of extra-large microporous compounds The ring number of the channel or pore system in extra-large microporous compounds is greater than 12. So far, more than ten extra-large microporous compounds have been synthesized (Table 5.18). However, these compounds in phosphate form have low thermal and hydrothermal stability, which limits their further application in industry. Recently, the successful synthesis of extra-large microporous silicate using special SDAs of large size and rigidity has attracted the interest of scientists working in this field. Some remarkable examples include UTD-1[89] and CIT-5[72,73] which contain a 14-membered ring in the framework. Therefore, it could be concluded that the SDA has a great influence on the synthesis of extra-large microporous compounds. The data in Table 5.18 show that: (1) the structures of the SDAs used in the synthesis of extra-large microporous aluminosilicates are very special. Most of them are large and have high rigidity, such as N(16)-methylsparteinium and [(Cp*)2Co]OH; (2) the SDAs used in the synthesis of extra-large microporous metal phosphates are ring-like amine or  chain-like multiamines, which could easily form H-bonds with P OH and P  O group in the inorganic frameworks. These H-bonds can stabilize the resulting structure. However, the removal of these organic amines usually leads to collapse of the frameworks; (3) small guest molecules could stay together to play a structure-directing role in the

Table 5.18 Extra-large microporous compounds and their corresponding SDAs Type material AlPO4-8 (AET) VPI-5 (VFI) Cloverite (CLO) JDF-20 ULM-5 UTD-1 (DON) ULM-16 CIT-5 (CFI) ND-1 FDU-4 NTHU-1 MIL-31

Ring size

Framework elements

14 18 20 20 16 14 16 14 24 24 24 18

Al, P Al, P Ga, P Al, P Ga, P Si, Al Ga, P Si Zn, P Ge, O Ga, P Ga, P

SDA DPA DPA, TBAOH quinine or piperidine, F TrEA 1,6-DHA, F [(Cp*)2Co]OH CPA, F N(16)-methylsparteinium DACH N(CH2CH2NH3)3 DETA H2N(CH2)9NH2, H2N(CH2)10NH2,

320

Chemistry of Zeolites and Related Porous Materials

synthesis of some extra-large microporous compounds. For example, JDF-20 contains four triethylamine molecules in its 20-membered-ring channel (see Figure 5.38 for details).[90] However, other amines with a size and shape similar to these four triethylamine molecules could not direct the framework of JDF-20. Similar examples include cloverite[91] which contains 8 quinuclidine molecules in its 20-membered-ring channel; ULM-5[92] contains 4 hexanediamine molecules; UML-16[47] contains 6 cyclopentylamine molecules; and MIL-31[93] contains 4 decamethylenediamine molecules. So far, this phenomenon was only observed in the structure of extra-large phosphates; (4) F could play a co-directing role in the synthesis of extra-large microporous gallophosphates. F ion F was used for the first time as mineralizer in the synthesis of zeolites under neutral or acidic conditions by Flanigen and Patton[94] Subsequently, Guth and Kessler further studied this new synthetic method.[95] Many high- or all-silica zeolites such as ZSM-5, ZSM-23, Theta-1, ferrierite, b, MTN, AST, UTD-1, ITQ-3, ITQ-4, and heteroatom (B, Al, Fe, Ga, and Ti)-substituted high-silica zeolites could be synthesized from an F synthetic system. In addition, most of the open-framework gallophosphates could be hydrothermally synthesized under the influence of F. In the synthesis of microporous compounds, F usually plays several roles as described below: (1) enabling crystallization to occur at low pH (3  10); (2) helping transition metal atoms enter the framework via the formation of a complex, which will lead to a higher content of heteroatom in the zeolite framework; (3) balancing the positive charge of SDA, which could reduce the framework defects created by the excessive positive charge of SDA; (4) changing the gel chemistry of the system: For example, TMAOH could direct to AlPO4-20 (SOD). The addition of HF to the same system results in the formation of UiO-7;[96] (5) stabilizing small structural units. In the framework of all- or high-silica zeolites, F was usually located in the relatively small cage. For example, F was located in cage [46], [415462], [435261], [4354], and [415262] for the structures of AST,[97] NON,[98] ITQ-4 (IFR),[99] SSZ-23 (STT),[100] and MFI,[101] respectively (see Figure 5.39 for details). Interestingly, at least one 4-membered ring was

Figure 5.38

20-Membered-ring channel of JDF-20 and 4 TEAþ

Crystallization of Microporous Compounds

321

Figure 5.39 The location of F in (a) AST [46]; (b) NON [415462]; (c) IFR [435261]; (d) STT [4354]; and (e) MFI [415262]

found in each of these 5 cages and F was located at a position close to this 4-membered ring. Therefore, it was suggested that F favors the formation of high-density frameworks which contain a small cage or 4-membered ring.[102] In addition, F was usually located in the double 4-membered ring of the gallophosphates such as CLO, LTA of gallophosphate, MU-1, MU-7, MU-2, and MU-15. Metal Complex In 1992, Balkus et al. successfully synthesized zeolite faujusite molecular sieve by using phthalein cyanogen dye as SDA.[103] Subsequently, they synthesized many molecular sieve structures by using dicyclopentadienylmetal and related derivatives as SDAs.[104,105] For example, they successfully synthesized AFI and AST by using Cp2Coþ as SDA and cloverite and UTD-10 by using [Co(NMe3)2sar]5þ (sar is 1,8-bis(trimethylammonio)3,6,10,13,16,19-hexaazabicyclo [6.6.6] jcosane) as SDA. The most remarkable example is the synthesis of UTD-1, which contains a 14-membered-ring channel, under the influence of cobaltocene.[89] In addition, some stable cobalt amine complexes have been used as SDAs, resulting in the formation of many new layered phosphate structures.[106–109] Metal complexes after consist of metal cations and organic ligands, which endues some special properties to the metal complex. However, only some of them, such as Cp*2Coþ, Cp2Coþ, Co(en)33þ, Co(NH3)63þ, and Co(tn)33þ (tn is propane-1, 3-diamine), could be used as SDAs in the synthesis of microporous compounds due to their general poor stability. Metal complexes mainly play the roles described below in the synthesis of microporous compounds: (1) balancing the charge of the framework; (2) filling the void and supporting the inorganic framework; (3) directing the structure (some special chiral metal complexes can direct to chiral structures. For example, trans-Co(dien)23þ and dCo(en)33þ directed the structure of trans-Co(dien)2 . Al3P4O16 . xH2O[108] and d-Co(en)3. Al3P4O16 . 3H2O,[109] respectively); and (4) transferring chirality of the metal complex to

322

Chemistry of Zeolites and Related Porous Materials

Figure 5.40 Enantiomorph of chiral metal complexes and inorganic chiral structural units in compound 1

the inorganic framework. [110] For example, a racemized cobalt-amine complex directed new zinc phosphates of CoII(en)3.Zn6P8O32H8 1 and CoIII(en)3.Zn8P6O24Cl.2H2O 2. [107] Structure analysis indicated that the enantiomers of Co(en)3Cl3 were separated as
and  configurations in the structures of both complexes 1 and 2, the chiral characters of which were believed to be transferred from the chiral metal complex. The inorganic layer of compound 1 features cap-like chiral structural units. It should be noted that the symmetry of the chiral cobalt-amine complex is D3 point group, while the cap-like chiral structural units have C3 symmetry, which is a sub-group of D3. In addition, the enantiomorph of the chiral structural units was found in the framework of compounds 1 and 2 (Figures 5.40 and 5.41). These results indicated that the chirality of the metal complex was transferred to the chiral structural units of the inorganic framework. Water, Anions, or Salts Water could play a structure-directing role in the synthesis of microporous compounds, but such examples are very rare. The most remarkable one is the synthesis of VPI-5 by using dipropylamine as SDA.[111] However, chemical analysis indicated that the organic amine content in the framework of VPI-5 is very small (about one dipropylamine molecule/2.5 cells). The channels of VPI-5 were filled with a large number of water molecules instead of dipropylamine molecules. There are 7 water molecules in one cell of VPI-5. Two of them (I, II) coordinated to the Al atom in the framework, and four of them (III–VI) formed three helical water molecule chains in the 18-membered-ring channel of VPI-5 via intermolecular H-bonds. These three helical water molecule chains connected all 6-coordinated Al atoms. The last water molecule (VII) was located in the center of the channel. Therefore, it was proposed that these water molecules played a structure-directing role in the creation of the framework of VPI-5. The synthesis of VPI-5

Crystallization of Microporous Compounds

323

Figure 5.41 Enantiomorph of chiral metal complexes and inorganic chiral structural units in compound 2

without the presence of any organic amines has not been reported so far. In addition, the low content of organic amine in the framework of VPI-5 made it difficult to study the SDE in the synthesis of VPI-5. Maybe the framework of VPI-5 was co-directed by water and organic amine molecules, but this needs to be confirmed by further experimental results. Olson pointed out that water molecules played an important role in the crystallization of high-alumina molecular sieves such as zeolite A and X.[112] In these cases, an instant water cluster could act as a template. Recent experimental results confirmed that dimers or tetramers of water molecules could be formed around the hydrophilic sites of the framework. Moreover, in the solvothermal synthesis of aluminophosphates, the addition of water to the synthetic system will affect the solubility of aluminum source and the hydrolysis of inorganic species. Sometimes the amount of water in the system could affect the formation of the final products. For example, for the Al2O3–P2O5–triethylaminetetraethylene glycol synthetic system, along with the increase of water, chain aluminophosphate of [AlP2O8H2], JDF-20, layered aluminophosphate, and AlPO4-5 were crystallized. Ozin concluded that the form of Al and P species existing in the solution is a chain-like structure with the composition [AlP2O8H2].[46,113] Along with the increase of water, [AlP2O8H2] started to hydrolyse and a new structure formed. The hydrolysis degree of this chain depended on the amount of water. Usually, the anions introduced by the addition of SDA would not significantly affect the synthesis of aluminosilicate zeolites. A few studies reported that some oxyacid anions could promote the nucleation and crystallization of zeolite. For example, it was reported that the addition of inorganic salts to the synthetic system affected the crystallization time for zeolite ZSM-5 and TS-1.[114] Inorganic salts sometimes acted as SDAs as well. For instance, cations and anions of the added inorganic salts were included in the SOD and CAN cages of zeolite sodalite and cancrinite, respectively. In the synthesis of zeolite A, the excessive salts from Al source could promote the formation of sodalite.

324

Chemistry of Zeolites and Related Porous Materials

The above discussion indicates that guest molecules or ions play different roles under different synthetic conditions. However, in most cases, it is very difficult to tell exactly the roles of inorganic or organic guest molecules in the synthesis. A large number of experiments still cannot be clearly interpreted. 5.3.2

Studies on the Interaction between Inorganic Host and Guest Molecules via Molecular Simulation

In recent years, many studies have been carried out to examine the nonbonded interaction between inorganic host and organic guest molecules via molecular simulation. For example, Stevens’s studies indicated that the surface shape and size of the SDAs played an important role in the creation of pore structure of the products. In some cases, these parameters are even more important than charge, deformability, and functionality of SDAs.[115] Zones et al.[116] tried to understand the role of organic templates with various sizes in the synthesis of high-silica molecular sieves by combining molecular simulation and experimental approaches. Their results showed that there is good correlation between the ‘appositeness’ of the energy of the SDA and the crystallization rate. In addition, three new microporous compounds, SSZ-35, SSZ-36, and SSZ-39, were successfully synthesized by using rigid SDAs designed via computer simulation. Lewis and Catlow and colleagues studied the roles of the SDAs by combining several computer simulation techniques. They first calculated the location and stability of the SDAs in the framework of ZSM-5, ZSM-11, b, EU-1, and ZSM-23 by combining Monte Carlo and energy-minimization methods.[117] By using this method, they studied the influence of the SDAs on the content of Co2þ in the framework of Co-AlPO4-5 and CoAlPO4-34.[118] They created a computer modeling approach for the structure and synthesis of microporous and mesoporous materials[119] as well as for nucleation, growth, and templating in hydrothermal synthesis.[120] Then they developed a method to perform de novo design of SDAs for the synthesis of microporous solids and wrote a program named ZEBEDDE (Zeolite By Evolutionary De Novo Design)[121] for the application of de novo techniques to microporous materials. According to this method, a molecule seed or primary fragment will be first placed (randomly or specifically) in the target area (the host). After that, the molecule or primary fragment grows by a number of random replacements of H atom by C or N atoms and the configuration of the resulting molecule is optimized by energy minimization. By using this method, they designed an organic template for a small-pore microporous material and successfully synthesized this structure with the designed organic molecule.[122] In addition, Cox et al. studied the roles of the SDAs in the synthesis of microporous materials by using CVFF (Consistent Valence Force Field) force field and the Biosym Discover program as well. Their studies mainly focused on the correlation between the structure of molecular sieves and that of the organic molecules, and revealed that the configuration of the channel of the final product was greatly affected by the shape of the organic molecules used in the synthesis.[123] They predicted the location and structure of the SDAs in the framework and on the surface of growth by using Monte Carlo and simulated anneal (MC-SA) methods.[124] Furthermore, they analysed the structure of the 160 SDAs used in the synthesis of 27 framework types of molecular sieves by using the TsarQSAR software (Quantitative Structure-Activity Relationship), concluding that

Crystallization of Microporous Compounds

325

there exists a clear correlation between the shape of the SDAs and the zeolite products.[125] Davis and coworkers designed a series of organic cations to synthesize high-silica molecular sieves.[126] They believed that the structure-directing ability of the organic cations they designed is related to their geometric configuration and hydrophobicity. The hydrophobicity of these SDAs in their iodide form was evaluated in terms of their phasetransfer behavior from water to chloroform, while their configuration was evaluated in terms of the number of tertiary and quaternary connectivity. When the SDA is extremely hydrophobic, it is difficult to obtain a molecular sieve. For hydrophobic SDAs, the introduction of a second charge into the molecule decreases the hydrophobicity and allows for structure-direction in molecular sieve synthesis. Thus, SDAs with intermediate hydrophobicity are found to be most useful for high-silica molecular sieve synthesis. In terms of SDA geometry, bulky, rigid molecules with limited conformational variability result in the formation of a great variety of new high-silica molecular sieves. We systematically studied the host–guest interaction of aluminophosphates with an Al/P ratio lower than unity by combining molecular kinetics, molecular mechanics, and Monte  Carlo methods.[127] There existed a large number of terminal P OH groups in  O or P the frameworks of aluminophosphate with an Al/P ratio lower than unity, which made the H-bond interaction between the inorganic framework and organic guest molecules much stronger than van der Waals’ interaction. Compared with the free state, the shape of the organic SDA molecules was usually changed due to the H-bond interaction. Based on our calculation of the interaction, we can rationally predict the possible SDAs for a given framework structure. Recently, Sastre et al. studied the role of cyclohexyl(alkyl)pyrrolidinium salts in the synthesis of Eu-1, ZSM-11, and ZSM-12 by combining computational and experimental approaches.[128] Their studies showed that the short-distance interaction between the inorganic host and organic guest (charged) played an important role at the nucleation stage. In the synthesis in which the N-methyl-N-cyclohexylpyrrole cation was used as SDA, along with the increase in Al content, the product was ZSM-12, EU-1, and b molecular sieve, respectively, which results from the interaction between the inorganic framework and SDAþ. The above-described studies show that the host–guest nonbonding interactions including static electricity, van der Waals’, and H-bonding interactions play extremely important roles in the synthesis of microporous materials. For the aluminosilicate zeolites, the main host–guest interactions are static electricity (dominant) and van der Waals’ force; for high- or all-silica zeolites with nearly neutral frameworks, the main host–guest interaction is van der Waals’ force; for microporous aluminophosphates with neutral frameworks, the main host–guest interaction is van der Waals’ force too, while for the anionic framework of aluminophosphates, the H-bond interaction between the inorganic host and the organic guest is much stronger than van der Waals’ interaction. 5.3.3

Conclusions and Prospects

Studies on the roles of templates or SDAs in the synthesis of microporous compounds have always attracted the attention of chemists in this field. Choosing an appropriate SDA for a specific framework is the key issue in the synthesis of new microporous structures.

326

Chemistry of Zeolites and Related Porous Materials

The roles of guest molecules (ions) such as metal cations, organic compounds, F, metal complexes, water, anions, and salts in the synthesis of microporous compounds have been summarized in this section. The dominant interactions, i.e., static electricity, van der Waals’, and H-bonds, between inorganic host and guest molecules (ions) varies with the synthetic system. Recently, studies on the calculation of the host–guest interaction and the rational design and screening of the SDAs assisted by molecular simulation have become a new hot topic. A large number of positive results have been achieved, which opens up a new direction for studies on the roles of SDAs in the synthesis of microporous materials. On the other hand, the roles of guest molecules or ions in the synthesis of microporous compounds and the crystallization mechanism of the resulting framework have not been well understood so far. Most of the present studies on this topic focused on the correlation between the structure of the final products and that of the SDAs. Studies on the roles of guest molecules (ions) in the gelation, nucleation, and crystallization process are still rare. Some experiments cannot be interpreted by the known templating or SDEs of guest molecules or ions. The use of the latest experimental analysis and detection techniques including liquid/ solid NMR, XRD, electron diffraction, and in situ analysis is very necessary for the studies on the roles of the SDAs in the crystallization process of microporous materials, which could help us gain a better understanding of the formation mechanism of the pore systems and reveal the real correlation between guest molecules or ions and the resultant frameworks. The above discussion indicates that the reaction conditions have a significant influence on the SDE of guest molecules or ions. The SDAs can only play their SDE under appropriate gel conditions. Therefore, studies on the roles of guest molecules or ions by applying advanced computational techniques such as data mining, statistical analysis, and neutral networking to the improved synthetic database of the microporous compounds should include other factors such as reaction temperature, time, gel composition, and pH. In addition, the SDE of guest molecules or ions is different in various synthetic systems. Compared with microporous aluminophosphates and all-silica zeolites with neutral frameworks, the frameworks of aluminosilicate zeolites are charged. The matching of the charge density between inorganic host and guest molecules or ions obviously plays an important role in the creation of the resulting framework. Moreover, the hydrophobicity of the frameworks of aluminosilicate zeolites will affect the SDE of  guest molecules or ions as well. The terminal P OH or P  O groups in the framework of aluminophosphates with an Al/P ratio less than unity will have a strong H-bonding interaction with the protonated organic amines. Therefore, investigation into the SDE of guest molecules or ions for different synthetic systems could provide us with a better understanding of the roles of SDAs.

5.4 Crystallization Kinetics of Zeolites Crystallization of zeolites under the usual conditions is distinguished by polycrystal growth. Therefore, a discussion of the crystallization of zeolites from macro-statistics

Crystallization of Microporous Compounds

327

analysis is very important for both theoretical and practical studies on zeolites. So far, studies on the crystallization kinetics of zeolites are mostly based on the description of the crystallization curves. The crystallization curve of the spontaneous system features an S shape, i.e., no X-raydiffraction-detectable crystals exist in the synthetic system at the early stage of the crystallization period for a relatively long time, which is called the induction period; subsequently, the crystallization rate gradually accelerates, which is called the autocatalytic process; finally, the crystallization rate gradually decreases at the last stage of the crystallization period, and the crystallization curve shows a gentle incline. Most of the kinetics equations established at the early stage of the studies were based on elementary functions. Therefore, the form of these equations varied from one study to another. Because the establishment of these equations was not based on the theoretical model, and most of them were represented by elementary functions, the theoretical and experimental crystallization curves just partially matched each other. Later, some crystallization kinetics equations based on designed theoretical models were established as well. Recently, the crystallization process was further simulated by the computational modeling approach. The crystallization kinetics described in this section is mainly based on the liquid-mediated transformation mechanism, which could be schematically represented by Figure 5.42. The data for the concentration change of the components and crystal growth could be experimentally obtained. Since the early 1980s, when the crystallization of zeolites started to be systematically studied, no appropriate measurement techniques have been available in the quantitative study of ‘nucleation’. All quantitative studies on the nucleation rate at the early stage were based on calculation of the parameters of the products. For example, Zhdanov calculated the nucleation rate according to the particle-size distribution of the final product.[129] Later, theoretical models used to simulate the crystallization process were proposed. The crystallization rates calculated according to these models were compared with the experimental data and the theoretical models were accordingly modified. The following are two main results achieved in the early 1980s. The first one was conducted by Zhdanov and Samuelevich in 1980.[129] They experimentally measured the crystal growth curve and the particle-size distribution curve (Figure 5.43)

Figure 5.42 Schematic presentation of the nucleaton rate, crystal growth, and nutrient concentration in zeolite synthesis. Reprinted from Nikolakis et al. [44]. Copyright (1998) Elsevier

328

Chemistry of Zeolites and Related Porous Materials

Figure 5.43 Crystallization of zeolite NaX. 1: crystal-growth curve; 2: particle-size-distribution curve. Reproduced with permission from [129]. Copyright (1980) John Wiley & Sons, Ltd

of zeolite NaX crystallized from the synthetic system 3.72 Na2O . Al2O3 . 3.5 SiO2 . 542 H2O at 90
C. The nucleation-rate curve of this crystallization system was accordingly determined (Figure 5.44). The second one is the solution-mediated phase-transition mechanism model established by Xu and coworkers at Jilin University in 1985 by investigating the crystallization of heteroatom-substituted Si-ZSM-5.[130] They quantificationally studied the crystallization kinetics of this system by using the mathematical calculation described below. Let g1, g2, and g3 be the molar fraction of zeolite crystal, the corresponding liquid phase, and the amorphous solid gel in the synthetic system, respectively, then, g1 þ g2 þ g3 ¼ 1. Let the saturated solubility of amorphous solid gel in the unit volume be c1 and that of the zeolite crystal be c0, then c1  g2  c0 is always true during the whole crystallization process, while (c1  g2 ) is one of the driving forces for the dissolution of solid gel and (g2  c0 ) is one of the driving forces for the nucleation and crystal growth. Compared with g2, the solubility of zeolites c0 is much smaller, which could be neglected in the calculation. Therefore, (g2  c0 ) could be replaced by g2 in the calculation. Let Z* be the number of nuclei, t be time of nucleation, and a and k1 be constants; then, the equation of nucleation rate could be represented as Equation (5.8). dZ ¼ k 1 t a g2 dt

ð5:8Þ

Crystallization of Microporous Compounds

329

Figure 5.44 Crystallization and nucleation of zeolite NaX. 1: crystal-growth curve; 2: nucleation curve; 3: crystallization curve. Reproduced with permission from [129]. Copyright (1980) John Wiley & Son, Ltd

To simplify the calculation, we introduce the assistant function Z. Given that Equation (5.9) holds: Z ¼ k11 Z

ð5:9Þ

Then we get Equation (5.10): dZ ¼ t a g2 ð5:10Þ dt Integrating Equation (5.10) from t0 , the beginning of nucleation, to time t gives Equation (5.11): ðt ð5:11Þ ZðtÞ ¼ ta g2 ðtÞdt 0

Let the radius and mass of the nucleus created at time t be rðt, tÞ and qðt, tÞ at time t, respectively, then, Equation (5.12) holds: qrðt; tÞ ¼ k 2 g2 qt Integrating Equation (5.12) from t to t gives Equation (5.13):

ð5:12Þ

ðt g2 ðsÞds

rðt; tÞ ¼ k2 t

ð5:13Þ

330

Chemistry of Zeolites and Related Porous Materials

Given that qðt; tÞ ¼ k3 rðt; tÞ3 , where k3 is a constant related to the geometric shape and density of the crystal, then, the increasing rate of crystal mass could be represented as Equation (5.14): qqðt; tÞ qrðt; tÞ ¼ 3k3 rðt; tÞ2 qt qt

ð5:14Þ

The mass of the nucleus of the zeolite in the period from t to ðt þ dtÞ could be represented by Equation (5.15): qðt; tÞk1 ta g2 ðtÞdt

ð5:15Þ

Then, the crystal-growth rate could be written as Equation (5.16): 2 ð t ð t dg1 ðtÞ 3 3 ¼ P k1 k2 k3 g2 ðsÞds g2 ðtÞta g2 ðtÞdt ð5:16Þ dt M ni t0 t P where M is the molecular weight of the resultant zeolite; ni is the sum of the mole quantities of the zeolite crystal, the corresponding liquid phase, and the amorphous solid gel per unit volume. Let Equation (5.17) hold: 3 K ¼ P k1 k23 k3 ð5:17Þ M ni where K is the crystal-growth constant. To simplify the calculation, we introduce assistant functions y1ðtÞ and y2ðtÞ, defined by Equations (5.18) and (5.19): ð t 2 ðt a y1 ðtÞ ¼ t g2 ðtÞ g2 ðsÞds dt ð5:18Þ t

t0

ðt y2 ðtÞ ¼ t0

ta g2 ðtÞ

ð t

 g2 ðtÞds dt

ð5:19Þ

t

then, Equations (5.20)–(5.22) hold: dg1 ðtÞ ¼ Kg2 ðtÞy1 ðtÞ dt dy1 ðtÞ ¼ 2g2 ðtÞy2 ðtÞ dt dy2 ðtÞ ¼ g2 ðtÞZðtÞ dt

ð5:20Þ ð5:21Þ ð5:22Þ

Thus, the dissolution rate of the solid gel could be represented as in Equation (5.23): 2 dg3 ðtÞ ¼ k4 ðc1  g2 Þð1  g1  g3 Þ3 ð5:23Þ dt where k4 is the gel-dissolution constant. Because the sum of g1, g2, and g3 is 1 (i.e., g1 þ g2 þ g3 ¼ 1), then Equation (5.24) holds:



2 dg2 ðtÞ dg1 ðtÞ dg3 ðtÞ ¼  ¼ kg2 ðtÞy1 ðtÞ þ k4 ðc  g2 Þð1  g1  g2 Þ3 dt dt dt

ð5:24Þ

Crystallization of Microporous Compounds

331

Finally, a group of differential equations can be obtained [Equations (5.24)–(5.29)]: dg1 ¼ Kg2 y1 dt 2 dg2 ¼ k4 ðc  g2 Þð1  g1  g2 Þ3  Kg2 y1 dt dz ¼ t a g2 dt dy1 ¼ 2g2 y2 dt dy2 ¼ g2 Z dt

ð5:25Þ ð5:26Þ ð5:27Þ ð5:28Þ ð5:29Þ

Boundary conditions are as in Equation (5.30): g1 ðt0 Þ ¼ Zðt0 Þ ¼ y1 ðt0 Þ ¼ y2 ðt0 Þ ¼ 0 g2 ðt0 Þ ¼ c

ð5:30Þ

The theoretical crystallization kinetics curves calculated by employing the simplex gradient algorithm were consistent with those experimentally obtained for B-, Al-, and Ga-Si-ZSM-5 zeolites as shown in Figure 5.45, and for Ti(III), V(III), Cr(III), Fe(III)-SiZSM-5 zeolites as shown in Figure 5.46. The calculated crystallization kinetics parameters for various M-Si-ZSM-5 zeolites and the SiO2/M2O ratio of the corresponding zeolites are summarized in Table 5.19. The data in Table 5.19 showed that B-Si-ZSM-5 has a remarkable nucleation rate constant k1, which might be caused by the similarity of the properties of B and Si. Along with a decrease in the ionic radii of subgroup elements M(III), the nucleation rate

Figure 5.45 Crystallization kinetics curves of B-, Al-, and Ga-Si-ZSM-5 zeolite. Reproduced with permission from [130]. Copyright (1985) Chem. J. Chin. Univ.

332

Chemistry of Zeolites and Related Porous Materials

Figure 5.46 Crystallization kinetics curves of Ti(III)-, V(III)-, Cr(III)-, and Fe(III)-Si-ZSM-5 zeolite. Reproduced with permission from [130]. Copyright (1985) Chem. J. Chin. Univ.

constants correspondingly decreased due to the possible condensation reaction shown in Equation (5.31): (5:31) The crystal-growth rate constants K of the main-group-element-substituted isomorphous zeolites is much higher than those of the subgroup-element analogs. This may be related to the dissolution of the amorphous gel. The data in Table 5.19 showed that the amorphous gel of the main-group-element-substituted-isomorphous zeolites have a much higher dissolution rate constant than do the subgroup-element-substituted zeolites. In a word, the formation process of zeolites is very complicated. The crystallization models and kinetics equations proposed during early studies are not very meticulous.

Table 5.19 Crystallization kinetics parameters and the SiO2/M2O ratio in the final products Zeolite M (M-Si-ZSM-5) B Al Ga Ti(III) V(III) Cr(III) Fe(III) a

Nucleation rate constant ka1

Crystal growth-rate constant K

Gel-dissolving rate constant k4

SiO2/M2O (mol/mol)

0.86 0.32 0.19 0.32 0.44 0.29 0.15

102.02 155.39 163.38 31.71 18.82 18.70 28.43

224.56 263.67 218.96 37.55 39.63 39.08 39.27

78b 42 76 96 86 80 66

K was obtained by the approximate calculation based on Equation (5.8). The percentage of B2O3 was obtained via chemical analysis.

b

Crystallization of Microporous Compounds

333

Nucleation and crystal growth are two main steps in the formation of zeolites. Crystal growth has been deeply studied, and well accepted points of view and conclusions have been obtained. However, theoretical and experimental studies on nucleation are still under development. Over the past 20 years, quantitative studies on the nucleation and crystallization of zeolites and other microporous materials have been improved in two aspects due to the development of in situ detection techniques and a deepening understanding of the crystallization process of zeolites. First, mathematical models and algorithms for the description of the nucleation and crystallization processes have been improved based on an in-depth understanding of the crystallization process of zeolites. For example, an S-shape time function was used to describe the crystallization kinetics curve of zeolites during the early stage. In 1992, by using a group of ordinary differential equations and the population-balance method, Subotic and Bronic studied the autocatalytic nucleation mechanism in the synthesis of zeolite A.[131] In 1994, Budd et al. described the appearance of the maximal nucleation rate in the crystallization of zeolite A by using the function shown in Equation (5.32).[132] In 2001, based on previous studies, Bronic and coworkers further improved the crystallization models and related algorithms for the population-balance method.[37,133] All these efforts have greatly improved the modeling and simulation studies, as well as the quantitative description, of the crystallization process of zeolites and related microporous materials. ZðxÞ ¼ S½xm  ðxm =tnm0  mÞ

ð5:32Þ

Secondly, the influence of the microstructure of the synthetic system on the nucleation and crystallization of zeolites has been gradually taken into consideration in the studies; for example, factors such as pore structure and components of the gel, the nutrient in the liquid phase within the gel, and degree of supersaturation. For example, Tsapatsis and coworkers studied the role of gel microstructure in crystallization kinetics in 1998. In their studies, they proposed that the precursor gel has a hierarchical structure involving micro-, meso-, and macro-pores, as shown in Figure 5.27. Zeolite nucleation takes place at the interface between the solution and the gel by adsorption and rearrangement of soluble precursors. Therefore, it is expected that different pore sizes will not contribute equally to nucleation and crystal growth. In order for crystal growth to occur, the nucleus needs to exceed a certain critical size. Therefore, it is expected that pores which are too small to accommodate a zeolite nucleus will not contribute to surface-catalysed heterogeneous nucleation. SAXS (Small Angle X-ray Scattering), TEM (Transmission Electron Microscope), and HRTEM (High Resolution Transmission Electron Microscope) analyses point to sizes of precursor entities of the order of 5  10 nm, providing a lower limit for the pore size of active porosity involved in nucleation and growth. Large pores, especially those exceeding several micrometers in diameter, are expected to play a minor role in interfacial nucleation owing to their small surface area compared with that of mesopores. Therefore, the most suitable pore size in favor of nucleation and crystallization is slightly bigger than that of the critical size of the nucleus (7 nm). After a nucleus exceeding the critical size has been formed, it continues to grow by a solutionmediated mechanism. Nucleation and crystal growth consume solution species while replenishment of the consumed solution species is provided by gel dissolution, which is

334

Chemistry of Zeolites and Related Porous Materials

also considered to be an interfacial process. Based on this physical picture and the population-balance equations, Tsapatsis proposed a mathematical model for the nucleation and crystallization of zeolites, as follows: (1) If r is the velocity with which a solid-gel surface element recedes owing to dissolution, then the consumption rate of the porous gel is given by Equation (5.33),   1=2 dEg B0 Eg ¼ 4pEg B1 r 1  ln ð5:33Þ dt Eg0 2pB21 where Eg0 and Eg are the initial and in situ fraction of the solid gel, and B1 and B0 are parameters of the initial gel pore structure. For a discrete pore size distribution, Equations (5.34) and (5.35) hold, X B0 ¼ li ð5:34Þ X B1 ¼ li Rio ð5:35Þ where li is the surface density of intersections of the axes of capillaries with radius Rio with a fixed plane. (2) The crystal growth is assumed to be linear with respect to the nutrients’ concentration, and to be independent of the crystal size, and so we can write Equation (5.36),   G zeol Q ¼ k1 v  Geq ð5:36Þ Ep

where k1 is the linear crystal growth constant, v is the zeolite molar density, and GEp and G zeol are the nutrients’ concentration and zeolite equilibrium concentration, respectively. eq (3) The nucleation rate is assumed to be proportional to the gel surface area and to the first power of the supersaturation of the nutrients’ concentration, as shown by Equation (5.37),   G BðtÞ ¼ k3  G zeol ð5:37Þ 4pEg ðB1 þ B0 qÞ eq Ep where k3 is the nucleation rate constant. (4) Thus, the changing rate of degree of supersaturation of the nutrients in the solution 3 can be represented as Equation (5.38), dm dt       dðG =Ep Þ 1 dEg G v0 n dm0 G v 1 G v ¼ 1  1  1    dt Ep v0 dt Ep dt Ep Ep v Ep Ep

ð5:38Þ

where m0 and m3 describe the number and crystal volume of zeolite particles, respectively. The above equations were solved using a Runge–Kutta method. Table 5.20 gives the parameter values used for the simulation that gave the computational results shown in Figure 5.47. According to the calculated nucleation and crystal-growth curves as shown in Figure 5.47(a), the nutrient concentration and gel consumption curves in Figure 5.47(b),

Crystallization of Microporous Compounds

335

Table 5.20 Values of the parameters used for the simulations of Figure 5.47 k1/(m s1) k2/(m s1) k3/(m s1) Geq of gel/ (mol m3) Geq of zeolite/ (mol.m3) Bidisperse pore-size distribution with total porosity (%) Average micropore size/nm Average macropore size/nm Calculated ‘active porosity’/% Calculated B0 =m2 Calculated B1 =m1 Nucleus size/nm n/(m3 mol1) n0 /(m3 mol1)

3:2  108 1:3  107 2:7  105 218 185 50 3.28 (90% of total porosity) 36.8 (10% of total porosity) 9 7:1  1013 8:0  105 7 2:9  105 1:3  104

Figure 5.47 Simulation results for (a) nucleation rate and crystallinity, (b) nutrient concentration and gel consumption, and (c) normalized nucleation rate and surface area. Reprinted from Nikolakis et al. [44]. Copyright (1998) Elsevier

336

Chemistry of Zeolites and Related Porous Materials

Figure 5.48 Effect of changes in gel microstructure (f), for 50% total porosity, on (a) gel consumption, (b) nutrient concentration, (c) crystallinity, (d) nucleation rate, and (e) particle size distribution. Reprinted from Nikolakis et al. [44]. Copyright (1998) Elsevier

and normalized nucleation rate and surface area in Figure 5.47(c), it can be concluded that the increase in the nucleation rate under constant supersaturation at the early stages of zeolite crystallization from precursor gels (30% of the gel has been converted into zeolite) can be partly due to a heterogeneous, solid-gel surface-catalysed nucleation mechanism, accompanied by an increase of the interfacial surface area involved. Figure 5.48 illustrates the effect of the active porosity of the gel on nucleation and crystal growth. As shown in Figure 5.48(f), the active porosity mainly refers to pores larger than the nucleus critical size (7 nm). The gels containing three different percentages of active porosity (9, 3, and 1%, respectively) are shown in Figure 5.48(f) and only the interfacial surface area created from the active pores was calculated during the simulation. As seen in Figure 5.48, even though the steady-state supersaturation value is not affected, there is a decrease or delay in the maximum nucleation rate as the active porosity decreases due to the decrease of the interfacial area available for nucleation. The overall time for the gelto-zeolite transformation increases with the decrease in active porosity. However, the crystal size is not affected much; only a relatively small increase in the size is observed with a decrease of the active porosity due to a corresponding decrease of the overall nucleation

Crystallization of Microporous Compounds

337

rate. The difference between the initial and maximum nucleation rates becomes more obvious with the decreasing active porosity due to the most obvious maximum of the surface area for the lower initial active porosity, as predicted from the capillary model. In a word, the results in Figure 5.48 indicate that small changes in the precursor gel microstructure can have a significant effect on the observed nucleation and growth kinetics of zeolites. Effects on the nucleation rates and induction period can be as important as are order-of-magnitude changes on rate constants such as the nucleation rate constant.

References [1] D.W. Breck, Zeolites Molecular Sieves, John Wiley & Sons Inc., New York, 1974. [2] A.V. McCormick, A.T. Bell, and C.J. Radke, Application of 29Si- and 27Al-NMR to Determine the Distribution of Anion in Sodium Silicate and Sodium Aluminosilicate Solutions. Proceedings of the 7th International Zeolite Conference, ed. Y. Murakami, A. Iijima, and J.W. Ward, Elsevier-Kodansha, Tokyo, 1986, 247–254. [3] R.K. Harris, C.T.G. Knight, and W.E. Hull, Nature of Species Present in an Aqueous Solution of Potassium Silicate. J. Am. Chem. Soc., 1981, 103, 1577–1578. [4] D. Hoebbel, G. Garzo, K. Ujszaszi, G. Engelhardt, B. Fahlke, and A. Vargha, Synthesis and Anion Constitution of Crystalline Tetramethylammonium Aluminosilicates and Aluminosilicate Solutions. Z. Anorg. Allg. Chem., 1982, 484, 7–21. [5] D. Hoebbel, G. Garzo, G. Engelhardt, and A. Vargha, On the Constitution and Distribution of Silicate Anions in Aqueous Tetramethylammonium Silicate Solutions. Z. Anorg. Allg. Chem., 1982, 494, 31–42. [6] D. Hoebbel, G. Garzo, G. Engelhardt, R. Ebert, E. Lippmaa, and M. Alla, On the Constitution of Silicate Anions in Tetraethylammonium Silicates and Their Aqueous Solutions. Z. Anorg. Allg. Chem., 1980, 465, 15–33. [7] R.K. Harris and C.T.G. Knight, Si-29 Nmr-Studies of Aqueous Silicate Solutions. 4. Tetraalkylammonium Hydroxide Solutions. J. Mol. Struct., 1982, 78, 273–278. [8] D. Hoebbel, A. Vargha, G. Engelhardt, and K. Ujszaszy, On the Anion Constitutions of Tetrabutylammonium Silicates and Their Aqueous Solutions. Z. Anorg. Allg. Chem., 1984, 509, 85–94. [9] S.D. Kinrade, C.T.G. Knight, D.L. Pole, and R.T. Syvitski, Silicon-29 NMR Studies of Tetraalkylammonium Silicate Solutions. 1. Equilibria, Si-29 Chemical Shifts, and Si-29 Relaxation. Inorg. Chem., 1998, 37, 4272–4277. [10] S.D. Kinrade, C.T.G. Knight, D.L. Pole, and R.T. Syvitski, Silicon-29 NMR Studies of Tetraalkylammonium Silicate Solutions. 2. Polymerization Kinetics. Inorg. Chem., 1998, 37, 4278–4283. [11] R.K. Iler, The Chemistry of Silica, John Wiley & Sons, Inc., New York, 1979. [12] G.B. Alexander and R.K. Iler, US Patent 2,601,235 (1952). [13] M.G. Sanchez, Canadian Patent 586,261 (1959). [14] M.F. Bechtold and O.E. Snyder, US Patent 2,574,902 (1951). [15] M.E. Winyall, US Patent 3,607,777 (1971). [16] M.E. Winyall, Fr. Patent 2,011,162 (1970). [17] E. Wagner and H. Brunner, Aerosil, Herstellung, Eigenschaften end Verhalten in organischer Flu¨ssigkeiten. Angew. Chem., 1960, 72, 744–750. [18] N. Petroleum, German Patent 1,208,741 (1955). [19] F.A. Cotton, G. Wilkinson, C.A. Murillo, and M. Bochmann. Advanced Inorganic Chemistry, John Wiley & Sons, Inc., New York, 1999.

338

Chemistry of Zeolites and Related Porous Materials

[20] R.M. Barrer and P.J. Denny, Hydrothermal Chemistry of Silicates. 9. Nitrogenous Aluminosilicates. J. Chem. Soc., 1961, 971–982. [21] B.D. McNicol, G.T. Pott, and K.R. Loos. Spectroscopic Studies of Zeolite Synthesis: Evidence for a Solid-state Mechanism, in ‘Molecular Series’. Adv. Chem. Ser., 1973, 121, 152–161. [22] W.Y. Xu, J.Q. Li, W.Y. Li, H.M. Zhang, and B.C. Liang, Nonaqueous Synthesis of ZSM-35 and ZSM-5. Zeolites, 1989, 9, 468–473. [23] M. Tsapatsis, M. Lovallo, and M.E. Davis, High-resolution Electron Microscopy Study on the Growth of Zeolite L Nanoclusters. Microporous Mater., 1996, 5, 381–388. [24] D.P. Serrano, M.A. Uguina, G. Ovejero, R. Van Grieken, and M. Camacho, Evidence of SolidSolid Transformations during the TS-1 Crystallization from Amorphous Wetness Impregnated SiO2-TiO2 Xerogels. Microporous Mater., 1996, 7, 309–321. [25] D.P. Serrano, R. Van Grieken, P. Sanchez, R. Sanz, and L. Rodriguez, Crystallization Mechanism of All-silica Zeolite Beta in Fluoride Medium. Microporous Mesoporous Mater., 2001, 46, 35–46. [26] M.A. Uguina, D.P. Serrano, G. Ovejero, R. Van Grieken, and M. Camacho, TS-2 Synthesis from Wetness-impregnated SiO2-TiO2 Xerogels. Zeolites, 1997, 18, 368–378. [27] S.P. Zhdanov. Some Problems of Zeolite Crystallization, in ‘Molecular Sieves.’ Adv. Chem. Ser., 1971, 101, 20–43. [28] R.R. Xu and X.S. Liu, The Mechanism of the Formation and the Crystal-Growth of Molecular-Sieve-Zeolite. 9. The Liquid-Phase Mechanism of the Formation of Zeolite KL. Acta Chim. Sin., 1984, 42, 227–232. [29] S.J. Ma, K.F. Liu, and M.Z. Cui, Formation Mechanism of Zeolite A. Chem. J. Chin. Univ., 1984, 5, 158–162. [30] S. Ueda, N. Kageyama, and M. Koizumi, Crystallization of Zeolite Y from Solution Phase. Proceedings of the 6th International Zeolite Conference, ed. D.H. Olson and A. Bidio, Butterworths, Guildford, 1984, 905–913. [31] W.Q. Pang, S. Ueda, and M. Koizumi, The Synthesis of Zeolite NaA from Homogeneous Solution. Proceedings of the 7th International Zeolite Conference, ed. Y. Murakami, A. Iijima, and J.W. Ward, Kodansha-Elsevier, Tokyo, 1986, 177–184. [32] W.Q. Pang, S.L. Qiu, and Q.B. Kan, Synthesis and Characterization of FAPO-5 from Clear Homogeneous Solutions. Stud. Surf. Sci. Catal., 1989, 49A, 281–289. [33] S. Kasahara, K. Itabaski, and K. Igawa, Clear Aqueous Nuclear Solution for Faujasite Synthesis. Proceedings of the 7th International Zeolite Conference, ed. Y. Murakami, A. Iijima, and J.W. Ward, Kodansha-Elsevier, Tokyo, 1986, 185–192. [34] C.J.Y. Honssian, B.L. Mojet, C.E.A. Kirschhock, V. Buschmann, P.A. Jacobs, J.A. Martens, and R.A. van Santen, Small Angel X-ray Scattering on TPA-Silicalite-1 Precursors in Clear Solutions: Influence of Silica Source and Cations. Stud. Surf. Sci. Catal., 2001, 135 (Zeolites and Mesoporous Materials at the Dawn of the 21st Century), 135–142. [35] K.A. Carlsson, J. Warzywoda, and A. Sacco, Jr., Modeling of Silicalite Crystallization from Clear Solution. Stud. Surf. Sci. Catal., 2001, 135 (Zeolites and Mesoporous Materials at the Dawn of the 21st Century), 255–262. [36] M. Smaihi, S. Kallus, and J.D.F. Ramsay, In-situ NMR Study of Mechanism of Zeolite A Formation. Stud. Surf. Sci. Catal., 2001, 135 (Zeolites and Mesoporous Materials at the Dawn of the 21st Century), 271–278. [37] J. Bronic, P. Frontera, F. Testa, B. Subotiae, R. Aiello, and J.B. Nagy, Study of Zeolite A Crystallization from Clear Solution by Hydrothermal Synthesis and Population Balance Simulation. Stud. Surf. Sci. Catal., 2001, 135 (Zeolites and Mesoporous Materials at the Dawn of the 21st Century), 358–365.

Crystallization of Microporous Compounds

339

[38] R. Grizzetti and G. Artioli, Kinetics of Nucleation and Growth of Zeolite LTA from Clear Solution by in situ and ex situ XRPD. Microporous Mesoporous Mater., 2002, 54, 105–112. [39] Y. Yu, G. Xiong, C. Li, and F.S. Xiao, Characterization of Aluminosilicate Zeolites by UV Raman Spectroscopy. Microporous Mesoporous Mater., 2001, 46, 23–34. [40] L.S. Dent Glasser and G. Harvey. NMR Studies of Aluminosilicates in Solution Proceedings of the 6th International Zeolite Conference, Butterworths, Guildford, 1984, 925–933. [41] G. Harvey. Properties of Aluminosilicate Solutions. PhD Thesis, University of Aberdeen, 1984. [42] M. Ogura, Y. Kawazu, H. Takahashi, and T. Okubo, Aluminosilicate Species in the Hydrogel Phase formed during the Aging Process for the Crystallization of FAU Zeolite. Chem. Mater., 2003, 15, 2661–2667. [43] C.G. Pope, Nucleation and Growth Theory in Zeolite Synthesis. Microporous Mesoporous Mater., 1998, 21, 333–336. [44] V. Nikolakis, D.G. Vlacho, and M. Tsapatsis, Modeling of Zeolite Crystallization: the Role of Gel Microstructure. Microporous Mesoporous Mater., 1998, 21, 337–346. [45] S.L. Burkett and M.E. Davis, Mechanism of Structure Direction in the Synthesis of PureSilica Zeolites. 2. Hydrophobic Hydration and Structural Specificity. Chem. Mater., 1995, 7, 1453–1463. [46] S. Oliver, A. Kuperman, and G.A. Ozin, A New Model for Aluminophosphate Formation: Transformation of a Linear Chain Aluminophosphate to Chain, Layer, and Framework Structures. Angew. Chem., Int. Ed., 1998, 37, 47–62. [47] C. Sassoye, J. Marrot, T. Loiseau, and G. Ferey, Utilization of Cyclopentylamine as StructureDirecting Agent for the Formation of Fluorinated Gallium Phosphates Exhibiting Extralarge-pore Open Frameworks with 16-Ring (ULM-16) and 18-Ring Channels (MIL-46). Chem. Mater., 2002, 14, 1340–1347. [48] L. Beitone, J. Marrot, T. Loiseau, G. Ferey, M. Henry, C. Huguenard, A. Gansmuller, and F. Taulelle, MIL-50, an Open-framework GaPO with a Periodic Pattern of Small Water Ponds and Dry Rubidium Atoms: a Combined XRD, NMR, and Computational Study. J. Am. Chem. Soc., 2003, 125, 1912–1922. [49] L.E. Iton, F. Trouw, T.O. Brun, J.E. Epperson, J.W. White, and S.J. Henderson, Small-Angle Neutron-Scattering Studies of the Template-Mediated Crystallization of ZSM-5-Type Zeolite. Langmuir, 1992, 8, 1045–1048. [50] R. Van Grieken, J.L. Sotelo, J.M. Menendez, and J.A. Melero, Anomalous Crystallization Mechanism in the Synthesis of Nanocrystalline ZSM-5. Microporous Mesoporous Mater., 2000, 39, 135–147. [51] R.J. Francis and D. O’Hare, The Kinetics and Mechanisms of the Crystallisation of Microporous Materials. J. Chem. Soc., Dalton Trans., 1998, 3133–3148. [52] E.M. Flanigen, R.L. Patton, and S.T. Wilson, Innovation Zeolite Material Science. Stud. Surf. Sci. Catal., 1988, 37, 13–19. [53] J.P. Arhancet and M.E. Davis, Systematic Synthesis of Zeolites that Contain Cubic and Hexagonal Stackings of Faujasite Sheets. Chem. Mater., 1991, 3, 567–569. [54] H. Robson, Synthesizing Zeolites. Chemtech, 1978, 8, 176–180. [55] E.M. Flanigen, Review and New Perspectives in Zeolite Crystallization. Adv. Chem. Ser., 1973, 121, 119–139. [56] P.K. Dutta, M. Pur, and D.C. Shieh, Mater. Res. Soc. Symp. Proc., 1988, 111, 101. [57] G.T. Kerr, Kokotail. Gt, Sodium Zeolite Zk-4, a New Synthetic Crystalline Aluminosilicate. J. Am. Chem. Soc., 1961, 83, 4675.

340

Chemistry of Zeolites and Related Porous Materials

[58] C. Baerloch and W.M. Meier, Synthesis and Crystal Structure of Tetramethylammonium Sodalite. Helv. Chim. Acta, 1969, 52, 1853. [59] R.R. Xu and W.Q. Pang, Inorganic Synthesis and Preparative Chemistry (Higher Education Press, Beijing, 2001). [60] J.C. Jansen and S.T. Wilson, The Preparation of Oxide Molecular Sieve. Stud. Surf. Sci. Catal., 2001, 137, 175–227. [61] W. Schmidt, F. Schuth, H. Reichert, K. Unger, and B. Zibrowius, VPI-5 and Related Aluminophosphates: Preparation and Thermal Stability. Zeolites, 1992, 12, 2–8. [62] B.M. Lok, T.R. Cannan, and C.A. Messina, The Role of Organic Molecules in MolecularSieve Synthesis. Zeolites, 1983, 3, 282–291. [63] J.H. Yu, R.R. Xu, and J.Y. Li, Structural Diversity of a Family of Aluminophosphates with Al/P Ratio of Non-unity. Solid State Sci., 2000, 2, 181–192. [64] S.L. Lawton and W.J. Rohrbaugh, The Framework Topology of ZSM-18, a Novel Zeolite Containing Rings of 3 (Si,Al)-O Species. Science, 1990, 247, 1319–1322. [65] J. Ciric, U.S. Patent 3,950,496 (1976). [66] K.D. Schmitt and G.J. Kennedy, Toward the Rational Design of Zeolite Synthesis – the Synthesis of Zeolite ZSM-18. Zeolites, 1994, 14, 635–642. [67] S.I. Zones and R.A. Vannordstrand, Novel Zeolite Transformations – the Template-mediated Conversion of Cubic-P Zeolite to SSZ-13. Zeolites, 1988, 8, 166–174. [68] S.I. Zones and R.A. Vannordstrand, Further Studies on the Conversion of Cubic P-Zeolite to High Silica Organozeolites. Zeolites, 1988, 8, 409–415. [69] M.E. Davis and R.F. Lobo, Zeolite and Molecular-Sieve Synthesis. Chem. Mater., 1992, 4, 756–768. [70] R.A. Vannordstrand, D.S. Santilli, and S.I. Zones, An All-Silica Molecular-Sieve That Is Isostructural with Alpo4-5. Acs Syn. Ser., 1988, 368, 236–245. [71] F. Delprato, L. Delmotte, J.L. Guth, and L. Huve, Synthesis of New Silica-Rich Cubic and Hexagonal Faujasites Using Crown-Ether-Based Supramolecules as Templates. Zeolites, 1990, 10, 546–552. [72] P. Wagner, M. Yoshikawa, M. Lovallo, K. Tsuji, M. Taspatsis, and M.E. Davis, CIT-5: a Highsilica Zeolite with 14-Ring Pores. Chem. Commun., 1997, 2179–2180. [73] M. Yoshikawa, P. Wagner, M. Lovallo, K. Tsuji, T. Takewaki, C.Y. Chen, L.W. Beck, C. Jones, M. Tsapatsis, S.I. Zones, and M.E. Davis, Synthesis, Characterization, and Structure Solution of CIT-5, a New, High-silica, Extra-large-pore Molecular Sieve. J. Phys. Chem. B, 1998, 102, 7139–7147. [74] L.D. Rollmann and E.M. Valyoosik, US Patent 4,108,881 (1978). [75] G.W. Noble, P.A. Wright, and A. Kvick, The Templated Synthesis and Structure Determination by Synchrotron Microcrystal Diffraction of the Novel Small Pore Magnesium Aluminophosphate STA-2. J. Chem. Soc., Dalton Trans., 1997, 4485–4490. [76] J.S. Chen, R.H. Jones, S. Natarajan, M.B. Hursthouse, and J.M. Thomas, A Novel OpenFramework Cobalt Phosphate-Containing a Tetrahedrally Coordinated Cobalt(II) Center – CoPO4 . 0.5 C2H10N2. Angew. Chem. Int. Ed. Engl., 1994, 33, 639–640. [77] T.M. Nenoff, W.T.A. Harrison, T.E. Gier, and G.D. Stucky, Room-Temperature Synthesis and Characterization of New Znpo and Znaso Sodalite Open Frameworks. J. Am. Chem. Soc., 1991, 113, 378–379. [78] M. Briend, A. Lamy, M.J. Peltre, P.P. Man, and D. Barthomeuf, Thermal-Stability of Tetrapropylammonium (TPA) and Tetramethylammonium (TMA) Cations Occluded in SAPO-37 Molecular-Sieves. Zeolites, 1993, 13, 201–211. [79] S.T. Wilson and L. King, EP Patent 293,938 (1988). [80] L. Sierra, J. Patarin, C. Deroche, H. Gies, and J.L. Guth, in Zeolites and Related Microporous Materials: State of the Art 1994, Elsevier Science Publishers B V, Amsterdam, 1994, 2237–2244.

Crystallization of Microporous Compounds

341

[81] S.J. Weigel, S.C. Weston, A.K. Cheetham, and G.D. Stucky, Cooperative Structure Direction in the Synthesis of Microporous Materials: Preparation and Crystal Structure of TRENGaPO. Chem. Mater., 1997, 9, 1293–1295. [82] J.P. Lourenco, M.F. Ribeiro, F.R. Ribeiro, J. Rocha, Z. Gabelica, N. Dumont, and E.G. Derouane, in Zeolites and Related Microporous Materials: State of the Art 1994, Elsevier Science Publishers B V, Amsterdam, 1994, 867–874. [83] D.E. Akporiaye, H. Fjellvag, E.N. Halvorsen, T. Haug, A. Karlsson, and K.P. Lillerud, UiO-6: A Novel 12-ring AlPO4, Made in an Inorganic-Organic Cation System. Chem. Commun., 1996, 1553–1554. [84] K.R. Franklin and B.M. Lowe, in Hydrothermal Crystallization of Silica Molecular Sieres and Clathrasils from Amine Containing Reaction Mixtures, Stud. Surf. Sci. Catal., 1988, 49, 179–188. [85] X.H. Bu, P.Y. Feng, and G.D. Stucky, Large-cage Zeolite Structures with Multidimensional 12-Ring Channels. Science, 1997, 278, 2080–2085. [86] I.D. Brown and D. Altermatt, Bond-Valence Parameters Obtained from a Systematic Analysis of the Inorganic Crystal-Structure Database. Acta Crystallogr., Sect. C-Struct. Commun., 1985, 41, 244–247. [87] I. Balakrishnan and S. Prasad, Synthesis of AlPO4-11 Using Long-Chain Secondary Amines. Appl. Catal., 1990, 62, L7–L11. [88] N.J. Tapp, N.B. Milestone, and D.M. Bibby, Synthesis of AlPO4-11. Zeolites, 1988, 8, 183– 188. [89] K.J. Balkus, M. Biscotto, and A.G. Gabrielov, in Progress in Zeolite and Microporous Materials, Pts A–C, Elsevier Science Publishers B V, Amsterdam, 1997, 415–421. [90] Q.H. Huo, R.R. Xu, S.G. Li, Z.G. Ma, J.M. Thomas, R.H. Jones, and A.M. Chippindale, Synthesis and Characterization of a Novel Extra Large Ring of Aluminophosphate JDF-20. J. Chem. Soc., Chem. Commun., 1992, 875–876. [91] M. Estermann, L.B. McCusker, C. Baerlocher, A. Merrouche, and H. Kessler, A Synthetic Gallophosphate Molecular-Sieve with a 20-Tetrahedral-Atom Pore Opening. Nature (London), 1991, 352, 320–323. [92] T. Loiseau and G. Ferey, Oxyfluorinated Microporous Compounds. 7. Synthesis and Crystal Structure of ULM-5, a New Fluorinated Gallophosphate Ga16(PO4)14(HPO4)2(OH)2F7, [H3N(CH2)6NH3]4,6H2O with 16-Membered Rings and both Bonding and Encapsulated F. J. Solid State Chem., 1994, 111, 403–415. [93] C. Sassoye, T. Loiseau, F. Taulelle, and G. Ferey, A New Open-framework Fluorinated Gallium Phosphate with Large 18-Ring Channels (MIL-31). Chem. Commun., 2000, 943– 944. [94] E.M. Flanigen and R.L. Patton, US Patent 4,073,865 (1978). [95] J.L. Guth, H. Kessler, and R. Wey, Stud. Surf. Sci. Catal., 1986, 28, 121–128. [96] D.E. Akporiaye, H. Fjellvag, E.N. Halvorsen, J. Hustveit, A. Karlsson, and K.P. Lillerud, UiO-7: A New Aluminophosphate Phase Solved by Simulated Annealing and High-Resolution Powder Diffraction. J. Phys. Chem., 1996, 100, 16641–16646. [97] P. Caullet, J.L. Guth, J. Hazm, J.M. Lamblin, and H. Gies, Synthesis, Characterization and Crystal-Structure of the New Clathrasil Phase Octadecasil. Eur. J. Solid State Inorg. Chem., 1991, 28, 345–361. [98] G. Vandegoor, C.C. Freyhardt, and P. Behrens, The Cobalticinium Cation [Co-III(ETA(5)C5H5)2]þ – a Metal-Organic Complex as a Novel Template for the Synthesis of Clathrasils. Z. Anorg. Allg. Chem., 1995, 621, 311–322. [99] P.A. Barrett, M.A. Camblor, A. Corma, R.H. Jones, and L.A. Villaescusa, Synthesis and Structure of as-prepared ITQ-4, a Large Pore Pure Silica Zeolite: The Role and Location of Fluoride Anions and Organic Cations. J. Phys. Chem. B, 1998, 102, 4147–4155.

342

Chemistry of Zeolites and Related Porous Materials

[100] M.A. Camblor, M.J. Diaz-Cabanas, J. Perez-Pariente, S.J. Teat, W. Clegg, I.J. Shannon, P. Lightfoot, P.A. Wright, and R.E. Morris, SSZ-23: An Odd Zeolite with Pore Openings of Seven and Nine Tetrahedral Atoms. Angew. Chem., Int. Ed., 1998, 37, 2122– 2126. [101] G.D. Price, J.J. Pluth, J.V. Smith, J.M. Bennett, and R.L. Patton, Crystal-Structure of Tetrapropylammonium Fluoride Containing Precursor to Fluoride Silicalite. J. Am. Chem. Soc., 1982, 104, 5971–5977. [102] H. Kessler, Mater. Res. Soc. Symp. Ser., 1991, 233, 47–55. [103] K.J. Balkus, C.D. Hargis, and S. Kowalak, Synthesis of Nax Zeolites with Metallophthalocyanines. ACS Symp. Ser., 1992, 499, 347–354. [104] K.J. Balkus, Jr, A.G. Gabrielv, and N. Sandler, Molecular Siere Synthesis using Metallocenes as Structure Directing Agents, Mater. Res. Soc. Symp. Proc., 1995, 368, 369–375. [105] K.J. Balkus, Jr., C.D. Hargis, and R. Szostak, Synthesis and Characterization of Zeolites Prepared using Metallocene Templates. Proceedings of 12th International Zeolite Conference, Materials Research Society, Warrendale, PA, 1999, 1931–1935. [106] K. Morgan, G. Gainsford, and N. Milestone, A Novel Layered Aluminum Phosphate [Co(en)3Al3P4O16 . 3H2O] Assembled About a Chiral Metal-Complex. J. Chem. Soc., Chem. Commun., 1995, 425–426. [107] J.H. Yu, Y. Wang, Z. Shi, and R.R. Xu, Hydrothermal Synthesis and Characterization of Two New Zinc Phosphates assembled about a Chiral Metal Complex: [CoII(en)3]2[Zn6P8O32H8] and [CoIIIen)3][Zn8P6O24Cl] . 2H2O. Chem. Mater., 2001, 13, 2972–2978. [108] D.A. Bruce, A.P. Wilkinson, M.G. White, and J.A. Bertrand, The Synthesis and Characterization of an Aluminophosphate with Chiral Layers; trans-Co(dien)2 . Al3P4O16 . 3H2O. J. Solid State Chem., 1996, 125, 228–233. [109] M.J. Gray, J.D. Jasper, A.P. Wilkinson, and J.C. Hanson, Synthesis and Synchrotron Microcrystal Structure of an Aluminophosphate with Chiral Layers containing Lambda tris(ethylenediamine)cobalt(III). Chem. Mater., 1997, 9, 976–980. [110] J.H. Yu, Y. Wang, Z. Shi, and R.R. Xu, Chiral Transference from the Chiral Metal Complex to the Zinc Phosphates. Chin. J. Inorg. Chem., 2002, 18, 51–55. [111] E. Fois, A. Gamba, and A. Tilocca, Structure and Dynamics of the Flexible Triple Helix of Water inside VPI-5 Molecular Sieves. J. Phys. Chem. B, 2002, 106, 4806–4812. [112] D.H. Olson, The Use of Water as a Probe Molecule in the Study of Zeolite, in Fuel Science Seminar (1996). Center for Applied Energy Research Seminars, University of Kentucky, Lexington. [113] S. Oliver, A. Kuperman, A. Lough, G.A. Ozin, J.M. Garces, M.M. Olken, and P. Rudolf, in Zeolites and Related Microporous Materials: State of the Art 1994, Elsevier Science Publishers B V, Amsterdam, 1994, 219–225. [114] R. Kumar, P. Mukherjee, R. Pandey, P. Rajmohanan, and A. Bhaumik, Role of Oxyanions as Promoter for Enhancing Nucleation and Crystallization in the Synthesis of MFI-type Microporous Materials. Microporous Mesoporous Mater., 1998, 22, 23–31. [115] A.P. Stevens, A.H. Gorman, C.M. Freeman, and P.A. Cox, Prediction of Template Location via a Combined Monte Carlo-simulated Annealing Approach. J. Chem. Soc., Faraday Trans., 1996, 92, 2065–2073. [116] S.I. Zones, Y. Nakagawa, L.T. Yuen, and T.V. Harris, Guest/Host Interactions in High Silica Zeolite Synthesis: [5.2.1.0(2.6)]tricyclodecanes as Template Molecule. J. Am. Chem. Soc., 1996, 118, 7558–7567. [117] D.W. Lewis, C.M. Freeman, and C.R.A. Catlow, Predicting the Templating Ability of Organic Additives for the Synthesis of Microporous Materials. J. Phys. Chem., 1995, 99, 11194– 11202.

Crystallization of Microporous Compounds

343

[118] D.W. Lewis, C.R.A. Catlow, and J.M. Thomas, Influence of Organic Templates on the Structure and on the Concentration of Framework Metal Ions in Microporous Aluminophosphate Catalysts. Chem. Mater., 1996, 8, 1112–1118. [119] D.W. Lewis, R.G. Bell, P.A. Wright, C.R.A. Catlow, and J.M. Thomas, in Progress in Zeolite and Microporous Materials, Pts A–C, Elsevier Science Publishers B V, Amsterdam, 1997, 2291–2298. [120] C.R.A. Catlow, D.S. Coombes, and J.C.G. Pereira, Computer Modeling of Nucleation, Growth, and Templating in Hydrothermal Synthesis. Chem. Mater., 1998, 10, 3249–3265. [121] D.W. Lewis, D.J. Willock, C.R.A. Catlow, J.M. Thomas, and G.J. Hutchings, De novo Design of Structure-directing Agents for the Synthesis of Microporous Solids. Nature (London), 1996, 382, 604–606. [122] D.W. Lewis, G. Sankar, J.K. Wyles, J.M. Thomas, C.R.A. Catlow, and D.J. Willock, Synthesis of a Small-pore Microporous Material using a Computationally Designed Template. Angew. Chem., Int. Ed. Engl., 1997, 36, 2675–2677. [123] P.A. Cox, A.P. Stevens, L. Banting, and A.M. Gorman, in Zeolites and Related Microporous Materials: State of the Art 1994, Elsevier Science Publishers B V, Amsterdam, 1994, 2115– 2122. [124] P.A. Cox, J.L. Casci, and A.P. Stevens, Molecular Modelling of Templated Zeolite Synthesis. Faraday Discuss., 1997, 106, 473–487. [125] R.E. Boyett, A.P. Stevens, M.G. Ford, and P.A. Cox, A Quantitative Shape Analysis of Organic Templates Employed in Zeolite Synthesis. Zeolites, 1996, 17, 508–512. [126] Y. Kubota, M.M. Helmkamp, S.I. Zones, and M.E. Davis, Properties of Organic Cations that Lead to the Structure-direction of High-silica Molecular Sieves. Microporous Mater., 1996, 6, 213–229. [127] J.Y. Li, J.H. Yu, W.F. Yan, Y.H. Xu, W.G. Xu, S.L. Qiu, and R.R. Xu, Structures and Templating Effect in the Formation of 2D Layered Aluminophosphates with Al3P4O163stoichiometry. Chem. Mater., 1999, 11, 2600–2606. [128] G. Sastre, S. Leiva, M.J. Sabater, I. Gimenez, F. Rey, S. Valencia, and A. Corma, Computational and Experimental Approach to the Role of Structure-directing Agents in the Synthesis of Zeolites: The Case of Cyclohexyl Alkyl Pyrrolidinium Salts in the Synthesis of Beta, EU-1, ZSM-11, and ZSM-12 Zeolites. J. Phys. Chem. B, 2003, 107, 5432–5440. [129] S.P. Zhdanov and N.N. Samuelevich, Nucleation and Crystal Growth of Zeolites in Crystallizing Aluminosilicate Gels. Proceedings of the 5th International Zeolite Conference, ed. L.V. Rees, Heyden and Son Ltd, London, 1980, 75–84. [130] S.H. Feng, S.G. Li, and R.R. Xu, Formation Mechanism and Crystal Growth of Zeolites Molecular Sieves (XIII)-M-Si-ZSM-5. Chem. J. Chin. Univ., 1985, 6, 855–860. [131] B. Subotic and J. Bronic, Modeling and Simulation of Zeolite Crystallization. Proceedings of the 9th International Zeolite Conference, ed. R. VonBallmoos, J.R. Higgins, and M.M.J. Treacy, Butterworth-Heinemann, Boston, 1992, 321–328. [132] P.M. Budd, G.J. Myatt, C. Price, and S.W. Carr, An Empirical Model for the Nucleation and Growth of Zeolites. Zeolites, 1994, 14, 198–202. [133] B. Subotic, T. Antonic, and J. Bronic, Population Balance: a Powerful Tool for the Study of Critical Processes of Zeolite Crystallization. Stud. Surf. Sci. Catal., 2001, 135 (Zeolites and Mesoporous Materials at the Dawn of the 21st Century), 319–326.

6 Preparation, Secondary Synthesis, and Modification of Zeolites 6.1 Preparation of Zeolites – Detemplating of Microporous Compounds As discussed in Chapters 3 and 4 (synthetic chemistry of microporous compounds I and II), various microporous compounds with particular frameworks, component elements, and channel structures (channel sizes, pore dimensionality, pore shapes and arrangements, etc.) can crystallize through hydrothermal and solvothermal synthetic routes. Except for some aluminosilicate zeolites, most microporous compounds contain guest species such as organic molecules or metal complexes in their structures as structuredirecting agents (SDAs) or templates. Because these guest molecules usually interact with zeolite frameworks through H-bonds or van der Waals’ forces, or sometimes coordination bonds, in order to extend the types of zeolites and related catalytic materials, it is very important to remove the templates from the guest-containing microporous compounds to form structurally stable, free-pore molecular sieves with particular surface properties. Depending on synthetic routes, the formation mechanisms of zeolites are different, and some frontier aspects will be discussed separately as follows. 6.1.1

High-temperature Calcination

The commonly used approach to drive off the guest molecules from the microporous frameworks is high-temperature (550
C in air) calcination, which oxidizes and decomposes the organic molecules. However, this process is highly exothermic, and if inappropriately handled, the zeolite structures would be destroyed by the calcination. For instance, Da[1] found that calcination of zeolite BEA at 550
C removes the template

Chemistry of Zeolites and Related Porous Material – Synthesis and Structure Ruren Xu, Wenqin Pang, Jihong Yu, Qisheng Huo and Jiesheng Chen # 2007 John Wiley & Sons, (Asia) Pte Ltd

346

Chemistry of Zeolites and Related Porous Materials

tetraethylammonium (TEA) but, in the meantime, the zeolite crystallinity decreases by 25–30%. In 1994, Corma et al.[2] reported their systematic investigation of hightemperature calcination effect on heteroatom-containing zeolites, and they found that during the detemplating process the heteroatoms or aluminum are driven off from the zeolite frameworks, leading to a decrease in surface acidity and related catalytic performance. Therefore, how to control and improve the calcination conditions is the frontier research subject in the field of zeolite molecular sieves. The control of calcination temperature and time, the selection of calcination atmosphere, the control of fluid dynamics[3], and the exploration of new calcination routes on the basis of detailed analysis of thermolysis processes all play an important role in the improvement of the calcination route for detemplating. Various approaches for the improvement of this detemplating route have been reported previously. For example, in one approach, the assynthesized materials are heated at a lower temperature to remove the adsorbed water molecules, followed by high-temperature calcination to remove the organic guest molecules; whereas in another approach, a slow temperature-increase rate is used for calcination of the as-synthesized zeolites at 500
C in O2-rich air to overcome the coking problem. Here we briefly describe recent progress in this aspect. Two-step Calcination In 2002, Duan et al.[1(b)] reported that for high-temperature calcination the temperatureincrease rate is usually large and, as a result, the guest organic molecules decompose dramatically, leading to abrupt increase of the inner pressure of the crystal lattice and to the destruction of the framework structure. Furthermore, the framework charge balance will possibly be broken under high-temperature calcination, resulting in the escape of aluminum and heteroatoms from the framework. To avoid these detrimental effects, Duan et al. proposed a new two-step calcination route in which the as-synthesized zeolites are calcined first at a lower temperature (Tc), followed by a gradual increase in the temperature to a higher level, and the zeolites are calcined at this higher temperature for a short time. They used zeolite beta (BEA) as a model and compared the effects of the two-step and the normal high-temperature (550
C in air for 2 h) calcination approaches. For the two-step calcination, Tc was established on the basis of the fact that the tetraethylammonium hydroxide (TEAOH) and tetraethylammonium (TEAþ) species in BEA decompose in the temperature range 200–290
C. In practice, BEA was heated at Tc 290
C for 2 h, and then the temperature was gradually increased to 550
C at a rate of 5
C/min. This 550
C calcination temperature was kept for 20 min. On the basis of crystallinity and surface acidity, the BEA treated using the two-step calcination method is superior to that obtained through the conventional high-temperature calcination technique. Furthermore, it was found that the surface acidity is correlated with the time the Tc was kept. These observations are conducive to the discovery of the optimal conditions for the two-step calcination technique. Two-step Calcination with Microwave Radiation In order to minimize the detrimental effects of high-temperature calcination on zeolite structures, Duan et al. suggested a two-step detemplating approach using microwave radiation. Again, BEA was used as the model for removal of the template TEA. First, the sample was heated at Tc 200
C for 3 h, followed by treatment under microwave radiation at room temperature for 40 min, and then the sample was subject to a programmed

Preparation, Secondary Synthesis, and Modification of Zeolites

347

temperature increase to 550
C at a rate of 5
C/min. The sample was further kept at 550
C for 20 min. The BEA thus obtained had a crystallinity and surface acidity higher than that obtained through the conventional high-temperature treatment. The two-step calcination technique is equally effective for the detemplating of ZSM-5 and MCM-41. 6.1.2

Chemical Detemplating

This technique involves the interactions between chemical reagents (liquid or gas) and the organic guest template in the pores and channels of a microporous crystal under mild conditions, and this detemplating technique may result in free pores with the whole structure of the compound remaining intact. The selection of the chemical reagents and the related chemical reactions should be based on the mildness of the reaction conditions and the ease of product separation. Oxidization Detemplating with Ozone under Mild Conditions (Oxidation Detemplating) In 1998 Keene, Denoyel, and Llewellyn proposed that, at room temperature, O3 treatment (UV lamp, 6.8 W, 254 and 180 nm) for 24 h can remove the surfactant CTABr (cetyltrimethylammonium bromide) from an MCM-41 sample.[4] It was revealed that in comparison with the MCM-41 sample obtained through calcination at 550
C in air, the material possesses a larger surface area and narrower pore-size distribution. In 2001 Mehn et al.[5] applied this technique to the detemplating of heteroatom-zeolites such as B-, Co-, CoAl-ZSM-5, and Ga-MCM-22 microporous compounds. The heteroatoms in the samples treated at 210
C for 3 h in an O2/O3 mixture stream remained intact and the samples were superior in property to those calcined at 550
C in air. Because of the strong oxidation ability of O3, this technique can be employed to decompose organic molecules under much milder conditions, even at room temperature, so that damage to the crystal structure will be minimized. In addition, this technique is convenient and the oxidation products are CO2 and H2O, which are environmentally benign, and therefore, this technique is rather promising. Detemplating through Ammonolysis at Medium Temperatures Quaternary ammonium species tetramethylammonium (TMAþ) is one of the important templates for the synthesis of high-silicon zeolites. In the high-silicon zeolites A (NaTMA-A) and Y (NaTMA-Y) synthesized using TMAþ as template, the TMAþ cation is located in the a cage, the supercage, and the b cage, and because the TMAþ ion is large, this species, especially when located in the b cage, which is small, finds it not easy to escape or to decompose inside the cages. Conventionally, this template was removed through long-time calcination at 550
C, which led to structural damage. In 2002 Ku¨hl and coworkers[6] reported a new detemplating technique that involved interactions of the small molecule NH3 with TMAþ at medium temperatures. Thus, the reaction of gaseous NH3 with TMAþ at 250
C forms CH3NH2 and (CH3)2NH that can escape from the supercages. The remaining problem for this technique lies in the fact that the TMAþ in the b cage decomposes to form bulkier (CH3)3NHþ which is difficult to be driven off from the zeolite structure. This problem is even severe for zeolite Na,TMAA because the opening of the a cage is composed of an 8-membered ring which is too small for bulkier molecules to escape through, leading to difficulties in complete

348

Chemistry of Zeolites and Related Porous Materials

Figure 6.1 Gas chromatogram of the ammonolysis products. (a) NaTMA-Y (SiO2/ Al2O3 ¼ 5.3) at 300
C; (b) NaTMA-A
(SiO2/Al2O3 ¼ 5.8) at 250
C. (*) NH3; (~) CH3NH2; (&) (CH3)2NH; (&) (CH3)3N. Reproduced with permission from [6]. Copyright (2002) Elsevier

detemplating [Figure 6.1(b)]. Further study is underway to improve the detemplating process using this technique. For zeolite Na,TMA-Y, the decomposition products can escape easily. As shown in Figure 6.1(a), at 250–300
C, ammonolysis for 3–5 h leads to nearly complete detemplating. Besides the respective advantages and disadvantages of the high-temperature calcination and chemical-reaction techniques for detemplating, these two methods have one shortcoming in common; that is, the destruction of the template during the detemplating process. If the experimental scale is large whereas the template used is expensive, the cost of production of the zeolite will be rather high. Furthermore, this will lead to environmental pollution. Therefore, recently, zeolite chemists have started to explore greener detemplating routes in which detemplating and recovery of templates are combined. The following subsection will describe the research progress on this aspect. 6.1.3

Solvent-extraction Method

The application of solvent extraction to removal and recovery of templates or SDAs from zeolite channels was initiated by Whitehurst[7] in the 1990s for the extraction of surfactant from mesoporous M41S materials. This method or its improved analogs have become one of the most important techniques to recover surfactants from mesoporous molecular sieves. However, it is still difficult to use this technique to remove and to recover the SDAs from microporous molecular sieves because, first, the size of SDA molecules is similar to that of the channel openings and the molecules are not able to diffuse out from the channels, and, secondly, there are usually strong interactions between the microporous frameworks and the SDA molecules that prevent the SDA molecules from being extracted solely by solvents. Modification of the conventional solvent-extraction technique, such as addition of chemical agents which can adjust the

Preparation, Secondary Synthesis, and Modification of Zeolites

349

acidity of the extraction system or may play a synergetic role in extraction or decrease the interactions between SDA and framework, and enlargement of the channel sizes have been applied to the detemplating of microporous molecular sieves and certain progress has been made. In 1998, Davis and coworkers[8] reported the successful extraction of tetraethylammonium fluoride (TEAF) from BEA using the solvent-extraction technique, and in 1999 they successfully extracted TEAOH from a microporous zincosilicate isostructural with BEA by using a similar method.[9] In 2001 Davis described[10] in detail the SDA extraction procedure from the medium-pore MFI and the large-pore BEA framework structures. Below we use two examples to demonstrate the application potential of the solvent-extraction technique for the recovery of SDA. Extraction of SDA from BEA A series of zeolites such as Si-beta-OH(F), B-beta-OH(F), and Al-beta-OH(F) can be synthesized by using TEAOH or TEAF as the SDA. In the structure of these zeolites, the TEAþ interacts with the O atoms of the SiO4 units through H-bonds. The TEAþ species can be removed through conventional calcination at 550
C. In 1998, Davis and colleagues suggested that by using acetic acid solution (50% H2O) as the extraction solvent through reaction-synergetic solvent extraction at 80
C for 12–24 h, the TEAþ could be extracted and recovered [Equation (6.1)][8] The extraction yield can reach 90% after separation and repeated extractions, and the zeolite structure is seldom damaged. In 1998, Davis et al. further investigated [8] in detail the TEA extraction from Si-BEA-OH(F), B-BEA-OH(F), and Al-BEA-OH(F) zeolites, and they found that for either TEAOH or TEAF-templated pure-silica BEA and B- and Al-containing BEA zeolites, the extraction yield is in the order Si-BEA-OH(F) (>99%) > B-BEA-OH(F) (75–85%) > Al-BEA-OH(F) (45–49%). It is believed that the TEAþ species are difficult to extract completely from B- and Al-containing BEA zeolites possibly because, in the M-BEA-OH(F) frameworks, the interactions between the heteroatoms (M ¼ B, Al) and the TEAþ are stronger than the H-bonding interactions between the Si(OSi)4 units and the TEAþ cations.

Si

O–

TEA+

H + Ac–

Si

OH

TEA +

Ac–

ð6:1Þ

Extraction of SDA Crown Ether from FAU and EMT [11] High-silica FAU (Si/Al ¼ 3.8) and EMT (Si/Al ¼ 3.8) zeolites can be synthesized using 15-crown-5 and 18-crown-6 as the SDA, and their compositions are Na40Al40Si152O384 (15-crown-5)8(H2O)120 and Na20Al20Si76O192(18-crown-6)4(H2O)60, respectively. In the zeolite structures, some of the Naþ cations form complexes with the 15-crown-5 and 18crown-6, whereas some Naþ are located in the b and the D6R cages. In order to recover the specious crown ether molecules from the zeolites through extraction, an exchangeextraction system involving quaternary ammonium or protonated ammonium cations was designed, and through the following exchange reaction [Equation (6.2)] with water or polar extracting agent the crown ether can be recovered: ½Naþ ;crown ether=Naþ zeolite þ½R3 Nþ ðor TAAþ Þextractant Ð ½Naþ ;crown ether=Naþ extractant þ½Naþ ;R3 Nþ ðor TAAþ Þzeolite

ð6:2Þ

350

Chemistry of Zeolites and Related Porous Materials

Et3N, Pr3N, TMAþ, Bu3N, TEAþ, where TAA ¼ tetraalhylammonium and tetrapropylammonium (TPAþ) solutions (pH ¼ 6) have been used for extraction, and, at 190
C for 6 h, the 15-crown-5 and 18-crown-6 are extracted almost completely to the aqueous phase, indicating that this technique is very promising for the recycling of SDA.

6.2 Outline of Secondary Synthesis The modification of zeolites mainly relies on secondary synthesis methods. The aim of modification is to reprocess the zeolites using suitable techniques to improve the properties and functions such as (1) acidity, (2) thermal and hydrothermal stability, (3) catalytic performance such as redox catalytic and coordination catalytic properties, etc., (4) channel structures, (5) surface properties and microporous frameworks and charge-balancing ions. Modification is also called secondary synthesis and can lead to new properties that cannot be achieved through direct synthesis. Let us consider the case of faujasite (FAU), the main component of the cracking catalyst, and its catalytic performance (represented by the catalytic activity K/K Std for n-hexane cracking). From Table 6.1 it is seen that the secondary synthesis affects the catalytic performance to a considerable degree. In fact, the modification of zeolites is the modification of zeolite structures and the further processing of zeolites, that is, the so-called secondary synthesis. Of course, the properties and functions of a zeolite depend on its framework and pore structure, but on the other hand the charge-balancing cations, their nature, their number, and their ionexchange capacity also influence the properties and function of the particular zeolite. Sometimes the cations may be the predominant factor that affects the zeolite properties and functions. For instance, acidity, porosity and channel-window size, thermal stability, and catalytic performance of a zeolite are usually correlated with cations present in the structure of the zeolite. In the 1970s Vogt and coworkers[13] pointed out that ‘Stability of the zeolite structure, both thermal and hydrothermal, could be achieved through ion exchange with multivalent cations.’ Therefore, since the middle of last century, quite a number of zeolite chemists have started to investigate the cation-exchange of zeolites, and the ion-exchange properties of zeolites have been thoroughly studied.[14] In this chapter, we will describe the secondary synthesis or modification of zeolites from four aspects: (1) cation-exchange and modification of zeolites; (2) dealumination and modification of zeolites; (3) isomorphous substitution of heteroatoms in zeolite Table 6.1 n-Hexane cracking activity (a value) of various ion-exchanged faujasite zeolites (secondary synthesis)[12] Catalyst Amorphous silica–alumina Faujasite Faujasite Faujasite Faujasite Faujasite Faujasite

Exchanging ion Naþ Ca2þ NH4þ La3þ Re3þ Re3þ, NH4þ

Temperature (T/
C) 540 540 530 350 270 <270 <270

a Value 1.0 1.2 1.1 6400 7000 >10000 >10000

Preparation, Secondary Synthesis, and Modification of Zeolites

351

frameworks; (4) channel and surface modification of zeolites. Before description of the modification and reprocessing of zeolite framework structures, we would like to discuss the cation-exchange first.

6.3 Cation-exchange and Modification of Zeolites As listed in Section 6.1, the development of cracking catalysts based on zeolite Y, generation after generation, are correlated with cation exchange of the zeolite. Below we would like to use some examples to discuss this subject. 6.3.1

Ion-exchange Modification of Zeolite LTA[14]

The typical composition of zeolite A is Na12[Al12Si12O48]27H2O and its Si/Al ratio ¼ 1, and as a result this zeolite is one of the zeolites with the largest ionic-exchange capacity. It is clear that, except for a few cations with a small radius and some protonated amine cations, other cations have a wmax of 1 in this zeolite. In addition, because the pore opening aperture ˚ , this zeolite falls into the category of small-pore zeolite molecular sieves, for NaA is 4 A and, after ion-exchange, the number, the size, and the location of the exchanged cations affect the pore size of zeolite LTA to a considerable extent. For instance, after exchange with ˚ , respectively. Therefore, Csþ, Kþ, and Ca2þ, the pore diameter becomes 2, 3, and 5 A commercially the KA, NaA, and CaA zeolites are called 3A, 4A and 5A zeolites; and because zeolite LTA possesses these ion-exchange features, it is possible to adjust the pore size of the zeolite to a great extent through ion-exchange. After ion-exchange, the size (radius), the charge, the polarizability, the deformation ability of the cations and their influence on the homogeneity of the framework electric field determine the adsorption and catalytic properties of the zeolite. In the next subsection the adsorption properties of ionexchanged zeolites will be described. We first discuss selectivity, and secondly the adsorption rate and capacity of adsorbates because, after ion-exchange, the zeolites can be widely used for gas drying, cleaning and separation, and shape-selective catalysis. Zeolite 5A (CaA) In the structure of the NaA zeolite after exchange with Ca2þ cations, the number, size, and location of the cations are changed because of the ion exchange of Naþ by Ca2þ, leading to variation of pore size and related charge distribution (Figure 6.2), as described below.

Figure 6.2 Distribution of sodium cations in the 4A zeolite. from [15]. Copyright (1978) Science Press

stands for Naþ. Reproduced

352

Chemistry of Zeolites and Related Porous Materials

In the unit cell there are 12 sodium cations, of which 8 are located at the eight 6-rings, and the remaining four cations are near the three 8-ring windows. That is, one of the 8-membered rings is possibly occupied by two sodium cations. Because the sodium cation on the 8-membered ring shifts toward one particular direction, part of the 8-ring pore is blocked and this makes the effective pore diameter ˚ . When Naþ is exchanged by Ca2þ, one Ca2þ cation can replace two Naþ cations. 4A Therefore, if in each unit cell four Naþ cations are replaced by two Ca2þ, the Naþ will be removed from one 8-membered ring, and, as a result, the pore diameter of this particular ˚ , and the corresponding zeolite is called 5A zeolite. 8-membered ring will expand to 5 A þ When 70% of the Na ions are exchanged in NaA, in each unit cell three 8-membered rings are empty, and the typical composition becomes Ca4Na4[Al12Si12O48]20H2O. In the dehydrated 5A zeolite, four of the eight 6-membered rings are occupied by Naþ ions whereas the other four are occupied by Ca2þ cations. The Ca2þ cations are located ˚ above the 6-ring plane, on the plane of the 6-ring, and the Naþ cation is about 0.4 A pointing toward the a-cage. This phenomenon indicates that, under there circumstances, the variation of the zeolite pore size depends on the exchange of Naþ by Ca2þ. Figure 6.3 shows adsorption properties of zeolite A after the Naþ is exchanged by Ca2þ. It is clearly seen that the pore size is varied after the Naþ is exchanged by Ca2þ. ˚ , and therefore CO2 (with a molecular diameter of 2.8 A ˚ ) is The pore size of NaA is 4 A ˚ ) and isobutene (with able to enter this zeolite, whereas n-butane (with a diameter of 4.9 A ˚ ) are not able to get inside. When one third of the Naþ ions are a diameter of 5.6 A 2þ exchanged by Ca , the amount of adsorbed n-butane increases dramatically because the decrease in the number of cations renders more space available and the pore size is ˚ . Nevertheless, the ion-exchanged zeolite still cannot adsorb isobutane, enlarged to 5 A suggesting that the determining factor for the pore diameter variation is the exchange of Naþ by Ca2þ.

Figure 6.3 Effect of degree of Ca2þ exchange on amount of adsorption. Reproduced from [15]. Copyright (1978) Science Press. Measurement conditions: 25
C, 700 mmHg (1 mmHg ¼ 1.33322  102 Pa) for this and Figures hereinafter, unless stated otherwise

Preparation, Secondary Synthesis, and Modification of Zeolites

353

Figure 6.4 Selective adsorption in 5A zeolite. Reproduced from [15]. Copyright (1978) Science Press

The pore size of a zeolite determines the accessibility of its channels and cages to a molecule with a particular size. Figure 6.4 shows the adsorption behavior of a mixture of ˚ ), benzene (diameter 5 A ˚ ), 1,2,3,4-tetrahydronaphthalene, and n-hexane (diameter 4.9 A methylcyclohexane in 5A zeolite. It is seen that the 5A zeolite selectively adsorbs n-hexane molecules. The adsorption amounts of hydrocarbons and alcohols in this zeolite can be compared with other adsorbents in Tables 6.2 and 6.3, and from the Tables we can see that the zeolite exhibits distinct adsorption selectivity for various molecules. From these results, it is concluded that, after exchange with Ca2þ, zejolite LTA can selectively adsorb n-butanol and higher normal alcohols, n-butene and higher normal alkenes, propane and C4  C14 normal alkanes, and cyclopropane, etc., which cannot be adsorbed by zeolite 4A. On the other hand, the presence of divalent Ca2þ promotes the adsorption selectivity of 5A zeolite for polar molecules and other unsaturated compounds. Pore size is not the only parameter that affects adsorption. Molecules containing polar  groups such as -OH, >CO, -NH2 and polarizable groups such as >CO  CO< and C6H5- strongly interact with the surface of zeolites; the cations in the zeolites generate

Table 6.2 Adsorption of hydrocarbons Amount adsorbed/% Adsorbate n-butane isobutene benzene

Temperature/
C 25 25 25

Pressure (mmHg) 47 98 50

5A

Silica gel

Active carbon

9.8 0.5 0.5

3.4 4.8 35

24 26 44

354

Chemistry of Zeolites and Related Porous Materials

Table 6.3 Adsorption of alcohols Amount adsorbed/% Adsorbate n-butanol sec-butanol Isobutyl alcohol

Temperature/
C 25 25 25

Pressure (mmHg) 2.0 3.0 1.5

5A 12.6 1.4 0.3

Silica gel Active carbon 27 21 25

39 21 24

strong local electrical fields, attracting the negative centers of polar molecules or else polarizing or deforming the polarizable molecules through static electric induction. The higher the polarity or the higher the polarizability of the molecules, the easier the adsorption of these molecules by zeolites. Figure 6.5 shows the adsorption curves for the polar molecule CO and the nonpolar atomic Ar in zeolite 5A. Although these two gases have similar diameters (both smaller ˚ ), and boiling points (CO 192
C; Ar 186
C), the adsorbed amount of CO is than 4 A distinctly larger than that of Ar. Because of the presence of Ca2þ in zeolite 5A, unsaturated molecules have considerable affinity toward this zeolite. Figure 6.6 displays the adsorption isotherms of acetylene in several adsorbents, and we can see that zeolite 5A exhibits an unsaturated hydrocarbon adsorption capacity higher than those of silica gel and active carbon. Water is a highly polar molecule, and zeolite 5A after Ca2þ exchange shows strong affinity toward water. In comparison with other drying agents, 5A is superior. At relatively low partial water vapor pressure, for example, 2  104 mmHg, zeolite 5A adsorbs 14.0% water, an amount higher than those adsorbed by other drying agents such as zeolite 4A (10.3%), 13X (11.7%) and Al2O3 (2.0%). At higher temperatures, zeolite 5A still adsorbs considerable amounts of water. For example, at 100
C, it adsorbs 13%, and even at 200
C it adsorbs 4% water, whereas at 100
C the adsorption capacity of

Figure 6.5 Adsorption isotherms for zeolite 5A (75
C). Reproduced from [15]. Copyright (1978) Science Press

Preparation, Secondary Synthesis, and Modification of Zeolites

355

Figure 6.6 Adsorption isotherms of acetylene (25
C). Reproduced from [15]. Copyright (1978) Science Press

silica gel, Al2O3, and other important drying agents is nearly zero. Furthermore, zeolite 5A shows a considerable adsorption capacity for water even in a high-speed gas stream. Therefore, zeolite 5A has been widely used as a dehydration agent for medium- and highpressure air and an important drying agent for important industrial feedstocks such as rare gases and permanent gases. Because of the aforementioned properties, zeolite 5A after Ca2þ exchange has been largely used as an important selective adsorbent industrially to separate or to purify various liquids and gases. Owing to its fast adsorption of gases, zeolite 5A can be used not only for isothermal adsorption separation such as N2–H2 separation, N2–He separation, production of O2-rich air and N2–O2 separation, and purification of H2, but also for temperature-variation separation. In the oil refinery industry, zeolite 5A is used as a major adsorbent for the dewaxing process. It can separate the normal alkanes from branched-chain and cyclic alkanes in the ˚, distillates. The critical diameter of the cross-section for normal alkane molecules is 4.9 A ˚. whereas that for isoalkanes, cycloalkanes, and aromatic molecules is larger than 5 A In addition, zeolite 5A is also an excellent adsorbent used for desulfurization of oil and oil gases. In the meantime, zeolite CaA and other modified materials can be used as catalysts for some shape-selective catalysis reactions. The preparation of zeolite 5A through exchange of NaA with Ca2þ will now be briefly described. The main procedure is as follows: The as-synthesized NaA zeolite is first filtered, and washed until the pH value of the filtrate is 1112. To the wet solid of the filtered NaA is added aqueous calcium chloride (concentration CaCl2 200 g/L), the amount of which is a bit more than one molar equivalent (one equivalent is the necessary amount of Ca2þ theoretically calculated on the basis that all the Naþ cations in the NaA zeolite are completely exchanged by the Ca2þ cations), and the mixture is stirred and heated to boiling, followed by further exchange for 3040 min. Afterwards, the solid is filtered off again and washed to remove the chloride ions, and the degree of Ca2þ exchange would be about 70%. The amount of Ca2þ added to the exchange system can be estimated according to the basicity of the filtrate solution. The excess of calcium chloride reacts with the base in the system to form calcium hydroxide, which is conducive to the strength of the product. The final product is dried at 110
C and calcined to be activated.

356

Chemistry of Zeolites and Related Porous Materials

Figure 6.7 Effect of degree of Kþ exchange on adsorption capacity curve. Reproduced from [15]. Copyright (1978) Science Press. 1. H2O (4.5 mmHg, 25
C); 2. CH3OH (4 mmHg, 25
C); 3. CO2 (700 mmHg, 25
C); 4. C2H4 (700 mmHg, 25
C); 5. C2H6 (700 mmHg, 25
C); 6. O2 (700 mmHg, 183
C)

3A (KA) zeolite KA zeolite, commercially called 3A zeolite, has a composition of 0.75 K2O  0.25 Na2O  Al2O3  2 SiO2  4.5 H2O and an effective pore diameter of 0.30  0.33 nm. Its surface area is large and it is thermally stable. This zeolite can also be used as an excellent adsorbent. It is prepared through exchange of Naþ in NaA by Kþ, according to the reaction Kþ þ NaA Ð KA þ Naþ, and the ion-exchange feature is shown in Figure ˚ ) is larger 6.7. The isotherm is of B-type with wmax ¼ 1. Because the radius of Kþ (1.33 A ˚ ), when Kþ enters the NaA zeolite through exchange, it replaces than that of Naþ (0.95 A the Naþ located at the opening of the 8-membered ring. Consequently, the Kþ blocks the 8-membered ring to a certain degree and decreases the window diameter of the zeolite ˚ to 3 A ˚ , as demonstrated in Figure 6.7. Therefore, KA is called 3A-type zeolite. from 4 A When 25% of the Naþ ions in 4A zeolite are exchanged by Kþ, the pore diameter of the zeolite decreases apparently, and as a result the adsorption capacity of the zeolite for most adsorbates is reduced, and finally only the polar water molecule can enter the zeolite. KA zeolite is mainly used for drying petroleum cracking gases and natural gas. Petroleum gases contain a large amount of hydrocarbons, especially alkenes, and, during the drying process, these molecules are adsorbed in the micropores of adsorbents, and tend to polymerize and to be cracked to coke, blocking the pores, reducing the zeolite’s adsorption capacity and shortening its life, whereas when used for drying petroleum gases, zeolite molecular sieves have no such drawbacks. Because the pore size of a zeolite is uniform, and especially when small-pore zeolites such as 3A-type zeolite are used, only water is adsorbed and larger molecules are not adsorbed. The drying effects of 3A zeolite and other agents such as silica gel, and low- and high-density alumina for petroleum cracking gas clearly show that the 3A zeolite is superior. Again, when used for drying the deeply cooled separation feedstocks of petroleum gases and alkenes, 3A zeolite exhibits its ability to distinguish between water and alkenes through adsorption. Therefore, 3A zeolite is an excellent drying agent. The preparation process for 4A, 3A, and 5A zeolites is presented in Figure 6.8.

Preparation, Secondary Synthesis, and Modification of Zeolites

357

Figure 6.8 Schematic of the preparation procedure for zeolite A. Reproduced from [15]. Copyright (1978) Science Press

6.3.2

Modification of FAU Zeolite through Ion-exchange

The faujasite zeolite (Y and X type) after exchange with a rare earth (generally denoted RE) is the major active component of the most important cracking catalyst. From Table 6.1 which lists the n-hexane cracking activity (a) of various ion-exchanged faujasite zeolites it is seen that the rare-earth faujasite obtained through rareearth-exchange of NaFAU shows an activity 10 000-times higher than that of NaFAU and CaFAU, and its activation temperature is decreased to a considerable extent. This is because the cracking catalytic activity mainly depends on the acidity of the catalyst, whereas after rare-earth-exchange the surface acid sites in the Y zeolite are more widely (even more widely than HY) distributed and the acid strength of the acid centers is increased (even acid centers with H0  12.8 appear) (H0 is Hammett acid strength). Secondly, after exchange with RE3þ trivalent cations, the obtained REY zeolite exhibits thermal and hydrothermal stabilities higher than those of monovalent- and divalentcation-exchanged Y zeolite. For instance, the lattice-destruction temperature for the REY is increased by 200
C in comparison with that for the NaY zeolite if the silicon/ aluminum ratio is 5 for the two zeolites. Therefore, since the 1960s, the Y zeolite after exchange with mixed rare earth salts (REY) has become the main active component of cracking catalysts. However, under

358

Chemistry of Zeolites and Related Porous Materials

Figure 6.9 The Naþ sites in the faujasite framework structure. Reproduced with permission from [17]. Copyright (1980) Chemical Journal of Chinese Universities

mild conditions the exchange capacity is limited and the wmax ¼ 0.69, and consequently re-exchange and calcination are needed for preparation of the catalyst. To gain insight into the exchange process, further investigation is required. The rare earth faujasite exchange system will now be described in a more detailed manner by taking the La3þNaY exchange system as an example. The purpose is to help to understand the formation of rare earth faujasite zeolite, and to provide a guide to the development of new cracking catalysts composed of REY as the active component. Sherry[16] investigated the hydrothermal ion-exchange reaction of La3þ-NaY zeolite. In 1968 Sherry proposed the exchange-reaction mechanism, and reached the conclusion that the maximum exchange degree (wmax) is 0.69 on the basis of the ionexchange isotherms measured at 25 and 82
C. Through calculation based on the NaY used (unit-cell composition Na51[(AlO2)51(SiO2)141]nH2O), the La3þ ions only replace the Naþ at the SII and SIII sites (that is, only the Naþ in the faujasite cage can be exchanged, see Figure 6.9), whereas the Naþ at the SI site is not able to be exchanged. It was believed that at temperatures lower than 100
C, La3þ cannot exchange the Naþ at the SI site because the hydrated La3þ ion is too large and its size exceeds the window diameter of the b cage, whereas the hydration enthalpy of La3þ is quite high. Under the conditions for investigation, the Naþ at the SI site is not exchangeable by La3þ. Because the highest temperature at which the ion-exchange reaction was conducted by Sherry was 100
C, it is not possible to confirm the mechanism proposed on the basis of the experimental results, and as a result the detailed hydrothermal exchange mechanism for the La3þ-NaY system is not available. The aforementioned exchange reaction is one of the most important exchange processes among zeolite exchange reactions, and it is generally significant. Xu et al.[17,18] determined the ion-exchange isotherms at higher temperatures (100 and 180
C) and investigated the relationship between the exchange degree of 13 individual rare earth (Ln) ions and the Ln3þ hydration enthalpies at 180
C, and the exchange reaction rates, reaction orders, and apparent activation energies at various temperatures. On the basis of their systematic investigation, they proposed a reaction mechanism for the La3þ-NaY hydrothermal exchange reaction system.

Preparation, Secondary Synthesis, and Modification of Zeolites

359

Figure 6.10 Ion-exchange isotherms for the La3þ-NaY system at 25, 82.2, 100, and 180
C. Reproduced with permission from [17] and [18]. Copyright (1980) Chemical Journal of Chinese Universities

Xu first determined the La3þ-NaY ion-exchange isotherms at 100 and 180
C, and plotted the curves in Figure 6.10 in combination with the exchange isotherms obtained by Sherry at 25 and 82.2
C. From the exchange isotherms it is seen that at 100
C or below, the maximum exchange degree is about 0.7. According to the location of the Naþ ions in the unit cell of NaY, there are 16 Naþ at SI, whereas there are 37 Naþ at SII and SIII sites. If the La(H2O)93þ can only exchange the Naþ at the SII and SIII sites of the faujasite cages, the maximum theoretical exchange degree is 0.7 as well. Therefore, it is believed that at 100
C or lower, La(H2O)93þ can only exchange the Naþ at the SII and SIII sites. That is, Equation (6.3): ðIÞLaðH2 O93þ þ NaY!LaðH2 OÞ93þ  NaðSI ÞY þ Naþ

ð6:3Þ

La(H2O)93þ

From the structural viewpoint this is reasonable because the ion has a radius ˚ , and the La(H2O)93þ diffused into the faujasite supercages can only exchange the of 7.92 A Naþ at the SII and SIII sites. It is hard for this hydrated ion to pass through b cage window to interact with the Naþ at the SI site, or at 100
C the reaction rate is very low. When the exchange-reaction temperature is increased, the exchange degree is enhanced accordingly. The exchange isotherm at 180
C clearly shows that the exchange degree distinctly exceeds 0.7, whereas when the LaS value is 0.9, the exchange degree even reaches 1.0. The ion-exchange isotherms at 100
C or below are of the D-type, whereas those at 180
C are of the A-type, indicating that the exchange dynamics changes distinctly with temperature; that is, further exchange reaction occurs when the temperature is increased, as shown in Equation (6.4): ðIIÞLaðH2 OÞ93þ þ LaðH2 OÞ93þ  NaðSI ÞY!LaðH2 OÞ93þ  LaY þ Naþ La(H2O)93þ

ð6:4Þ

The at the SII, and SIII sites of the faujasite supercage after initial exchange are subject to electrostatic interactions, and when the temperature is increased the La3þ-H2O

360

Chemistry of Zeolites and Related Porous Materials

vibration frequency is increased accordingly, and, as a result, some of the La(H2O)93þ ions are able to overcome the dehydration energy barrier to become naked La3þ ions ˚ ] which can enter the b cage to exchange the Naþ ions located at the SI [d(La3þ) ¼ 2.3 A sites. Therefore, it is believed that the secondary exchange reaction is, in fact, a combination of the following two successive reactions [Equations (6.5) and (6.6)]: II-A : LaðH2 OÞ9 3þ ! La3þ þ 9H2 O II-B : La3þ þ LaðH2 OÞ9

3þ



NaðSI ÞY! LaðH2 OÞ93þ

ð6:5Þ

 LaY þ Naþ

ð6:6Þ

Because the hydration enthalpy (
Hhydration) for La(H2O)93þ in reaction II-A is quite large (788.5 kcal 1/mol), whereas in reaction II-B the speed at which the Naþ cations diffuse from the b cage is high, the rate-determining step should be the La3þ-NaY exchange in the II-A reaction. If this viewpoint is in agreement with the experimental observation, we can conclude that: (1) at the same temperature and for the same reaction time, rare earth ions with different
Hhydration values should have exchange degrees corresponding to their
Hhydration; (2) as the rate-controlling step, the dehydration reaction (II-A) is a first-order reaction, and consequently the whole exchange reaction (II) is also a first-order reaction. The experimental results confirmed this conclusion. Through detailed investigation, Xu et al. believed that when the RE3þ represented by La3þ exchange the Naþ in NaY zeolite, Lahydration3þ þ NaY(solid) ! LaY(solid) þ Naþ(liquid), the reaction mechanism can be considered to be as shown in Equations (6.7)– (6.10): (1) at 100
C or above, the exchange reaction proceeds through two steps:

II:

LaðH2 OÞ93þðFAU cageÞ

I: LaðH2 OÞ93þ þ NaYðsolidÞ ! LaðH2 OÞ93þ NaðSI ÞYðsolidÞ þ Naþ

ð6:7Þ

LaðH2 OÞ93þ

ð6:8Þ

þ

NaðSI ÞYðsolidÞ ! LaðH2 OÞ9

3þ

þ

þ LaYðsolidÞ þ Na

(2) the secondary exchange reaction (Equation II) consists of the following two successive reactions: II-ALaðH2 OÞ93þðFAU cageÞ ! La3þ þ 9 H2 O

ð6:9Þ

II-BLa3þ þ LaðH2 OÞ93þ NaðSI ÞYðsolidÞ ! LaðH2 OÞ9 3þ LaYðsolidÞ þ Naþ

ð6:10Þ

And it is confirmed that reaction II-A is the rate-controlling step for the total exchange reaction. This result explains why, under normal conditions for the preparation of REY catalyst, the highest exchange degree of the exchange reaction between rare earth and NaY zeolite is only 0.7, and to exchange all the Naþ in the NaY zeolite with RE3þ a secondary exchange reaction is necessary. The corresponding process is the so-called double exchange (exchange twice)–double calcination (calcination twice) technique. In addition, to increase the initial exchange degree, high temperature and high pressure or applied fields such as microwave radiation are needed in the exchange-reaction process.

1

1 cal ¼ 4.184 J.

Preparation, Secondary Synthesis, and Modification of Zeolites

361

6.4 Modification of Zeolites through Dealumination The properties and functions of a zeolite depend on its framework composition and its pore structure. The former is a reflection of the Si/Al ratio, which is highly related with the thermal and hydrothermal stability, the chemical stability, the adsorption properties, the acidity and the catalytic activity of the zeolite. For a particular zeolite there is a fixed silicon/aluminum (represented by Si/Al or SiO2/Al2O3) ratio range. In general, a zeolite with high Si/Al ratio is more able to resist heat, water vapor, and acid, and furthermore, with variation of Si/Al ratio zeolite molecular sieves usually exhibit different catalytic performances for some catalytic reactions. For instance, the catalytic activity of some multivalent-cation-exchanged faujasite zeolites (X- and Y-type) normally increases with Si/Al ratio when they are used for cracking, isomerization, and other catalytic reactions which involve carbocations formed through Bro¨nsted acid centers. Whereas for some other catalytic reactions, such as that involving mordenite as a catalyst to crack n-decane, the activity first increases with Si/Al ratio and reaches a maximum value, but then it decreases with further increase of the Si/Al ratio. On the other hand, with increase of Si/ Al ratio, the surface hydrophilicity and hydrophobicity vary accordingly, leading to variation of adsorption properties of the zeolite. The phenomena described above indicate that the Si/Al ratio of zeolites is closely related with the physicochemical properties, acidity, catalytic activity, and adsorption properties of the zeolites, and sometimes the relationship is predictable. Therefore, to vary the Si/Al ratio of zeolites through direct synthesis and modification so as to adjust and control their properties and functions is a major research subject in the field of zeolite science. On the basis of previous investigations, it is believed that the Si/Al ratio can be varied only over a certain range if direct synthetic approaches are used. For instance, it is very difficult to directly synthesize NaY zeolite with an Si/Al ratio larger than 3. Examples like this are very common, and as a result, to obtain high-silicon NaY (Si/Al > 3), it is necessary to reprocess the initially synthesized zeolite product through specially designed routes and methods and to control the formed extraframework aluminum (EFAL). Dealumination is the major technique to enhance zeolite framework Si/Al ratios, and different dealumination routes and conditions can lead to different properties and functions of zeolites. In the following subsection we will describe the dealumination routes and methods for zeolites, and in the meantime we will discuss the effects of dealumination on structures, properties and functions of zeolites. 6.4.1

Dealumination Routes and Methods for Zeolites

Zeolite dealumination is one of the most important subjects in the field of zeolite secondary synthesis and modification. For decades, zeolite scientists have been investigating dealumination routes and techniques in order to optimize the properties and functions of zeolites. Generally, there are three dealumination routes presented as follows: (1) Dealumination and ultra-stabilization of zeolites through high temperature thermal treatment and hydrothermal treatment. (2) Chemical dealumination route. This technique has been used for decades, and it involves acids (including inorganic and organic acids), alkalis, salts in solution which react with the zeolite for dealumination, or else involves inorganic ligands such as F

362

Chemistry of Zeolites and Related Porous Materials

and chelates such as ethylenediaminetetra-acetic acid (EDTA) and acetylacetonate (ACAC) that bind aluminum. In addition, gas–solid reactions such as reactions with F2 and CoCl2, etc. can also remove aluminum from zeolite frameworks to dealuminate the corresponding zeolites. Among the gas–solid reaction routes, the SiCl4 reaction method is the most commonly used for dealumination and siliconenrichment. (3) Optimal combination of high-temperature hydrothermal and chemical dealumination routes. Zeolite dealumination not only increases the Si/Al ratio, but also results in various EFAL atoms in the pores, channels, and surfaces of zeolites. Furthermore, dealumination may also lead to framework defects, local structure collapse, or even blocking of some channels. As a result, the properties and functions of zeolites, especially their acidity, channel structures, and thermal stability and catalytic performance, may be changed. These changes depend on the routes, methods, and conditions adopted for dealumination. In general, because of the respective advantages and disadvantages of different dealumination routes, zeolite chemists have been striving to find new dealumination methods through practice on the basis of the available routes to improve and to optimize the dealumination effects. In the next subsection we will describe the aforementioned dealumination routes in more detail by taking zeolite Y as an example. The dealumination procedure and conditions and their influence on structure, properties and function of the dealuminated zeolites will be discussed. 6.4.2

High-temperature Dealumination and Ultra-stabilization[19]

Usually the starting material of this dealumination route is ammonium zeolite. Taking NH4Y for example, this zeolite is formed by repeated exchange of NaY of a certain Si/Al ratio with NH4þ solution. The ammonium salts used for exchange include ammonium chloride, ammonium sulfate, ammonium nitrate, and so on, and 2/3 of the Naþ ions in NaY zeolite can be easily exchanged. When the Naþ content in the zeolite reaches 1025%, it is difficult to exchange more sodium ions by NH4þ. To reduce the Na2O content in the zeolite to lower than 0.3%, one can calcine (at 200600
C) the corresponding zeolite when the Na2O content is reduced to 4% (amounting to 1/3 of the Naþ remaining in the zeolite) and further exchange the zeolite. In this way, the Na2O content in the zeolite can be lowered to below 0.1%, because the Naþ cations in NaY move from the unexchangeable sites (see SI in Figure 6.9) to the exchangeable ones SII and SIII. The redistribution of cations makes the exchange reaction easier to proceed. When using NH4þ for exchange, addition of acid will reduce the pH of the exchange solution, in favor of the removal of Naþ ions, whereas some organic acids can act as a buffer to keep the pH value of the system low and the exchange reaction will be more effective. Heating of ammonium zeolite to 260
C results in decomposition of NH4þ and the formation of an Hþ-containing intermediate called H-type zeolite. Because Hþ is very small and it has a strong polarizing ability, this cation can form hydroxyl with the zeolite framework oxygen. At high temperatures, these hydroxyls can be removed as water, resulting in decationed zeolites. There is no boundary line between H-type and decationed zeolites, and what usually forms within a certain temperature range is a mixture of the two. Dealumination through calcination of NH4Y to above 500
C leads to

Preparation, Secondary Synthesis, and Modification of Zeolites

363

structure destruction. Therefore, high temperature calcination of NH4Y alone cannot be used to obtain dealuminated product. Usually it requires high temperature (600  900
C) water-vapor treatment of NH4Y zeolite to achieve dealumination and frameworkstructure stabilization. The product thus obtained through this high-temperature hydrothermal route is USY (ultra-stable Y zeolite). Scherzer.[20] proposed a dealumination and ultra-stabilization process on the basis of the investigation carried out by Kerr and Maher, and the process proceeds as shown in Equations (6.11)–(6.13): (A) framework dealumination Si

Si (1)

O Si

O

Al

O

H –

H

High temperature O

Si

Water vapor

Si

O

O

H

O

H

Si

ð6:11Þ

Al(OH)3

H O

Si

Si

ð2ÞAlðOHÞ3 þ Hþ ½Z!AlðOHÞ2 þ ½Z þ H2 O

ð6:12Þ

(B) framework stabilization Si

Si

O

O

H Si

O

H

H H

O

Si

+ SiO2 Water vapor

Si

O

O Si

O

Si

ð6:13Þ

Si

O Si

McDaniel, Maher, Kerr and Shipman have studied[21,22] framework dealumination under high-temperature conditions for (A) in detail, and they concluded that the dealumination product is complicated and in the extraframework channels, cages, and on the surface there are various extraframework aluminum species (generally denoted EFAL) (Figure 6.11). Scherzer concluded that in the USY formed from NH4Y through high-temperature water-vapor treatment, the aluminum exists in three forms: (1) in the USY framework, (2) outside the framework as 6-coordinated octahedral Al, (3) on the USY surface as differently coordinated Al which results in surface Al enrichment. The Al in the USY framework produces B acid centers, and although the number of the acid centers is smaller than that of the parent Y zeolite, the strength and distribution of these centers may favor cracking catalytic reactions more. It was also pointed out by Scherzer that for dealumination and framework stabilization in (B), at high temperatures, water vapor interacts with the O

Si

O

Al

O

bonds in

364

Chemistry of Zeolites and Related Porous Materials

Figure 6.11 Presence of framework (1) and nonframework (2) aluminum species(cations and neutral species) in USY. Reproduced with permission from [20]. Copyright (1984) American Chemical Society

the framework, resulting in dealumination and formation of hydroxyl nests, and in the meantime, unavoidably small parts of framework will collapse to release SiO2, Si(OH)4 Si O

which subsequently polymerize to fill the hydroxyl nests to form

Si O

Si

O

Si

O Si

structure or

Si

O

Si

through condensation of adjacent -Si OH bonds.

˚ ) is shorter than that (1.75 A ˚ ) of Al Since the Si O bond length (1.66 A O, the crystal lattice of zeolites is shrunken and the structure is stabilized after dealumination and silicon enrichment, as confirmed as follows. Of course, unavoidably there exist silicondeficient hydroxyl nests after both dealumination and ultra-stabilization, and some mesopores will be generated in the USY framework. 6.4.3

Chemical Dealumination and Silicon Enrichment of Zeolites

Liquid Phase Dealumination and Silicon Enrichment In solution, zeolite framework dealumination can be realized through interactions of zeolites with acids, salts, and chelates. 1. Acid treatment Usually, inorganic and organic acids can be used for framework dealumination of zeolites, and the acids include hydrochloric acid, nitric acid, formic acid, acetic acid, and so on. According to its acid-resistance ability, hydrochloric acid can be used for highsilica zeolites such as mordenite, clinoptilolite, erionite, etc. We will take mordenite as an example to describe this dealumination method (Table 6.4). The first step in the treatment of mordenite using hydrochloric acid is to convert the zeolite into H-type, and further acid treatment can then enlarge the pore diameter through dealumination. After partial dealumination, the Si/Al ratio of the zeolite is increased and the heat-resistance, waterresistance, and acid-resistance abilities are enhanced.

Preparation, Secondary Synthesis, and Modification of Zeolites

365

Table 6.4 Acid-treatment conditions of zeolite mordenite Hydrochloric acid-treatment Raw material conditions mordenite composition Concentration Temperature Time /(mol/L) (T/
C) (t/h) No. Na2O:Al2O3:SiO2:H2O

Composition of product after dealumination Na2O:Al2O3:SiO2:H2O

1 2 3 4 5

0.38 0.53 0.61 0.31 0.46

0.78 0.88 0.97 1.03 1.01

1 9.43 1 8.9 1 9.9 1 8.8 1 10.6

5.3 7.0 3.6 4.9 5.1

12 12 6 6 4.7

100 100 20 100 100

12 1 16 20 1

1 1 1 1 1

15.7 14.8 10.0 9.7 12.7

10.7 8.4 4.5 4.1 7.3

In addition, for mordenite after acid treatment, the acid usually dissolves the amorphous species that are blocking the channels and decreases the stacking defects, and the small-radius protons replace the cations with larger radius, leading to pore-size enlargement and adsorption-capacity enhancement. For zeolite Y and other zeolites with a low Si/Al ratio, acid treatment is inappropriate because this will generate a large amount of hydroxyl nests and will lead to lattice destruction. 2. Chelating dealumination Zeolite framework dealumination using chelates in solutions was initiated by Kerr in 1968.[22] With EDTA as chelating agent, 70% at most of the aluminum in a zeolite framework can be removed, whereas the zeolite’s crystallinity is largely retained. Since the 1970s, on the one hand the EDTA-chelating dealumination approach has been improved, and, on the other, numerous other chelating agents such as acetylacetone, tartaric acid, oxalic acid, and oxalate and aminoacid chelates have been tested for dealumination under various conditions. Below we will take the dealumination of NaY using H4EDTA as an example to describe the elementary process of chelating dealumination. First, H4EDTA interacts with NaY, converting the latter into the H-form [Equation (6.14)], and Equation (6.15) represents the hydrolysis of AlO4 in the vicinity of Hþ to form aluminum hydroxide. In Equation (6.16), Al(OH)3 interacts with proton centers to form cationic Al(OH)2þ and H2O, whereas in Equation (6.17), the Al(OH)2þ is exchanged by Naþ and chelated by EDTA. This guarantees the dealumination. For NaY, the dealumination degree ranges from 25 to 70%, and if it surpasses 70% the crystal lattice will be destroyed to a considerable extent.

Si ONa 2 Si

O

–

Al O

Si

Si +

O

O O

Si

O

O

+ H4EDTA

2 Si

O

Al O Si

O H O

Si

O

+ Na2H2EDTA

O

ð6:14Þ

366

Chemistry of Zeolites and Related Porous Materials Si Si

O Si

O

O

H Al O O

+ 3H2O

O

Si

O

OH Si

OH

HO

ð6:15Þ

+ Al(OH)3

O

Si

OH

O

O Si

Si Si O Si

O

Si Al(OH) 2+ O

O

H Al O O

Si

O

+ Al(OH)3

Si

O

Al O Si

Si

O

+ H2O

ð6:16Þ

O

Si

Si Al(OH) 2+ O O O

O

O

Si

Si

Si

Al O

O

–

–

O

Si O

ONa O

+ NaH2EDTA

Si

O

Al O

+

O O

Si

O

+ NaA1EDTA*H2O + H2O

O

Si

ð6:17Þ The chelating dealumination method proposed by Kerr and Shipman[22] takes advantage of the interactions between H4EDTA and NaY, and the procedure is as follows. A certain amount of NaY zeolite is boiled in water and to the system is added dropwise a suitable quantity of H4EDTA solution until the ratio H4EDTA/zeolite cation ¼ 0.250.50 (based on the amount of aluminum). This process is accomplished in 18 h under reflux. After reaction, the product is filtered off, dried, and finally calcined at 800
C in an inert atmosphere. The dealumination product after calcination contracts its unit-cell dimensions by about 1%. The dealumination by EDTA method usually leads to surface dealumination of Y zeolite, and in the meantime a large number of hydroxyl groups or silanol nests may be generated on the surface. 3. Dealumination and silicon-enrichment reaction of (NH4)2SiF6 (AHFS) with zeolites Besides the hydrothermal method for preparation of ultra-stable Y zeolite (USY), Breck and Skeels[23] in 1983 invented a new secondary synthesis method for silicon-enriched zeolites. This method uses an ammonium hexafluorosilicate solution to remove the aluminum atoms from the framework structure of Y zeolite to the solution, and to insert silicon atoms back into the Al-removal vacancies in the framework so as to form a more or less perfect Y zeolite with a high Si/Al ratio. In comparison with the USY prepared by the hydrothermal method, the framework silicon-enriched Y zeolite obtained through the current technique possesses fewer framework hydroxyl vacancies, and the resulting zeolite has an ideal crystal lattice, and hence higher structural stability. Meanwhile, there

Preparation, Secondary Synthesis, and Modification of Zeolites

367

are no extraframework aluminum fragments in the resulting zeolite, and therefore the nonselective catalytic reactions will be limited and the coke selectivity is improved if the zeolite is used as a catalyst. Therefore, the zeolite obtained through this dealumination method is considered to be a better active component of residual oil cracking catalyst than is the USY zeolite. Nevertheless, since this approach involves ammonium hexafluorosilicate as the Al-removal agent, it produces fluorine-containing waste that requires treatment. This is the shortcoming of the technique. Min investigated the reaction mechanism of this dealumination approach.[24] Breck and Skeels had found that, in the ammonium hexafluorosilicate solution, the extraframework silicon atoms could substitute for the aluminum atoms in the crystal lattice of the zeolite framework, but the mechanism of the whole reaction process was not clear. Through 29Si,27Al-MAS-NMR and IR characterization techniques, Min and coworkers investigated the removal of Al atoms from the zeolite framework, the chemical environment of the extraframework silicon atoms, and the species of the silicon that can be inserted into the Al-removal framework vacancies during the reaction process, and finally they proposed the reaction mechanism [Equations (6.18)–(6.25)] as follows. ðNH4 Þ2 SiF6 !2NH4þ þ SiF6 2 SiF6

2

þ H2 O!SiF5 ðOHÞ

SiF5 ðOHÞ

2

2

ð6:18Þ 

þF þH

þ

2



þ H2 O!SiF4 ðOHÞ2

SiF4 ðOHÞ2

2

SiF3 ðOHÞ3

2

þ H2 O!SiF3 ðOHÞ3

ð6:19Þ

þF þH

2



þ þ

ð6:20Þ

þF þH

ð6:21Þ



ð6:22Þ

þ

þ H2 O!SiðOHÞ4 þ 3F þ H

SiF6 2 þ 4H2 O!SiðOHÞ4 þ 6F þ 4Hþ

ð6:23Þ

ONH4+ 6F- +

O

Al

O

+ 4H+

OH

OH HO

+ AlF 63– + NH4+

(6:24)

OH

O

(Zeolite)

(Framework vacancy)

O OH Si(OH)4 +

OH

HO OH

O

Si

O

+ 4H2O

(6:25)

O

(FSY zeolite)

In the reaction process of dealumination and silicon-addition for zeolites with ammonium hexafluorosilicate solution, the ammonium hexafluorosilicate is first hydrolysed to form Si(OH)4, and the released F interacts with the framework Al atoms and

368

Chemistry of Zeolites and Related Porous Materials

remove the latter from the framework as AlF63. Afterward, the Si(OH)4 molecule is inserted into the vacancy, resulting in silicon-rich zeolite, and the isomorphous substitution reaction of framework Al by extraframework Si is accomplished. On the basis of the research results, Min and coworkers developed a series of patent techniques.[12] They invented the method for the preparation of Si-rich zeolites using the side-product silicon hydrofluorosilicic acid as the Al-removing agent (CN 1,048,836A); the preparation method using NH4F as the Al-removing agent (CN 1,011,881A); the preparation method using HY as the starting material (CN 1,088,247A); the preparation method using H2SiF6 as the starting material (CN 1,102,400A); and the preparation method using silicic acid instead of hydrofluorosilicic acid for Si-enrichment agent (CN 1,121,484A). All of these patent techniques can be used to prepare Si-rich zeolites, and they provide new vistas for industrial production. At present, the aforementioned patent techniques have been industrialized for preparation of Si-rich zeolites. The industrial equipment based on the CHZ-3 residual oil cracking catalyst prepared using the corresponding zeolite as the active component has been in operation for heavy-oil catalytic cracking (RFCC). From the operation results of the RFCC industrial apparatus it is seen that in comparison with the commonly used residual oil cracking catalyst CHZ-2, the CHZ-3 catalyst with the Si-rich zeolite as the active component increases the content of lowpressure residual oil by 8.02%, whereas it decreases the oil pulp yield by 1.34% under circumstances where the coke yield remains constant. Meanwhile, the light-oil component yield increases by 1.10%, whereas the combined yeild of liquefied gas þ light oil increases by 1.73% if CHZ-3 catalyst is used, indicating that this catalyst has excellent activity-stability as well. In 1983, Skeels and Breck [23] successfully performed dealumination and Si-addition for zeolite Y (NH4þ- and Naþ-type) and mordenite (NH4þ-type) using (NH4)2SiF6 solution under mild conditions. This dealumination and Si-addition method is based on the following chemical reactions: If the zeolite is H-mordenite or HY, because the Hþ reacts with the F released from hydrolysis to form HF which can undergo the second-step reaction shown in Equation (6.26): O

H 3O + _ O

+ H 2O + AlF 3

+ 3HF

Al O

OH HO

O

(6:26)

OH HO

As a result, more Al will be complexed to leave the framework, increasing the Si/Al ratio of the zeolite. Table 6.5 lists the chemical and physical properties of the product formed by the zeolite after interaction with the fluorosilicate solution. The Si/Al ratio of the zeolite sample is 4.84. Under the first reaction conditions, the NH4Y zeolite reacts with excess of (NH4)2SiF6 and 37% of the framework Al is replaced by Si, leading to a product designated LZ-210 (9.31). Under the second reaction conditions, NH4Y reacts with excess of (NH4)2SiF6 and 57% of the framework Al is replaced by Si, leading to a product called LZ-212 (14.84). The Si/Al ratio of H-type mordenite is 14.00, and when it

Preparation, Secondary Synthesis, and Modification of Zeolites

369

Table 6.5 Examples of dealumination through the (NH4)2SiF6 (AHFS) method Raw AHFS AHFS dealumination product Raw material material dealumination NH4Y LZ-210 LZ-212 H-MOR product LZ-211 SiO2/Al2O3 4.84 9.31 F content (%) 0 0.05 Crystallinity I/Is (%) 100 106 Unit-cell dimension (A˚) 24.67 24.49 Lattice-collapse temperature (T/
C) 860 1037 SiO2/Al2O3 ratio in the dealuminated product 1. Theoretical SiO2/ 9.30 Al2O3 calculated on the basis of the (NH4)2SiF6 amount added 2. Experimental 9.31 SiO2/Al2O3 of the product

14.84 0.15 93 24.39

14.00 0 10

27.70 0.01 106

1128 12.41

13.85

reacts with excess of (NH4)2SiF6 solution, 49% of the framework Al is replaced by Si, the corresponding product being called LZ-211. The conditions for the preparation of LZ-210 zeolite are as follows: NH4Y (1015 g) is added to 100 mL of water and the mixture is pre-heated at 7595
C. To this mixture is added 1 mol of (NH4)2SiF6 solution at a rate of 0.005 mol per mol of framework Al per min, and the pH value of the solution should be kept at about 6 during the reactant addition. Alternatively, the pH of the solution can be controlled through addition of buffer compounds such as salts like NH4Ac, and the (NH4)2SiF6 solution is added to the buffered system followed by heating of the mixture at 7595
C for 13 h. In the reaction, the amount of (NH4)2SiF6 added is determined by the Si/Al ratio of the product and the complexing degree. After thorough investigation of the properties and functions of products obtained by calcination of various dealuminated zeolites at 550
C, it was found that the technique has the following advantages: compared with the parent zeolite, the dealuminated product retains its crystallinity and adsorption capacity; it has a contracted unit cell and higher thermal stability; and the framework defects and vacancies are scarce, as confirmed by various characterizations. It is worthwhile to point out that the dealumination and Si-addition can be based on calculations according to the Si/Al ratio required. Below we will describe NH4Y (with 80% of the Naþ being exchanged by NH4þ) and HMOR zeolites.[23] Because the aforementioned technique shows excellent effects for dealumination and silicon-addition, this technique has been applied to the adjustment of Si/Al ratio for various zeolites such as erionite, zeolite L, clinoptilolite, and chabazite.

370

Chemistry of Zeolites and Related Porous Materials

Figure 6.12 Apparatus for SiCl4 dealumination. Reproduced with permission from [25]. Copyright (1985) Royal Society of Chemistry. 1, Thermocouple. 2, Zeolite powder bed. 3, Quartz wool. 4, Fused quartz plate. 5, N2 gas saturated with SiCl4 at room temperature

Dealumination and Si-addition through vapor phase methods It has been reported that F2, phosgene COCl2, and SiCl4 can be used for gas-phase dealumination. The commonly used method to obtain dealuminated zeolites involves SiCl4 (gas)–zeolite (solid) replacement. In 1985, Beyer[25] reported the preparation of high-silica Y-zeolite and even pure-silica Y-zeolite through interaction of SiCl4 with NaY for dealumination and Si-addition. This technique is based on the fact that at high temperatures SiCl4 can easily react with zeolites to dealuminate, and the main reaction is given by Equation (6.27). high temperature

Na½AlO2 ðSiO2 Þx ðsolidÞ þ SiCl4 ðgasÞ ! ½ðSiO2 Þxþ1 ðsolidÞ þ AlCl3 ðgasÞ " þNaClðor NaAlCl4 Þ

ð6:27Þ

This dealumination route uses the laboratory apparatus shown below, and the corresponding procedure is as follows: NaY zeolite (25 g) was placed in the quartz reactor shown in Figure 6.12 dry N2 gas (10 L/h) was passed into the system. The temperature is increased to 620
C at a increasing rate of 10 K/min and this temperature is kept for 2 h. When the zeolite is completely dry, the temperature of the reactor is reduced to 520 K (247
C), and a N2 gas stream saturated with SiCl4 is passed again into the reactor (5 L/h). The temperature is now adjusted to the reaction-required value (TR), which is kept for about 40 min. Afterward, the SiCl4 stream is stopped, and the system is blown for another 15 min, then cooled to room temperature. The product is washed with distilled water until no Cl is present. The final product is dried at 400 K (127
C), and if NH4þ-type zeolite is required, the product is subject to exchange with 1 mol/L aq. NH4Cl at room temperature several times. The dealumination degree is closely correlated with reaction temperature (TR) (Table 6.6). To prevent the crystal lattice from destruction or collapse, it is required that the reaction temperature be kept below 770 K (497
C), and the rate of temperature increase

Preparation, Secondary Synthesis, and Modification of Zeolites

371

Table 6.6 Reaction temperature (TR) and dealumination of product Initial reaction temperature/K 520 521 521 521 520

Normal reaction temperature (T R) 520 600 675 720 770 900

Si/Al of product 4.8 6.2 12.0 53.0 lattice collapse

should be carefully controlled because the interaction of SiCl4 with zeolite NaY is highly exothermic. To remove the deposited NaAlCl4 in the Y zeolite channels, thorough washing and appropriate acid treatment of the product is necessary. To avoid incompleteness of dealumination reaction which causes silanol nests and defects, the SiCl4dealuminated product needs to be further treated by high-temperature water vapor. Combination of High-temperature Hydrothermal and Chemical Dealuimination Routes As discussed above, either the hydrothermal or the chemical dealumination route alone has both advantages and disadvantages. Therefore, usually these two routes are optimally combined in practice; that is, on the basis of hydrothermal treatment the zeolite is further dealuminated using acids such as HCl or HNO3, or alkalis such as NaOH, or salts such as AHFS or chelates such as EDTA. Alternatively, on the basis of chemical dealumination, the zeolite is treated further by high-temperature water vapor. The combination of these two methods may achieve good results, and various combinations can be conducted on the basis of comparison for further improvement. In the former combination, the purpose is to chemically remove the residual EFAL in the USY framework, and the vacancies and defects (which affect the crystallinity and structural stability) formed by rapid dealumination rate after high-temperature hydrothermal ultra-stabilization. Lynch has studied the combination and its effect in detail[26]. He summarized the side-effects of hydrothermal treatment on the generation of EFAL species and catalytic and transport properties reported in the literature since Scherzer [20]. On the basis of summarization, Lynch investigated the state of EFAL, and the effect of acid, AHFS, and EDTA on the removal of the EFAL. The investigation provided vistas for further improvement of dealumination effect through combination of high-temperature hydrothermal and chemical dealumination routes. The following text describes the results achieved by Lynch. The USY[26] obtained by high-temperature 923 K (650
C) water vapor treatment is further treated with 0.1 3 mol/L HNO3, 0.4 mol/L AHFS, and 0.05 g/mL Na2H2EDTA at different leaching strengths, and the results are presented in Tables 6.7, 6.8, and 6.9. In the three Tables presented above, the Al concentration in the unit cell is calculated on the basis of the total 192 T (Si or Al) atoms in the cell, whereas the EFAL quantity is calculated from the total Al quantity after subtraction of the framework Al quantity. Figure 6.13 shows the EFAL-removal curves based on the data in the three Tables, and the purpose of this Figure is to compare the EFAL-removal effects from the USY using HNO3, AHFS, and H2Na2EDTA.

372

Chemistry of Zeolites and Related Porous Materials

Table 6.7 Dealumination reaction results of hydrothermally treated (923 K; 650
C ) USY with dilute HNO3 of various leaching strengths HNO3 treatmenta

0

0.12

0.31

0.61

Crystallinity (%) Total Si/Al Framework Si/Al

83 2.8 11

94 5.9 10

95 6.0 11

94 6.2 12

1.22 94 6.8 15

1.83

2.32

85 6.9 13

60 10.8 17

2.75

3.66

69 64 21.8 60 >50 >50

a

Leaching strength represents the HNO3/Al molar ratio.

Table 6.8 Dealumination reaction results of hydrothermally treated (923 K; 650
C) USY with AHFS of various leaching strengths AHFS treatmenta 0

0.13

0.22

Crystallinity (%) 83 75 75 Total Si/Al 2.8 2.95 3.2 Framework Si/Al 11 10 10

0.32

0.42

0.52

0.64

0.75

0.80

0.90

1.0

76 80 5.3 6.0 10 9

79 6.9 9

89 7.5 10

77 8.4 11

80 89 9.4 10.9 12 12

68 14.0 14

a

Leaching strength represents the (NH4)2SiF6/Al molar ratio.

From the three curves shown in Figure 6.13 it is seen that the EFAL-removal mechanisms for treatment of USY samples using dilute HNO3, AHFS, and EDTA are somewhat different. The residual EFAL states after dealumination with HNO3, AHFS, and EDTA at various leaching strengths characterized by Lynch using XRD and NMR techniques are presented in Table 6.10. The unit cell of USY after high-temperature hydrothermal dealumination (crystallinity ¼ 83%) contains 34 EFAL, and the states of the EFAL detected by means of XRD Rietvied analysis and NMR (3Q-NMR and MAS-NMR) characterization are presented in Table 6.10. If using the AHFS method with a leaching strength of 0.420.52 for further treatment, 5–7 EFAL will remain in the solid unit cell whereas the state of the AFAL will be AlIV (distorted AlO4 tetrahedron). The result is similar but a bit inferior if the EDTA method is used (see Figure 6.13). When the HNO3 treatment method is used, the effect is even worse; not only does the unit cell contain 1/3 EFAL that cannot be removed, but also the existent states of the EFAL species are complicated (possibly there are amorphous materials). These results provide information for further improvement of high-temperature hydrothermal-chemical dealumination combination.

Table 6.9 Dealumination reaction results of hydrothermally treated (923 K) USY with EDTA of various leaching strengths EDTA treatment

0

0.46

0.56

0.60

0.91

2.05

2.09

Crystallinity (%) Total Si/Al Framework Si/Al

83 2.8 11

94 3.0 9

92 3.5 7

86 4.7 9

93 7.8 11

91 8.5 9

91 10 10

a

Leaching strength represents the H2Na2EDTA/Al molar ratio.

Preparation, Secondary Synthesis, and Modification of Zeolites

373

Figure 6.13 The EFAL-treatment effects using different agents. Reproduced with permission from [26]. Copyright (2000) Elsevier

6.5 Isomorphous Substitution of Heteroatoms in Zeolite Frameworks Heteroatom zeolites are prepared through partial isomorphous substitution of framework Si, Al, and P by heteroatoms. The elements entering the zeolite frameworks can be some main group elements or variable-valence transition metals. In general, one kind of heteroatom is introduced into the zeolite but it is also possible for more than one kind of heteroatoms to enter the structure of a zeolite. In heteroatom zeolites, because of the introduction of special metals or nonmetallic elements, the acidity, redox properties, catalytic activity, and other functions of the parent zeolite may be adjusted. Through modification of heteroatom introduction, the resulting zeolites can become excellent catalysts and functional materials. In the 1970s, Ueda and Flanigen, respectively, reported the synthesis of beryllium- and phosphorus-containing zeolites at the American Zeolite Conference, and these reports attracted attention to the study of heteroatomzeolite synthesis. In 1978, Laszlo reported the synthesis of V-, Cr-, Fe- and As-containing zeolites, whereas in 1980. Taramasso reported at the fifth International Zeolite Conference the synthesis of four types of boron-silica zeolites. In 1982, Whittam synthesized

Table 6.10 States of residual EFAL XRD method AlIV (EFI) Parent Parent Parent Parent

USY USY þ HNO3 USY þ AHFS USY þ EDTA

þ þ þ

Al(OH)2þ (EF2) þ þ

NMR method AlIV

AlV

AlVI

þ þ þ þ

þ

þ þ

NMR-undetectable þ þ þ

374

Chemistry of Zeolites and Related Porous Materials

Nu-5-, Nu-13-type heteroatom zeolites. In the aforementioned zeolites, Ge can replace Si, whereas Ga, B, Fe, Cr, V, Mo, As, Ti, etc. can partially substitute for aluminum and silicon in zeolite frameworks. In 1986, Flanigen[27] systematically reported the formation of heteroatom phosphate zeolites such as SAPO-n (MeAPO-n), ElAPO-n, and AlAPSO-n through substitution of Si, metals (Me), and main group elements (EL) into the frameworks of AlPO4-n molecular sieves. Pang started investigation in synthesis of boron-silica MFI zeolites since 1982, and later on the synthesis was extended to Cr, Ti, Zr, Fe, Co, V, Ga, Ge, Sn, Mo, and W-containing zeolites. A series of significant results were achieved by Pang et al.[28] As for aluminosilicate and aluminophosphate zeolites, the preparation of heteroatom zeolites is also through hydrothermal or solvothermal methods. The heteroatom sources can be oxides, salts, and complexes. The heteroatom zeolites can be directly prepared through crystallization of the mixture of reactants at a certain temperature. The synthesis of heteroatom zeolites can also be realized through isomorphous substitution; that is, modification of parent zeolites or secondary synthesis. As in the case of enhancement of the Si/Al ratio of aluminosilicate zeolites through gas– solid reaction of zeolites with SiCl4 vapor at a certain temperature, heteroatom zeolites can be prepared through gas–solid and liquid–solid substitution reactions in a similar way to the preparation of Si substitutes for Al in zeolite frameworks. For example, the gas– solid substitution reaction of BCl3 or B2H6 with ZSM-5 at a particular temperature can form B-ZSM-5. Similarly, using TiCl4 to react with ZSM-5 results in Ti-ZSM-5, whereas liquid–solid substitution reaction of an alkaline solution of gallium salt or gallium fluoride solution with zeolites leads to gallium-substituted zeolites. This kind of secondary synthesis is, in fact, isomorphous substitution of framework elements. Barrer described in detail the isomorphous substitution of zeolite framework elements in his monograph ‘Hydrothermal Chemistry of Zeolites’ published in 1982.[29] In addition, it should be pointed out that the isomorphous substitution approach has its own advantages and disadvantages, just as the direct hydrothermal method does. Its advantage lies in that it can be used to obtain heteroatom zeolites that cannot be prepared through direct hydrothermal synthesis. Furthermore, in the zeolites synthesized though direct hydrothermal methods, usually the heteroatom content is low (<5% and in most cases about 3%). However, through the secondary synthesis by isomorphous substitution a high content of hetetoatoms in zeolite frameworks can be achieved, and it is possible to obtain stoichiometric amounts of heteroatoms to meet application requirements through control of the hydrothermal reaction conditions. In this section, we describe only recent research results concerning the isomorphous substitution secondary synthesis of heteroatomcontaining zeolites. 6.5.1

Galliation of Zeolites – Liquid–Solid Isomorphous Substitution

In 1986, Liu and Thomas[30] reported the systematic study of the preparation of galliumcontaining zeolites through reaction of Ga(OH)4 with high-silica zeolites in an alkaline solution. They invented the ‘galliation’ concept in their publication. We will take silicalite-II for example to describe zeolite galliation. Silicalite-II has a very high Si/ Al ratio (>1000). The silicalite-II (1 g) is treated with 30 mL of 0.1 mol gallate solution (0.0278 mol/L Ga2O3, 0.10 mol/L of NaOH) at 20100
C under stirring for 24 h (the ratio of zeolite over solution is 1:30). The treated silicalite-II is filtered off, washed, and

Preparation, Secondary Synthesis, and Modification of Zeolites

375

Table 6.11 Galliation of several zeolites with various Si/Al ratios OH Zeolite concentration/(mol/L) ZSM-5 MAZ OFF LTL FAU FAU FAU FAU

0.100 0.220 0.366 0.440 0.280 0.280 0.100 0.260

Si/Al Ga2O3 concentration/(mol/L) 0.0267 0.0200 0.0285 0.0280 0.0175 0.0175 0.0265 0.0385

Ba

Ab

16.45 4.25 3.77 2.94 2.50 2.00 1.54 1.23

15.05 4.03 3.54 2.72 2.36 1.83 1.48 1.22

Si/Ga of galliated zeolite 13.9 35.1 43.8 19.7 56.7 65.3 73.3 90.0

a

B stands for the Si/Al ratio of the zeolite before galliation; A for the Si/Al ratio of the zeolite after galliation.

dried, and a galliated silicalite product is obtained. The galliation degree is represented by the ratio Si/Ga of the zeolite product. With an increase of galliation temperature, the degree of galliation or isomorphous substitution increases. In the aforementioned reaction, the Si/Ga ratios in the silicalite-II products galliated at 20, 45, 75, and 100
C for 24 h are 30.1, 26.4, 9.5, and 9.8, respectively. It is confirmed by powder X-ray diffraction, IR, scanning electron microscopy, solid-state high-resolution 29SiNMR, electron probe microanalysis, chemical analysis, and adsorption and surface acidity analysis that the gallate ions in the reaction system substitute for Si in the framework of the zeolite during the treatment process. To gain further insight into the galliation mechanism, Liu.[31] conducted additional investigations into Ga(OH)4 galliation of zeolites with various Si/Al ratios, and the results are listed in Table 6.11. Liu[31] reached the following conclusions for the galliation isomorphous substitution and its mechanism on the basis of the aforementioned investigation in zeolite galliation: (1) A variety of zeolites, especially those with high silica contents, can be galliated with Ga(OH)4 under mild conditions; (2) The nature of the galliation reaction is the substitution of Si4þ in the zeolite framework by Ga3þ, and the substitution site is Si(0Al) whereas the reaction route is shown by Equation (6.28). Si

Si

O Si

O

Si

O O

Si

+

Ga(OH)4

–

Si

O

Ga

O

O

Si

Si

O

Si

+

Si(OH)4

(6:28Þ (3) The Lowenstein rule is followed by the isomorphous substitution.

376

Chemistry of Zeolites and Related Porous Materials

Figure 6.14 Plot of mechanism of gallium–aluminum substitution. Reproduced with permission from [32]. Copyright (1991) Royal Society of Chemistry

In 1991, Dwyer and Karim[32] reported another galliation route, which uses ammonium hexafluorogallate to react with zeolite Y under mild conditions, and the reaction proceeds as shown in Equation (6.29). 3x GaF3x þ ½AlO4 ðzeoliteÞ !½GaO4  x ðzeoliteÞ þ AlFx

ð6:29Þ

The Ga substitutes for Al isomorphously and the resulting zeolite is galliated to contain Ga (Figure 6.14). The galliation mechanism in this isomorphous substitution is different from that proposed by Liu for the replacement of framework Si4þ in high-silica zeolites by Ga3þ of Ga(OH)4. In the system containing fluoride, the (NH4)3GaF6 in solution is readily hydrolysed to form F and GaO45 and this affects the Ga3þ –Al3þ(zeolite) substitution to a great extent. F is a strong complexing agent, and hence in the presence of a large amount of F in the solution the framework Al can be complexed by the F to enter the liquid phase, leaving vacancies in the framework. These vacancies are subsequently occupied by [GaO4] species entering the framework, facilitating lattice stabilization and the formation of galliated zeolite. This mechanism is similar to that for dealumination of zeolites using (NH4)2SiF6. In order to further confirm the Ga3þ– Al3þ(zeolite) substitution mechanism, Dwyer and Karim synthesized NH4Y zeolite with various Si/Ga ratios, and they confirmed the gallium–aluminum substitution mechanism on the basis of unit-cell dimensions and IR variation results. The Ga O bond length ˚ ) is larger than that of the Al ˚ ), and therefore the unit-cell (1.72 A O bond length (1.69 A dimensions of GaY are increased in comparison with the zeolite before galliation. In contrast, when (NH4)2SiF6 is used for dealumination and Si-addition, the unit-cell dimensions will be proportionally reduced. At present, this liquid–solid substitution approach has been extended to the secondary synthesis of zeolites containing Si, Fe, Sn, Ti, Cr, and other heteroatoms. An exception is that BF4 cannot be used for liquid-phase dealumination and boron-addition of zeolites to prepare boron-containing zeolites.[33] In the following galliation of NH4Y will be taken as an example to discuss the isomorphous substitution secondary synthesis technique.

Preparation, Secondary Synthesis, and Modification of Zeolites

377

NH4Y zeolite (Si/Al ¼ 2.5) is mixed with ammonium acetate solution (3.4 mol/L), and a homogeneous paste is formed. The paste (10 g of zeolite in 100 mL of solution) is heated to 7080
C, followed by slow addition of a hexafluorogallate solution prepared by mixing Ga(NO3)3 and NH4F (this addition needs 34 h generally). The reaction product is filtered off, washed with (NH4)2SO4 (1.5 mol/L) at 353358 K (80–85
C) for 2.5 h, and this process is repeated. Afterwards, the product is further washed to completely remove the fluoride, dried, and calcined to obtain the final product. The galliation degree of the zeolite is controlled through varying the fluorogallate concentration.[34] 6.5.2

Secondary Synthesis of Titanium-containing Zeolites – Gas–Solid Isomorphous Substitution Technique

Because of valence variability of titanium, titanium-containing zeolites have been widely used as oxidation catalysts. Among the Ti-containing zeolites, the most important ones are TS-1, Ti-beta, TS-2, Ti-MCM-22, and so on. Titanium-containing mesoporous and macroporous molecular sieves have also started to be applied to some catalytic reactions. Ti-containing zeolites such as TS-1 are usually prepared through hydrothermal methods. ˚ and the Si/Ti ratio falls within the range 30–40. TS-1 has a pore diameter of 5.3  5.5 A ˚ ) epoxidation catalyst, and it Ti-beta is a very promising large-pore (diameter 7.6  6.4 A is normally prepared through Ti-substitution of b-zeolite. However, when (Ti,Al)-beta is used for epoxidation catalysis, the presence of framework Al will influence the acidity and hydrophilicity of the catalyst, resulting in decrease of activity and selectivity. Therefore, in order to obtain a worthwhile catalyst, the isomorphous substitution using TiCl4 must be complete. In 1994, Rigutto et al.[35] proposed two routes for the preparation of Ti-beta zeolite. In one route, a boron-containing zeolite, H-[B]-b, is first subjected to gas–solid reaction with TiCl4 to form a substituted intermediate, and this intermediate is subsequently hydrolysed or alcoholized to remove boron and to add titanium (see Figure 6.15). In another route, the B-containing zeolite is used as the starting material again, but dilute acid or methanol is used for removal of boron to generate hydroxyl vacancies, followed by addition of titanium with TiCl4 under particular conditions. The Ti-b zeolites prepared through these two routes are effective catalysts for epoxidation reactions of 1-hexene or 1-octene. In 1999, Krijnen et al.[36] proposed another approach for the preparation of an effective Ti-b epoxidation catalyst. This approach is similar to the gas-solid isomorphous substitution technique, and two steps are involved. The first step is the dealumination, that is, to generate vacancies such as hydroxyl vacancies, and the second step is titanation

Figure 6.15 Deboronation and titanium addition of B-containing zeolite b through hydrolysis or alcoholysis. Reproduced with permission from [35]. Copyright (1994) Elsevier

378

Chemistry of Zeolites and Related Porous Materials

Table 6.12 Composition of Ti-b zeolite Catalyst number Ti-56BEA-450 Ti-43BEA-160 Ti-32BEA-267 Ti-30BEA-32 Ti-27BEA-72

Si/Ti (AAS)

Si/Ti (XPS)

Ti/unit cell (in product)

Al/unit cell (after dealumination)

56 43 32 30 27

49 41 30 31 31

1.1 1.4 1.9 2.1 2.3

1.5 1.3 1.4 1.5 0.8

TiO2 Raman signal no no weak weak intense

through reaction of titanium sources such as TiCl4 which enters the vacancies and undergoes condensation to add Ti in the framework and to stabilize the framework. In the first step, oxalic acid (0.251.5 mol/L) or HNO3 (17 mol/L) solution is refluxed with zeolite b (Si/Al ¼ 37.5) to dealuminate for 5 h; the product is washed, dried, and calcined at 550
C for 5 h, and 8090% of the framework Al can be removed in this way. In the second step, the dealuminated zeolite b is titanated for 30 min in a chemical-vapor deposition (CVD) fluid bed using TiCl4 þ N2 gas, and the product is blown with dry N2 to remove residual TiCl4 followed by calcination at 550
C for 4 h. From Table 6.12, we can see that the Al in the parent b zeolite framework are completely replaced by Ti so that the framework becomes hydrophobic. This hydrophobicity is important for the successful application of the Ti-containing zeolite to liquid phase oxidation catalysis. In addition, the control of titanation after dealumination to form hydrophobic, high-titanium, and highly active liquid-phase oxidation catalyst is also very important. This titanation approach is clearly superior to the direct titanation method which involves no pre-dealumination of zeolites such as FAU and MOR molecular sieves. The ability to control the formation of amorphous TiO2 makes this titanation route superior to the Rigutto method (see Table 6.12) because amorphous TiO2 affects the reactant diffusion, the activity and selectivity, and the decomposition of the Ti-containing zeolites. The two-step dealumination gas–solid substitution route has been widely applied to isomorphous substitution of MFI, MEL, FAU, MOR, and BEA zeolites. This substitution route is suitable for less metallic heteroatoms with a high oxidation state, such as B, Si, Al, Ga, In, Sb, As, Ti, Zr, V, Mo, W, etc. These elements form highly volatile chlorides such as BCl3, SiCl4, AlCl3, GaCl3, InCl3, SbCl5, AsCl5, TiCl4, VOCl3, MoCl3, WOCl4 and so on,[37] and oxides as the heteroatom vapor-phase sources for substitution reactions. Yashima et al.[38] used the term ‘atom planting’ to describe this substitution technique. 6.5.3

Demetallation of Heteroatom Zeolites through High-temperature Vapor-phase Treatment

The structures and properties of heteroatom zeolites vary due to the introduction of heteroatoms in the zeolite frameworks. When treated in high-temperature water vapor, zeolites Y, mordenite, and ZSM-5 are dealuminated to be ultra-stabilized, whereas heteroatom zeolites undergo ‘demetallation’; that is, the heteroatoms are removed from the frameworks to form extraframework species, and the resulting zeolites become catalytically active for special reactions. For instance, when treated in water vapor at

Preparation, Secondary Synthesis, and Modification of Zeolites

379

Figure 6.16 Catalytic cracking product distribution of n-butane over zeolite catalysts with various heteroatoms. Reproduced with permission from [19]. Copyright (2001) Elsevier. (a) H-Al-ZSM-5, (b) H-Fe-ZSM-5, (c) H-Ga-ZSM-5

elevated temperatures, the framework Fe is removed from zeolite Fe-ZSM-5 to form extraframework Fe-O-Fe species, and at 550
C these species are dispersed on the ZSM-5 framework surface as iron oxide particles, whereas at 770
C the particles coalesce to form larger ones. Similarly, when Ga-ZSM-5 is treated at elevated temperatures in water vapor, degalliation occurs. For some catalytic reactions involving carbocations, the heteroatom zeolites after demetallation show enhanced catalytic performance and the catalytic selectivity is as varied as in the dealuminated ZSM-5 and zeolite Y. In the following paragraph the catalytic cracking of n-butane is used as an example to compare the catalytic performances of the demetallated zeolite with H-Al-ZSM-5 after similar water vapor treatment, and the results are listed in Figure 6.16. From Figure 6.16, it is clearly seen that the extraframework gallium oxide generated by demetallation at elevated temperatures in water vapor enhances the activity, and the yield of alkenes is increased, whereas the formation of extraframework titanium dioxide leads to the yield of a large amount of methane in the cracking product.

6.6 Channel and Surface Modification of Zeolites Zeolites are important materials for the manufacture of adsorbents and catalysts because these materials possess special shape selectivities. The concept of ‘shape selectivity’ was first proposed by Weisz and Frilette in 1960[39] to describe the phenomenon that only those molecules with shapes and sizes matching the zeolite pores are able to be adsorbed

380

Chemistry of Zeolites and Related Porous Materials

or catalysed. The shape selectivity of an unmodified zeolite is mainly determined by its crystal structure. The pore opening of a zeolite is composed of an oxygen ring and the maximum diameter of the pore opening depends on the number of the oxygen atoms in the ring. The well known pore openings are the 8-, 10-, and 12-membered oxygen rings and the corresponding zeolites have a pore maximum diameter of 0.45, 0.63, and 0.80 nm, respectively.[40] The pore diameter increases by about 0.18 nm when the number of oxygen is increased by 2. The ‘molecular sieving’ effect can be realized on the basis of the difference in pore size and channel structure among zeolites. However, in practice, the difference in dynamic diameter among molecules which require recognition and differentiation is far smaller than 0.1 nm, and therefore only after fine adjustment of the zeolite pore size can the shape-selective goal be achieved. Because the variation of the pore size of zeolites themselves is not continuous, it is not possible to finely adjust the pore diameters through direct synthesis, and alternative routes must be adopted. In addition, in order to improve the shape selectivity of adsorbents and catalysts, besides adjustment of pore size, the external surface of zeolites also needs to be modified to deactivate the adsorption sites and catalytically active sites present on the external surface which shows no shape selectivity. The external surface modification is especially important for small-particle and nano-sized zeolites. At present, the techniques to modify zeolite channels and surface can be classified as three categories; that is, cation-exchange method, channel modification, and surfacemodification methods. 6.6.1

Cation-exchange Method

In a zeolite crystal, the number and nature of the cations near pore openings affect the pore size of the zeolite, and as a result, cation-exchange can vary the pore size of zeolites.[41] A typical example is zeolite A. The pore diameter of NaA zeolite is about ˚ , whereas when Naþ cations are exchanged by Ca2þ, the sites occupied by the Naþ are 4A emptied by half, and the pore diameter is enlarged accordingly. As a result, the pore ˚ . On the other hand, when the Naþ ions in NaA are diameter of CaA becomes 5 A ˚ in exchanged by the bulkier monovalent Kþ, the zeolite pore size is reduced to about 3 A KA due to the increase of the volume of the exchanged cations near the pore opening. Iwamoto et al.[42] used the ion-exchange approach to finely adjust the pore diameter of zeolite A, and they successfully separated O2 from N2 using the modified zeolite. The ˚, sizes of O2 and N2 are very similar (dynamic diameters are 3.46 and 3.64 A respectively), and KA adsorbs neither molecule because of its small pore size, whereas NaA adsorbs both due to its larger pore diameter but the N2-adsorption amount is larger than that of O2 because the polarity of N2 is larger than that of O2. Therefore, it is necessary to adjust the zeolite pore size in order to adsorb O2 instead of N2. Figure 6.17 shows the O2- and N2-adsorption properties of KA exchanged with Zn2þ to various degrees. The KA zeolite without Zn2þ exchange adsorbs almost no O2 and N2 due to its small pore size. With the introduction of Zn2þ, the pore diameter of the zeolite increases, and the amount of O2 adsorption is increased gradually. When the degree of Zn2þ exchange reaches 41%, the zeolite adsorbs a certain amount of O2 whereas no N2 is adsorbed, indicating that on this occasion the pore diameter of the zeolite is between that of O2 and that of N2. Further increase of Zn2þ-exchange degree increases the amount of

Preparation, Secondary Synthesis, and Modification of Zeolites

381

Figure 6.17 O2- and N2-adsorption properties of KA zeolite with various. Reproduced with permission from [42]. Copyright (1984) American Chemical Society. Zn2þ-exchange degrees. (a), (c), (e) Are for oxygen adsorption; (b), (d), (f) are for nitrogen adsorption

N2 adsorption whereas when the exchange degree reaches 58%, the adsorption amount of N2 exceeds that of O2. Besides oxygen and nitrogen separation, the zeolite adsorbents prepared through cation-exchange can also be used for adsorption dewaxing of petroleum distillates, for mixed xylene separation, for dimethylnaphthalene separation, and so on. However, there are disadvantages and shortcomings for cation-exchange techniques: (1) This technique is not suitable for high-silica zeolites; (2) there is no linear relationship between zeolite pore size and degree of cation-exchange, and the degree of ion-exchange is difficult to control so that it is hard to finely adjust the pore diameter of zeolites; (3) the cationexchange influences the property of the zeolite. 6.6.2

Channel-modification Method

This method is to insert guest molecules or clusters into zeolite channels so as to make the channels narrower and to achieve the goal of pore-size adjustment. This method is also called internal surface modification to stress the variation of the inside of channels, but in fact both the internal and external surfaces are modified during the treatment process.[43] The modifying agents initially used were oxides, and alkaline earth metal salts were impregnated on HZSM-5, and, after calcination, the oxide entered into the channels of

382

Chemistry of Zeolites and Related Porous Materials

Table 6.13 Adsorption properties of oxide-modified HZSM-5 zeolite Adsorption amount (mg/g)

Sample HZSM-5 P-ZSM-5 (P ¼ 0.72 mmol/g) B-ZSM_5 (B ¼ 0.35 mmol/g) Mg-ZSM-5 (MgO ¼ 0.20 g/g)

Total parao-Methylphenol m-Methylphenol p-Methylphenol amount Selectivity 31.8 10.5

37.2 15.4

39.0 29.3

108 55.2

0.361 0.530

14.6

22.2

35.3

72.0

0.489

10.9

17.1

51.6

79.6

0.644

the zeolites, reducing the acid centers on the zeolite surface and resulting in a decrease in effective pore size of the zeolites.[44] Table 6.13 presents the competitive adsorption results for methylphenol isomers on oxide-modified zeolites.[45] After oxide modification, the total adsorption decreases but the para-selectivity increases distinctly. In early 1980s, Vansant proposed the silylation method to modify zeolite channels.[46–52] The principle is to use silicanes to react with the surface hydroxygroups in H-form zeolites, and after hydrolysis the formed oxides narrow the channels so as to achieve the goal of pore-size adjustment. Figure 6.18 shows the Xe-adsorption-dynamic curves for the H-form mordenite samples silylated to various degrees. From the Figure

Figure 6.18 Xe-adsorption-dynamic curves of H-form mordenites with various silylation degrees. Reproduced with permission from [46–52]. Copyright (1978–1989) Royal Society of Chemistry. * Unsilylated zeolite; } zeolite silylated after thermal treatment; & 373 K (100
C);
473 K (200
C); 573 K (300
C); 673 K (400
C); hydrolysed and dehydrated zeolite after silylation

Preparation, Secondary Synthesis, and Modification of Zeolites

383

Table 6.14 Gas-separation effects of mordenite after silylation treatment

Gas mixture Ar þ Kr O2 þ Kr N2 þ Kr N2 þ O2 þ Kr Kr þ N2 Kr þ N2 þ O2

Initial concentration (%) 90.57 þ 9.43 90.04 þ 9.96 89.35 þ 10.65 70.46 þ 18.39 þ 11.15 89.88 þ 10.12 10.28 þ 69.69 þ 20.03

Gas separated Ar O2 N2 N2/O2 Kr Kr

Final concentration (%) 99.996 99.994 99.999 99.992 99.996 99.994

Treatment temperature/ (T/K) 195 195 195 195 273 273

it is seen that with an increase in silylation degree, the Xe-adsorption capacity of the zeolite apparently decreases, indicating that the silylation narrows the zeolite channels and the effective pore diameter becomes smaller, achieving the goal of pore-diameter adjustment. Because the activity of silicanes is very high, the zeolite-adjustment range of the silylation method, subject to repeated treatments of the zeolite channels, is much larger than that of oxide modification. In addition, the silylation degree can be controlled through variation of reaction temperature, time, and silicane pressure, so that the accuracy for zeolite pore-size adjustment is enhanced enormously. Therefore, the silylation method is an ideal channel-modification technique. The zeolites appropriately modified through this technique can be effectively used for separation of gas mixtures. Table 6.14 presents the separation effects of mixed gases of Kr, N2, O2, and Ar on H-form mordenite after silylation treatment. Similar results can be obtained if boranes or boranes and amines instead of silicanes are used for zeolite channel modification. However, the channel-modification method has its own disadvantages. Because the modifier interacts with the whole channel system, then, besides the pore diameter, the properties of the zeolite internal surface can also be varied. This may affect the adsorption and catalytic properties of the zeolite involved. In addition, because a large amount of modifier enters the zeolite channels, the void volume of the zeolite becomes smaller and consequently the adsorption capacity and the space available for reaction are reduced accordingly. 6.6.3

External Surface-modification Method

In order to overcome the shortcomings of the internal surface-modification method, it is necessary to modify the zeolite channels using molecules larger than the pore opening of the zeolite under the circumstance that the pore size is adjusted without affecting the internal channels. Because on this occasion the modifying agent cannot enter the channels of zeolites but instead interacts with the external surface of the zeolites, this method is called the external surface-modification method. This approach was initially proposed by Niwa[53–59], et al.and they used Si(OCH3)4 as the modifying agent to modify zeolites through a CVD route. Because the dynamic diameter of an Si(OCH3)4 molecule ˚ , larger than that of the zeolite pore opening, it cannot enter the channels of is about 8.9 A the zeolite but instead interacts with the hydroxy groups on the zeolite external surface

384

Chemistry of Zeolites and Related Porous Materials

Figure 6.19 Ref. 53–59

Schematic illustration of the vapor-deposition process of Si(OCH3)4 on zeolite.

and on the pore openings. After calcination in air the formed SiO2 layer is deposited on the zeolite external surface and the pore openings, narrowing the pore opening and achieving the goal of effective pore-size control. Figure 6.19 schematically shows the process of vapor deposition of Si(OCH3)4 on zeolite. Fine adjustment of zeolite pore-opening size with the CVD technique may enhance the shape-selective separation ability of zeolites distinctly. HZSM-5 zeolite modified through the CVD route has been successfully used for shape-selective separation of xylene and methylphenol isomers.[60] Figure 6.20 shows the dependence of p-xylene and m-xylene adsorption amounts on the SiO2 deposition quantity on CVD-modified SiHZSM-5 zeolite. Because the m-xylene molecule (0.71 nm) is larger than the pore opening (0.54  0.56 nm) of HZSM-5, whereas p-xylene (0.58 nm) is similar in size to the HZSM-5 pore opening, the unmodified HZSM-5 exhibits a certain shape selectivity, and the adsorption amount of p-xylene is 85% of the total. As the SiO2 deposition amount is increased, the total adsorption of xylenes is slightly reduced, but the adsorption of m-xylene drops quickly, and when the SiO2 deposition amount reaches 2.3%, the zeolite pore opening is so narrowed that no m-xylene is accommodated, whereas the p-xylene selectivity reaches 100%. From the Figure it is known that if the deposition amount is in the range 2.3 – 3.0%, the two isomers can be effectively separated through shapeselective adsorption.

Figure 6.20 Shape-selective adsorption of m-/p-xylene on SiHZSM-5. Reproduced with permission from [60]. Copyright (1996) Acta Chimica Sinica

Preparation, Secondary Synthesis, and Modification of Zeolites

385

Figure 6.21 Shape-selective adsorption of m-/p-methylphenol on SiHZSM-5. Reproduced with permission from [60]. Copyright (1996) Acta Chimica Sinica

Another successful example is the application of CVD-modified HZSM-5 to the separation of m-methylphenol and p-methylphenol.[60] The difference in size of mmethylphenol (0.64 m) and p-methylphenol (0.58 nm) molecules is smaller than that of the two xylene isomers, and the unmodified HZSM-5 zeolite has little shape selectivity for adsorption of the two phenol isomers. The adsorption amount of p-methylphenol on HZSM-5 is only 57% of the total (Figure 6.21). With an increase in the SiO2 deposition amount, the pore opening of the zeolite decreases gradually, and the m-methylphenol adsorption amount drops rapidly, whereas in contrast the p-methylphenol adsorption amount increases. When the SiO2 deposition amount falls in the range 3.5 – 4.2%, the p-methylphenol adsorption amount remains high whereas its selectivity reaches 95– 100% so that the two methylphenol isomers can be effectively separated through the shape-selective adsorption approach. For catalytic reactions, the fine adjustment of pore size of zeolite catalysts through the CVD technique can improve the shape selectivity of reactants and products. For instance, the pore size of mordenite catalyst is larger than the molecular size of the three reactants n-octane (0.43 nm), 3-methylheptane (0.55 nm), and 2,2,4-trimethylpentane (0.62 nm) in the octane isomer mixture usually used for industrial cracking, and therefore the crackingreaction rates for the three isomers are similar (see Figure 6.22).[56] However, as the SiO2 deposition amount increases, the selectivity of the three isomer reactants is evidently enhanced. When the SiO2 deposition amount is 3.2%, the reaction of 2,2,4-trimethylpentane is completely inhibited, whereas if the SiO2 deposition amount is increased to 3.4%, the 3-methylheptane reaction is also inhibited. Further increase of the SiO2 deposition amount to 3.7% results in n-octane undergoing no reaction either, indicating that the reactant selectivity can be controlled through the shrinkage of zeolite pore size. In Table 6.15 the reaction data of toluene disproportionation on the CVD-modified mordenite zeolite catalyst at 400
C are presented.[61] As the SiO2 deposition amount increases, the toluene conversion decreases gradually. From the product distribution it is seen that the amount of nonaromatics increases slightly whereas the amount of trimethylbenzene is reduced to zero, and the xylene/benzene ratio drops as well,

386

Chemistry of Zeolites and Related Porous Materials

Figure 6.22 Cracking reaction of C8 isomers on mordenite zeolite with various SiO2 depositions. Reproduced with permission from [56]. Copyright (1985) Royal Society of Chemistry. * n-Octane; ~ 3-methylheptane; } 2,2,4-trimethylpentane. (a) PtHM; (b) SiPtHM (3.2%); (c) SiPtHM (3.4%); (d) SiPtHM (3.7%) (W/F) is defined as the amount of zeolite (W) divided by the flow rate (F ¼ 0.091 h-1) is shown on the abscissa

indicating that the xylene-disproportion reaction is inhibited but that the dealkylation is enhanced. Meanwhile, in the xylene product the content of p-xylene is increased apparently to surpass the equilibrium concentration, suggesting that the shrinkage of zeolite pore size enhances the para-selectivity of the catalytic reaction. The CVD technique can be effective for zeolite pore-size adjustment, and the zeolites after modification through this technique exhibit distinctly enhanced shape-selective adsorption separation and catalytic performances. However, this technique requires vacuum apparatus and the monetary investment for this technique is large. In addition,

Table 6.15 Reaction data of toluene disproportion on CVD-modified SiHM catalyst (400
C) Sample

HM

SiHM(0.7) SiHM(1.6) SiHM(2.6)

SiHM(3.7)

SiHM(4.0) HZSM-5

NA B T p-X o-X m-X TMB X/B p-X/X NA/(1  T) TMB/(1  T) C/%

0.005 0.166 0.647 0.039 0.087 0.036 0.020 0.98 0.241 0.014 0.056 35.3

0.005 0.170 0.636 0.041 0.090 0.037 0.020 0.99 0.244 0.014 0.055 36.4

0.006 0.085 0.827 0.023 0.043 0.016 0 0.96 0.280 0.035 0 6.9

0.003 0.028 0.946 0.012 0.009 <0.001 0 0.75 0.571 0.058 0 5.2

0.004 0.143 0.710 0.035 0.078 0.031 0 1.00 0.243 0.014 0 29.0

0.006 0.085 0.827 0.023 0.043 0.016 0 0.96 0.280 0.035 0 17.3

0.001 0.005 0.990 0.003 0.002 0 0 1.00 0.600 0.010 0 1.0

C stands for conversion; NA for nonaromatics; B for benzene; T for toluene; X for xylene; TMB for trimethylbenzene.

Preparation, Secondary Synthesis, and Modification of Zeolites

387

the operation of this technique is complicated, rendering it difficult to be applied widely. Gao et al. proposed a chemical-liquid deposition (CLD) method to replace the CVD method for zeolite pore size adjustment[62–67] and achieved effective results, and the approach has been widely recognized and applied.[68–75] The principle of CLD method for zeolite pore-size adjustment is similar to that of the CVD method. Through this method, the modifier in solution interacts with the hydroxy groups on the zeolite external surface and the pore openings to form an SiO2 layer deposited on the external surface and the pore openings, achieving the effects of zeolite pore-size adjustment. The CLD method is simple, and the detailed procedure is as follows: A certain amount of pre-treated zeolite is added to a nonpolar organic solvent such as hexane, and to the mixture is added a certain amount of modifier such as Si(OCH3)4, SiCl4, etc., followed by reaction at room temperature for a particular time. The reaction system is then dried under an infrared lamp and calcined at 550
C until the sample becomes white in color. The advantage of CLD lies in that it is suitable for various zeolites and not only for H-form zeolites. In addition, it does not require special apparatus, and the reaction temperature is mild so that the technique is applicable industrially on a large scale. Furthermore, for the CLD method, the zeolite pore-size adjustment is realized through varying the concentration of the modifier, and the accuracy (<0.05 nm) of adjustment is higher than that of the CVD method. The CLD approach for zeolite modification has almost no effect on the total surface area, the pore volume, or the surface acidity of the zeolites involved because the SiO2 deposition layer covers the zeolite external surface and pore openings only, and, as a result, the zeolites are nearly not affected except for the diminishing size of the pore openings. Table 6.16 lists characterization results of the HZSM-5 zeolite after CLD modification. O’Conner and colleagues characterized the HZSM-5 modified through Si(OCH3)4 liquid-phase deposition, and similar results were obtained.[68,69] After CLD modification, the zeolites exhibit excellent shape selectivity for separation and purification of various isomers. For instance, the pore-size-adjusted NaY zeolite after Si(OCH3)4 modification is very effective for the separation of methylnaphthalene and trimethylbenzene isomers. Because the pore size of NaY itself is large, before modification the zeolite shows no shape selectivity for the two methylnaphthalene isomers, and the adsorption capacities for 1-methylnaphthalene and 2-methylnaphthalene are similar. However, as the pore size decreases, the zeolite exhibits increasing adsorption capacity

Table 6.16 Characterization results of CLD-modified HZSM-5 zeolite

Sample

Depositing agent

Surface area (m2/g)

Pore volume (mL/g)

HZSM-5 SiHZSM-5(0.1) SiHZSM-5(0.2) SiHZSM-5(0.1) SiHZSM-5(0.2)

Si(OCH3)4 Si(OCH3)4 SiCl4 SiCl4

538 511 507 507 497

0.206 0.197 0.182 0.194 0.185

NH3 desorption amount (mmol/g) I (
C)

II (
C)

I þ II

0.32 (244)

0.38 (372)

0.70

0.31 (243) 0.31 (243) 0.30 (242)

0.38 (374) 0.38 (374) 0.37 (378)

0.69 0.69 0.67

388

Chemistry of Zeolites and Related Porous Materials

Figure 6.23 Shape-selective adsorption of a/b-methylnaphthalene (MN) on SiNaY zeolite. 1. b-MN; 2. a-MN; 3. MN; 4. b/. Reproduced with permission from [64]. Copyright (1996) Acta Chimica Sinica

for 2-methylnaphthalene, whereas the adsorption capacity for 1-methylnaphthalene is significantly reduced. The adsorption selectivity for 2-methylnaphthalene is larger than 90%, achieving an ideal separation effect as shown in Figure 6.23.[64] The separation effect for the mixture of 1,2,4-trimethylbenzene and 1,3,5-trimethylbenzene is similar, and as the amount of Si(OCH3)4 used increases, the pore size of NaY zeolite is narrowed gradually, and the adsorption selectivity of the zeolite for 1,2,4trimethylbenzene increases rapidly. When the Si(OCH3)4 amount used is 0.15 mL/g, the adsorption capacity of the zeolite for 1,2,4-trimethylbenzene remains almost unchanged, but the adsorption for 1,3,5-trimethylbenzene is very small, and the 1,2,4-trimethyl benzene adsorption selectivity is exceeds 90%, achieving an ideal separation effect as shown in Figure 6.24.[64] Comparison of the two systems reveals that to achieve an ideal separation effect, more modifier is required; that is, the zeolite pore size should be narrower for separation of trimethylbenzene mixture because the size of the 1,3,5trimethylbenzene molecule is smaller that that of 1-methylnaphthalene.

Figure 6.24 Shape-selective adsorption of 1,2,4-/1,3,5-trimethylbenzene on SiNaY zeolite. 1. 1,2,4-TMB; 2. 1,3,5-TMB; 3. TMB; 4. 1,2,4-TMB/. Reproduced with permission from [64]. Copyright (1996) Acta Chimica Sinica

Preparation, Secondary Synthesis, and Modification of Zeolites

389

Taking into account the high cost of Si(OCH3)4, Gao et al. proposed use of a series of hydrolysable chlorides such as SiCl4 to replace the orthosilicates as modifying agents,[62] and they achieved excellent pore-size-adjustment effects. Because the reaction activity of SiCl4 with zeolite surface hydroxy groups and adsorbed water is higher than that of Si(OCH3)4, the zeolite pore-size-adjustment range is extended and the amount of modifying agent used is reduced, resulting in a decrease in cost. Table 6.17 presents the separation effects of SiCl4-modified HZSM-5 for various disubstituted benzene isomers such as m-xylene/p-xylene, o-dichlorobenzene/p-dichlorobenzene, and mmethylphenol/p-methylphenol.[63] From the results in the Table it is seen that as the deposition amount increases, the SiO2 deposited on the zeolite surface and the pore openings increases as well and the pore size is narrowed. As a result, the adsorption amount of the larger meta- and ortho-isomers drops rapidly whereas the adsorption amount of the smaller para-isomer remains almost unchanged, the para-isomer adsorption selectivity being enhanced significantly. Careful comparison of the three systems reveals that their requirements for zeolite pore-size adjustment are different. With an increase in the amount of deposition agent used, the adsorption of the largest xylene (m-xylene) is first inhibited, and the amount of SiCl4 used to achieve a p-xylene adsorption selectivity of 100% is only 0.05 mL/g. The second-most inhibited molecule is o-dichlobenzene. The amount of SiCl4 used is 0.10 mL/g when the adsorption selectivity of p-dichlorobenzene reaches 100%, indicating that because the molecular size of o-dichlorobenzene is smaller than that of m-xylene, to prevent the odichlorobenzene from being adsorbed more SiO2 is needed to deposit on the zeolite pore openings to narrow the pore diameter. If the adsorption of m-methylphenol needs to be inhibited completely, the SiCl4 amount used must be increased to more than 0.15 mL/ g. The aforementioned experimental results indicate that with the CLD method the zeolite pore size can be finely controlled, and molecular isomers with a size difference smaller than 0.05 nm can be separated. On the basis of Gao’s work, Tan et al. loaded an SiCl4-modified HZSM-5 adsorbent in a high-pressure adsorption apparatus, and conducted the separation of equal quantities of m-methylphenol and p-methylphenol using propane as the carrier and desorption gas. They acquired m-methylphenol and pmethylphenol with purities higher than 98% whereas the recovery rate of the feedstock reached 100%. Table 6.17 Separation performance of CLD-modified HZSM-5 zeolite for various isomer mixtures Adsorption amount (%) Samplea

p-X

HZSM-5 SiHZ (0.025) SiHZ (0.050) SiHZ (0.075) SiHZ (0.100) SiHZ (0.150)

7.55 7.03 6.68 6.16

m-X X p-X/ p-DCB o-DCB DCB p-DCB/ p-C 1.20 0.91 0 0

8.75 7.94 6.68 6.16

86 89 100 100

5.70 6.09 6.38 6.97 5.82

1.60 1.42 0.96 0.53 0

7.30 7.51 7.34 7.50 5.82

78 81 87 93 100

4.16 4.53 5.67 4.68 4.68 4.25

m-C

C

4.50 4.14 2.73 1.46 0.95 0

8.66 8.67 8.39 6.14 5.63 4.25

p-C/ 48 52 67 76 83 100

a The data in parentheses are the SiCl4 amounts used (mL/g zeolite); X stands for xylene; DCB for dichlorobenzene; C for methylphenol.

390

Chemistry of Zeolites and Related Porous Materials

Table 6.18 p-Xylene disproportionation reaction data over Si-HZSM-5 zeolite Product distribution Catalysta

Conversion (%)

HZSM-5 SiHZ (0.05) SiHZ (0.075) SiHZ (0.10) SiHZ (0.20) SiHZ (0.50)

12.4 12.1 12.5 11.3 10.6 10.3

NA

B

T

p-X

m-X

o-X

p-X/

0.51 0.20 0.00 0.00 0.00 0.00

5.18 5.27 5.47 5.11 4.94 4.65

87.6 87.9 87.5 88.7 89.4 89.7

1.98 1.96 2.28 2.47 2.25 2.34

3.53 3.34 3.37 2.80 2.74 2.86

1.58 1.19 1.34 0.92 0.71 0.75

0.28 0.30 0.33 0.40 0.40 0.41

a

The data in parentheses are the SiCl4 amount used (mL/g zeolite); NA stands for nonaromatics; B for benzene; T for toluene; X for xylene.

The CLD modification of zeolites can also effectively enhance the shape selectivity of catalytic reactions. Gao and coworkers used this method to adjust the pore size of HZSM5 and effectively enhanced the shape selectivity of the zeolite for toluene-disproportionation reaction (see Table 6.18).[67] It is seen that as the SiO2 deposition amount is increased, the pore size of the zeolite decreases gradually, and the toluene conversion reduces slightly but the p-X/X value increases from 0.28 to 0.41, that is, the paraselectivity is enhanced by 46%; the p-X concentration exceeds the thermodynamic equilibrium value. Recently, Weitkamp and coworkers[75] applied the CLD-modified HZSM-5 zeolite catalyst to the hydroconversion of methylcyclohexane (MCH) to obtain high-quality water vapor cracking feedstock; that is, Cþ 2 normal alkanes. The modifying agent they used was Si(OC2H5)4 (TEOS), and the solvent was n-heptane. The preparation procedure is similar to that invented by Gao et al., and the reaction results are listed in Table 6.19. Table 6.19 Product distribution of methylcyclohexane hydroconversion over CLD-modified HZSM-5 zeolite catalyst (400
C) Catalyst MCH conversion(%) CH4(%) C2H6(%) C3H8(%) C4H10(%) Cþ 2 normal alkanes (%) Isoalkanes (%) Cycloalkanes (%) Aromatics (%) a

HZSM-5 (4.8)a 100

HZSM-5 TEOS (25)b/ (6.8)a HZSM-5 (4.7)a 100

TEOS (70)b/ TEOS (70)b/ HZSM-5 (4.8)a HZSM-5 (6.5)a

99.3

99.7

100

2.6 14.3 50.3 11.4 76.8

3.2 13.9 50.4 12.0 77.0

5.0 20.0 51.2 7.8 79.4

5.2 20.0 52.1 7.5 80.0

6.6 25.7 51.2 5.7 82.7

18.8 0.1 1.7

19.7 0.02 0.02

14.4 0.1 0.4

14.2 0.1 0.2

10.7 0 0

Values in parentheses are nAl/nAl þ nS. Values in parentheses are the reaction temperatures of TEOS with the zeolite; MCH stands for methylcyclohexane.

b

Preparation, Secondary Synthesis, and Modification of Zeolites

391

From the Table it is seen that as the deposition temperature increases, the SiO2 deposition amount increases and the methylcyclohexane conversion remains almost unchanged, but the yield of the C2þ normal alkanes is distinctly enhanced whereas that of the isoalkanes decreases quite apparently. Because the cracking of normal alkanes in water vapor produces ethylene and propylene whereas the main product from iso alkanes is methane, the CLD modification of HZSM-5 catalyst is very conducive to the improvement of alkene yield. This new technique has been suggested to be applied to the conversion of the excessive amounts of aromatics in gasoline to the water-vapor cracking feedstock (highly required alkenes in the market) through hydrogenation and hydro decomposition. The HZSM-5 catalyst prepared through SiCl4 liquid deposition has also been successfully applied to the synthesis of p-isopropylmethylbenzene (p-cymene), which is an important intermediate for the production of pesticides, bactericides, and perfumes, from toluene and propan-2-ol.[72,74] After modification, the para-selectivity of the catalyst increases by 20%, and at 270
C under the conditions of weight hourly space velocity (WHSV) 6.5 g/g h and toluene/propan-2-ol ratio 11, the selectivity and yield of p-isopropylmethylbenzene reach 94 and 84%, respectively. In addition, adjustment of the pore size of ferrierite through SiCl4 liquid-phase deposition can also enhance the selectivity of isomerization of 1-butene to isobutene, because when the pore size of the zeolite is reduced, the formation of octane intermediate is inhibited.[71] Under suitable conditions, the isobutene selectivity can be increased from 48.8% to 82.5%. From the aforementioned examples, it is seen that through zeolite pore and surface modification, the shape selectivity of adsorbents and catalysts can be effectively improved, and consequently many practical problems can be solved. Especially, the discovery of the CLD method overcomes the shortcomings of the CVD method, clearing a series of obstacles for large-scale industrial applications, and therefore this new technique is very promising.

References [1] (a) Z.J. Da, Fen Zi Shai Kuai Xun, 1995, 1, 5. (b) J. He, X.B. Yang, D.G. Evans, and X. Duan, New Methods to Remove Organic Templates from Porous Materials. Mater. Chem. Phys., 2002, 77, 270–275. [2] A. Corma, V. Fornes, M.T. Navarro, and J. Perez-Pariente, Acidity and Stability of MCM-41 Crystalline Aluminosilicates. J. Catal., 1994, 148, 569–574. [3] O. Pachtova´, M. Kocirik, A. Zika´nova´, B. Bernauer, S. Miachon, and J.-A. Dalmon, A Comparative Study of Template Removal from Silicalite-1 Crystals in Pyrolytic and Oxidizing Regimes. Microporous Mesoporous Mater., 2002, 55, 285–296. [4] M.T.J. Keene, R. Denoyel, and P.L. Llewellyn, Ozone Treatment for The Removal of Surfactant to form MCM-41 Type Materials. Chem. Commun., 1998, 2203–2204. [5] D. Mehn, A. Kukovecz, I. Kiricsi, F. Testa, E. Nigro, R. Aiello, G. Daelen, P. Lentz, A. Fonseca, and J.B. Nagy, The Effect of Calcination on the Isomorphously Substituted Microporous Materials Using Ozone. Stud. Surf. Sci. Catal., 2001, 135, 215–251. [6] O. Kresnawahjuesa, D.H. Olson, R.J. Gorte, and G.H. Ku¨hl, Removal of Tetramethylammonium Cations from Zeolites. Microporous Mesoporous Mater., 2002, 51, 175–188.

392

Chemistry of Zeolites and Related Porous Materials

[7] D.D. Whitehurst, Method to Recover Organic Templates from Freshly Synthesized Molecular Sieves. US Patent 5,143,879 (1992). [8] C.W. Jones, K. Tsuji, and M.E. Davis, Organic-Functionalized Molecular Sieves as Shapeselection catalysts. Nature (London), 1998, 393, 52–54. [9] T. Takewaki, L.W. Beck, and M.E. Davis, Synthesis of CIT-6, a Zincosilicate with the BEA topology. Top. Catal., 1999, 9, 35–42. [10] C.W. Jones, K. Tsuji, T. Takewaki, L.W. Beck, and M.E. Davis, Tailoring Molecular Sieve Properties During SDA Removal Via Solvent Extraction. Microporous Mesoporous Mater., 2001, 48, 57–64. [11] F. Dougnier and J.L. Guth, Possible Recovery of Crown Ethers Occluded in FAU and EMTType Zeolite. Microporous Mater., 1996, 6, 79–88. [12] E.Z. Min, Preparation and Development of Industrial Catalysts, China Petrochemical Press, Beijing, 1997. [13] H. Bremer, W. Morke, R. Schodel, and F. Vogt, Influence of Cations on The Thermal Stability of Modified Y Zeolites. Adv. Chem. Ser., 1973, 121, 249–257. [14] D.W. Breck, Ion Exchange Reactions in Zeolites. In Zeolite Molecular Sieves, ch. 7, 529–592, John Wiley & Sons, Inc., New York, 1974. [15] Molecular Sieve Group in Dalian Institute of Chemical Physics of Chinese Academy of Science, Zeolite Molecular Sieves, Science Press, Beijing, 1978. [16] H.S. Sherry, The Ion Exchange Properties of Zeolities III. Rare Earth Ion Exchange of Synthetic Faujasites. J. Colloid Interface Sci., 1968, 28, 288. [17] R.R. Xu, G.Z. Yu, Y.Q. Lu, S.H. Feng, and W.C. Chang, Hydrothermal Ion Exchange Reactions of La3þ-NaY. Chem. J. Chin. Univ., 1980, 1(1), 1–8. [18] R.R. Xu, Y.Q. Lu, S.J. Ma, G.Z. Yu, S.G. Li, and J.M. Zhang, The Physico-Chemical Properties of the High Temperatures Ion-exchanged Individual Rare Earth Y Type Zeolites, Chem. J. Chin. Univ., 1980, 1(2), 39–47. [19] R. Szostak, Secondary Synthesis Methods. Stud. Surface Sci. Catal., 2001, 137, 261–297. [20] J. Scherzer, The Preparation and Characterization of Aluminum Deficient Zeolites. In Catalytic Materials, ACS Symp. Ser., 1984, 248, 157–200. [21] C.V. McDaniel and P.K. Maher, Zeolite Chemistry and Catalysis, ed. J.A. Rabo, American Chemical Society, Washington DC, 1974, 285. [22] G.T. Kerr and G.F. Shipman, The Reaction of Hydrogen Zeolite Y with Ammonia at Elevated Temperatures. J. Phys. Chem., 1968, 72, 3071–3072; G.T. Kerr, Chemistry of Crystalline Aluminosilicates. V. Preparation of Aluminum-Deficient Faujasites. J. Phys. Chem., 1968, 72, 2594–2596. [23] G.W. Skeels and D.W. Breck, Zeolite Chemistry V – Substitution of Silicon for Aluminum in Zeolite Via reaction with Aqueous Fluorosilicate. Proceedings of the 6th International Zeolite Conference, Butterworths Guildford, 1983, 87–96. [24] Y.G. He, C.Y. Li, and E.Z. Min, A Mechanism Study of Framework Si-Al Substitution in Y Zeolite During Aqueous Fluorosilicate Treatment. Stud. Surf. Sci. Catal., 1989, 49A, 189–197. [25] H.K. Beyer, I.M. Belenykaja, F. Hange, M. Tielen, P.J. Grobet, and P.A. Jacobs, Preparation of High-Silica Faujasite by Treatment with Silicon Tetrachloride. J. Chem. Soc., Faraday Trans. 1, 1985, 81, 2889–2901. [26] A. Gola, B. Relhours, E. Milazzo, J. Lynch, E. Benazzi, S. Lacombe, L. Delevoye, and C. Fernandez, Effect of Leaching Agent in The Dealumination of Stabilized Y-Zeolites. Microporous Mesoporous Mater., 2000, 40, 73–83. [27] E.M. Flanigen, B.M. Lok, R.L. Patton, and S.T. Wilson, Aluminophosphates Molecular Sieves and Periodic Table, In New Developments in Zeolite Science and Technology. Proceedings of the 7th International Zedite Conference, ed. Y. Murakami, A. Lijima, and J.W. Ward, Kodansha-Elsevier, Tokyo, 1986, 103–112.

Preparation, Secondary Synthesis, and Modification of Zeolites

393

[28] W.Q. Pang, S.L. Qiu, and F.Q. Zhou. Progress in Synthesis of Heteroatom-substituted Zeolites, J. Jilin Univ. (special issue), Sci. Ed., 1992, 78–84. [29] R.M. Barrer, Isomorphous Replacements in the Framework of Zeolites and Other Tectosilicates, in Hydrothermal Chemistry of Zeolites, Academic Press London, 1982, 251–305. [30] X.S. Liu and J.M. Thomas, Gallosilicate Zeolites from Porotectosilicate Precursors: the Ready Preparation of Gallo-ZSM-11 from Silicalite-II. J. Chem. Soc., Chem. Commun., 1985, 1544–1545. [31] X.S. Liu, Gallosilicate Zeolites. PhD Thesis, University of Cambridge (1986). [32] J. Dwyer and K. Karim, The Incorporation of Heteroatoms into Faujasitic Framework by Secondary Synthesis Using Aqueous Fluoride Complexes. J. Chem. Soc., Chem. Commun., 1991, 905–906. [33] X.S. Liu and R.R. Xu, Dealumination of Zeolites Using an Aqueous Solution of Ammonium Tetrafluoroborate. J. Chem. Soc., Chem. Commun., 1989, 1837–1839. [34] K. Karim, J. Dwyer, D.J. Rawlence, M. Tariq, and A. Nabham, Secondary Synthesis of Gallium-Substituted Wide Pore Zeolites. J. Mater. Chem., 1992, 2, 1161–1166. [35] M.S. Rigutto, R. de Ruiter, J.P.M. Niederer, and H. Van Bekkum, Titanium-Containing Large Pore Molecular Sieves from Boron-Beta: Preparation, Characterization and Catalysis. Stud. Surf. Sci. Catal., 1994, 84C, 2245–2252. [36] S. Krijnen, P. Sa´nchez, B.T.F. Jakobs, and J.H.C. van Hooff, A Controlled Post-Synthesis Route to Well-Defined and Active Titanium Beta Epoxidation Catalysts. Microporous Mesoporous Mater., 1999, 31, 163–173. [37] J.P.M. Niederer and W.F. Hoelderich, Isomorphous Substituted Early Transition Metal Containing BEA Vin Post-Synthesis Modification of H-[B]-BEA. Proceedings of the 12th International Zeolite Conference, Part III, ed. M.M.J. Treacy, B.K. Marcus, M.E. Bisher, and J.B. Higgins, MRS, Warrendale, PA, 1998, 1893–1900. [38] T. Yashima, K. Yamagishi, and S. Namba, Preparation of Metallosilicates with MFI Structure by Atom-Planting Method. Stud. Surf. Sci. Catal., 1991, 60, 171–178. [39] P.B. Weisz and V.J. Frilette, Intracrystalline And Molecular-Shape-Selective Catalysis by Zeolite Salts. J. Phys. Chem., 1960, 64, 382–383. [40] R.R. Xu, W.Q. Pang, and K.G. Tu, Structures and Synthesis of Zeolites, Jilin University Press, Changchun, 1987, 5. [41] D.W. Breck, W.G. Eversole, R.M. Milton, T.B. Reed, and T.L. Thomas, Crystalline Zeolites. I. The Properties of a New Synthetic Zeolite, Type A. J. Am. Chem. Soc., 1956, 78, 5963–5971. [42] M. Iwamoto, K. Yamaguchi, Y. Akutagawa, and S. Kagawa, Partially Zinc Ion-Exchanged K-A Zeolites as Molecular Sieves Distinguishing Oxygen from Nitrogen. J. Phys. Chem., 1984, 88, 4195–4197. [43] T. Yashima, Proceedings of the 7th International Congress on Catalysis, ed. T. Seiyama and K. Tanabe, Tokyo, Kodansha, 1980, 739. [44] W.W. Kaeding, C. Chu, L.B. Young, B. Weinstein, and S.A. Butter, Selective Alkylation of Toluene with Methanol to Produce para-Xylene. J. Catal., 1981, 67, 159–174. [45] S. Namba, Y. Kanai, H. Shoji, and T. Yashima, Separation of p-Isomers from Disubstituted Benzenes by Means of Shape-selective Adsorption on Mordenite and ZSM-5 Zeolites. Zeolites, 1984, 4, 77–80. [46] R.M. Barrer, E.F. Vansant, and G. Peeters, Sorption Behaviour of Silanated H-Mordenite. J. Chem. Soc., Faraday Trans. 1, 1978, 74, 1871–1878. [47] A. Thijs, G. Peeters, E.F. Vansant, I. Verhaert, and P.D. Bie`vre, Encapsulation of Gases in H-Mordenite Modified with Silane and Diborane. J. Chem., Soc., Faraday Trans. 1, 1983, 79, 2835–2842.

394

Chemistry of Zeolites and Related Porous Materials

[48] A. Thijs, G. Peeters, E.F. Vansant, I. Verhaert, and P.D. Bie`vre, Purification of Gases in H-Mordenite Modified with Silane and Diborane. J. Chem. Soc., Faraday Trans. 1, 1983, 79, 2821–2834. [49] A. Thijs, S. Peeters, E.F. Vansant, G. Peeters, and I. Verhaert, Modification of H-Mordenite with Silane and Diborane. A Comparative Study of the Reaction Parameters. J. Chem. Soc., Faraday Trans. 1, 1986, 82, 963–975. [50] J. Philippaerts and E.F. Vansant, Silanes, Surf. Interfaces, 1987, 2, 271. [51] Y. Yan, J. Verbiest, P.D. Hulsters, and E.F. Vansant, Modification of Mordenite Zeolites by chemisorption of Disilane and its Influence on the Adsorption Properties. Part 1. – A Modification Parameter Study. J. Chem. Soc., Faraday Trans. 1, 1989, 85, 3087–3094. [52] Y. Yan, J. Verbiest, P.D. Hulsters, and E.F. Vansant, Modification of Mordenite Zeolites by Chemisorption of Disilane and its Influence on the Adsorption Properties. Part 2. – An Adsorption Study. J. Chem. Soc., Faraday Trans. 1, 1989, 85, 3095–3105. [53] M. Niwa, H. Itoh, S. Kato, T. Hattori, and Y. Murakami, Modification of H-Mordenite by a Vapour-Phase Deposition Method. J. Chem. Soc., Chem. Commun., 1982, 819– 820. [54] M. Niwa, S. Morimoto, M. Kato, T. Hattori, and Y. Murakami, Fine control of pore size of Hmordenite by Vapor-phase deposition of Si(OCH3)4, Proceedings of the 8th International Congress on Catalysis, Verlag Chemie, Weinheim, 1984, 701–711. [55] M. Niwa, S. Kato, T. Hattori, and Y. Murakami, Fine Control of the Pore-Opening Size of the Zeolite Mordenite by Chemical Vapour Deposition of Silicon Alkoxide. J. Chem. Soc., Faraday Trans. 1, 1984, 80, 3135–3145. [56] M. Niwa, K. Kawashima, and Y. Murakami, A Shape-Selective Platinum-Loaded Mordenite Catalyst for the Hydrocracking of Paraffins by the Chemical Vapour Deposition of Silicon Alkoxide. J. Chem. Soc., Faraday Trans. 1, 1985, 81, 2757–2761. [57] M. Niwa, M. Kato, T. Hattori, and Y. Murakami, Fine Control of the Pore-Opening Size of Zeolite ZSM-5 by Chemical Vapor Deposition of Silicon Methoxide. J. Phys. Chem., 1986, 90, 6233–6237. [58] M. Niwa and Y. Murakami, CVD Zeolite: Preparation, Characterization and Molecular Shapeselectivity. Mater. Chem. Phys., 1987, 17, 73–85. [59] M. Niwa and Y. Murakami, CVD Zeolites with Controlled Pore-opening Size. J. Phys. Chem. Solids, 1989, 50, 487–496. [60] Y.H. Yue, Y. Tang, and Z. Gao, Studies on the Control of Pore-opening Size of HZSM-5 and its Shape-selective Adsorption. Acta Chim. Sin., 1996, 54, 248–252. [61] Y. Tang, L. Lu, and Z. Gao, Studies on the Control of the Pore-opening Size of HM Zeolite and its Para-selectivity. Acta Physico-Chim., Sin. 1994, 10, 514–520. [62] Z. Gao, Y. H. Yue, and Y. Tang, Chinese Patent ZL 95 1 11,520.0. [63] Y.H. Yue, Y. Tang, Y. Liu, and Z. Gao, Chemical Liquid Deposition Zeolites with Controlled Pore-opening Size and Shape-selective Separation of Isomers. Ind. Eng. Chem. Res., 1996, 35, 430–433. [64] Y.H. Yue, Y. Tang, Y.Z. Kan, and Z. Gao, Pore Size Control of NaY Zeolite by Chemical Liquid Deposition and Shape-selective Separation. Acta Chim. Sin., 1996, 54, 591–597. [65] Y.H. Yue, Y. Tang, and Z. Gao, Zeolite Pore Size Engineering by Chemical Liquid Deposition, Progress in Zeolite and Microporous Materials. Stud. Surf. Sci. Catal., Part C, 1996, 105, 2059–2065. [66] Y.H. Yue, Y. Tang, and Z. Gao, Fine Control of Zeolites Pore Size by Chemical Liquid Deposition. Acta Petrolei Sin. (Petroleum Processing Section), 1997, 13, 30–35. [67] Q. Xu, Y.H. Yue, and Z. Gao, Chemical Liquid Deposition and Shape Selectivity of Zeolite Catalysts. Chin. J. Catal., 1998, 19, 349–353.

Preparation, Secondary Synthesis, and Modification of Zeolites

395

[68] R.W. Weber, K.P. Mo¨ller, M. Unger, and C.T. O’Conner, The Chemical Vapour and Liquid Deposition of Tetraethoxysilane on the External Surface of ZSM-5. Microporous Mesoporous Mater., 1998, 23, 179–187. [69] R.W. Weber, K.P. Mo¨ller, and C.T. O’Conner, The Chemical Vapour and Liquid Deposition of Tetraethoxysilane on ZSM-5, Mordenite and Beta. Microporous Mesoporous Mater., 2000, 35–36, 533–543. [70] K.R. Lee and C.S. Tan, Separation of m- and p-Cresols in Compressed Propane Using Modified HZSM-5 Pellets. Ind. Eng. Chem. Res., 2000, 39, 1035–1038. [71] P. Can˜izares, A. Carrero, and P. Sa´nchez, Isomerization of n-Butene over Ferrierite Zeolite Modified by Silicon Tetrachloride Treatment. Appl. Catal. A., 2000, 190, 93–105. [72] T.W. Kuo and C.S. Tan, Alkylation of Toluene with Propylene in Supercritical Carbon Dioxide over Chemical Liquid Deposition HZSM-5 Pellets. Ind. Eng. Chem. Res., 2001, 40, 4724–4730. [73] S.R. Zheng, H.R. Heydenrych, A. Jentys, and J.A. Lercher, Influence of Surface Modification on the Acid Site Distribution of HZSM-5. J. Phys. Chem. B, 2002, 106, 9552–9558. [74] T.C. Chiang, J.C. Chan, and C.S. Tan, Alkylation of Toluene with Isopropyl Alcohol over Chemical Liquid Deposition Modified HZSM-5 under Atmospheric and Supercritical Operations. Ind. Eng. Chem. Res., 2003, 42, 1334–1340. [75] C. Berger, A. Raichle, R.A. Rakoczy, Y. Traa, and J. Weitkamp, Hydroconversion of Methylcyclohexane on TEOS-modified H-ZSM-5 Zeolite Catalysts: Production of a Highquality Synthetic Steamcracker Feedstock. Microporous Mesoporous Mater., 2003, 59, 1–12.

7 Towards Rational Design and Synthesis of Inorganic Microporous Materials 7.1 Introduction The most important goal of chemistry is to create new materials. Up to now, more than 30 million compounds have been created by scientists. However, the methods for their synthesis and preparation are still mainly based on a trial-and-error approach. With the development of science and the progress of social civilization, developing atomeconomic and highly selective, efficient, and versatile synthetic methods to create new materials based on functional requirements have become key issues for the development of chemistry in the new century. As compared with traditional chemistry,[1] one of the most significant features of rational synthesis is that it directs scientists to pay more attention to the relationship between function, structure, and synthesis and, furthermore, to explore these rules and principles at the molecular level. This will facilitate the coming age of molecular machines. Molecular design and rational synthesis of inorganic microporous crystalline materials are frontier subjects in the fields of zeolites science and molecular engineering. Zeolite synthesis is an active field of research because zeolites with uniform micropores are important in many industrial processes in catalysis, adsorption, and separation, and are finding new applications in electronics, magnetism, chemical sensors, and medicine, etc.[2–9] Synthesis of such materials typically involves crystallization from a gel medium under hydrothermal/solvothermal conditions in the presence of organic amines as

Chemistry of Zeolites and Related Porous Material – Synthesis and Structure Ruren Xu, Wenqin Pang, Jihong Yu, Qisheng Huo and Jiesheng Chen # 2007 John Wiley & Sons, (Asia) Pte Ltd

398

Chemistry of Zeolites and Related Porous Materials

structure-directing agents (SDAs). The crystallization kinetics is governed by various factors, such as the reactant source, gel composition, pH value, cations, organic templates, solvents, crystallization temperature, and crystallization time, etc. So far, 167 types of zeolite frameworks[10] and thousands of zeolite-related open frameworks have been identified. However, the crystallization mechanism by which the inorganic frameworks are assembled around the cations and the organic SDAs is still not clearly understood on the molecular level. There remain significant challenges in the synthesis of zeolites and related porous materials associated with the emerging demands of new materials with predicted structures and properties. Despite the problems mentioned above, considerable effort has been made to establish ways towards the rational design and synthesis of such materials with predetermined structures, compositions, and properties.[11,12] Knowledge is progressing to the point where scientists are beginning to tailor zeolite structures and their syntheses. One of the significant developments in recent years is the use of computational methods, which has greatly enhanced researchers’ understanding of the synthesis of target materials. This chapter will present the current developments in the design and synthesis of inorganic microporous and related materials which is mainly focused on the work by the Jilin group, and others.

7.2 Structure-prediction Methods for Inorganic Microporous Crystals An important prerequisite for the rational synthesis of inorganic microporous materials is the proper design of framework structures based on functional requirements. Therefore, the development of crystal-structure-design methods has aroused considerable attention amongst chemists. Recently, computational simulation techniques used in the design of crystal structures have developed rapidly.[13–39] Several approaches to the structural design of microporous crystals have been described. For example, Smith developed a mathematical theory, and systematically assembled building units into networks of microporous compounds;[15] Akporiaye and Price used operator sequences and sheet coding to systematically enumerate the zeolite frameworks;[16] Newsam and Deem developed a method to determine the 4-connected framework crystal structures by using a simulated annealing method;[17,21] Draznieks et al. predicted inorganic crystal structures through automated assembly of the secondary building units (SBUs) method;[35,36] the Jilin group developed a method to enumerate a series of 2-D 3connected nets systematically,[34] a method to generate a zeolite framework with defined pore geometry by constrained assembly of atoms,[37,38] and a method to generate open-framework aluminophosphates with specified stoichiometries by constrained assembly of building units.[38] By using these methods, one can predict the framework of microporous crystals effectively and better understand the structural chemistry of microporous materials. More importantly, these methods lay a strong foundation for achieving the rational synthesis of crystalline materials with predicted structures and properties.

Towards Rational Design and Synthesis of Inorganic Microporous Materials

7.2.1

399

Determination of 4-Connected Framework Crystal Structures by Simulated Annealing Method

The frameworks of molecular sieves are constructed from 4-connected TO4 tetrahedra. Deem and Newsam developed an approach to optimize an initially arbitrary T-atom configuration with respect to a ‘cost function’ based on the T

T distances, T

T

T angles, and number of first-neighbor T-atoms, by simulated annealing using the Monte Carlo method.[17,21] This method could be used to solve 4-connected crystal structures, as well as to predict unknown hypothetical structures. The simulation procedure is described as below: 1. Define the total energy of a given T-atom configuration based on (1) the T

T distances, (2) T

T

T angles, (3) the number of first neighbors N1, and (4) the number of symmetry operations that reproduce the original T-atom (Figure 7.1). The energy terms for T

T distances and for T

T

T angles are derived from histograms of observed data from a representative number of known zeolite structure types. Appropriate relative weights for these four terms are derived mainly by trial and error. 2. Achieve the optimized configuration of T-atoms by minimizing the total energy through simulated annealing based on a random arrangement of T-atoms. During annealing, low-energy 4-connected configurations generated are stored for further evaluation. 3. Examine visually the representations (projections down the three principal crystallographic directions) of the full unit-cell contents for each of the saved models, followed by addition of framework oxygen atoms between neighboring T-atoms, and DLS (Distance Least Squares) optimization. Results for a plane cell with P6mm symmetry (with 12 symmetry operators) and ˚ , for nunique ¼ 2 and nT ¼ 24, and for nunique ¼ 2 and nT ¼ 18, are shown in a ¼ 18:4 A Figure 7.2(b)–(f), compared with the projection of the T-atom configuration in zeolite L [Figure 7.2(a)]. A single run with nunique ¼ 2 and nT ¼ 24 gave the ltl net [Figure 7.2 (b)], and nets labeled tsv [Figure 7.2 (d)] and twy [Figure 7.2 (e)] by Smith[40] (the projections of the 3-D 4-connected nets numbered 313[41] or 81(2),[42] and 318,[41] respectively). The 2-D nets labeled eoo[40] [Figure 7.2 (f), the projection of the 81(1) or 520[43] net] and tfn[40] [Figure 7.2 (c)] were produced in a run based on nunique ¼ 2 and nT ¼ 18.

Figure 7.1 Energy as a function of the T

T distance (a), the T

T

T angle (b), and the merging distance of two symmetry-related atoms (c) used in the simulation procedure. Reproduced with permission from [17]. Copyright (1989) Nature Publishing

400

Chemistry of Zeolites and Related Porous Materials

Figure 7.2 The projection of the T-atom configuration in zeolite L (a) compared with twodimensional modeling results (b)–(f) based on a hexagonal plane cell, P6mm, with a ¼ 18.4 A˚

The full 3-D case involves a large number of independent variables. Annealing parameters that have been adjusted to facilitate convergence include the temperature decrement between successive cooling stages, and the temperature dependence of the maximum T-atom displacement in the Monte Carlo jiggle step. Typical results are shown in Figure 7.3(a) for a hexagonal cell with P6/mmm symmetry (24 symmetry operators),

Figure 7.3 Eight distinct structure types produced in 3-D based on a hexagonal unit cell, P6/ mmm, with a ¼ 18.4 A˚, c ¼ 7:5 A˚, nunique =2 and nT =36. The structures are shown following addition of the framework oxygen atoms and optimization of the atomic coordinates and the unit-cell dimensions by conventional DLS. Reproduced with permission from [17]. Copyright (1989) Nature Publishing

Towards Rational Design and Synthesis of Inorganic Microporous Materials

401

˚ , c ¼ 7:5 A ˚ , and with nunique ¼ 2 and nT ¼ 36. A total of ten different runs a ¼ 18:4 A generated the LTL-framework (a) and several hypothetical frameworks initially proposed for zeolite omega (c), frameworks corresponding to nets numbered 313 (b), 318 (d), and 315 (e), a related framework with a circuit symbol (4363)1(4363)2 (f), and two sheet structures (g that has nT ¼ 48, and h). These topologies correspond in each case to one of the two-dimensional nets deduced in the projection approximation [a, c, and g correspond to Figure 7.2 (b); b to Figure 7.2 (f), d and f to Figure 7.2 (d); e and h to Figure 7.2 (e)]. The simulated annealing method based on atom assembly can predict the theoretical structures of known and unknown frameworks. In general cases, nunique < 6. 7.2.2

Generation of 3-D Frameworks by Assembly of 2-D Nets

The frameworks of zeolites can be viewed as being constructed from the stacking of simple 3-connected 2-D nets. Akporiaye and Price developed an approach to enumerate systematically the structures of zeolites based on the analysis of the frameworks in terms of component sheets.[16] This method uses simple operator sequences in describing the repeating pattern of the component sheets, which is efficient to describe known zeolite structures and enumerate theoretical structures systematically. Layer Structural Units Different stacking sequences of layer structural units could result in different structure types, which can be illustrated by considering the structures of zeolites ZSM-5 and ZSM11. Both ZSM-5 and ZSM-11 structures are constructed from the same basic structural layer, as shown in Figure 7.4(a). Two different symmetry transformations define the two different structures of ZSM-5 and ZSM-11. In the case of ZSM-5, adjacent layers are related by an inversion center (I) [Figure 7.4 (b)], whereas for ZSM-11, adjacent layers are related by a mirror (M) [Figure 7.4 (c)]. Other structures may also be formed from combinations of two transformations. In this case, there is an infinite number of stacking arrangements available. The simplest one is the ZSM-5/ZSM-11 intermediate with strictly alternating inversion and mirror, as shown in Figure 7.4(d).

Figure 7.4 (a) Projection of ZSM-5 layer; (b) construction of ZSM-5 using the inversion operator; (c) construction of ZSM-11 using the mirror operation; (d) the intermediate ZSM-5/ ZSM-11 structure

402

Chemistry of Zeolites and Related Porous Materials

Smith has already described the zeolite frameworks in terms of 3-connected 2-D nets.[15] The 3-D framework is formed by the addition of a fourth bond to each node of the planar net. The nature of the inter-sheet bonding is automatically defined by the stacking operator, with bonds being formed between nodes within the Si

O

Si-allowed bonding distance. Operators Basic operators that describe the construction of the framework from the component sheet include: (1) Translation (T): This operator describes the vertical translation of the sheet along the stacking direction. Zeolite theta-1 is constructed in this manner, which is illustrated in Figure 7.5. (2) Mirror (M): This operator has already been demonstrated in the above example of ZSM-11. (3) Mirror (Mo): This operator defines an alternative mirror operator for which the mirror plane passes through some of the nodal points of the sheet, effectively fusing two sheets together. As shown in Figure 7.6, ferrierite is constructed in this manner. (4) Mold (Mz): Different from the previous three operations, this operator is not a symmetry transformation. The upper layer is molded down to generate the lower sheet. Figure 7.7 shows an example of this operation, in which the single sheet is used to generate the ‘double sheet’ building unit for the structure of ZSM-5. ZSM-5 structure can be produced from this sheet by a combined MzIMzI operator sequence.

Figure 7.5 The use of T operator to construct theta-1. Reproduced from [16]. Copyright (1989) Elsevier

Towards Rational Design and Synthesis of Inorganic Microporous Materials

403

Figure 7.6 The use of Mo operator to construct FER. Reproduced from [16]. Copyright (1989) Elsevier

The above four operators are the ones that occur most commonly in known zeolite structures. Other operators, such as glide or rotation, are also possible. Sheet Conformation An important prerequisite to this approach is the ability to specify the conformation of the component sheet. Akporiaye and Price developed a coding system to define the 3-D conformation of 3-connected 2-D sheet. This notation is related to that used by Smith, but

Figure 7.7 The use of MzIMzI operator to construct ZSM-5. Reproduced from [16]. Copyright (1989) Elsevier

404

Chemistry of Zeolites and Related Porous Materials

Figure 7.8 Examples using the hexagonal net, showing the relationship of the coding notation from the three groups to the conformation of the hexagonal sheet; filled circles ¼ U, unmarked nodes ¼ D. (a) Group (i); (b) group (ii); (c) group (iii). Reproduced from [16]. Copyright (1989) Elsevier

is employed in describing the conformation of the sheet rather than the interlayer bonding. The symbols U, D, and are assigned to the nodes of the net and appropriately determine the distortion of the net. If all nodes of a sheet are assigned in this way, the sheet is considered to be ‘fully coded’. This can be illustrated by considering the conformation of sheets produced from a hexagonal net using the symbols assigned, in turn, from each of the three groups: (i) U, D; (ii) ; (iii) U, D, . (i) This group uses combinations of the symbols U (upward) and D (downward), defining the perturbation of nodes in the net. An example is shown in Figure 7.8(a). (ii) This group uses the ‘arrow’ symbol , enabling the specification of a diagonal ‘step’ in the net, with the tail and head assigned to adjacent nodes across a linkage [Figure 7.8 (b)]. In assigning this symbol to the net, two rules must be obeyed: (a) ‘arrows’ introduced in this way may not share the same node, and (b) in a circuit around each closed ring there must be an equal number of ‘steps up’ and ‘steps down’ for reasons of parity. (iii) This group uses combinations of all three symbols. An example of a coded sheet is shown in Figure 7.8(c). Allowed Operators Each coded sheet, in conjunction with an operator sequence, can be used to generate structures. However, there are restrictions on the types of permissible operator and on their order in the operator sequence due to the need to produce feasible structures having effective bonding between adjacent sheets and achieving the proper coordination. For example, viable structures may be generated from any of the sheets derived from the first coding group (U, D) using the mirror M operator, but the T operator cannot be used in conjunction with the UD group sheet conformation. Allowed operator combinations for each group are given in Table 7.1. Despite these restrictions on coding arrangements and the use of operators, there are still an infinite number of structures available for each net. A unique structure will be generated by specification of the net, the coding that describes the sheet conformation, and the sequence of operators. Enumeration of Structures The effectiveness of this approach in enumerating zeolite frameworks is illustrated by the ‘ferrierite’ net. Group (i): By applying all possible arrangements of U and D symbols on the net,[24] coded sheet structures are generated. To reduce the number of sheets, equal numbers of U

Towards Rational Design and Synthesis of Inorganic Microporous Materials

405

Table 7.1 Allowed operator combinations for a triple operator sequence. Reproduced with permission from [16]. Copyright (1989) Elsevier Coding group and symbol

First operator

(i) [U,D]

Second operator

M

M

M

(ii) ½ 

Mo T I

Third operator M

M

M

Mo

Mo

T

T

I

I

M

M

Mo

Mo

T

T

M

M

I

I

M Mo

(iii) ½D;



Mz

Mz

Mz

T I

(iv) [U,



M

M

Mo

Mo

Mz

Mz

Mz

T

T

I

I

and D nodes have been restricted. These sheets have the coding arrangements with the highest symmetry. The coded sheets in Figure 7.9 show the set of five arrangements derived from the ferrierite net on this basis. Group (ii): For this group, owing to the restrictions on coding arrangements, there are only two sets of coding schemes available for the ‘ferrierite’ net, as shown in Figure 7.10 (a) and (b). The asterisk on the linkage indicates which pair of nodes may be coded by the ‘arrow’, without specifying its direction. From these two schemes, a number of sheets can be specified which are determined by the relative orientation of the arrows. For the

406

Chemistry of Zeolites and Related Porous Materials

Figure 7.9 The five least-strained structures from group (i) for the ferrierite net, derived from high-symmetry coding arrangements; filled circles ¼ U, unmarked nodes ¼ D. Reproduced from [16]. Copyright (1989) Elsevier

translation operator T, only two different frameworks can be isolated, as illustrated by the arrangements in Figure 7.10(c) and (d): one is that of zeolite theta-1; the other is a related structure. For the Mo operator, only three different frameworks can be isolated; these are illustrated by the arrangements shown in Figure 7.10(e)–(g). One of these structures is that of the zeolite ferrierite [Figure 7.10 (f)], the others being related types. Alternatively, the M operator and combinations of MMo and MT sequences can be applied. Group (iii): A variety of coding arrangements can be specified. The first sheet [Figure 7.11(a)] generates ZSM-5 and ZSM-11, and related structures may be produced from the other two sheets [Figure 7.11(b) and (c)] using similar operators. A different set of structures can be obtained from these three nets by repeating the Mz sequence. This is the only operator allowed for other three nets [Figure 7.11 (d)–(f)], which are variants of the ZSM-5/ZSM-11 coding arrangement in Figure 7.11(a), obtained by replacing some of the D nodes by U. Therefore, related structures can be derived from the same net, the same operator sequence, or the same coding group. Out of the three groups mentioned, group (ii) has been found to produce the greatest variety of frameworks. Group (i) may generate a large set of coding arrangements, but it has only a very limited number of allowed operators. In contrast, groups (ii) and (iii) have only a limited number of coding arrangements available for a particular net, but the variety of allowed operators enables the construction of a variety of structural types and an infinite number of stacking combinations. To sum up, this approach can be used to generate a number of known structures and enumerate related unknown structures. 7.2.3

Automated Assembly of Secondary Building Units (AASBU Method)

Generally, SBUs can be used to describe the structures of microporous inorganic crystals. Draznieks et al. developed a computational method for the prediction of inorganic crystal structures through automated assembly of secondary building units (AASBU).[35,36] This

Towards Rational Design and Synthesis of Inorganic Microporous Materials

407

Figure 7.10 The allowed coding arrangements from group (ii) for the FER net. Reproduced from [16]. Copyright (1989) Elsevier

method has lots of advantages, including: (1) no constraints on the nature or size of the SBUs involved, (2) no constraints on cell parameters and symmetry, (3) the possibility of using various types of SBUs, (4) the possibility of introducing different kinds of linkages between SBUs; for example, corner-, edge-, and face-sharing modes, and (5) flexibility in the definition of the linkage points. Construction of SBU SBU can be extracted from known inorganic crystal structures. Examples of several SBUs are shown in Figure 7.12. Figure 7.12 (a) shows examples of simple SBUs MxLy (M: central metal atoms; L: coordinating or ‘ligand’ atoms), including tetrahedra, octahedra, and corner-, edge-, and face-sharing tetrahedra and/or octahedra. Figure 7.12 (b) shows the D4R unit, M8B12L8 (M: central metal atoms; L: ligand atoms; B: bridging atoms), which is commonly found in a series of inorganic structures. Here only the assembly of D4R units will be described. Ligand atoms are the connection points between the D4Rs units.

408

Chemistry of Zeolites and Related Porous Materials

Figure 7.11 The allowed coding arrangements from group (iii) for the FER net; filled circles ¼ U, empty circles ¼ D. Reproduced from [16]. Copyright (1989) Elsevier

Cost Function and Force Field Parameters During the simulations, the SBUs are treated as rigid bodies. The rule that controls the possible assembly of D4R units is a force field with some Lennard-Jones expressions that essentially favor the attraction between ligand atoms. In other words, this force field favors the formation of L   L ‘sticky-atom’ pairs and allows two D4R units to assemble

Figure 7.12 Examples of building units used in the AASBU method. (a) Simple building units, and (b) D4R unit

Towards Rational Design and Synthesis of Inorganic Microporous Materials

409

Table 7.2 Lennard-Jones potential parameters used in the simulations for assembling D4R units. Reproduced with permission of John Wiley & Sons, Ltd and the American Chemical Society Atom pairs Li  Lj M i  Mj Li  Mj Li  Bj

eij =ðkcal=molÞa

˚ rij =A

400 1 50 1

0.2 3.4 1.8 2.8

a

1 cal ¼ 4.184 J.

through L   L linkages. The Lennard-Jones expression for the energy of interaction between pairs of atoms i and j is defined in Equation (7.1): Eij ¼ eij ½ðrij rij
1 Þ12
2ðrij rij
1 Þ6 

ð7:1Þ

There exist several linkages of the SBUs which should be avoided, for example, the short separation of two metal atoms (Mi   Mj), one ligand atom and one bridging atom (Li   Bj), and Li   Mj in SBUs unlinked. To prevent these undesirable linkages, other potential functions have to be considered, including the repulsive potential between Mi   Mj pairs, the attractive potential between Li    Mj pairs, and the repulsive potential between Li   Bj pairs. A repulsive potential between Mi    Mj pairs prevents SBUs from ˚ overlapping with each other. The distance between the Mi    Mj pair is limited to 3.4 A for D4R. The Lennard-Jones potential parameters used in assembling the D4R are provided in Table 7.2. The total energy (Etotal) of all the D4Rs in one unit cell can be defined as: X Etotal ¼ ðEL...L þ EM...L þ EM...M þ EL...B Þ ð7:2Þ The magnitude of this cost function provides an estimate of the degree of connectivity of a given arrangement of SBUs. In Equation (7.2), the weight of each term is given by the depth of the Lennard-Jones potential well (eij ). Simulation Steps for the Generation of Candidate Structures (a) The first step consists of a simulated annealing procedure where periodic trial arrangements of D4R units are randomly generated[29] within a specified space group and a specified number of D4R units per asymmetric unit. Here only one D4R unit per asymmetric unit is selected as an example to illustrate this method. The angular degrees of freedom of the D4R units are sampled by a Metropolis Monte Carlo algorithm.[44] During the simulated annealing, configurations of lower cost are stored. Simulations are performed in various space groups, typically within a temperature range of 300106 K. (b) Redundant arrangements of D4R units are removed through the comparison of radial distribution functions and simulated diffraction patterns.[30] (c) Each set of D4R unit arrangements is minimized with respect to the cost function Equation (7.2) within the original defined space group. This is the crucial step of the

410

Chemistry of Zeolites and Related Porous Materials

simulations where the D4R units are allowed to connect through the predefined linkage points. (d) After the minimization step, the redundant arrangements are eliminated in a way similar to step (b). (e) Finally, the pairs of sticky atoms, that is, L   L pairs at very short separation ˚ ), are merged into single atoms to form realistic crystal structures. distances (0.20 A The true symmetry of each simulated structure is then redetermined automatically by the find_symmetry algorithm.[45] A series of structures are successfully generated during the simulations, including known zeotypes ACO, AFY, and LTA. The ACO topology is obtained in various space groups: P1, P-1, C2, C2/c, P21, P21/c, C2221, and Pna21. Its tilted derivative is also generated in other space groups: P2, P21, C2, Cc, P21/c, P222, P212121, and Pna21. The symmetry of their SiO2 energy-minimized isotypes is determined after lattice energy minimizations, which lead to Im-3m for the ACO topology in agreement with the experimentally known structure and to P4/mnc for the tilted topology. The AFY topology is generated in P-1 space group exclusively with a final symmetry of P-31m in its SiO2

Figure 7.13 Hypothetical frameworks made of D4R units as generated by simulations. The known ACO topology is successfully predicted together with a tilted variant ACOtilted. Reproduced with permission from [36]. Copyright (2002) American Chemical Society

Towards Rational Design and Synthesis of Inorganic Microporous Materials

411

Table 7.3 Lennard-Jones potential parameters used in the simulations for assembling ML4 units Atom pairs Li    Lj Mi    Mj Mi    Lj

eij /(kcal/mol)

˚ rij =A

400 1 1

0.2 3.9 1.9

form. The LTA zeotype is generated in R3 exclusively, with a final symmetry Pm-3m in its SiO2 form. Except for the known zeolite frameworks mentioned above, a number of hypothetical structures can be also generated during the simulations (structures T1–T10), which are shown in Figure 7.13. Each structure possesses a 3-D framework containing cages or channels. More details could be found in Reference [36]. The above description has illustrated the computational design of new inorganic frameworks constructed from D4R units. Other types of SBUs are also allowed in this ˚ ) is used as the SBU (Lennard-Jones potential parameters method. If ML4 (M

L ¼ 1.65 A are given in Table 7.3), the potential function of SBU in the unit cell is calculated by Equation (7.3):[35] X Etotal ¼ ðEL...L þ EM...L þ EM...M Þ ð7:3Þ SBUs

During the simulation, one or two SBUs are used per asymmetric unit with a selected set of space groups. The known structure types are generated, including GME, FAU, RHO, and LTL. New hypothetical frameworks are also predicted (Figure 7.14). They are the two LTL-related structures found in space group P6/mmm. The first structure [Figure 7.14(a)] is constructed by a 3-D arrangement of gmelinite cage interconnected through their 6-rings, in contrast to the cancrinite cages in the LTL framework but has a

Figure 7.14 (a) Comparison between the LTL-framework and a hypothetical 12-ring channel zeotype-structure predicted by the AASBU method. (b) A second zeotype structure is predicted, which contains GME cages and has channels surrounded by 24 tetrahedra. Reproduced with permission from [35]. Copyright (2000) Wiley-VCH

412

Chemistry of Zeolites and Related Porous Materials

similar 12-ring channel system. The second structure [Figure 7.14(b)] also contains gmelinite cages and has channels running along the [001] direction surrounded by 24 ˚ . Each 24-ring channel connects tetrahedra with free-aperture dimensions of 17.2  19.4 A with its six neighbors through gmelinite cages and through six double-8-rings. The unit˚ , c ¼ 10:58 A ˚. cell content is Si48O96, and the cell parameters are a ¼ 24:73 A The AASBU method provides an effective way for the structural design on the basis of predetermined SBUs. 7.2.4

Prediction of Open-framework Aluminophosphate Structures by using the AASBU Method with Lowenstein’s Constraints

Recently, the Jilin group described an approach to generate hypothetical open-framework aluminophosphate structures with specified Al/P stoichiometry using the AASBU method developed on the basis of the AASBU method as described above.[39] Compared with the existing AASBU method, additional constraints based on Lowenstein’s rule are introduced in this approach; that is, no Al

O

Al or P

O

P linkages are allowed. For each specified Al/P stoichiometry, all the possible combinations of Al and P atoms with different coordination states could be calculated according to Lowenstein’s rule. The Al and P atoms with different coordination states, together with the clusters constructed by them, could be selected as the building units for the structure generation. To satisfy Lowenstein’s rule, additional constraints are introduced into the simulation. The force field that controls the assembly of the building units is parameterized to favor the formation of Al

O

P linkages while avoid the formation of Al

O

Al or P

O

P linkages. This method has been applied to generate aluminophosphates with a specified stoichiometry. According to Lowenstein’s rule, two criteria must be satisfied during the calculation of the possible combinations for a specified stoichiometry: 1) The sum of the numbers of each kind of Al atoms must be equal to the number of Al atoms in the unit cell, and similarly, the sum of the numbers of each kind of P atoms must be equal to the number of P atoms in the unit cell; 2) The number of Al

Ob bonds must be equal to the number of P

Ob bonds. These two rules could also be summarized as Equation (7.4): I X i¼1

mi ¼

J X

nj

ð7:4Þ

j¼1

where I (J) is the number of Al (P) atoms in the unit cell; mi (nj ) is the number of Ob atoms connected with the ith (jth) Al (P) atom. All the possible combinations of different kinds of Al and P atoms can be calculated for a specified stoichiometry. Here the prediction of open-framework aluminophosphates with Al8P10O40 stoichiometry is taken as an example. For instance, 6 Al(Ob)4, 2 Al(Ob)5, 4 P(Ob)4, and 6 P(Ob)3(Ot) atoms could be a possible combination for an aluminophosphate structure with such a stoichiometry. Assuming a space group with two symmetric operations, such as space group P-1, only half of the atoms, i.e., 3 Al(Ob)4, 1 Al(Ob)5, 2 P(Ob)4 and 3 P(Ob)3(Ot) atoms, are needed to be introduced in the asymmetric unit. All the building units used in this example are shown in Figure 7.15 together with the labels

Towards Rational Design and Synthesis of Inorganic Microporous Materials

413

Figure 7.15 The building units for the generation of aluminophosphate structures with 6 Al(Ob)4, 2 Al(Ob)5, 4 P(Ob)4, and 6 P(Ob)3(Ot) atoms in a unit cell. The lighter polyhedra denote the Al-centered polyhedra; the darker tetrahedra denote the P-centered tetrahedra. Reproduced with permission from [39]. Copyright (2005) American Chemical Society

of the force-field atom types for the oxygen atoms. The simulation is carried out in a unit ˚ and a ¼ b ¼ g ¼ 90 . The generation of hypothetical cell with a ¼ b ¼ c ¼ 10A structure H-1 is taken as an example to illustrate the general assembly procedure. First, the building units are randomly introduced in the unit cell [Figure 7.16 (a)]; then, they are assembled through the attraction between the O_al and O_p atoms under the

Figure 7.16 The generation process of H-1 using the AASBU simulation. (a) The initial building units are randomly generated in the unit cell; (b) the building units are assembled together through the attraction of O_al and O_p atoms; (c) the ‘sticky atom’ pairs are ‘glued’ together to form the structure. The lighter polyhedra denote the Al-centered polyhedra, the darker tetrahedra denote the PO4 tetrahedra, and the balls denote the oxygen atoms. Reproduced with permission from [39]. Copyright (2005) American Chemical Society

414

Chemistry of Zeolites and Related Porous Materials

control of the force field including a Lenard–Jones expression and a repulsive exponential expression [Figure 7.16(b)]; at the end of the simulation, the building units are assembled together and the ‘sticky atom’ pairs are formed [Figure 7.16(c)]. After the simulation terminates, ‘sticky atom’ pairs are glued into single Ob atoms, and the hypothetical structure H-1 is formed. Both the atomic coordinates and the cell parameters of H-1 are further refined using a Burchart force field. As shown in Figure 7.17, this structure has a 3-D open framework. Along the [100] direction, there are distorted 16ring channels in which pairs of terminal O atoms protrude [Figure 7.17(a)]. In addition to the 16-ring channels, there exist 8-ring channels running along the [010] direction [Figure 7.17(b)]. This structure could also be viewed as being exclusively constructed by the Al4P5 building blocks [Figure 7.17(c)]. This method will not only aid the prediction of hypothetical aluminophosphate frameworks, but also serve as a tool to set up the initial structural models for the solution of unknown aluminophosphate structures.

Figure 7.17 The hypothetical aluminophosphate structure H-1. This structure has a distorted 16-MR channel along [100] direction (a) and an 8-MR channel along [010] direction (b). This structure could also be seen as constructed by the Al4P5 building blocks (c). The lighter polyhedra denote the Al-centered polyhedra; the darker tetrahedra denote the P-centered tetrahedra. Reproduced with permission from [39]. Copyright (2005) American Chemical Society

Towards Rational Design and Synthesis of Inorganic Microporous Materials

415

Figure 7.18 Forbidden zones defined as cylinders in one unit cell (the thickness of the pore wall is 0.05 A˚ ). Reproduced with permission from [37]. Copyright (2003) American Chemical Society

7.2.5

Design of Zeolite Frameworks with Defined Pore Geometry through Constrained Assembly of Atoms

In 2003, the Jilin group developed a computational methodology for the design of open frameworks with predefined pore geometry.[37] The concept of forbidden zone is introduced. The forbidden zone corresponds to a porous pattern inside which no Tatom can be placed. Compared with previous simulation methods, this method is much more straightforward and efficient, especially for the design of zeolite frameworks with desired pore geometry. As illustrated in Figure 7.18, the pore structure, represented as four cylinders called forbidden zones, is first defined in a unit cell, and then atoms are placed outside the forbidden zones on the basis of specified symmetry and distance constraints. Two constraint conditions must be satisfied when placing the atoms: (i) no T atom is allowed inside a forbidden zone, and (ii) the distance between any two T atoms should not be less ˚ (Si
than 3.0 A
Si distance). This method allows a user to specify the pore size, the number and site symmetry of unique atoms, the unit cell, and the space group. The generation procedure includes the following major steps: (1) A unit cell and the space group are first defined. (2) The pore dimensions and the corresponding number of unique atoms are selected in terms of zeolite framework density. (3) Unique atoms are randomly placed outside the forbidden zones in the unit cell, followed by automatic generation of their equivalent atoms based on symmetry operations. Placed atoms violating conditions (i) and (ii) above will be removed from further consideration; otherwise, a data set containing unique T-atom coordinates will be saved. For distance calculation, the crystal coordinates are converted into Cartesian coordinates. (4) Multiple data sets will be generated and saved by repeating step (3) until certain criteria are met (e.g., a specified number of different data sets have been generated). Coordination sequences (CS) [46,47] are used to distinguish different framework types, as well as to judge whether the generated configuration is a viable one. (5) For each data set, a structural model is built using the Cerius2 software package.[45] Bridging atoms are added using Structure Solve_Bridging atom in Cerius2. Bridging

416

Chemistry of Zeolites and Related Porous Materials

oxygen atoms can be automatically added on the basis of distances between the nearest atoms. (6) The idealized cell parameters are calculated using a DLS-refinement.[48] The unit cell and atom positions of the models are allowed to change. The refinement is carried out assuming a SiO2 composition (interatomic distances: dSi-O ¼ ˚ dO-O ¼ 2:629 A, ˚ and dSi-Si ¼ 3:07 A) ˚ using the weights of 2.0, 0.61, and 1:61 A; 0.23, respectively. The final determination of the symmetry of each generated structure is performed through a ‘Find-Symmetry’ analysis. (7) The potential energy of each structural model is calculated using the Burchart 1.01 force field.[49,50] Here the design of zeolite frameworks with hexagonal space group P63/mmc (No. ˚ for the unit cell, is taken as an example to illustrate this method. 194), a ¼ b ¼ c ¼ 15 A Through combinations of different pore sizes, different numbers of unique T-atoms, and different site symmetries, various structural topologies can be generated. The generation conditions and results are listed in Table 7.4. The pore radii are defined as 3.0, 4.0, 5.0, ˚ , respectively. The numbers of unique T-atoms range from 1 to 3. The atoms are and 6.0 A Table 7.4 Generation conditions and results assuming a P63/mmc space group (No. 194) with unit cell a ¼ b ¼ c ¼ 15 A˚. Reproduced with permission from [37]. Copyright (2003) American Chemical Society Defined pore ˚ ) radius (A 6.0 6.0 6.0 6.0 6.0 6.0 6.0 6.0 6.0 6.0 6.0 5.0 4.0 4.0 4.0 4.0 4.0 3.0 3.0 3.0 3.0 3.0 3.0 3.0 3.0 3.0

Unique T-atoms l lj ll l l ljj ljj ljj jlj ljj ljj ljj jlj ij il lj jl jlj ij jl il lj ll ll jli jli ilj

Refined unit-cell a, c (A˚) 13.7, 12.5, 16.8, 16.8, 19.6, 16.9, 17.4, 17.6, 19.6, 17.6, 17.4, 17.6, 12.6, 13.2, 13.1, 12.6, 17.6, 12.6, 12.6, 13.2, 13.1, 13.7, 12.9, 12.5, 16.8, 12.5,

9.8 15.7 14.2 14.2 5.3* 14.4 5.3* 5.3* 5.3* 5.3* 5.3* 5.3* 10.3 15.0 15.2 15.6 5.3* 10.3 15.6 15.0 15.2 19.8 20.0 20.8 10.5 20.8

Formula Si24O48 Si36O72 Si48O96 Si48O96 Si24O48 Si48O96 Si24O48 Si24O48 Si24O48 Si24O48 Si24O48 Si24O48 Si24O48 Si36O72 Si36O72 Si36O72 Si24O48 Si24O48 Si36O72 Si36O72 Si36O72 Si48O96 Si48O96 Si48O96 Si48O96 Si48O96

Structure type GME H1 H2 H2 H3 H2 H4 H5 H3 H5 H4 H5 LOS EAB ERI H1 H5 LOS H6 EAB ERI AFX H7 AFG H8 AFG

Bold: the unique atoms confined to the pore wall. *: Space group changes to P6/mmm (No.191) after refinement.

Towards Rational Design and Synthesis of Inorganic Microporous Materials

417

Table 7.5 Atomic coordinates of hypothetical zeolite structures. Reproduced with permission from [37]. Copyright (2003) American Chemical Society Structure H1 H2 H3 H4 H5 H6 H7 H8 H9

Atom

X

Y

Z

T1 T2 T1 T2 T1 T2 T1 T2 T1 T2 T1 T2 T1 T2 T1 T2 T3 T1 T2 T3

0.3349 0.9233 0.1452 0.5814 0.6687 0.9080 0.2681 0.3628 0.1580 0.7282 0.9984 0.3352 0.2370 0.6652 0.1853 0.2954 0.5146 0.8070 0.8770 0.0000

0.9227 0.5898 0.4783 0.6074 0.4941 0.3723 0.3730 0.8939 0.6692 0.6289 0.7466 0.9178 0.2373 0.0937
0.0022 0.0000 0.6657 0.8070 0.6180 0.1192

0.0851 0.2500 0.5193 0.1466 0.0000 0.5000 0.0000 0.5000 0.0000 0.5000 0.2500 0.9150 0.3271 0.0606 0.2500 0.0000 0.8976 0.2500 0.6168 0.1238

placed in the general site symmetry l and/or special site symmetry k, j, and i.[51] Each general position corresponds to 24 equivalent atoms, whereas each special position corresponds to 12 equivalent atoms. A number of known zeolite frameworks, including GME, ERI, EAB, LOS, AFX, and AFG, are successfully produced by this program. Their refined unit cells are identical to the idealized cell parameters given in the Atlas of Zeolite Framework Types.[52] In addition, a number of hypothetical structural models have been generated, which have comparable potential energies ranging from
1814.5 to
1824.6 kJ/mol/T to known zeolites. Their atomic coordinates are listed in Table 7.5. If the 24 atoms satisfy conditions (i) and (ii), i.e., all atoms outside the forbidden zones ˚ , the atomic coordinate of the first unique atom T1 is and T

T distances no less than 3.0 A saved [Figure 7.19(a)]. The second unique atom T2 is placed in the special position j followed by generation of its 11 equivalent atoms [Figure 7.19(b)]. The distance between any two atoms of the 36 atoms is calculated. If the atoms satisfy conditions (i) and (ii), the position of the second unique atom T2 is kept; otherwise, the atoms generated by the second unique atom will be replaced until all the atoms satisfy the conditions. For each set of unique T-atom coordinates, a structural model is built using Cerius2. Bridging atoms are ˚ and c ¼ 15:7 A ˚ added [Figure 7.19(c)]. The optimized structure has a unit cell a ¼ 12:5 A with a space group P63/mmc (No. 194) and an empirical formula Si36O72. As shown in Figure 7.19 (d), the structure of H1 consists of 1-D 12-ring channels along the [001] direction. It contains columns of can cages along the c axis; six such columns enclose a 12ring channel. Its simulated X-ray powder diffraction pattern is also shown in Figure7.19(d). Figures 7.20–7.24 illustrate the hypothetical frameworks of H2, H3, H4, H5, and H6, respectively.

418

Chemistry of Zeolites and Related Porous Materials

Figure 7.19 Illustration of the generation process for a hypothetical framework, denoted as H1. The pore radius is set at 6.0 A˚. Two unique atoms are placed, one at the general position l and the other at the special position j. The first unique atom T1 is constrained to the pore wall, and the position on the wall is randomly selected. Then its 23 equivalent atoms are generated using the symmetry operation [Figure 7.19(a)]. The distance between any two atoms is calculated. Reproduced with permission from [37]. Copyright (2003) American Chemical Society

Figure 7.20 Hypothetical framework structure of H2 and its simulated XRD pattern. Reproduced with permission from [37]. Copyright (2003) American Chemical Society

Towards Rational Design and Synthesis of Inorganic Microporous Materials

419

Figure 7.21 Hypothetical framework structure of H3 and its simulated XRD pattern. Reproduced with permission from [37]. Copyright (2003) American Chemical Society

A few examples have been presented which are designed under the P63/mmc space group with the same unit cell. This method certainly allows design of zeolite frameworks under any crystal system, including the cubic, hexagonal, rhombohedral, tetragonal, orthorhombic, monoclinic, and triclinic systems with unit cells of various dimensions. Cross-linked channels can be obtained under a specified space group.[39] For example, assuming a tetragonal space group I4/mmm (No. 139), to generate the crosslinked channels, a forbidden cylinder should be defined along the [001] direction and a forbidden cylinder perpendicular to the [001] direction is also required.

Figure 7.22 Hypothetical framework structure of H4 and its simulated XRD pattern. Reproduced with permission from [37]. Copyright (2003) American Chemical Society

420

Chemistry of Zeolites and Related Porous Materials

Figure 7.23 Hypothetical framework structure of H5 and its simulated XRD pattern. Reproduced with permission from [37]. Copyright (2003) American Chemical Society

Figure 7.25 illustrates two ways to define extra forbidden zones. One way is by defining a forbidden zone along the [100] direction, other channels which are parallel to [010] direction can be generated by 4-fold operation naturally [Figure 7.25(a)]; another way is by defining a forbidden zone along the [110] direction, and the channels which are parallel to that [1
10] direction will be generated naturally [Figure 7.25(b)]. Thus, channels along the [001], [100], and [010] directions can be generated by the first method, and the second method can generate the channels along the [001], [110], and ˚ , using a pore size of [1
10] directions. By defining the cell parameters from 10 to 30 A

Figure 7.24 Hypothetical framework structure of H6 and its simulated XRD pattern. Reproduced with permission from [37]. Copyright (2003) American Chemical Society

Towards Rational Design and Synthesis of Inorganic Microporous Materials

421

Figure 7.25 Two ways to obtain the extra forbidden zones: (a) define a channel along the [100] direction; (b) define a channel along the [110] direction. The solid lines represent the defined channels, and the dashed lines represent the generated channel by symmetry operations. Reproduced from [37]. Copyright (2004) Elsevier

˚ and the number of unique atoms as 2, 3, and 4 respectively, a number 3.0, 4.0, and 5.0 A of structures can be generated, including the known structure types UFI,[53] MER,[52] SBS,[52] and KFI.[52] A series of hypothetical structures can be also generated, for example H9–H17. Table 7.6 and Table 7.7 list their structural data. ˚ , c ¼ 20 A ˚ , using a pore size of 4.0 A ˚ , and By defining the cell parameters of a ¼ 15 A the number of unique atoms of 2, H9 and H10 can be generated by the first method. As Figure 7.26 illustrates, the framework of H9 contains three types of cage: 44488284 cage, 486482 cage, and 4882 (D8R) cage. There are two types of channel along the [001] ˚ ) with round window, direction. One is an 8-ring channel (O   O distance: 4.4  4.3 A ˚ ). Furthermore, it and another is a smaller 8-ring channel (O   O distance: 3.8  3.6 A also has 8-ring channels along the [100] and [010] directions. ˚ ), [100] Figure 7.27 shows that H10 has 8-ring channels along the [001] (4.2  4.2 A ˚ ˚ (3.9  3.8 A), and [010] (3.9  3.8 A) directions. The 8-ring channel is composed of three kinds of cage: 444848828488 cage, D8R cage, and D4R cage, which are connected together alternately. Figure 7.28–Figure 7.34 illustrate the frameworks of H11–H17. For details see Reference [39]. Table 7.6 The space group, cell parameters and framework energy after energy minimization. Reproduced with permission from [37]. Copyright (2004) Elsevier Structure

Formula

H9 H10 H11 H12 H13 H14 H15 H16 H17

Si64O128 Si64O128 Si96O192 Si64O128 Si48O96 Si64O128 Si16O32 Si24O48 Si54O108

Space group and cell parameters/A˚ I4=mmm a ¼ 3:2975; b ¼ 20:1271 I4=mmm a ¼ 4:8141; b ¼ 20:2563 I4=mmm a ¼ 6:4476; b ¼ 24:1577 I4=mmm a ¼ 21:3316; b ¼ 9:3152 I4=mmm a ¼ 2:1506; b ¼ 9:3119 I4=mmm a ¼ 3:5364; b ¼ 20:6527 P4=mmm a ¼ 1:7886; b ¼ 7:1905 P4=mmm a ¼ 2:4582; b ¼ 9:4306 P4=mmm a ¼ 3:7212; b ¼ 7:6444

Energy/[(kJ/mol)/T]
1829.42
1822.86
1821.69
1826.16
1826.70
1822.90
1822.98
1826.70
1828.58

422

Chemistry of Zeolites and Related Porous Materials Table 7.7 Atomic coordinates of hypothetical zeolite structures. Reproduced with permission from [37]. Copyright (2004). Elsevier Structure H9 H10 H11 H12 H13 H14 H15 H16 H17

Atom T1 T2 T1 T2 T1 T2 T3 T1 T2 T1 T2 T1 T2 T1 T2 T1 T2 T1 T2 T3 T4

Atomic coordinates 0.116 0.223 0.250 0.101 0.408 0.205 0.407 0.344 0.327 0.194 0.130 0.115 0.116 0.316 0.500 0.703 0.500 0.133 0.113 0.769 0.230

0.280 0.385 0.396 0.394 0.274 0.340 0.274 0.240 0.428 0.374 0.500 0.274 0.389 0.132 0.133 0.123 0.126 0.272 0.389 0.388 0.613

0.191 0.076 0.075 0.176 0.274 0.167 0.062 0.338 0.165 0.176 0.422 0.197 0.072 0.500 0.207 0.663 0.156 0.296 0.148 0.587 0.000

In addition, a series of chiral zeolite frameworks can be generated by defining chiral space groups.[38] For example, assuming a hexagonal chiral space group P6122 (No. 178), and defining a forbidden zone (channel) along the 61 screw axis in the unit cell, as shown in Figure 7.35, the TO4 tetrahedra would form a helical chain by the 61 operation and further convolve around the forbidden zone to form a chiral channel. Among the 167 zeotype frameworks collected in the IZA online structure database, only four are chiral, namely BEA (P4122), CZP (P6122), GOO (C2221), and OSO

Figure 7.26 Hypothetical framework structure of H9 and its simulated XRD pattern, (a) viewed along the [001] direction; (b) viewed along the [100] direction; (c) the cage unit in H9. Reproduced from [37]. Copyright (2004) Elsevier

Towards Rational Design and Synthesis of Inorganic Microporous Materials

423

Figure 7.27 Hypothetical framework structure of H10 and its simulated XRD pattern, (a) viewed along the [001] direction; (b) viewed along the [100] direction; (c) the cage unit in H10. Reproduced from [37]. Copyright (2004) Elsevier

(P6222). Here the hexagonal space group P6122 (No. 178) is chosen as an example for the simulation to illustrate the efficacy of this method for generating frameworks with chiral channels. During simulation, the forbidden zones are defined as a cylinder running along the 61 screw axis, i.e., the [001] direction, and the cell parameters vary from 5 to ˚ . The cylinder’s radii are defined at 3.0, 4.0, 5.0, and 6.0 A ˚ , respectively. The 25 A numbers of unique T-atoms range from 1 to 5. Under these conditions, a variety of hypothetical structures with chiral channels are successfully generated. The symmetry of each structure is re-checked after geometry optimization to confirm its chirality. Very few

Figure 7.28 Hypothetical framework structure of H11 and its simulated XRD pattern, (a) viewed along the [001] direction; (b) viewed along the [100] direction; (c) the cage unit in H11. Reproduced from [37]. Copyright (2004) Elsevier

424

Chemistry of Zeolites and Related Porous Materials

Figure 7.29 Hypothetical framework structure of H12 and its simulated XRD pattern, (a) viewed along the [001] direction; (b) viewed along the [100] direction; (c) the cage unit in H12. Reproduced from [37]. Copyright (2004) Elsevier

Figure 7.30 Hypothetical framework structure of H13 and its simulated XRD pattern, (a) viewed along the [001] direction; (b) viewed along the [110] direction; (c) the cage unit in H13. Reproduced from [37]. Copyright (2004) Elsevier

Figure 7.31 Hypothetical framework structure of H14 and its simulated XRD pattern, (a) viewed along the [001] direction; (b) viewed along the [110] direction; (c) the cage unit in H14. Reproduced from [37]. Copyright (2004) Elsevier

Towards Rational Design and Synthesis of Inorganic Microporous Materials

425

Figure 7.32 Hypothetical framework structure of H15 and its simulated XRD pattern, (a) viewed along the [001] direction; (b) viewed along the [110] direction; (c) the cage unit in H15. Reproduced from [37]. Copyright (2004) Elsevier

Figure 7.33 Hypothetical framework structure of H16 and its simulated XRD pattern, (a) viewed along the [001] direction; (b) viewed along the [110] direction; (c) the cage unit in H16. Reproduced from [37]. Copyright (2004) Elsevier

Figure 7.34 Hypothetical framework structure of H17 and its simulated XRD pattern, (a) viewed along the [001] direction; (b) viewed along the [110] direction; (c) the cage unit in H17. Reproduced from [37]. Copyright (2004) Elsevier

426

Chemistry of Zeolites and Related Porous Materials

Figure 7.35 Generation of the chiral channel in the space group P6122. The polyhedra represent TO4 tetrahedra, the cylinder represents the predefined forbidden zone. Assembly of TO4 tetrahedra outside the forbidden zone by the 61 operation will generate a chiral channel. Reproduced with permission from [38]. Copyright (2005) American Chemical Society

structures with acquired mirror planes, glide planes, or rotatory-inversion axes are removed. Some examples are shown in Figure 7.36 and Figure 7.37. As compared with previous simulation methods, this method is much more straightforward and efficient. It lays a significant foundation for the rational design and synthesis of zeolite materials with specified channels. 7.2.6

Design of 2-D 3.4-Connected Layered Aluminophosphates with Al3P4O163
Stoichiometry

2-D net sheet structures which are closely related to the 3-D microporous frameworks show rich structural diversity. For example, 2-D 3.4-connected layered aluminophosphates with Al3P4O163
stoichiometry exhibit various sheet structures.[54–67] Except for

Figure 7.36 (a) The chiral channel of the hypothetical framework H178-1 (cylinder diameter of 9.5 A˚), (b) the supertetrahedron T2 building block, (c) the super-5MR constructed by T2 building blocks, and the framework viewed along (d) the [001] direction and (e) the [010] direction. Reproduced with permission from [38]. Copyright (2005) American Chemical Society

Towards Rational Design and Synthesis of Inorganic Microporous Materials

427

Figure 7.37 (a) The left-handed chiral channel in the hypothetical framework H178-2 (cylinder diameter of 8.4 A˚), (b) the chiral channel formed by super ESC in the hypothetical framework H178-7 (cylinder diameter of 5.2 A˚), and (c) the chiral channel formed by double helical chains in the hypothetical framework H180-1 (cylinder diameter of 6.9 A˚). Reproduced with permission from [38]. Copyright (2005) American Chemical Society

Al3P4O16H2C3H5N2,[65] these 2-D networks of Al3P4O163
are all built up from
alternating AlO4 and PO3(


O) units. All Al tetrahedra share four oxygen atoms with adjacent P atoms, whereas P tetrahedra share only three oxygen atoms with adjacent
Al atoms, leaving a terminal P


O bond. Such a connection mode constrains the 3
Al3P4O16 stoichiometry, giving rise to 2-D 3.4-connected nets. In 1999, the Jilin group developed a computational method to systematically enumerate the 2-D 3.4connected Al3P4O163
nets.[34] The method includes the following steps: 1. Generation of the 2-D 3.4-connected meshes. The meshes are generated in a hexagonal array of nodes [Figure 7.38(a)].[68] The generation of 2-D 3.4-connected meshes must obey the following rules in topology: (i) There are two types of points: 3connected points (P-atom) and 4-connected points (Al-atom); (ii) the same type of points cannot directly be connected, according to Lowenstein’s rule;[69] (iii) the mesh must be expanded into a periodic 2-D net. In the 2-D 3.4-connected nets, each node has three kinds of possible linkages; that is, connecting with three neighboring nodes representing 3-connected points, connecting with four neighboring nodes representing 4-connected points, and connecting with no neighboring nodes. Thus, each node may have C63 þ C64 þ 1 ¼ 36 possible configurations. To avoid repeatedly enumerating and reducing the exponential time complexity, the divide-and-conquer algorithm [70,71] is used to reduce the degree of combination explosion. In the first step, the child meshes are generated based on rules (i) and (ii). The primary child meshes are 1  1 meshes with a total number of 36 connections. Child meshes of larger size can be derived from these primary child meshes. However, the two child meshes that must have self-consistent linkages can be incorporated into a bigger child mesh [Figure 7.38(b)]. Then, the generated child meshes are incorporated into a bigger periodic mesh [Figure 7.38(c)]. By combining four child meshes,

428

Chemistry of Zeolites and Related Porous Materials

Figure 7.38 (a) Axial system and unit cell of nodes in mesh generation; (b) the generation of a 2  2 mesh; (c) the generation of a 5  4 mesh by combining four child meshes (1) 3  2, (2) 2  2, (3) 3  2, (4) 2  2 (in b); (d) the 4.6.12 net which is expanded from the 5  4 mesh in (c); (e) the relaxed 4.6.12 net. Reproduced with permission from [34]. Copyright (1999) American Chemical Society

(1) 3  2 mesh, (2) 2  2 mesh, (3) 3  2 mesh, and (4) 2  2 mesh, a final 5  4 mesh (5) is obtained, which can be further expanded into the mesh of a 4.6.12 net [Figure 7.38(d)]. 2. Identify the same topological mesh.[46] 3. Relax the meshes by a genetic algorithm.[72–74] The genetic algorithm is adopted to find the global minimum of the generated structure, which consequently alters the coordinates of the points in the mesh and the shape of the mesh, making the angle whose vertex is the 3-connected point close to 120 , and the angle whose vertex is the 4-connected point close to 90 . Figure 7.38(e) shows the relaxed 4.6.12 net. 4. After the series of computer procedures mentioned above are performed, a large number of 3.4-connected meshes with periodic arrangements are generated. The 2-D Al3P4O163
nets are obtained by substituting the 3- and 4-connected points in the meshes with P and Al atoms, respectively. Bridging oxygen atoms are added between P and Al atoms, and one terminal oxygen atom is attached to each P atom. The energy minimization of these 2-D nets is performed using the Cerius2 package[45] with a Burchart 1.01 force-field.[49,50] The generated nets include all the known 3.4connected sheets and some hypothetical nets. Figure 7.39 shows four typical hypothetical nets. This method can also be applied in the enumeration of 2-D net sheets or even 3-D frameworks with other stoichiometries.

Towards Rational Design and Synthesis of Inorganic Microporous Materials

429

Figure 7.39 Generated 2-D nets with Al3P4O163
stoichiometry: G1–G4 (a–d). Reproduced with permission from [34]. Copyright (1999) American Chemical Society

7.2.7

Hypothetical Zeolite Databases

With considerable efforts on the prediction of hypothetical zeolites having been made, enormous zeolite frameworks have been generated. Gathering all these frameworks together, setting up a database, and analysing their structures will help people understand the rich structural chemistry of zeolites. In particular, a database with a large amount of hypothetical structures is so useful when unknown structures need to be solved while single crystals are unavailable. Moreover, the hypothetical database will play an important role in assisting the synthetic oriented chemists to realize the rational synthesis of zeolite materials with desired structure. Recently, several hypothetical zeolite databases have been built up. In this section, we will introduce some of them, which are freely accessible through the internet from all over the world. Hypothetical Zeolites – Enumeration Research (http://www.hypotheticalzeolites.net/) This database was recently built up by Martin Foster and Michael Treacy at Arizona State University. They have generated over 2 million 4-coordinated frameworks using the Symmetry-Constrained Intersite Bonding Search (SCIBS) method, which are divided into three sub-databases according to specified cost functions. In this database, there are lots of query tools. Users could search specified frameworks by choosing space group, number of T-atoms, framework density, unit-cell volume, TD10, framework energy, and cell parameters, etc. Furthermore, other topological parameters, such as vertex symbol and coordination sequences, could also be used to search a framework. It should also be noted that besides all these features, there are 3 analysis tools that are useful for a user to understand each framework in this database. TOTOPOL is a topological analysis tool; CIF-2-POWD is an online program to calculate the simulated powder XRD pattern; and Sphere Viewer is a program that computes and displays the spheres which can fit inside a framework structure. A user could use these tools directly

430

Chemistry of Zeolites and Related Porous Materials

when they find some interesting frameworks in this database, or upload one’s own structures for analysis and characterization. Hypothetical Zeolite Database (http://mezeopor.jlu.edu.cn/hypo/) This database was recently built up by the Jilin group. General structural parameters, such as space group, cell parameters, number of T-atoms, atomic coordinates, coordination sequences, vertex symbols, loop configurations, framework density, and topological density, etc, could be queried. In this database, both known and hypothetical structures could be found. Data for known structures are obtained from the IZA Database of Zeolite Structures.[10] Except for these known frameworks, all the hypothetical structures are generated through constraint assembly of atoms around a predefined ‘forbidden zone’ that corresponds to a specified pore geometry. The hypothetical structures in this database are featured by specified channel structures. Users can use accessible volume to search structures with pores of certain size. Furthermore, there is a list for all the 3-connected cavities/cages in this database, and users can select one or several of them to search for structures with such building units. For structures with 1-, 2-, or 3-dimensional channels, users can use sizes and orientations as the query conditions. Simulated XRD reflections have been calculated for each framework. Users can use the positions of the first three peaks to search if their unknown samples could be found in this database. It should be noted that all the structures in this database have been fully refined assuming a SiO2 composition. Only structures with low energies comparable with known frameworks are stored in this database. To make sure that all the structures in this database are unique, coordination sequences and vertex symbols are used to get rid of duplicated frameworks.

7.3 Towards Rational Synthesis of Inorganic Microporous Materials At present, rational design and synthesis of microporous crystalline materials has aroused considerable interest amongst zeolite scientists. Molecular engineering of inorganic microporous crystalline materials has become an important frontier subject in the field of inorganic synthesis chemistry. To date, the synthesis of inorganic microporous materials remains more of an art than a science because of the lack of a full understanding of their crystallization mechanisms on the molecular level. The rational synthesis of zeolite materials with predetermined pore architectures and properties is a long-term goal. Although there has not been a great deal of success, researchers are making efforts to get close. In this section, we will present several theoretical methods and experimental approaches towards the rational synthesis of zeolites and related microporous compounds with specified structures. 7.3.1

Data Mining-aided Synthetic Approach

It is well known that the crystallization of microporous crystalline compounds is manipulated by various factors: gel composition, reactant sources, solvent, cation, type of template, pH value, crystallization temperature, and crystallization time, etc. Systematically summarizing the experimental data and extracting the essential rules and theories

Towards Rational Design and Synthesis of Inorganic Microporous Materials

431

will be important for directing the rational synthesis of such materials. In order to well understand the relationship between various synthetic factors and the resulting products with the aid of computational methods, the Jilin group has built up a synthesis database of zeolites and related microporous materials since the 1990s. The database comprises thousands of reaction data about their synthesis. Recently, their preliminary work demonstrated that the knowledge discovery data mining (KDD) method will potentially serve as a useful guidance in aiding the rational synthesis of microporous materials.[75] Through data-mining techniques, the synthetic factors in determining the synthesis of microporous aluminophosphates, an important family of microporous crystalline materials, have been investigated. Several suitable templates for the formation of aluminophosphates with 12-ring channel structures have been theoretically predicted. Introduction to Data Mining Data mining is a new methodology for improving the quality and effectiveness of business and scientific decision-making processes.[76] There are currently several data mining techniques available. The decision-tree method is one of the most important techniques.[77] The attribute-splitting criteria of C5.0 decision tree is entropy, which is used to describe uncertainty of a stochastic experiment. X ¼ p1 ð1Þ; p2 ð2Þ; . . . ; pn ðnÞ is used to describe a stochastic experiment, in which pi denotes the probability of one experiment whose result is i. So the entropy function can be expressed as Equation (7.5): Hðp1 ; p2 ; . . . ; pn Þ ¼


n X

pi 1gpi

ð7:5Þ

i
1

Hðp1 ; p2 ; . . . ; pn Þ is called entropy by Shannon.[78] Decision tree is a technique for partitioning a training file into a set of rules. A decision tree consists of nodes and branches. The starting node is called the root node. Depending upon the results of a test the training files are partitioned into two or more sub-sets. The end result is a set of rules covering all possibilities. In the decision-tree-construction phase, the C5.0 decision tree chooses an attribute that makes the entropy decline most rapidly. The chosen attributes and their order in the decision tree show the importance and the relative importance among attributes in the reaction. So, a decision tree is useful to help analyse the relationship among reaction factors and products with specified structures in the synthesis database of microporous materials. Data-mining Process In the data-mining process, data clean, code, computation, and verification are the four main steps. In the code step, some parameters are used to describe the organic amine template, which plays an important role in the formation of a specified framework in the reaction. There are eight parameters describing the nature of the template, including the longest distance, second longest distance, and shortest distance in the template geometry, the ratio of C to N in the template molecule, the dipole moment, the heat enthalpy, the maximum Hþ number in the template molecule, and the ratio of maximum Hþ number to N in the template molecule. There are six other input attributes, which include the ratio of P to Al in the reaction, the type of solvent, the ratio of solvent to Al, the crystallization temperature, the ratio of template to Al, and the type of template including primary amine, secondary amine, tertiary amine, quarternary amine, and other templating molecules.

432

Chemistry of Zeolites and Related Porous Materials

Table 7.8 The nonbonding interaction energy between AlPO4-5 framework and previously used organic templates. Reproduced with permission from [75]. Copyright (2003) Chemistry Journal of the Chinese University No.

Previously used template

1 2 3 4 5 6 7 8 9

Cyclohexylamine Triethylamine Tetraethylammonium bromide 2-Methylpyridine 4-Methylpyridine 3-Methylpyridine N, N-Dimethylethanolamine Tripropylamine N, N-Diethylethanolamine

Number of templates/unit 2 2 2 2 2 2 2 1 2

Energy/(kJ/mol)
178.75
172.03
172.03
148.89
145.19
145.07
127.76
117.60
96.35

The rules to predict AlPO4-5 (AFI) synthesis are preliminarily built by data mining. The results associate with six attributes, involving the longest atomic distance 0.496 nm, the secondary distance 0.765 nm, the ratio of the number of protons acceptable by the template to the number of N atoms 8, and the formation enthalpy

421.41 kJ/mol. The reliability of the constraint is 178/190 ¼ 93.7%, and the supportability of that is 190/549 ¼ 34.6%. Notably, the six attributes are all related to the template, implying the importance of the template for the formation of AlPO4-5 as is well acknowledged in the synthesis of microporous materials. From Table 7.8 and Table 7.9, it can be seen that the nonbonding interaction energies between the AlPO4-5 framework and theoretical templates are very close to those between the AlPO4-5 framework and previously used templates. Therefore, it is possible to obtain AlPO4-5 by using the predicted seven kinds of templates listed in Table 7.9. In order to confirm this prediction, a series of experiments have been designed. For example, AlPO4-5 could be synthesized by using diethanolamine as the template, and its X-ray powder diffraction pattern is in good agreement with the standard pattern of AlPO4-5, as shown in Figure 7.40. This example showed that extracting the synthetic Table 7.9 The nonbonding interaction energy between AlPO4-5 framework and theoretical templates. Reproduced with permission from [75]. Copyright (2003) Chemistry Journal of the Chinese University No.

Theoretical template

1 2 3 4 5 6 7

N, N-Dimethylcyclohexylamine 1,1-Dimethylpropylamine Trimethylamine Tributylamine Tributyl (benzyl) ammonium bromide Pyridine Diethanolamine

Number of templates/unit 2 2 3 1 2 3 1

Energy/(kJ/mol)
202.10
176.86
172.41
157.25
152.5
122.60
93.68

Towards Rational Design and Synthesis of Inorganic Microporous Materials

433

Figure 7.40 The XRD pattern of AlPO4-5 synthesized by using diethanolamine as the template. Reproduced with permission from [75]. Copyright (2003) Chemistry Journal of the Chinese University

rules or characteristics from the synthesis database by using the decision-tree technique through the data-mining method, and finding out new synthetic conditions according to these rules or characteristics, will greatly assist the rational synthesis. At present, the data-mining technique is still in its infancy in the application to reaction database of microporous crystals, which greatly depends on the user’s experience and the quality of the data sets. However, we believe that it will open up a new route for the rational synthesis of microporous crystals. Finally some experiments are carried out based on the predicted reaction conditions. 7.3.2

Template-directed Synthetic Approach

The choice of organic templates, or SDAs, is acknowledged to be of crucial importance in the formation of a particular inorganic porous structure. Davis et al. summarized the roles of organic SDAs as follows based on the specificity of the inorganic host and organic guest: i) space-filling, ii) structure-directing, and iii) true templating effect.[79] A templating agent is also called an SDA. A better understanding of the host–guest relationship, in particular the role that organic SDAs play in the crystallization, will aid in the design of microporous materials. Host–guest chemistry has given some direction to the template synthesis of zeolite materials. The guest SDA stabilizes the host framework through Coulombic, van der Waals’, and H-bonding interactions, etc. Each may dominate under a selected synthesis system such as low-silica zeolites, highsilica zeolites, aluminophosphate molecular sieves, and interrupted-framework metal phosphates where N   O-type H-bonds of the host–guest play an important role in the formation of the structures. The empirical evidence is that for a template to be successful there must be a ‘good fit’ between the guest molecule and the host framework formed. Recently, computational simulation techniques have greatly facilitated the development of rational synthetic chemistry.[80–87] Computational techniques have been successfully utilized to study the nucleation, growth, and templating effect during hydrothermal synthesis.[84] The Jilin group has studied the templating abilities of different organic amines in specified layered and microporous aluminophosphate structures through a molecular-simulation method.[88,89] Suitable organic templates for the formation of a specified structure can be predicted in terms of host–guest nonbonding interaction

434

Chemistry of Zeolites and Related Porous Materials

energy, which efficiently assists the rational synthesis. Furthermore, on the basis of molecular recognition and chirality transfer between the guest chiral complex molecules and the inorganic host framework, a series of open-framework metal phosphates can be designed and synthesized.[90] These investigations will play an important role in the rational design of inorganic microporous compounds with specified structures. Designing Templates for the Synthesis of Microporous Solids Using de novo Molecular Design Method Lewis et al. developed a method for de novo design of template molecules, which are computationally ‘grown’ within the confines of the pore system for specific microporous materials.[82,83] Growth is initiated from a seed molecule placed in the host. The molecule then grows by a number of random actions using the fragment library as the source of new atoms. The actions include: (i) build, (ii) rotate, (iii) shake, (iv) rock, (v) bond twist, (vi) ring formation, and (vii) energy minimization. Successful SDAs, on the whole, must effectively fill the void space of the framework. A cost function based on overlap of van der Waals’ spheres is used to control the development of new templates; as described in Equation (7.6): X fc ¼ CðtzÞ=n ð7:6Þ t

where C(tz) signifies the closest contact between a template atom t and any host atom z, and n is the number of atoms in the template. During actions (ii)–(vi), movements of the template are made as a series of small steps; the action only continues if fc has increased. Such a strategy allows the template to locate rapidly the largest available space in the host, and to achieve this there its optimum geometric complementarity with the host. This method demonstrates its potential in developing new templates on different types of zeolite topologies, such as LEV, MFI, EU-1 and CHA, etc. Figure 7.41 illustrates the

Figure 7.41 (a) The growth processes involved in the formation of the disubstituted cyclohexane derivative in the levyne structure. Starting from a methane seed, the alkyl chain grows inside the restricted void of the levyne structure (I–V). High bias of the ring formation action results in the formation of methylcyclohexane in step VI, when an atom with a fifth-order neighbor within a distance of 3 A˚ is selected. Further substitution of this ring then occurs. (b) The calculated position of the template in the levyne cage. Reproduced with permission from [83]. Copyright by permission of Nature Publishing

Towards Rational Design and Synthesis of Inorganic Microporous Materials

435

growth process involved in the formation of a candidate template, 1,2-dimethylcyclohexane in the LEV structure starting from a methane seed. It shows a high binding energy with the host framework. Experimentally, using the amino analogue of this template, 2-methylcyclohexylamine, as the SDA, a microporous cobalt-aluminophosphate DAF-4 with LEV structure type is formed. The method of de novo design of SDAs should be applicable to producing candidate templates for the synthesis of new microporous materials with desired interesting framework structures. Predicting the Templating Ability of Organic Amines in the Formation of 2-D Layered Aluminophosphates with Al3P4O163
Stoichiometry Layered aluminophosphates with Al3P4O163
stoichiometry exhibit rich structural diversity,[54–67] the characteristics of which are summarized in Table 7.10. The inorganic
sheets are constructed from alternating AlO4 and PO3(


O) tetrahedra, in which all the four vertices of AlO4 tetrahedra and three-quarters of the PO4 tetrahedra are shared,
leaving a P


O terminal group. Different connection modes result in different layer structures, such as 4.6.8(a)-, 4.6.12-, 4.6(a)-, 4.6(b)-, 4.6(c)-, and 4.6.8(b)-nets. The topologies and secondary building units (SBUs) of six representative 2-D networks are shown in Figure 7.42. These 2-D sheets are found to stack in different sequence, i.e., AAAA, ABAB, ABCABC and ABCDEFABCDEF. The organic template molecules located in the
interspace of inorganic layers form N

H   O hydrogen bonds with P


O groups protruding into the interspace and with bridging O-atoms in the layers. It is found that hydrogen-bonding interaction of the host–guest plays an important role in the construction of a specified structure. The hydrogen-bond interaction between the template agent and inorganic layer is illustrated in Figure 7.4.3. According to the nonbonding interaction energy of host–guest [the hydrogen bond and the van der Waals’ (VDW) energy], a theoretical method to explore the templating ability of organic molecules has been developed by the Jilin group.

Table 7.10 2-D Layered aluminophosphates with Al3P4O163
stoichiometrya L-n L-1 L-2 L-3 L-4 L-5 L-6 L-7 L-8 L-9 L-10 L-11

Formula [Al3P4O16][NH3(CH2)2NH3][OH2(CH2)2OH][OH(CH2)2OH] [Al3P4O16]3[CH3CH2NH3] [Al3P4O16][NH3(CH2)5NH3][C5H10NH2] [Al3P4O16]3[CH3CH2CH2NH3] [Al3P4O16][C5H10NH2]2[C4H7NH3] [Al3P4O16]1.5[NH3(CH2)4NH3] [Al3P4O16]3[CH3(CH2)3NH3] [Al3P4O16]1.5[NH3CHCH3CH2NH3]0.5[H2O] [Al3P4O16]2[C5N2H9][NH4] [Al3P4O16][TETAH3] [Al3P4O16][Co(en)3]3H2O

Structure feature

Stacking sequence

SBU

Ref.

4.6.8(a)-net

ABAB

SBU1

54

4.6.8(a)-net 4.6.8(a)-net 4.6.8(a)-net 4.6.8(a)-net 4.6.12-net 4.6.12-net 4.6(a)-net

AAAA AAAA AAAA AAAA ABAB AAAA ABAB

SBU1 SBU1 SBU1 SBU1 SBU1 SBU1 SBU2

55 60 64 63 56 57 59

4.6(b)-net 4.6.8(b)-net 4.6(c)-net

AAAA ABAB ABAB

SBU3 SBU3 SBU4

59 66 61

a Abbreviations: TETA ¼ NH2(CH2)2NH(CH2)2NH(CH2)2NH2; SBU1, capped 6-MR; SBU2, double-diamond SBU; SBU3, branched double edge-sharing 4-MR; SBU4, the Al3P2 trigonal-bripyrimidal core. (Ref. [88], ACS)

436

Chemistry of Zeolites and Related Porous Materials

Figure 7.42 The topologies and SBUs of six distinct 2-D networks of L-n: (a) 4.6.8(a)-net, (b) 4.6.12-net, (c) 4.6(a)-net, (d) 4.6(b)-net, (e) 4.6(c)-net, and (f) 4.6.8(b)-net. Reproduced with permission from [88]. Copyright (1999) American Chemical Society

Methodology: (1) The Force Field and Parameter. This method is based on the Burchart 1.01–Dreiding 2.21 force field in the Cerius2 package.[45] The Burchart force field is used to treat the frameworks, and the Dreiding II force field is used to treat the intra- and inter-molecular interactions. On the basis of known structure parameters, some parameters which are not addressed in the force field are added. (2) Building Models. The structural models of experimental 2-D layered aluminophosphates are built up according to their crystal structure data using the Cerius2 package.[45] Adding various templates to the experimental sheets involves the following steps: (i) choose several experimental sheets as hosts; (ii) decrease their crystal symmetries to P1 in order to fit all kinds of organic molecules; (iii) determine the number of template molecules in one unit cell based on charge balance; and (iv) define the N

O atom at the position which is favorable to form an H-bond to a terminal P

group, then add the C atoms and H atoms in turn; and finally (v) energy minimization is employed to optimize the configuration of the template. The addition of template agent can be also achieved by using the Monte Carlo method in the Cerius2 package[45] i.e., the organic molecules are randomly added to the specified inorganic layers, and the configuration with the lowest energy is selected.

Towards Rational Design and Synthesis of Inorganic Microporous Materials

437

Figure 7.43 Schematic representation of the interaction of template–host inorganic layers. (a) L-1, (b) L-6, (c) L-7, (d) L-8, (e) L-9, (f) L-10, and (g) L-11 (the number of H-bonds provided by the templates to each SBU is indicated by 1, 2, 3; the horizontal lines indicate the layers). Reproduced with permission from [88]. Copyright (1999) American Chemical Society

(3) Energy Minimization and Calculation. After the model is built, energy minimization is carried out to find the configuration with the lowest energy. Energy optimization is done by using energy minimization to roughly optimize the structure first, and then the Anneal Dynamics-NPH ensembles of Molecular Dynamics are used to make global optimization. To obtain more precise results, a multicycle calculation method is adopted, i.e., the calculation is not considered to have been completed until the difference of the last two calculated total energies of the whole structure is lower than 1 kcal/mol (4.184 kJ/mol). In this work, mainly the nonbonding interaction energies (Einter) between the inorganic sheets and organic templates are studied. It is assumed that Einter ¼ E
ES
ER , where E is the total energy of the whole structure, ES is the steric energy of the inorganic sheets, which includes four kinds of energies: bonds, Urey–Bradley, VDW, and H-bonding energies, and ER is the energy of the organic template itself. The interaction energy Einter, including the energies of VDW and H-bond forces, could then be calculated. Because nonbonding interaction with lower energy reflects the characteristics of the template, the electrostatic energy is omitted in the calculation.

438

Chemistry of Zeolites and Related Porous Materials

Table 7.11 Experimental Es and its components per Al3P4O163
Unit (kcal/mol)a. Reproduced with permission from [89]. Copyright (1999) American Chemical Society L-n

Bonds

L-1 L-2 L-3 L-4 L-5 L-6 L-7 L-8 L-9 L-10 L-11


2681.10
2682.77
2681.53
2683.48
2680.79
2681.58
2682.29
2643.21
2681.78
2669.38
2667.28

Urey– Bradley 2.50 2.72 3.20 5.49 3.49 2.45 2.54 224.31 2.92 12.01 22.44

VDW
13.61
13.44
12.67
12.68
12.46
13.39
12.54
14.08
13.40
13.81
14.51

Electrostatic
34.04
34.16
32.22
32.18
32.17
35.60
35.62
31.72
32.17
31.81
35.78

Note

Es
2726.28
2727.65
2723.20
2722.85
2721.93
2728.13
2727.88
2465.34
2724.43
2703.00
2695.13

Ref.

4.6.8(a)-net, ABAB 4.6.8(a)-net, AAAA 4.6.8(a)-net, AAAA 4.6.8(a)-net, AAAA 4.6.8(a)-net, AAAA 4.6.12-net, ABAB 4.6.12-net, AAAA 4.6(a)-net, ABAB 4.6(b)-net, AAAA 4.6.8(b)-net, ABAB 4.6(c)-net, ABAB

54 55 60 64 63 56 57 59 59 66 61

a

1 cal ¼ 4.184 J.

In fact, the host–guest electrostatic energy is very similar for the compounds with specified Al3P4O163
stoichiometry. The Coulombic interactions between the framework and template are ignored in the calculation. The calculated ES, Einter, and their components of known compounds L-n (n ¼ 1–11) are listed in Table 7.11 and Table 7.12, respectively. The results in Table 7.11 show that the calculated experimental ES values of L-n (n ¼ 1–7, and 9) are similar. The energy difference between them is lower than 7.0 kcal/mol. L-8, L-10, and L-11 exhibit relatively higher Es values, and this is because that their experimental layers are more highly puckered than other normal 2-D nets. As a consequence, their Bond and Urey– Bradley energies are much higher as compared with others. The results in Table 7.12 show that the interaction energies of L-n vary from
11.49 to
33.22 kcal/mol, with the exception of L-8 and L-11, which exhibit higher interaction energies of
4.51 and
4.10 kcal/mol, respectively. As shown in Table 7.12, H-bonding energy can stabilize the 2-D layered structures, and VDW interaction is often an unfavorable factor in these Table 7.12 Experimental interaction energies of templates-inorganic sheets (Einter) per Al3P4O163
Unit (kcal/mol)a and the interlayer spacing D/A˚. Reproduced with permission from [89]. Copyright (1999) American Chemical Society L-n L-1 L-2 L-3 L-4 L-5 L-6 L-7 L-8 L-9 L-10 L-11 a

D 8.852 9.363 9.801 9.597 9.799 9.207 9.633 7.368 10.402 9.955 10.797

1 cal ¼ 4.184 J.

VDW

H-bond

Einter/Al3P4O16

15.36 9.75 8.20
6.02
19.07 18.99 14.08 17.13
1.88 2.33 17.60


31.71
26.50
25.58
27.20
11.91
30.48
28.99
21.64
20.58
16.53
21.70


16.36
16.75
17.38
33.22
30.98
11.49
14.91
4.51
22.46
14.20
4.10

Note 4.6.8-net, ABAB 4.6.8- net, AAAA 4.6.8- net, AAAA 4.6.8- net, AAAA 4.6.8- net, AAAA 4.6.12- net, ABAB 4.6.12- net, AAAA 4.6- net, ABAB 4.6- net, AAAA 4.6.8- net, ABAB 4.6- net, ABAB

Towards Rational Design and Synthesis of Inorganic Microporous Materials

439

structures. This is because the template molecules need to exist in a special geometrical configuration so as to favor the formation of H-bonds to the inorganic sheets. Table 7.12 also shows that the experimental interlayer spacing D of L-n does not change very much for those layered compounds with the same structures while directed by different templates with various sizes and shapes. This means that the size and shape of template molecule does not affect the interplane spacing much. In terms of the optimized interaction energy of host–guest, it shall be able to determine whether a template molecule can stabilize a particular structure. Table 7.13 lists the interaction energies (optimized energies) of the seven types of inorganic sheets and some typical organic amines including mono-, di-, and cyclic amines. From the calculated results, it is found that the experimental structures have lower host–template interaction energies, ranging from
37.07 to
29.02 kcal/mol. This means that experimental template–host network combinations are energetically favorable. On comparing the interaction energies (Einter) of a given template with different inorganic networks, suitable templates can be predicted to have lower interaction energies with a given inorganic network, as indicated by data in italics in Table 7.13. For example, if the experimental sheets of L-2 are chosen as a host, calculated results give the interaction energy of L-2 and T-2 to be
33.27 kcal/mol, which is close to the Table 7.13 Interaction energies (Einter) of templatesa–inorganic sheets per Al3P4O163
unit (kcal/mol).b Reproduced with permission from [89]. Copyright (1999) American Chemical Society Code Template T-1 T-2 T-3 T-4 T-5 T-6 T-7 T-8 T-9 T-10 T-11 T-12 T-13 T-14 T-15 T-16 T-17 T-18 a

C2H5NH2 C3H7NH2 CH3CH(CH3) CH2NH2 C4H9NH2 H2NC2H4NH2 H2NCHCH3 CH2NH2 H2NC3H6NH2 H2NCH2CH (CH3)CH2NH2 H2NC4H8NH2 H2NC5H10NH2 C5H9NH2 C6N4H18 H2N(CH2)6NH2 H2NC5H10NH2: C5H10NH C4H7NH2 C5H10NH:2 C4H7NH2 Co(en)3 H2NC2H4NH2: 2(CH2OH)2

L-7

L-6

L-2

L-1

L-10

L-8


26.66
32.36
28.99


26.33
33.11
27.77
33.27
27.25
30.78


30.64
25.12
25.05
28.86
25.38
21.83
22.39
25.64
22.96
25.19
21.13
32.47


37.07
17.85
21.22


29.16
28.69
18.25
23.26
21.32
23.31

28.91
25.53
21.62
23.72
24.68
28.28
23.36
22.71
17.00
29.11
29.02
26.32


19.02
23.87


19.66
22.30
29.47
23.08


22.68
21.28
23.57
13.19
27.48
21.61
26.11
22.15


22.18
22.49
27.85
29.14
23.16


32.54
26.83
22.37
29.85
24.85


20.57
20.85
29.98
27.76
21.26


22.00
25.33
28.60
13.17
31.68
36.36


23.89
21.69
21.78
30.94
20.23


24.03
23.13
17.98
29.84
22.49


21.21
25.91
23.74
26.21
22.87


21.63
34.38
31.36
36.88

The experimental templates are emboldened; templates predicted as suitable templates are italicized. 1 cal ¼ 4.184 J.

b

L-11

440

Chemistry of Zeolites and Related Porous Materials

Figure 7.44 (a) Theoretical template CH3CH2CH2NH2 (T-2) packing in the interlayer region of the sheets of L-2. (b) Experimental template CH3CH2CH2NH2 packing in the interlayer region of L-4. Reproduced with permission from [88]. Copyright American Chemical Society

optimized experimental Einter value of
33.11 kcal/mol. Therefore, CH3CH2CH2NH2 (T-2) is a suitable template to stabilize the sheets of L-2. Figure 7.44(a) shows the predicted position of T-2 in the interlayer region of L-2. The predicted template position and the manner of H-bonding interaction in the theoretical structure are in agreement with those in the experimental structure of L-4 as shown in Figure 7.44(b). Another example is that of 1,6-hexanediamine (T-13), predicted to be a suitable template for the formation of the 4.6.8(a)-net sheet of L-2. The calculated interaction energy of L-2 and T-13 is
31.68 kcal/mol. This value is close to the optimized experimental interaction energy of
33.11 kcal/mol in L-2. Therefore, 1,6-hexanediamine is predicted to stabilize sheets of L-2. This is confirmed by the experimental fact that 1,6-hexanediamine can direct the formation of the 4.6.8(a)-net ˚ , b ¼ 9.381 A ˚, sheets of L-12 [Al3P4O16]  1.5[NH3(CH2)6NH3] (P-1, a ¼ 8.952 A    ˚ c ¼ 14.840 A, a ¼ 91.87 , b ¼ 91.46 , and g ¼ 102.39 ). Figure 7.45(a) shows the

Figure 7.45 (a) Theoretical template 1,6-hexanediamine (T-13) packing in the interlayer region of the sheets of L-2. (b) Experimental template 1,6-hexanediamine packing in the interlayer region of L-12. Reproduced with permission from [88]. Copyright (1999) American Chemical Society

Towards Rational Design and Synthesis of Inorganic Microporous Materials

441

theoretical template 1,6-hexanediamine packing in the interlayer region of the sheets of L-2. The predicted template position and the manner of H-bonding interaction in the theoretical structure are in good agreement with those in the experimental structure of L-12 [Figure 7.45(b)]. Thus, this methodology will serve as a powerful tool for the rational synthesis of target materials with specified sheets and microporous frameworks. Predicting the Templating Ability of Organic Amines for the Synthesis of Aluminophosphates with an Open Framework Analogous to AlPO-HDA The previous section presents the methodology for investigating the templating ability of organic amines in the formation of 2-D layered compounds. This method can be further applied to 3-D microporous aluminophosphate systems to study the templating ability of organic amines.[89] The modeling method is the same as above and the addition of template to the channel is attained by using Monte Carlo method. AlPO-HDA ([Al4P5O20H][C6H18N2])[91] is made up of alternating Al units (AlO4 and

O)(OH)] forming an interrupted open-framework AlO5) and P units [PO4 and PO2(

structure with interconnected 12- and 8-ring channels along the [010] and [100] direction,

O groups protrude into the channel. The template respectively. Terminal P

OH and P

agent, diprotonated 1,6-hexanediamine (HDA) molecule, is located in the main 12-ring channels and interacts with the framework through H-bonds as shown in Figure 7.46 (a). As a comparison, Figure 7.46(b) shows an isolated HDA molecule at its lowest-energy state. It can be seen that the conformations of the isolated HDA molecule and the encapsulated HDA molecule in the channel are quite different. The energy of the encapsulated HDA (14.03 kcal/mol) is much higher than that of the isolated HDA (
1.09 kcal/mol). However, the configuration of the encapsulated HDA ensures the lowest host–guest interaction energy. This implies that host–guest interaction is important for stabilization of the framework of AlPO-HDA.

Figure 7.46 (a) Open-framework structure of AlPO-HDA with 1,6-hexanediammonium cation located in the 12-membered ring. (b) Isolated 1,6-hexanediamine molecule at its lowest-energy state. Reproduced with permission from [89]. Copyright (2000) American Chemical Society

442

Chemistry of Zeolites and Related Porous Materials

Table 7.14 Interaction energies (Einter) of template–host per unit of [Al4P5O20H]2
(kcal/ mol).a Reproduced with permission from [89]. Copyright [2000] American Chemical Society No.

Templates

1 2 3 4 5 6 7 8 9 10 11 12 13

H2N(CH2)6NH2 HNC6H10NH H2NCHC6H10CHNH2 H2N(CH2)5NH2 H2N(CH2)4NH2 H2N(CH2)7NH2 H2N(CH2)3NH2 H2N(CH2)2NH2 TETA DETA H2N(CH2)8NH2 H2NCH(CH3)CH2NH2 H2NCH2CH(CH3)CH2NH2

EVDW

EH-bond


18.34
16.21
12.94
19.03
15.67
15.21
11.94
8.72
15.99
19.61 1.16
10.78
11.82

0.00
2.12 0.00
2.37
3.48
0.17
2.31
3.33
2.61
1.69
0.28
3.74
1.83

Einter
18.34
18.33
12.94
21.40
19.15
15.39
14.26
12.05
18.60
21.30 0.88
14.52
13.36

a

TETA: H2N(CH2)2NH(CH2)2NH(CH2)2NH2; DETA: H2N(CH2)2NH(CH2)2NH2.

Theoretical calculations show that the experimental and optimized interaction energies Einter between the framework and template molecules in AlPO-HDA are
17.83 and
18.34 kcal/mol (per unit of [Al4P5O20H]2
), respectively. To find suitable templates that can potentially direct the formation of the framework of AlPO-HDA, some typical organic diamines and polyamines, such as 1,5-pentanediamine, 1,4-butanediamine, 1,3propanediamine, diethylenetriamine (DETA), triethylenetetramine (TETA), and so forth, are placed in the main 12-membered ring opening followed by energy optimization. The optimized interaction energies of the host framework and the theoretical templates are presented in Table 7.14. Some suitable templates can be predicted that have lower interaction energies with the host framework, of which 1,5-pentanediamine and DETA exhibit the lowest host–guest interaction energies of
21.40 and
21.30 kcal/mol, respectively. On the basis of energy-minimization results, these two organic amines are selected as the templates to carry out the syntheses of aluminophosphates with openframework structures analogous to AlPO-HDA. By using 1,5-pentanediamine as a template agent, large single crystals of [C5H16N2][Al4P5O20H] (AlPO-PDA) were crystallized in the gel mixture 1.0 Al(iPrO)3 : 2.4 H3PO4 : 1.3 1,5-pentanediamine : 96 H2O at 180 C for 7 days. Single-crystal analysis ˚, showed that it crystallizes in triclinic space group P1 with a ¼ 9.2450(9) A ˚ , c ¼ 5.0657(5) A ˚ , a ¼ 96.02(1) , b ¼ 105.89(1) , and g ¼ 102.88(1) . b ¼ 12.688(2) A As with AlPO-HDA, the structure of AlPO-PDA consists of an open-framework macroanion [Al4P5O20H]2
, which is constructed from the alternation of Al polyhedra

O)(OH)] to form an open framework (AlO4 and AlO5) and P tetrahedral [PO4 and PO2(

with interconnected 12- and 8-MR channels. The diprotonated 1,5-pentanediamine cations occupy the main 12-membered ring, as seen in Figure 7.47(a). The manner of H-bonding between the template and framework is in good agreement with the theoretical prediction, as shown in Figure 7.47(b). By using diethylenetriamine as the template, [C4H15N3]  [Al4P5O19(OH)] was also successfully prepared, whose framework structure was similar to that of AlPO-HDA.

Towards Rational Design and Synthesis of Inorganic Microporous Materials

443

Figure 7.47 (a) Experimental 1,5-pentanediamine located in the 12-membered ring of AlPOHDA. (b) Predicted atomic positions of 1,5-pentanediamine inside the 12-membered ring. Reproduced with permission from [89]. Copyright (2000) American Chemical Society

The experimental structures have confirmed the theoretical predicted results. On the other hand, when using organic amines that have higher interaction energies, the results are that no open frameworks analogous to AlPO-HDA could be synthesized. For example, energy-minimization calculations showed that the interaction energies of H2N(CH2)8NH2 (No. 11) and H2N(CH2)2NH2 (No. 8) are 0.88 and
12.05 kcal/mol, respectively, thereby indicating that they are not favorable template candidates for the formation of the open-framework structure of AlPO-HDA. In fact, using H2N(CH2)8NH2 as a template, only a layer phase [Al3P4O16H3] . 1.5[H3N(CH2)8NH3] analogous to SCS22 could be obtained in the gel system of Al(iPrO)3/H3PO4/NH2(CH2)8NH2/H2O;[92] using H2N(CH2)2NH2 as a template, 1-D chains, [H3N(CH2)2NH3][NH4][AlP2O8][93] and [H3N(CH2)2NH3][AlP2O8H],[94] and a 2-D layer, [H3N(CH2)2NH3][H2O(CH2)2OH] [Al3P4O16],[54] could be obtained in the gel system Al(iPrO)3/H3PO4/H2N(CH2)NH2/EG. Varying the synthetic conditions by using these templates led to no crystalline phase with structure analogous to AlPO-HDA being obtained. This method is also applicable to the synthesis of other microporous compounds, and it will help synthetically oriented chemists to realize their goal of rational synthesis of desired materials. Chirality Transfer from Guest Chiral Metal Complexes to Inorganic Framework Chiral zeolites are particularly desirable for asymmetric catalysis and separation.[79,95] Theoretically, a diverse range of chiral zeolite frameworks have been predicted,[37] however, there is only a single example of zeolite beta (BEA) that was slightly enriched in chiral polymorph A.[79] It is believed that to synthesize a chiral zeolite, the use of a chiral SDA is necessary. In fact, a number of organic SDAs with chiral centers have been used in the synthesis of silica-based zeolites, such as SSZ-24 (AFI), CIT-5 (CFI), CIT-1 (CON), and ZSM-12 (MWW), etc., but none of the resulting frameworks was chiral.[95] The reason why chiral organic SDAs have not been able to transfer their chirality to zeolites is perhaps due to the lack of sufficient noncovalent interactions (e.g., H–bonds) between the host-guest so that SDA is rotationally ‘free’ in the void space of zeolites. It is well known that multiple hydrogen bonds cooperatively exert dramatic influences on the supramolecular assemblies in chemical and biological system.[96–98] Therefore multipoint cooperative noncovalent interactions stronger than van der Waals’ forces are perhaps necessary for transfer of chirality. The Jilin group has observed that multiple

444

Chemistry of Zeolites and Related Porous Materials

H-bonding interactions exert dominant roles in the chirality transfer of guest chiral SDAs to host open-framework metal phosphates based on their systematic work by using chiral cationic cobaltammine complexes as the SDAs.[90] In 1995, Morgan et al. synthesized a layered aluminophosphate compound by using a chiral cobaltammine complex as the template for the first time.[61] Recently, the Jilin group has synthesized a number of 2-D layered and 3-D open-framework metal phosphates by using a racemic mixture or an optically pure chiral metal complex as the template, and has systematically studied the chirality transfer from the guest chiral complex templates to the host inorganic open frameworks.[90] Table 7.15 lists some metal phosphates and oxides with open-framework structures templated by optically pure or racemic cobalt ammine complexes. Chiral metal complex cations commonly used in the synthesis include Co(en)33þ, Co(tn)33þ, Co(dien)23þ, Ir(en)33þ, and Ir(chxn)33þ [where en ¼ 1,2-diaminoethane, tn ¼ 1,3-diaminopropane, dien ¼ bis-(2-aminoethyl)amine, and chxn ¼ 1,2-trans-cyclopentanediamine]. These octahedral metal complexes are chiral, existing in
or  enantiomers. Examination of these chiral metal complex-templated structures reveals that there exists a stereospecificity between the chiral template and the inorganic host, i.e., a common chiral configuration and symmetry in the inorganic micro environments as a result of chirality transfer from chiral metal complexes through hydrogen bonding.[99–110] The followings are some structures to exemplify the chirality transfer. [d-Co(en)3][Al3P4O16] . 3H2O was the first chiral layered aluminophosphate templated by an optically pure chiral complex.[102] As shown in Figure 7.48(a), the inorganic layer is a 4.6-net sheet, which features a series of [3.3.3] propeller-like chiral motifs consisting of three 4-rings exclusively with the  configuration. The inorganic layers are stacked in an ABAB sequence, and the complex cations occluded in the interlayer region are exclusively of the  configuration. It is noted that the inorganic chiral motif has the same C2 symmetry as with the complexed metal cation. The stereospecific correspondence between the chiral complex template and chiral inorganic structural motif reveals the molecular recognition and chiral-transfer phenomenon, which can be well understood by H-bond interactions in the host–guest complex. When the number of hydrogen bonds is reduced from the original ten to eight, the hydrogen-bond energy of the host–guest is increased to 44.85 (kJ/mol)/Co(en)33þ upon replacing the metal complex with  configuration by its enantiomer in the framework. [trans-Co(dien)2][Al3P4O16] . 3H2O (Gtex-3), a chiral layered aluminophosphate, [106] also contains a chiral 4.6-net sheet structure stacked helically in an ABCDEF sequence (Figure 7.49). Its 4.6-net sheet features a series of [3.3.3] propellerlike chiral motifs exclusively with the  configuration, the symmetry of which is C2. If metal complexes with the ‘wrong’ configuration are inserted into the experimental lattice, no hydrogen bonds will be formed between the complex template and the inorganic host. The difference in hydrogen-bond energy of the host–guest between the experimental and the reversed hypothetical frameworks is
44.85 kJ/mol per complex. This explains why a structural motif with the  configuration is induced instead of the
configuration in the crystal lattice. [CoII(en)3]2 . [Zn6P8O32H8] (JLU-7)[107] is a layered zinc phosphate compound with 4.6.8-net sheet structure. Figure 7.50 shows a pair of enantiomers with chiral complexes induced by a pair of enantiomers with [3.3.3] propeller-like chiral structural motifs in the

Towards Rational Design and Synthesis of Inorganic Microporous Materials

445

Table 7.15 Open-framework compounds templated by chiral complexes Dimensionality

Molecular formula

Metal complex 3þ

0-D

[Co(en)3][V3P3BO19] [HPO4] . 4H2O(JLU-1)

Co(en)3

1-D

[Co(en)3][B2P3O11(OH)2]

Co(en)33þ

[Co(en)3][V3O9] . H2O (JLU-2)

Co(en)33þ

[Co(dien)2][V3O9] . H2O (JLU-3) Co(dien)23þ

2-D

Ir(chxn)3[Al2P3O12] . xH2O

Ir(chxn)33þ

[d, l-Co(en)3][Al3P4O16] xH2O(Gtex-1)

Co(en)33þ

[d-Co(en)3][Al3P4O16] . 3H2O

d-Co(en)33þ

[Co(en)3][Ga3(H2PO4)6(HPO4)3] Co(en)33þ (JLU-4)

[Co(en)3][Ga3P4O16] .3H2O(JLU-5)

Co(en)33þ

[Co(dien)2][Ga3P4O16] .3H2O(JLU-6)

Co(dien)23þ

[Ir(en)3][Al3P4O16]  xH2O

Ir(en)33þ

Space group and crystal parameter

Ref.

C2/c 99 ˚ a ¼ 32.8492(14) A b ¼ 11.9601(3) A˚ ˚ c ¼ 22.6001(7) A b ¼ 108.9630(8) C2/c 100 ˚ a ¼ 10.6029(6) A b ¼ 13.8831(7) A˚ c ¼ 12.5583 (4) A˚ b ¼ 92.173(3) 99 P212121 ˚ a ¼ 8.1587(16) A ˚ b ¼ 12.675(3) A c ¼ 18.046(4) A˚ P21/c 99 ˚ a ¼ 16.1663(10) A ˚ b ¼ 8.7028(3) A ˚ c ¼ 13.9773(5) A b ¼ 103.1340(18) P-1 101 a ¼ 9.649(5) A˚ ˚ b ¼ 12.365(9) A c ¼ 16.038(8) A˚ a ¼ 100.02(6) b ¼ 101.64(5) g ¼ 104.75(6) 61 Pna21 a ¼ 8.521(5) A˚ b ¼ 13.775 (5)A˚ c ¼ 21.594 (8) A˚ 102 C2221 a ¼ 8.502(6) A˚ ˚ b ¼ 14.620(6) A ˚ c ¼ 20.890(11) A P21/m 103 a ¼ 9.2103(3) A˚ b ¼ 22.0936(8) A˚ c ¼ 9.5458(4) A˚ b ¼ 108.278(2) Pnna 104 a ¼ 8.6618(2)A˚ b ¼ 21.6071(5) A˚ ˚ c ¼ 13.7426(4) A P6522 104 a ¼ 8.5152(7) A˚ ˚ b ¼ 8.5152(7) A c ¼ 63.278(8) A˚ 101 Pnna a ¼ 8.548(5) A˚ b ¼ 21.983(14) A˚ ˚ c ¼ 13.970 (9)A (Continued)

446

Chemistry of Zeolites and Related Porous Materials

Table 7.15 (Continued) Dimensionality

3-D

Molecular formula

Metal complex 3þ

[Co(tn)3] . [Al3P4O16] . 2H2O(Gtex-2)

Co(tn)3

[trans-Co(dien)2][Al3P4O16] .3H2O(Gtex-3)

Co(dien)23þ

[CoII(en)3]2[Zn6P8O32H8] (JLU-7)

CoII(en)32þ

[Co(en)3][Zn8P6O24Cl] .2H2O (JLU-8)

Co(en)33þ

[d-Co(en)3][H3Ga2P4O16]

d-Co(en)33þ

[CoII(en)3][Zn4(H2PO4)3 (HPO4)2(PO4)(H2O)2](JLU-9)

CoII(en)32þ

[Co(dien)2][Zn2(HPO4)4] . H3O(JLU-10)

Co(dien)23þ

Space group and crystal parameter P21 a ¼ 8.862(4) A˚ ˚ b ¼ 14.703(3) A c ¼ 11.402(5) A˚ b ¼ 108.87(4) P6522 a ¼ 8.457(3) A˚ c ¼ 63.27(2) A˚ C2/c a ¼ 16.623(2) A˚ ˚ b ¼ 30.589(5) A c ¼ 17.441(2) A˚ b ¼ 90.352(10) P-31c a ¼ 8.8775(7) A˚ ˚ b ¼ 8.8775(7) A c ¼ 23.775(3) A˚ I2 a ¼ 9.580(2) A˚ b ¼ 12.6789(5) A˚ c ¼ 9.9631(6) A˚ b ¼ 97.85(2) Pbcn ˚ a ¼ 10.4787(8) A ˚ b ¼ 20.0091(14) A ˚ c ¼ 14.9594(10) A Fdd2 a ¼ 9.271(4) A˚, ˚ b ¼ 19.781(9) A c ¼ 27.045(8) A˚

Ref. 105

106 107

107

108

90

109

Figure 7.48 (a) The 4.6-net sheet in [d-Co(en)3][Al3P4O16] . 3H2O. The [3.3.3] propellanelike chiral motif ( configuration) is also shown. (b) The Co(en)33þ cation with  configuration occluded in the interlayer region. Reproduced with permission from [91]. Copyright (2003) Wiley-VCH

Towards Rational Design and Synthesis of Inorganic Microporous Materials

447

Figure 7.49 The [3.3.3] propellane-like chiral motif and the chiral [Co(dien)2]3þ ions occluded in the interlayer region in Gtex-3. Reproduced with permission from [91]. Copyright (2003) Wiley-VCH

layered structure of JLU-7. The complex cations have C2 and C1 symmetry, and the structural motif has C1 symmetry. [CoIII(en)3][Zn8P6O24Cl] . 2H2O (JLU-8)[107] is a 3-D open-framework zinc phosphate. As shown in Figure 7.51, a pair of enantiomers (D3 symmetry) of chiral complexes induce a pair of enantiomers (C3 symmetry) of caplike chiral structural motifs in the 3-D open-framework structure of JLU-8, which is similar to that of JLU-7. The above

Figure 7.50 A pair of enantiomers of [3.3.3]propellane-like chiral structural motifs in the inorganic layer, and a pair of enantiomers of chiral complexes alternately residing on the interlayer region in JLU-7. Reproduced with permission from [91]. Copyright (2003) Wiley-VCH

448

Chemistry of Zeolites and Related Porous Materials

Figure 7.51 A pair of enantiomers of caplike chiral structural motifs in the inorganic framework and a pair of enantiomers of chiral complexes alternately residing in different channels in JLU-8. Reproduced with permission from [91]. Copyright (2003) Wiley-VCH

examples suggest that the template can impose its individual symmetry constraint onto the inorganic structural motif. A remarkable stereospecificity between the chiral templates and the chiral inorganic structural motifs is illustrated in open-framework zinc phosphate [CoII(en)3][Zn4(H2PO4)3(HPO4)2(PO4)(H2O)2] (JLU-9).[90] It crystallizes in the orthorhombic space group ˚ , b ¼ 20.0091(14) A ˚ , and c ¼ 14.9594(10) A ˚ . The asymPbcn with a ¼ 10.4787(8) A metric unit, as shown in Figure 7.52, contains two unique Zn atoms and four unique P atoms. Both Zn(1) and Zn(2) are tetrahedrally coordinated, making four Zn

O

P

Figure 7.52 Thermal ellipsoid plot (50%) showing the labeling scheme used in JLU-9. Reproduced with permission from [91]. Copyright (2003) Wiley-VCH

Towards Rational Design and Synthesis of Inorganic Microporous Materials

449

Figure 7.53 The open-framework structure of JLU-9 viewed along the [001] direction. Reproduced with permission from [91]. Copyright (2003) Wiley-VCH

linkages. Zn(2) is positionally disordered over two sites, with a lattice water molecule contributing to the coordination of Zn(2)0 . Four P atoms are all 4-coordinated. Of the four distinct P atoms, P(1) and P(2) lie on a twofold axis, while P(3) and P(4) occupy general positions. Three types of phosphate groups, H2PO4
, HPO42
, and PO43
, are found to share their nonhydroxy-group oxygen atoms with Zn atoms. In addition, each asymmetric unit also contains one unique Co atom lying on the twofold axis. The Zn- and P-centered tetrahedra alternate to form a 3-D open framework with a channel running along the [001] direction (Figure 7.53). The channels are partially blocked by the H2PO42
groups that protrude into the channel. A pair of enantiomeric [Co(en)3]2þ ions reside within the channel. It is helpful to view the inorganic framework as being built up from a simple structural motif composed of three 4-rings. These structural motifs are stacked along the [001] direction and are linked together through bridging oxygen atoms, O(3) and O(4), along the [100] and [010] directions to form the 3-D open framework. Notice that the motifs can twist in either the right- or left-handed direction along their symmetry axis. The former is denoted as
configuration and the latter as  configuration. They have the same C2 symmetry as the chiral complex cation. Notably, each chiral structural motif is associated with a chiral metal-complex cation in such a way that the metal complex with the
configuration is in close contact with the chiral motif with the  configuration, and vice versa. This remarkable stereospecific correspondence between the metal-complex template and the structure of the inorganic host clearly indicates that molecular recognition between the guest and the host exists; this allows the configuration and symmetry information of the guest template to be passed onto the inorganic framework. The H-bonding between the inorganic host framework and the guest molecules can help us to understand the observed chiral molecular-recognition phenomenon. Figure 7.54 shows the H-bonding arrangement between the complex cations and the nearby chiral structural motifs in JLU-9. The hydrogen bonds are all of N

H   O type, and the N   O distances involved in the hydrogen bonding are in the range of 2.917– ˚ . These hydrogen-bond distances are within the range expected for this type of 3.066(8) A bonding. Each [Co(en)3]2þ forms ten hydrogen bonds with four chiral inorganic structural motifs of the nearby inorganic host; these all have the same chirality. Namely,

450

Chemistry of Zeolites and Related Porous Materials

Figure 7.54 The hydrogen bonds between the complex templates and nearby chiral structural motifs in JLU-9. Reproduced with permission from [91]. Copyright (2003) Wiley-VCH

the chiral structural motifs that have  configuration form hydrogen bonds with chiral complex cations that have
configuration. Hydrogen bonds can be formed only when chiral structural motifs and chiral complexes have the same configuration. This implies a chirality-selective recognition; in other words, a chiral discrimination effect. Having examined the hydrogen-bonding network, it is readily recognized that if a metal complex with  configuration is to be inserted into the lattice position of the metal complex with
configuration, or vice versa, the number of hydrogen bonds per cobalt complex will be reduced to eight in the reversed framework. The difference between the experimental structure and the reversed hypothetical structure of the hydrogen-bond energy of the host–guest is
27.13 kJ/mol per [Co(en)3]2þ. As a consequence, the complex cations and the chiral structural motifs are related by a twofold axis. Therefore, the hydrogen bonding imposes the C2 symmetry operation of the chiral complex template onto the chiral structural motif. This demonstrates that chiral molecular recognition between the guest and host occurs through hydrogen bonds. Zinc phosphate [Co(dien)2][Zn2(HPO4)4] . [H3O] (JLU-10) [109] with multidirectional intersecting helical channels is another good example showing the chirality transfer from guest template to host framework. It crystallizes in the orthorhombic space group Fdd2 ˚ , b ¼ 19.781(9) A ˚ , and c ¼ 27.045(8) A ˚ . The alternation of (No. 43) with a ¼ 9.271(4) A ZnO4 and HPO42
tetrahedra forms the inorganic framework. Each Zn atom shares four bridging O atoms with adjacent P atoms, and each P atom shares two bridging O atoms with adjacent Zn atoms, leaving two terminal oxygen atoms. The framework of JLU-10 consists solely of 12-rings. Such 12-rings are connected together to form a very open framework with a multidirectional helical pore system. The framework density is 9.7 T ˚ 3 (T ¼ tetrahedrally coordinated atoms Zn or P). Figure 7.55 (a) shows the per 1000 A framework viewed along the [100] direction. It contains 12-ring channels that run along this direction. Each 12-ring accommodates one [Co(dien)2]3þ ion. Note that the [Co(dien)2]3þ ions in alternating rows I and II are a pair of enantiomers that are related by the d glide-plane operation. Interestingly, each 12-ring channel is enclosed by two intertwined helices of the same handedness, which are connected through Zn

O

P linkages [Figure 7.55(b)]. Such chiral interpenetrating double helices are particularly rare in inorganic materials. Figure 7.56(a) shows the framework of JLU-10 viewed along the [110] direction. Besides the 12-ring channels, the channels which appear to have an 8-ring opening can be

Towards Rational Design and Synthesis of Inorganic Microporous Materials

451

Figure 7.55 (a) The framework of JLU-10 viewed along the [100] direction. The [Co(dien)2]3þ ions reside in the 12-membered-ring channels. (b) A space-filling diagram of two intertwined helices that enclose the 12-MR channel. Reproduced with permission from [110]. Copyright (2003) Wiley-VCH

easily seen. In fact, they are enclosed by two types of helical chains as seen in Figure 7.56(b). Furthermore, there are helical channels along the [1
1 0], [4 1 1], [4 1
1], [0 1 1], [0
1 1], [0 3 1], [0
3 1], [1 0 1], and [1 0
1] directions. Alternatively, it is helpful to view the framework of JLU-10 as being built up from a simple structural motif composed of a Zn(1)-centered tetrahedron with four dangling PO4 groups [Figure 7.57(a)]. These units are connected through Zn(2) atoms to form the 3-D open framework. It is notable that such a structural motif is chiral and that it has the same

Figure 7.56 (a) The framework of JLU-10 viewed along the [110] direction showing the 12-MR channels and two types of helical channels. (b) The right-handed (R) and left-handed (L) helical channels. Reproduced with permission from [110]. Copyright (2003) Wiley-VCH

452

Chemistry of Zeolites and Related Porous Materials

Figure 7.57 (a) Zn(1)-centered tetrahedral structural motif. (b) H-bonding arrangement between the chiral complex and the chiral structural motif. Reproduced with permission from [110]. Copyright (2003) Wiley-VCH

C2 symmetry as the chiral complex cation [Co(dien)2]3þ. Figure 7.57(b) shows the Hbonding arrangement between the complex cations and the host framework. Each [Co(dien)2]3þ ion forms a total of 10 H-bonds to the water molecules and the bridging and terminal oxygen atoms in the structural motif. The N   O separations are in the ˚ . Furthermore, the H-bonding between the chiral metal range 2.939(9)–3.124(8) A complex and the chiral structural motifs is related by a twofold symmetry axis. This suggests that the H-bonding imposes the C2 symmetry operation of the chiral complex template onto the chiral structural motif. Upon investigation of the structures of the above examples, the following conclusions are obtained: (1) An asymmetric microenvironment can be invariably induced in the inorganic framework as a result of chirality transfer from chiral metal complexes. (2) There exists molecular recognition between the host framework and the chiral template guest; this allows the symmetry and configuration information of the guest template to be passed onto the inorganic structural motif. (3) The remarkable stereospecific correspondence between the complex template and the inorganic host is attributed to the hydrogen bonding between the host framework and the guest molecules. It is therefore believed that the approach employing a rigid and optically active metal complex or organic amine as a template will become an efficient method for the synthesis of chiral microporous compounds. However, developing a route to the preparation of chiral zeolites, in particular the design of appropriate chiral SDAs, remains a challenging task. Rational Synthesis of Zeolites through Template Design Host–guest chemistry has also given some direction to the template synthesis of zeolite materials. Notable examples are the synthesis of extra-large zeolites SSZ-53 (SFH) and SSZ-59 (SFN) by means of bulky rigid quaternary ammonium SDAs.[111] Burton et al.

Towards Rational Design and Synthesis of Inorganic Microporous Materials

453

Table 7.16 Stabilization energies calculated for each SDA within the frameworks of SSZ-53 and SSZ-59. Reproduced with permission from [111]. Copyright (2003) John Wiley & Sons, Ltd

Molecule

N[a]

Essz-53[b] [KJ per mol SDA]

Essz-59[c] [KJ per mol SDA]

Essz-53[b] [KJ per mol T atoms]

Essz-59[c] [KJ per mol T atoms]

53SDA1

2


192


174


8.0


7.3

53SDA2

2


194


174


8.1


7.3

53SDA3

2


167


149


7.0


6.2

53SDA4

2


181


130


7.5


5.4

53SDA1

2


170


171


5.3


5.3

53SDA2

2


175


186


5.5


5.8

53SDA3

2


197


191


6.2


6.0

53SDA4

2


186


183


5.8


5.7

[a] N ¼ number of SDA molecules per supercell. [b] Essz
53 ¼ Stabilization energy in SSZ-53. [c] Essz
59 ¼ Stabilization energy in SSZ-59.

rationalized the structure-directing effects (SDEs) by using energy-minimization calculations (Table 7.16). Theoretical calculations showed that all the SDA molecules are favorable for the stabilization of both SSZ-53 and SSZ-59. This work demonstrated that the design of SDA molecules in terms of their interaction with the inorganic frameworks was crucial in the successful preparation of desired structures. Another remarkable example is the synthesis of LTA as all-silica polymorph (ITQ-29) by Corma et al. by using a supramolecular organic SDA obtained via the self-assembly through p–p stacking interactions of two identical organic cationic moieties.[112] The selection of the SDAs to synthesize high-silica LTA is based on the following criteria:

454

Chemistry of Zeolites and Related Porous Materials

Figure 7.58 Formation of the LTA structure from the supramolecular self-assembly of the organic SDA molecules. Reproduced with permission from [112]. Copyright (2004) Nature Publishing

i) the bulky organic SDA should fit the inner surface of the a cage of the LTA structure with as many van der Waals’ contacts as possible, but with the least deformation; ii) the SDA should have a weak tendency to form complexes with the solvent, as well as maintain adequate hydrophobicity. However, it is difficult to choose a bulky SDA to fulfill all these requirements simultaneously. Therefore Corma et al. started from the idea that a very large SDA could be achieved by supramolecular self-assembly of two moieties through p–p stacking interactions, each with proper rigidity and polarity properties. 4-Methyl-2,3,6,7-tetrahydro-1H,5H-pyrido: [3.2.1-ij]: quinolinium iodide was judiciously selected because it could form self-assembled dimers in aqueous solution. Figure 7.58 shows the formation of the LTA structure from the supramolecular self-assembly of the organic SDA molecules. By using SDAs in a new way that involves the supramolecular self-assembly of two organic moieties, it has been possible to synthesize the Al-free as well as the pure-silica zeolites with the LTA structure. It is believed that innovations in template design, coupled with new synthetic chemistry, will further lead to the discovery of many novel zeolites with new structures, compositions, and properties. 7.3.3

Rational Synthesis through Combinatorial Synthetic Route

Combinatorial chemistry is the production of libraries of compounds that represent permutations of a set of chemical or physical variables. In recent years, combinatorial chemistry has attracted considerable attention in materials science.[113] Originating from the discovery of new drugs by pharmaceutical companies, combinatorial methods have been employed in the areas of organic, biochemical, and inorganic chemistry, etc. In recent years, the combinatorial approach has been successfully applied to the hydrothermal synthesis of zeolites and related materials.[114] The Jilin group presented a strategy toward the rational synthesis of zeolite materials by combination of computational and combinatorial approaches.[115] In terms of

Towards Rational Design and Synthesis of Inorganic Microporous Materials

455

nonbonding interaction energies of host–guest calculated by molecular simulations, the templating abilities of various organic amines in the formation of aluminophosphate molecular sieve AlPO4-21 (AWO) have been evaluated. Through rational selection of the predicted suitable templates, such as ethanolamine, ethylamine, dimethylamine, and n-propylamine, etc., AlPO4-21 has been successsfully synthesized by hydrothermal combinatorial approach in the reaction system with molar composition 1.0 Al(iOPr)3 : x H3PO4 : y R: 255.0 H2O (R: amines). This work further demonstrates that the combination of computational simulation approach and combinatorial approach will be powerful for the rational synthesis of zeolite materials with predicted structures or properties. Recently, Corma and coworkers synthesized pure silica ITQ-24 (IWR) as well as borosilicate polymorphs with a tridimensional intersecting 10- and 12-ring channel system up to an Si/TIII ratio of 10 by combining a rational design of SDAs and highthroughput synthetic techniques.[116] A set of experiments were designed by using hexamethonium (HM) as SDA in F
media and varying the Si/Ge and TIII/TIV ratios in the gel system (1
x) SiO2 : x GeO2 : y B2O3 : 0.25 HM(OH)2 : 0.5 NH4F : 3 H2O at 175 C for 14 days. 7.3.4

Building-block Built-up Synthetic Route

Construction of new solid crystalline materials through molecular design methods has attracted much attention in recent years. The rational synthesis of metal– organic frameworks (MOFs) through the assembly of SBUs has achieved much progress.[117–119] By this approach, specified rigid molecule building blocks are assembled to ideal ordered structures through chemical bonds. The method can not only modify the structure, but also tune the function of materials, and this opens a new route to the rational synthesis of solid crystalline materials with specified structures, compositions, and functions. However, the prerequisite of design is to maintain the rigid SBU structure throughout the construction process, which is a great challenge for the synthesis of inorganic microporous crystalline materials. Although SBUs, such as D4R, D6R, and various cage units, have been commonly used to describe microporous frameworks, they probably do not exist at all during crystallization of the final product. Construction of Open-framework Metal Phosphates through Building-block Built-up Approach Open-framework metal phosphates display rich structural diversity. For example, aluminophosphate family includes 0-D clusters, 1-D chains, 2-D layers and 3-D frameworks. Upon investigating the structures of aluminophosphates, it is found that 1-D AlP2O83
chain (AlPO-CSC) made up of corner-sharing of Al2P2 4-ring is the parent chain, i.e., building unit, for the construction of various complicated aluminophosphate frameworks.[92] Ozin and coworkers proposed that the linear aluminophosphate chains might be the precursor species during the crystallization of the framework. Under appropriate conditions, it may form other complicated structures through hydrolysis– condensation (Figure 7.59).[120] Experimentally, using transition metal ions, the linear 1-D aluminophosphate chains have been assembled into 3-D frameworks, with 1-D AlP2O83
chains being kept intact in the structure.[121,122]

456

Chemistry of Zeolites and Related Porous Materials

Figure 7.59 Illustration of the formation of one-, two-, or three-dimensional structures by a hydrolysis–condensation self-assembly pathway starting from a chain aluminophosphate. Reproduced with permission from [120]. Copyright (1998) Wiley-VCH

The experimental process is as follows. First, 1-D aluminophosphate chain [H3N(CH2)2NH3][NH4][AlP2O8] is prepared in the reaction mixture with the molar composition Al(iPrO)3/5.0 H3PO4/6.0 EN/82.4 EG in which ethylenediamine (en) is used as the template. As shown in Figure 7.60, the 1-D AlP2O83
anionic chain is built up by

Figure 7.60 1-D AlP2O83
anionic chains made up of corner-sharing Al2P2 4-MRs. Reproduced with permission from [121]. Copyright (2000) American Chemical Society

Towards Rational Design and Synthesis of Inorganic Microporous Materials

457

Figure 7.61 3-D open-framework [C2N2H9][NiAlP2O8] (intact AlPO-CSC chains are retained in the structure). Reproduced with permission from [121] . Copyright (2000) American Chemical Society

corner-sharing Al2P2 4-rings, which are in turn made up of alternating AlO4 and

O)2 tetrahedra. Diprotonated [H3NCH2CH2NH3]2þ and NH4þ cations reside in PO2(



O groups protrude into these regions. Second, the interchain regions, and the P

transition metal cation species M2þ (M ¼ Ni2þ, Co2þ, Fe2þ) are introduced into the reactant mixture in which [H3N(CH2)2NH3][NH4][AlP2O8] has already formed. After crystallization for 15 days, 3-D open-framework compounds [C2N2H9][MAlP2O8] (M ¼ Ni2þ, Co2þ, Fe2þ) are crystallized. The structure of [C2N2H9][NiAlP2O8] is shown in Figure 7.61. It has 8-ring channels along the [001] direction. As shown in Figure 7.61, the 1-D AlP2O83
chains are assembled through octahedral Ni atoms,

O groups and one N atom from coordinated to five terminal oxygen atoms from the P

the template molecules, forming

Ni

O

Ni

chains running along the [001] direction. Strikingly, the AlPO-CSC chain structure is retained (parallel to the [001] direction). This work demonstrates that 1-D AlPO-CSC chain can be used as a building unit to construct complicated aluminophosphate structures, thereby providing an important method for the rational synthesis of inorganic microporous compounds through the building-block builtup approach. Recently, through the building-unit approach, Rao systematically studied the construction of open-framework zinc phosphates by using 1-D zinc phosphate chains similar to AlPO-CSC and D4R units.[123] Construction of D4R-containing Zeolites by using Ge as a Silica Substituent Recently, the use of germanium as a silica substituent coupled with the use of novel SDAs by Corma et al. has led to the discovery of many new D4R-containing zeolite

458

Chemistry of Zeolites and Related Porous Materials

Table 7.17 Some novel structures from use of Ge in the synthesis Zeolite

Code

ITQ-15 IM-12 ITQ-17 ITQ-21 ITQ-22 ITQ-24 ITQ-29

UTL UTL BEC IWW IWR LTA

Channel systema 2-D, 2-D, 3-D, 3-D, 3-D, 3-D, 3-D,

14-R* $ 14-R* $ 12-R*** 12-R*** 12-R* $ 12-R* $ 8-R ***

12-R* 12-R* 10-R* $ 10-R* $ 8-R* 10-R* $ 10-R*

a

The number of asterisks in the notation indicates the channel system being one-, two-, or three dimensional; $ : Interconnecting channel systems are separated by a double arrow.

matertials.[11] The key feature of the introduction of Ge is its SDE on the formation of double 4-rings (D4Rs) building unit. For example, zeolite ITQ-21 was synthesized in the gel with molar composition 0.33 GeO2 : 0.67 SiO2 : 0.50 MSPTOH : 0.5 HF : 20 H2O at 175 C for 5 d using N-methylsparteinium hydroxide (MSPTOH) as SDA.[124] Without Ge in the reaction mixture, CIT-5 (CFI) [125] was formed, which did not have D4Rs in the structure. The structure-directing role of Ge towards the D4R-containing zeolites is also clearly shown by the synthesis of polymorph C of zeolite beta (BEC) by Corma A et al. [126] The syntheses were carried out by using different organic SDAs in the gels with molar compositions (1
x) SiO2 : xGeO2 : (0.5–0.25) SDAOH : 0.5 HF : w H2O at 135  175 C for 15120 h. Without Ge, zeolite beta (BEA) or pure silica ZSM-12 (MTW) were obtained. Table 7.17 summarizes some novel structures resulting from the Ge-directed synthetic approach. Studies have shown that there was a preferential location of Ge in D4R position.[127] The SDE of Ge towards the formation of D4R-containing zeolites is due to the fact that the smaller Ge

O

Ge angles compared with the Si

O

Si angles can relax the geometric constraints in the D4R units and thus stabilize the resulting structures.[128–130]. Remarkably, new zeolites with intersecting or interconnected channels have been prepared by using this approach. It has been a challenge to explore silica-based zeolites with intersecting channels with regard to the demands of shape-selective catalytic reactions in the past decade. In particular, interconnected channels with different dimensions (e.g., 12/10, 14/12, etc.) are very important in the aspects of molecular diffusion and shape-selective catalytic reactions. The first zeolite with interconnected 10and 12-ring pores was the CON family, formed by SSZ-26/SSZ-33/CIT-1. However, they are intergrown materials of two polymorphs A and B. Zeolite ITQ-24 (IWR) with the proposed structure of D4R-containing polymorph C of the CON family, has been successfully prepared with hexamethonium dihydroxide as the SDA by using a Gedirected synthetic approach.[131] It contains a tridirectional pore system, including a ˚ ) running perpendicular to the ab plane, a 12-ring 12-ring straight channel (7.5  5.6 A ˚ ) running along the a axis, and a 10-ring channel sinusoidal channel (7.7  6.2 A ˚ ) intersecting perpendicularly both 12-ring channel systems. The synthesis (5.7  4.8 A involves the hydrothermal reaction of a mixture with molar composition 5.0 SiO2 : 1.0 GeO2: 0.15 Al2O3 : 1.5 hexamethonium dihydroxide: 30 H2O at 175  C for 15 days. Zeolite ITQ-22 (IWW) has been prepared with interconnected 8-ring (4.52  ˚ ), 10-ring (5.86  4.98 A ˚ ), and 12-ring (6.66  6.66 A ˚ ) channels by using 3.32 A

Towards Rational Design and Synthesis of Inorganic Microporous Materials

459

1,5-bis(methylpyrrolidinum)pentane as SDA.[132] A typical synthesis of ITQ-22 involves heating a gel with the molar proportions 0.66 SiO2 : 0.33 GeO2 : 0.01 Al2O3 : 0.2 SDA(OH)2 : 15 H2O at 175 C for 12 d. IM-12 (UTL) is the first zeolite possessing a two-dimensional channel system formed by extra-large pore and large pore channels.[133] It was prepared in a gel with molar composition 0.8 SiO2 : 0.4 GeO2 : 0.3 ROH: 30 H2O at 170  C for 6 d using (6R,10S)6,10-dimethyl-5-azoniaspiro[4,5]decane hydroxide (ROH) as SDA. Its structure contains ˚ ) and 12-ring (8.5  5.5 A ˚ ) intersecting channels parallel to the c and 14-ring (9.5  7.1 A b axes, respectively. Ge atoms occupy the D4Rs. According to chemical analysis, the average Si/Ge ratio in the D4Rs is 7. ITQ-15 (UTL), having the same framework topology as IM-12, was prepared by using 1,3,3-trimethyl-6-azoniatricyclo[3.2.1.46,6]dodecane hydroxide as the SDA.[134] The synthesis of ITQ-15 involved the hydrothermal reaction of a gel with molar composition 0.91 SiO2 : 0.09 GeO2 : 0.01 Al2O3 : 0.5 [C14H26N]OH : 10 H2O at 448 (175 C) K for 18 d. Therefore, the Ge-directed approach has facilitated the formation of a range of D4Rcontaining zeolites, and more significantly, the formation of zeolites with interconnected channel systems. To sum up, the building-unit approach will open a new gate to the rational synthesis of inorganic microporous crystalline materials.

7.4 Prospects Oriented by the function of materials, the design, tailoring, and rational synthesis of target materials is a desired goal for synthetic chemists. Zeolite synthetic chemistry has manifested itself as an exciting area of great importance in both applied and fundamental research. Molecular engineering of zeolite materials has become the key focus in the field of zeolites science in the 21st century. Although great strides have been made towards the rational design and synthesis of zeolitic materials with the assistance of computational simulations and the development of new synthetic methodologies, there remain significant challenges towards rationalization of the synthesis associated with the emerging demands of new zeolite materials with predicted structures and properties. Meanwhile, it must be borne in mind that the rational synthesis of zeolites and related microporous materials is a long-term goal, which requires a better understanding of the crystallization mechanism, and a better understanding of the relationship between function, structure, and synthesis, as well as researchers’ continuous efforts.

References [1] Y. Tang and L. Gui, Studies on Molecular Engineering of Functional Systems. Hunan Science and Technology Press, Changsha, 1997. [2] S.I. Zones and M.E. Davis, Current Zeolite Materials: Recent Discoveries and Future Prospects. Curr. Opin. Solid State Mater. Sci., 1996, 1, 107–117. [3] A. Corma, Microporous to Mesoporous Molecular Sieve Materials and Their Use in Catalysis, Chem. Rev., 1997, 97, 2373–2419. [4] J.M. Thomas, Design, Synthesis, and in situ Characterization of New Solid Catalysts. Angew. Chem., Int. Ed., 1999, 38, 3588–3628.

460

Chemistry of Zeolites and Related Porous Materials

[5] F. Schu¨th and W. Schmidt, Microporous and Mesoporous Materials. Adv. Mater., 2002, 14, 629–638. [6] M.E. Davis, Ordered Porous Materials for Emerging Applications. Nature (London), 2002, 417, 813–821. [7] A. Corma, A. State of the Art and Future Challenges of Zeolites as Catalysts. J. Catal. 2003, 216, 298–312. [8] A. Stein, Advances in Microporous and Mesoporous Solids – Highlights of Recent Progress, Adv. Mater., 2003, 15, 763–775. [9] D. Bruhwiler and G. Calzaferri, Molecular Sieves as Host Materials for Supramolecular Organization, Microporous Mesoporous Mater., 2004, 72, 1–23. [10] International Zeolite Association Structure Commission Website: www.iza-structure.org/ databases. [11] A. Corma, Recent Advances in the Science and Technology of Zeolites and Related Materials. Stud. Surf. Sci. Catal., 2004, 154, 25–40. [12] J. Yu and R. Xu, Recent Advances in the Science and Technology of Zeolites and Related Materials. Stud. Surf. Sci. Catal., 2004, 154, 1–13. [13] S. Kirkpatrick, C.D. Gelatt, and M.P. Vecchi, Optimization by Simulated Annealing. Science, 1983, 220, 671–680. [14] J. Maddox, Crystals from 1st Principles. Nature (London), 1988, 335, 201. [15] J.V. Smith, Topochemistry of Zeolites and Related Materials. 1. Topology and Geometry. Chem. Rev., 1988, 88, 149–182. [16] D.E. Akporiaye and G.D. Price, Systematic Enumeration of Zeolite Framework. Zeolites, 1989, 9, 23–32. [17] M.W. Deem and J.M. Newsam, Determination of 4-Connected Framework Crystal-Structures by Simulated Annealing. Nature (London), 1989, 342, 260–262. [18] M. O’Keeffe and B.G. Hyde, Plane Nets in Crystal Chemistry, Philos. Trans., R. Soc. London 1981, A295, 553–618. [19] C. Catlow and G.D. Price, Computer Modeling of Solid-State Inorganic Materials. Nature (London), 1990, 347, 243–248. [20] J. Pannetier, Prediction of Crystal-Structures from Crystal-Chemistry Rules by Simulated Annealing. Nature (London), 1990, 346, 343–345. [21] M.W. Deem and J.M. Newsam, Framework Crystal-Structure Solution by Simulated Annealing – Test Application to Known Zeolite Structures. J. Am. Chem. Soc., 1992, 114, 7189– 7198. [22] M. O’Keeffe and N.E. Brese, Uninodal 4-Connected 3d Nets. 1. Nets without 3-Rings or 4-Rings. Acta Crystallogr., Sect. A: Found. Crystallogr., 1992, 48, 663–669. [23] N.J.M. Zeolite Structural Problems from a Computational Perspective. Proceedings of the 9th International Zeolite Conference, Butterworth-Heinemann, Montreal, 1992. [24] C.M. Freeman, Inorganic Crystal-Structure Prediction Using Simplified Potentials and Experimental Unit Cells – Application to the Polymorphs of Titanium-Dioxide. J. Mater. Chem., 1993, 3, 531–535. [25] C.R.A. Catlow, New Methods for Modeling Processes within Solids and at their Surfaces. Oxford University Press, London, 1993. [26] M.B. Boisen, G.V. Gibbs, and M.S.T. Bukowinski, Framework Silica Structures Generated Using Simulated Annealing with a Potential-Energy Function-Based on an H6Si2O7 Molecule. Phys. Chem. Miner., 1994, 21, 269–284. [27] D.M. Teter, First-Principles Study of Several Hypothetical Silica Framework Structures. Phys. Rev. B: Condens. Matter 1995, 52, 8064–8073. [28] J.C. Schn and M. Jansen, First Step towards Planning of Synthesis in Solid-state Chemistry. Angew. Chem., Int. Ed. Engl., 1996, 35, 1286–1304.

Towards Rational Design and Synthesis of Inorganic Microporous Materials

461

[29] M.J. Treacy, Enumeration of Periodic Tetrahedral Frameworks. 2. Kristallogr., 1997, 212, 768–791. [30] P. Verwer, F.J.J. Leusen, Computer Simulation to Predict Possible Crystal Polymorphs, Rev. Comput. Chem., 1998, 12, 327–365. [31] M.B. Boisen, A Generation of Framework Structures for the Tectosilicates using a Molecularbased Potential Energy Function and Simulated Annealing Strategies. Microporous Mesoporous Mater., 1999, 29, 219–266. [32] S.M. Woodley, The Prediction of Inorganic Crystal Structures using a Genetic Algorithm and Energy Minimisation. Phys. Chem. Chem. Phys., 1999, 1, 2535–2542. [33] O.D. Friedrichs, Systematic Enumeration of Crystalline Networks. Nature (London), 1999, 400, 644–647. [34] B. Zhou, J. Yu, J. Li, and R. Xu, Rational Design of Two-dimensional Layered Aluminophosphates with [Al3P4O16]3
Stoichiometry. Chem. Mater., 1999, 11, 1094–1099. [35] C.M. Draznieks, De novo Prediction of Inorganic Structures developed through Automated Assembly of Secondary Building Units (AASBU method). Angew. Chem., Int. Ed., 2000, 39, 2270–2275. [36] C.M. Draznieks, S. Girard, and G. Ferey, Novel Inorganic Frameworks constructed from Double-four-ring (D4R) Units: Computational Design, Structures, and Lattice Energies of Silicate, Aluminophosphate, and Gallophosphate Candidates. J. Am. Chem. Soc., 2002, 124, 15326–15335. [37] Y. Li, J. Yu, D. Liu, W. Yan, and R. Xu, Design of Zeolite Frameworks with Defined Pore Geometry through Constrained Assembly of Atoms. Chem. Mater., 2003, 15, 2780–2785. Y. Li, M. Guo, J. Yu, J.Y. Li, and R. Xu, Design of Zeolite Frameworks with Cross-linked Channels through Constrained Assembly of Atoms. Stud. Surf. Sci. Catal., 2004, 154, 308–316. [38] Y. Li, J. Yu, Z. Wang, J. Zhang, M. Guo, and R. Xu, Design of Chiral Zeolite Frameworks with Specified Channels through Constrained Assembly of Atoms. Chem. Mater., 2005, 17, 4399– 4405. [39] Y. Li, J. Yu, J. Jiang, Z. Wang, J. Zhang and R. Xu, Prediction of Open-Framework Aluminophosphate Structures Using the Automated Assembly of Secondary Building Units Method with Lowenstein’s Constraints. Chem. Mater., 2005, 17, 6086–6093. [40] J.V. Smith, in Zeolites: Facts, Figures, Future, ed. R.A. van Stanten and P.A. Jacobs 1989, 29–47. [41] J.V. Smith and W. J. Dytrych, Enumeration of 4-Connected 3-Dimensional Nets and Classification of Framework Silicates – 3d-Nets Based on Double-Bifurcated Chains and the 4.8(2) and 4.6.12 3-Connected Plane Nets. Z. Kristallogr., 1986, 175, 31–36. [42] J.V. Smith and W.J. Dytrych, Nets with Channels of Unlimited Diameter. Nature (London), 1984, 309, 607–608. [43] J.W. Richardson, J.V. Smith, and J.J. Pluth, Theoretical Nets with 18-Ring Channels – Enumeration, Geometrical Modeling, and Neutron-Diffraction Study of AlPO4-54. J. Phys. Chem., 1989, 93, 8212–8219. [44] N. Metropplis, A. Rosenbluth, and M. Rosenbluth, Equation-of-state Calculations by Fast Computing Machines. Chem. Phys., 1953, 21, 1087–1092. [45] Cerius2. 1995, @Molecular simulations/Biosym corporation, San Diego. [46] W.M. Meier and H.J. Moeck, Topology of 3-Dimensional 4-Connected Nets – Classification of Zeolite Framework Types Using Coordination Sequences. J. Solid State Chem., 1979, 27, 349–355. [47] M. O’Keeffe and S.T. Hyde, Vertex Symbols for Zeolite Nets. Zeolites, 1997, 19, 370– 374.

462

Chemistry of Zeolites and Related Porous Materials

[48] Ch. Baerlocher, H. A. and W.M.M., DLS-76, a Program for the Simulation of Crystal Structures by Geometric Refinement. Lab. Fur Kristallographie, Zu¨rich, 1978. [49] E. de Vos Burchart, Studies on Zeolites; Molecular Mechanics, Framework Stability and Crystal Growth. Technische Universiteit Delft, 1992. [50] E. de Vos Burchart, A Consistent Molecular Mechanics Force-Field for Aluminophosphates. J. Chem. Soc., Faraday Trans., 1992, 88, 2761–2769. [51] International Tables for Crystallography, ed. Hahn T. 1983. [52] C. Baerlocher, W.M. Meier, and D.H. Olson, Atlas of Zeolite Framework Types, Elsevier, London, 2001. [53] C.S. Blackwell, R.W. Broach, M.G. Gatter, J.S. Holmgren, D.Y. Jan, G.J. Lewis, B.J. Mezza, T.M. Mezza, M.A. Miller, J.G. Moscoso, R.L. Patton, L.M. Rohde, M.W. Schoonover, W. Sinkler, B.A. Wilson, and S.T. Wilson, Open-framework Materials Synthesized in the TMAþ/TEAþ Mixed-template System: The New Low Si/Al Ratio Zeolites UZM-4 and UZM-5. Angew. Chem., Int. Ed., 2003, 42, 1737–1740. [54] R.H. Jones, J.M. Thomas, R. Xu, Q. Huo, A.K. Cheetham, and A.V.J. Powell, Synthesis and Structure of a Novel Aluminum Phosphate Anion-(Al3P4O16)3
. J. Chem. Soc., Chem. Commun., 1991, 1266–1268. [55] Q. Gao, B. Li, J. Chen, S. Li, R. Xu, I.D. Williams, J. Zheng, and D. Barber, Nonaqueous Synthesis and Characterization of a New 2-Dimensional Layered Aluminophosphate [Al3P4O16] . 3[CH3CH2NH3]. J. Solid State Chem., 1997, 129, 37–44. [56] J.M. Thomas, R.H. Jones, R. Xu, J. Chen, A.M. Chippindale, S. Natarajan, and A.K. Cheetham, A Novel Porous Sheet Aluminophosphate Al3P4O163
1.5[NH3(CH2)4NH3]2þ. J. Chem. Soc., Chem. Commun., 1992, 929–931. [57] A.M. Chippindale, A.R. Cowley, Q. Huo, R.H. Jones, and A.D. Law, Synthesis and Structure of a New Layered Aluminium Phosphate: [BuNH3]3[Al3P4O16]. J Chem. Soc., Dalton Trans., 1997, 2639–2643. [58] I.D. Williams, Q. Gao, J. Chen, L. Nagi, Z. Lin, and R. Xu, Organic-template Control of Inorganic Structure: a Low-symmetry Two-dimensional Sheet Aluminophosphate 3[NH3CHMeCH2NH3][Al6P8O32] . H2O. Chem. Commun., 1996, 1781–1782. [59] J. Yu, J. Li, K. Sugiyama, N. Togashi, O. Terasaki, K. Hiraga, B. Zhou, S. Qiu, and R. Xu, Formation of a New Layered Aluminophosphate [Al3P4O16][C5N2H9]2[NH4]. Chem. Mater., 1999, 11, 1727–1732. [60] R.H. Jones, A.M. Chippindale, S. Natarajan, and J.M. Thomas, A Reactive Template in the Synthesis of a Novel Layered Aluminum Phosphate (Al3P4O16)3
[NH3(CH2)5NH3]2þ (C5H10NH2)þ. J. Chem. Soc., Chem. Commun., 1994, 565–566. [61] K. Morgan, G. Gainsford, and N. Milestone, A Novel Layered Aluminum Phosphate [Co(en)3Al3P4O16 . 3H2O] Assembled About a Chiral Metal-Complex. J. Chem. Soc., Chem. Commun., 1995, 425–426. [62] P.A. Barrett and R.H. Jones, The Pore-size Modification of a Layered Aluminophosphate [Al3P4O16C9H24N3] by Rational Selection of the Intercalated Template Cation. J. Chem. Soc., Chem. Commun., 1995, 1979–1981. [63] S. Oliver, A. Kuperman, and A. Lough, Synthesis and Crystal Structures of two Novel Anionic Aluminophosphates: A One-dimensional Chain, UT-7 ([Al3P5O20H]5
[C7H13NH33þ]5), and a Layer containing two Cyclic Amines, UT-8 ([Al3P4O16]3
[C4H7NH33þ]2[C5H10NH22þ]). Inorg. Chem., 1996, 35, 6373–6380. [64] N. Togashi, Synthesis and Structure of a 2-D-Aluminophosphate Al3P4O16 . 3CH3CH2CH2NH3, J. Meter. Chem., 1998, 8, 2827–2830. [65] J.H. Yu and I.D. Williams, Two Unusual Layer Aluminophosphates Templated by Imidazolium Ions; [N2C3H5][AlP2O8H2 . 2H2O] and 2[N2C3H5][Al3P4O16H]. J. Solid State Chem., 1998, 136, 141–144.

Towards Rational Design and Synthesis of Inorganic Microporous Materials

463

[66] Y. Yao, S. Natarajan, and J. Chen, Synthesis and Structural Characterization of a New Layered Aluminophosphate Intercalated with Triply-protonated Triethylenetetramine [C6H21N4][Al3P4O16]. J. Solid State Chem., 1999, 146, 458–463. [67] L. Vidal, V. Gramlich, and J. Patarin, Synthesis, Characterization and Structure of Mu-7, a New Layered Aluminophosphate with a 4  8 Network. Chem. Lett., 1999, 201–202. [68] I.G. Wood and G.D. Price, A Simple, Systematic Method for the Generation of Periodic, 2-Dimensional, 3-Connected Nets for the Description of Zeolite Frameworks. Zeolites, 1992, 12, 320–327. [69] W. Lowenstein, Am. Mineral., 1954, 39, 92–96. [70] J.D. Becker, I. Eisele, F.W.M. Demann, Implementation of Divide-and-Conquer Algorithms on Multiprocessors, Artif. Intell., 1991, 565, 121–136. [71] D.T. Lee and B.J. Schachter, Two Algorithms for Constructing a Delaunay Triangulation. Int. J. Comput. Info. Sci., 1980, 9, 219–242. [72] J.H. Holland, Adaptation in Natural and Artificial Systems. University of Michigan Press, Ann Arbor, 1975. [73] J. Grefenaatette and J. Baker, in Proceedings of the 3rd International Conference on G. A., 1989. [74] J.D. Schaffer, R.A. Carunan, L.J. Eshelman, and R. Das, in Proceedings of the 3rd International Conference on G. A., 1989 [75] X. Liu, Y. Xu, J. Yu, and R. Xu, Rational Design and Synthesis Aided by Data Mining(I) – Microporous Aluminophosphates with Specified Structures. Chem. J. Chin. Univ., 2003, 24, 949–952. [76] J. Han and K. Micheline, Data Mining: Concepts and Techniques. Higher Education Press, 2001, 68–74. [77] J.R. Quinlan, Induction of Decision Trees. Machine Learning, 1986, 1, 81–106. [78] C.E. Shannon, Communication Theory of Secrecy Systems. Bell Systems Technical Journal, 1949, 656. [79] M.E. Davis and R.F. Lobo, Zeolite and Molecular Sieve Synthesis. Chem. Mater., 1992, 4, 756. [80] D.W. Lewis, C.M. Freeman, and C.R.A. Catlow, Predicting the Templating Ability of Organic Additives for the Synthesis of Microporous Materials. J. Phys. Chem., 1995, 99, 11194–11202. [81] P. Demontis and G.B. Suffritti, Structure and Dynamics of Zeolites Investigated by Molecular Dynamics. Chem. Rev., 1997, 97, 2845–2878. [82] D.W. Lewis, D.J. Willock, C.R.A. Catlow, J.M. Thomas, and G.J. Hutchings, De novo Design of Structure-directing Agents for the Synthesis of Microporous Solids. Nature (London), 1996, 382, 604–606. [83] D.W. Lewis, G. Sankar, and J.K. Wyles, Synthesis of a Small-pore Microporous Material using a Computationally Designed Template. Angew. Chem., Int. Ed. Engl., 1997, 36, 2675–2677. [84] C.R.A. Catlow, D.S. Coombes, and J.C.G. Pereira, Computer Modeling of Nucleation, Growth, and Templating in Hydrothermal Synthesis. Chem. Mater., 1998, 10, 3249– 3265. [85] A.M. Fogg, A.L. Rohl, and G.M. Parkinson, Predicting Guest Orientations in Layered Double Hydroxide Intercalates. Chem. Mater., 1999, 11, 1194–1200. [86] C.M. Freeman, D.W. Lewis, T.V. Harris, A.K. Cheetham, N.J. Henson, P.A. Cox, A.M. Gorman, S.M. Levine, J.M. Newsam, E. Herna´ndez and C.R.A. Catlow, Simulating the Behavior of Organic-molecules in Zeolites. Computer-Aided Mol. Design., 1995, 326–340. [87] P. Wagner, Y. Nakagawa, and G.S. Lee, Guest/host Relationships in the Synthesis of the Novel Cage-based Zeolites SSZ-35, SSZ-36, and SSZ-39. J. Am. Chem. Soc., 2000, 122, 263–273.

464

Chemistry of Zeolites and Related Porous Materials

[88] J. Li, J. Yu, W. Yan, and R. Xu, Structures and Templating Effect in the Formation of 2D Layered Aluminophosphates with Al3P4O163
Stoichiometry. Chem. Mater., 1999, 11, 2600–2606. [89] J. Yu, J. Li, K. Wang, and R. Xu, Rational Synthesis of Microporous Aluminophosphates with an Inorganic Open Framework Analogous to Al4P5O20H . C6H18N2. Chem. Mater., 2000, 12, 3783–3787. [90] Y. Wang, J. Yu, Y. Li, and R. Xu, Chirality Transfer from Guest Chiral Metal Complexes to Inorganic Framework: The Role of Hydrogen Bonding. Chem.-Eur. J., 2003, 9, 5048– 5055. [91] J. Yu, K. Sugiyama, S. Zheng, and R. Xu, Al16P20O80H4 . 4C6H18N2: A New Microporous Aluminophosphate containing Intersecting 12- and 8-Membered Ring Channels. Chem. Mater., 1998, 10, 1208–1211. [92] B. Kraushaarczarnetzki, W.H.J. Stork, and R.J. Dogterom, Novel Aluminophosphate-Based Compounds with a Layered Structure and Intercalation Behavior. Inorg. Chem., 1993, 32, 5029–5033. [93] Q. Gao, J. Chen, S. Li, and R. Xu, Synthesis and Structure of a Chain Aluminophosphate Filled with [NH4]þ and [H3NCH2CH2NH3]2þ Cations. J. Solid State Chem. 1996, 127, 145–150. [94] I.D. Williams, J. Yu, Q. Gao, and R. Xu, New Chain Architecture for a One-dimensional Aluminophosphate, [H3NCH2CH2NH3][AlP2O8H]. Chem. Commun., 1997, 1273–1274. [95] M.E. Davis, Reflections on Routes to Enantioselective Solid Catalysts, Top. Catal., 2003, 25, 3–7. [96] G.A. Jeffrey. An Introduction to Hydrogen Bonding 1st Ed, Oxford University Press. 1922. [97] J. Isrealachvili, Intermolecular and Surface Forces, 2nd Edn, Academic Press, London, 1992. [98] N.B. Bowden, Molecule-mimetic Chemistry and Mesoscale Self-assembly. Acc. Chem. Res., 2001, 34, 231–238. [99] Y. Wang, J. Yu, Y. Li, and R. Xu, Synthesis and Structural Characterization of 0D Vanadium Borophosphate [Co(en)3]2[V3P3BO19][H2PO4] . 4H2O and 1D Vanadium Oxides [Co(en)3] [V3O9] . H2O and [Co(dien)2][V3O9] . H2O Templated by Cobalt Complexes: Cooperative Organization of the Complexes and the Inorganic Networks. Inorg. Chem., 2004, 43, 559–565. [100] G. Yang and S.C. Sevov, [Co(en)3][B2P3O11(OH)2]: A Novel Borophosphate Templated by a Transition-metal Complex. Inorg. Chem., 2001, 40, 2214–2215. [101] I.D. Williams, J. Yu, Q. Gao, and R. Xu, Iridium(III) Amine Complexes as High-stability Structure-directing Agents for the Synthesis of Metal Phosphates. Chem. Mater., 1999, 11, 2241–2249. [102] M.J. Gray, J.D. Jasper, and A.P. Wilkinson, Synthesis and Synchrotron Microcrystal Structure of an Aluminophosphate with Chiral Layers containing Lambda Tris(ethylenediamine)cobalt(III). Chem. Mater. 1997, 9, 976–980. [103] Y. Wang, J. Yu, Z. Shi, and R. Xu, Synthesis and Characterization of a New Layered Gallium Phosphate [Co(en)3][Ga3(H2PO4)6(HPO4)3] Templated by Cobalt Complex. J. Solid State Chem., 2003, 170, 176–181. [104] Y. Wang, J. Yu, Y. Du, and R. Xu, Two Layered Gallophosphates Assembled about the Chiral Metal Complexes: Co(en)3 . Ga3P4O16 . 5H2O and trans-Co(dien)2 . Ga3P4O16 . 3H2O. J. Solid State Chem., 2004, 177, 2511–2517. [105] D.A. Bruce, The Synthesis and Structure of a Chiral Layered Aluminophosphate Containing the Template Co(tn)33þ. J. Chem. Soc., Chem. Commun., 1995, 2059–2060. [106] D.A. Bruce, The Synthesis and Characterization of an Aluminophosphate with Chiral Layers; trans-Co(dien)2 . Al3P4O16 . 3H2O. J. Solid State Chem. 1996, 125, 228–233.

Towards Rational Design and Synthesis of Inorganic Microporous Materials

465

[107] J. Yu, Y. Wang, Z. Shi, and R. Xu, Hydrothermal Synthesis and Characterization of Two New Zinc Phosphates Assembled about a Chiral Metal Complex: [CoII(en)3]2[Zn6P8O32H8] and [CoIII(en)3][Zn8P6O24Cl] . 2H2O. Chem. Mater., 2001, 13, 2972–2978. [108] S.M. Stalder and A.P. Wilkinson, Synthesis and Characterization of a Chiral 3D-Framework Material: d-Co(en)3[H3Ga2P4O16]. Chem. Mater., 1997, 9, 2168–2173. [109] Y. Wang, J. Yu, M. Guo, and R. Xu, [{Zn2(HPO4)4}{Co(dien)2}] . H3O: A Zinc Phosphate with Multidirectional Intersecting Helical Channels. Angew. Chem., Int. Ed., 2003, 42, 4089–4092. [110] Y. Du and J. Yu, [Co(en)3][In3(H2PO4)6(HPO4)3] . H2O: A New Layered Indium Phosphate Templated by Cobalt Complex. J. Solid State Chem., 2004, 177, 3032– 3037. [111] A. Burton, S. Elomari, C.-Y. Chen, R.C. Medrud, I.Y. Chan, L.M. Bull, C. Kibby, T.V. Harris, S.I. Zones, and E.S. Vittoratos, SSZ-53 and SSZ-59: Two Novel Extra-Large Pore-Zeolites, Chem. Eur. J., 2003, 9, 5737–5748. [112] A. Corma, F. Rey, J. Rius, M.J. Sabater, and S. Valencia, Supramolecular Self-assembled Molecules as Organic Directing Agent for Synthesis of Zeolites. Nature (London), 2004, 431, 287–290. [113] B. Jandeleit, D.J. Schaefer, T.S. Powers, H.W. Turner, and W.H. Weinberg, Combinatorial Materials Science and Catalysis. Angew. Chem., Int. Ed., 1999, 38, 2494–2532. [114] J.M. Newsam, T. Bein, J. Klein, W.F. Maier, and W. Stichert, High Throughput Experimentation for the Synthesis of New Crystalline Microporous Solids. Microporous Mesoporous Mater., 2001, 48, 355–365. [115] Y. Song, J. Li, J. Yu, K. Wang, and R. Xu, Towards Rational Synthesis of Microporous Aluminophosphate AlPO4-21 by Hydrothermal Combinatorial Approach, Top. Catal., 2005, 35, 3–8. [116] A. Cantin, A. Corma, M.J. Diaz-Cabanas, J.L. Jorda, and M. Moliner, Rational Design and HT Techniques allow the Synthesis of New IWR Zeolite Polymorphs. J. Am. Chem. Soc., 2006, 128, 4216–4217. [117] O.M. Yaghi, M. O’Keeffe, and M. Kanatzidis, Design of Solids from Molecular Building Blocks: Golden Opportunities for Solid State Chemistry – Preface. J. Solid State Chem., 2000, 152, 1–2. [118] O.M. Yaghi, M. O’Keeffe, and N.W. Ockwing, Reticular Synthesis and the Design of New Materials. Nature (London), 2003, 423, 705–714. [119] C. Livage, N. Guillou, J. Chaigneau, P. Rabit, M. Drillon, and G. Ferey, A Three-dimensional Metal-organic Framework with an Unprecedented Octahedral Building Unit. Angew. Chem., Int. Ed., 2005, 44, 6488–6491. [120] S. Oliver, A. Kuperman, and G.A. Ozin, A New Model for Aluminophosphate Formation: Transformation of a Linear Chain Aluminophosphate to Chain, Layer, and Framework Structures. Angew. Chem., Int. Ed., 1998, 37, 47–62. [121] B. Wei, J. Yu, and Z. Shi, Three-dimensional Open-framework Nickel Aluminophosphate [NiAlP2O8][C2N2H9]: Assembly of One-dimensional AlP2O83
Chains through [NiO5N] Octahedral. Chem. Mater., 2000, 12, 2065–2067. [122] K.X. Wang, J. Yu, Y. Song, and R. Xu, Assembly of One-dimensional AlP2O83
Chains into Three-dimensional MAlP2O8 . C2N2H9 Frameworks through Transition Metal Cations (M ¼ Ni2þ, Co2þ and Fe2þ). Dalton Trans., 2003, 1, 99–103. [123] C.N.R. Rao, Aufbau Principle of Complex Open-framework Structures of Metal Phosphates with Different Dimensionalities, Acc. Chem. Res., 2001, 34, 80–87. [124] A. Corma, M.D. Cabanas, J. Martinez-Triguero, F. Rey, and J. Rius, A large-cavity Zeolite with Wide Pore Windows and Potential as an Oil Refining Catalyst. Nature (London), 2002, 418, 514–517.

466

Chemistry of Zeolites and Related Porous Materials

[125] P. Wagner, M. Yoshikawa, M. Lovallo, K. Tsuji, M. Taspatsis, and M.E. Davis, CIT-5: A High-silica Zeolite with 14-Ring Pores. Chem. Commun., 1997, 2179–2180. [126] A. Corma, M.T. Navarro, F. Rey, J. Rius, and S. Valencia, Pure Polymorph C of Zeolite Beta Synthesized by Using Framework Isomorphous Substitution as a Structure-directing Mechanism. Angew. Chem., Int. Ed., 2001, 40, 2277–2280. [127] G. Sastre, J.A. Vidal-Moya, T. Blasco, J. Rius, J.L. Jorda, M.T. Navarro, F. Rey, and A. Corma, Preferential Location of Ge Atoms in Polymorph C of Beta Zeolite (ITQ-17) and Their Structure-directing Effect: A Computational, XRD, and NMR Spectroscopic Study. Angew. Chem., Int. Ed., 2002, 41, 4722–4726. [128] A. Corma and M.E. Davis, Issues in the Synthesis of Crystalline Molecular Sieves: Towards the Crystallization of Low Framework-Density Structures. Chem. Phys. Chem., 2004, 5, 304–313. [129] A. Burton, Zeolites: Porous Architectures, Nature Mater. (London), 2003, 2, 438–440. [130] T. Blasco, A. Corma, M.J. Diaz-Cabanas, F. Fey, J.A. Vidal-Moya, and C. Zicovich-Wilson, Preferential Location of Ge in the Double Four-Membered Ring Units of ITQ-7 Zeolite, J. Phys. Chem. B., 2002, 106(10) 2634–2642. [131] R. Castaneda, A. Corma, V. Fornes, F. Rey, and J. Rius, Synthesis of a New Zeolite Structure ITQ-24, with Intersecting 10- and 12-Membered Ring Pores. J. Am. Chem. Soc., 2003, 125, 7820–7821. [132] A. Corma, F. Rey, S. Valencia, J.L. Jorda, and J. Rius, A Zeolite with Interconnected 8-, 10and 12-Ring Pores and its Unique Catalytic Selectivity. Nature Mater. (London), 2003, 2, 493–497. [133] J.L. Paillaud, B. Harbuzaru, J. Patarin, and N. Bats, Extra-large-pore Zeolites with Twodimensional Channels formed by 14- and 12-Rings. Science, 2004, 304, 990–992. [134] A. Corma, M.J. Diaz-Cabanas, F. Rey, S. Nicolooulas, and K. Boulahya, ITQ-15: The First Ultralarge Pore Zeolite with a Bi-directional Pore System formed by Intersecting 14- and 12Ring Channels, and its Catalytic Implications. Chem. Commun., 2004, 1356–1357.

8 Synthesis, Structure, and Characterization of Mesoporous Materials The ordered mesoporous materials (or crystalline mesoporous materials) such as MCM41 (MCM stands for Mobil composite of matter), MCM-48 and SBA-15 (SBA stands for University of California, Santa Barbara) are a new generation of materials that are different from nonordered (amorphous) mesoporous materials. They are amorphous and not ordered at the atomic level from a classical crystallographic view point, but their regular channels or pores are ordered at the nanometer level. Because of this, these materials have certain characteristics of crystalline solids. Their structural information can be obtained by diffraction methods and other structural analysis techniques. The discovery of periodic mesoporous structures is a major advance in composite organic– inorganic materials synthesis. Although studies on the subject of ordered mesoporous materials were started about 15 years ago, the unique structure and the properties of these materials attracted many scientists in different fields of research. Their efforts resulted in fruitful results that have been reported in thousands of publications. The flexibility and complexity of their synthesis and structure, and the extensive application potentials of mesoporous materials, create a huge opportunity for researchers and developmental scientists. This chapter will summarize the research results on mesoporous materials from syntheses, structures, formation mechanisms, compositions, morphologies, pore-size control, modifications, applications, challenges, and so on. This chapter reviews the state of study on mesoporous materials based on a literature survey. However, this survey could not cover all aspects because this field is too broad. In the beginning of the reference list for this chapter, we have listed some popular and valuable research papers[1–35] and review articles.[36–51] (according to ISI, over 300 and

Chemistry of Zeolites and Related Porous Material – Synthesis and Structure Ruren Xu, Wenqin Pang, Jihong Yu, Qisheng Huo and Jiesheng Chen # 2007 John Wiley & Sons, (Asia) Pte Ltd

468

Chemistry of Zeolites and Related Porous Materials

200 citations, respectively, the latest update being in November 2006). Please keep in mind that this is not an ISI official learning list and the author does not believe that citation is the best index to evaluate a research paper.

8.1 Introduction According to the IUPAC classification, porous solids can be arranged in three main categories, depending on their pore size: micropore ð< 2 nmÞ, mesopore ð2
50 nmÞ, and macropore ð> 50 nmÞ. Here, the prefix meso- describes a state between the microand the macro-. Mesopore sizes are in the nanometer region; therefore, ‘nanoporous’ is frequently used in its place in the literature. Sometimes the term ‘ultra-micropore’ is used for pore sizes smaller than 0.7 nm. The pore sizes discussed here represent the diameter or the width of the pore, not the radius. In fact, the pore sizes of some mesoporous materials discussed in this chapter are slightly smaller than 2 nm. However, we still call them mesoporous materials even though they are beyond that official IUPAC definition. Based on the diffraction characteristics of solids, the porous materials may be divided into three types: amorphous, sub-crystal, and crystal. The amorphous solid does not give any diffraction peaks. The sub-crystal gives no diffraction peaks or very few broad diffraction peaks. The crystalline solid can produce a set of characteristic diffraction peaks which correspond to its crystallographic system and symmetry. Amorphous and sub-crystalline porous materials have already been used in various chemical industries for many years, such as amorphous silica gel and alumina gel. They lack long-range order (possibly ordered locally). Their channels or pores are irregular; therefore the pore-size distribution is very broad, although the sub-crystalline material includes many small ordered regions. The crystalline porous materials give a very narrow pore-size distribution because the channels or pores are determined by their crystalline or ordered structure. The pore size and pore shape can be controlled by selecting or modifying different structures. Therefore, the crystalline porous materials have many superiorities. The porous materials are extensively applied as catalysts (or catalyst supports) and adsorbents due to their open structure and huge surface area. Zeolite is an excellent member of the crystalline microporous materials and is already used in various applications: catalysis, adsorption, separation, environmental protection, biological technology, functional material, and so on. Regardless of the great amount of work dedicated to zeolites and related crystalline molecular sieve materials, the dimensions and accessibility of pores are restrained to the sub-nanometer scale ð<
1:3 nmÞ. This limits their applications to small molecules, and not larger organic or biological molecules. During the past few decades, one major effort in microporous material research was to make molecular sieves with larger pore sizes. The amorphous (nonordered) mesoporous materials such as ordinary SiO2 aerogel and porous glass possess mesopores, but the channels or pores are irregular and the pore sizes distribute over a wide range. Most macroporous materials such as ceramics and cement have the same characteristics: irregular pores and wide pore-size distribution. The ordered mesoporous and macroporous materials can overcome the above shortcomings. These materials may be organic, inorganic, or organic–inorganic hybrids.

Synthesis, Structure, and Characterization of Mesoporous Materials

469

Common characteristics among them are that the pores or channels are regular and three dimensionally (3-D) ordered over long range. Because of the scope of this book, this chapter will focus only on the ordered mesoporous and macroporous inorganic and inorganic–organic hybrid materials. All synthetic mesoporous and macroporous silicas made by traditional methods, such as aerosol and aerogel, are not ordered (or crystalline) porous materials, because the preparation process is difficult to control. Some zeolites have mesopores after dealumination treatments, but the size and quantity of the pores are also very difficult to control. Certain clay and layered phosphates can be interclated by using large inorganic species (e.g., polymeric cations or silicon-containing compounds) to produce mesoporous material. Although the clay or the phosphate sheet (or layer) is crystalline, the pillars are not ordered. Thus, the mesopores in the products are not unique and are also not ordered, although their pore-size distribution may be narrower than that of other nonordered mesoporous materials. Since scientists in the former Mobil Corporation (Kresge et al.)[2,3] utilized a nanostructural self-assembly technique to synthesize mesoporous SiO2 molecular sieves with uniform channels (e.g., MCM-41) in the early of 1990s, all shortcomings of the nonordered (amorphous) mesoporous materials have been overcome gradually. To date, the ordered mesoporous materials and their synthesis methods have become a ‘hot’ research area in material science fields all over the world. The synthesis of mesoporous materials has already extended into various inorganic compositions with ordered pore systems (unique pore size in
1:5--30 nm range for mesoporous, and >
50 nm for macroporous materials). The synthesis method is generalized into a sol–gel process with supra-molecule (or assembly of molecules), small liquid or solid, or porous solid as template by controlling the interface between organic and inorganic species. The discovery of ordered mesoporous material not only expands the molecular sieves from micropore materials to mesopore materials, but also fills a gap in the porous material family. Some illustrative porous material examples are given in Figure 8.1. The synthesis of ordered mesoporous material was started as early as the 1970s. A research group in Japan also started synthesis work in 1990. Only the report of the M41S

Figure 8.1 Comparison of pore size for typical porous materials

470

Chemistry of Zeolites and Related Porous Materials

Figure 8.2 Structures for M41S mesoporous materials

family (MCM-41, MCM-48, etc.) in 1992 started to attract the attention of many scientists all over the world, and the discovery of MCM-41 is believed to be the starting point of ordered-mesoporous-material research. The members of M41S include MCM-41 (hexagonal phase), MCM-48 (cubic phase), and MCM-50 (lamellar phase). Figure 8.2 shows their structures. The synthesis of the M41S family of mesoporous silica breaks through the traditional principle of zeolite synthesis: use of single molecules or cations as templates. Organized assemblies of molecules (micelles) were used as templates for the synthesis of mesoporous materials. This discovery should be an important milestone in the history of molecular sieve research. This success could be comparable to the great achievement of Mobils’ in the 1970s: the synthesis of ZSM-5. These two successful syntheses have the same characteristic: both gain the unique molecular sieve property by controlling the pore (or channel) size and pore shape. Mesoporous materials with pore sizes in the range 1.5–30 nm overcome the pore-size limitation of zeolites and microporous molecular sieves. The mesopores allow many reactions on ordered porous materials to be possible, such as modification by utilizing larger organic or biological molecules. Mesoporous materials also provide new opportunities for both fundamental research (e.g., gas adsorption modeling, biomolecular catalysis) and practical application (e.g., adsorption, separation, and purification of gas and liquid, catalyst, biological material, semiconductor, optics component, sensors, drugdelivery carrier, material for environmental protection, energy-storage host, and so on). Ordered mesoporous silica has been developed on the basis of traditional zeolites and microporous molecular sieves. The basic synthesis principle is still the formation of porous inorganic solid through the template mechanism. The commonest template for mesoporous silica is a surfactant. In contrast with the small and simple molecular (or cationic) template for zeolites, the typical surfactant is a larger molecule (or cation) which has one hydrophilic headgroup and one hydrophobic tail. The hydrophilic part likes water, and the hydrophobic part does not like water. Surfactant molecules will assemble into micelles to keep the lowest energy in solution. The hydrophobic parts of surfactants define the internal region of micelle (almost all or all do not contact water), while the hydrophilic parts of surfactants define the outer region of the micelle (contact water). Therefore, the micelles look like a huge soluble (hydrophilic) molecule (or cation) which may be the template for the synthesis of porous silica. The initial templates for the M41S family materials are quaternary ammonium surfactants with positive charge and long tail (hydrocarbon chain). The synthesis was believed to proceed via the formation of a lyotropic liquid crystal, which is an ordered assembly of micelles (details will be discussed in the mechanism section). The

Synthesis, Structure, and Characterization of Mesoporous Materials

471

surfactants in the mesopore of silica materials can be removed by calcination at high temperature or extraction with a solvent. The resultant product is the mesoporous silica material. Since the discovery of M41S materials, various synthesis systems and synthesis pathways for ordered mesoporous silica materials have been reported. The core principle is still the same: use of surfactant or other species as templates through a sol–gel process under the guidance of the interface between inorganics–organics. The synthesis media may vary from very acidic to very basic. The synthesis temperatures may be applied in a wide range: lower than room temperature to
150  C (>200  C for extreme hightemperature synthesis). The templates may be cationic, anionic, neutral, multiple charges, multiple alkyl chains, and polymer surfactant. The structures and compositions of materials are also extended, including silica-based materials: FSM-16 (FSM stand for the folded-sheets mesoporous material), SBA-1 and SBA-6 (Pm3n), SBA-2 and SBA-12 (P63/mmc), SBA-11 (Pm3m), SBA-16 (Im3m), SBA-8 (c2mm), FDU-n (FDU represents Fudan University), and disordered (low-ordered) HMS (HMS represents hexagonal mesoporous silica), MSU-n (MSU represents Michigan State University material), KIT-1 (KIT represents Korea Advanced Institute of Science and Technology), and so on. The attractive parts of the ordered mesoporous materials are that they possess some exclusive outstanding properties which other porous materials do not have. These properties include: (1) (2) (3) (4) (5) (6) (7)

well defined pore sizes and shape, narrow pore-size distribution; highly ordered pore (or channel) structure system at the nanometer level; adjustable pore size in the range of
1:3 to
30 nm; various structures, wall compositions, and pore shapes; high thermal and hydrothermal stability if properly prepared or treated; high surface area, high porosity; various controllable regular morphologies at different scale from nanometers to micrometers; (8) existence of micropores in the amorphous wall (for those thicker-wall materials); (9) application potentials, such as large-molecule catalysis, biological process, selective adsorption, functional material. The big advantage of the ordered mesoporous material is that it has a high surface area and a large pore volume. The big disadvantage may be that its wall is amorphous, and is not ordered at the atomic level. This results in its poor performances: low hydrothermal stability and low catalytic acid strength, although there are many methods available to improve those performances. Simultaneously, the amorphous wall makes the modification chemistry on the wall much easier, which allowed the various wall compositions to be synthesized successfully. In contrast, the extension of zeolite framework compositions is difficult because of the limits of their critical crystalline structures at the atomic level. The history of mesoporous material synthesis is unintentionally or intentionally duplicating the development of zeolites and microporous molecular sieve. It starts from silicate and aluminosilicate, through heteroatom substitution, to other oxide compounds and sulfides. It is worth mentioning that many unavailable compositions for zeolite (e.g., certain transition metal oxides, even pure metals and carbon) can be made in mesoporous material form.

472

Chemistry of Zeolites and Related Porous Materials

Table 8.1 Common chemicals (or groups) for mesoporous silica synthesis Name

Abbreviation

Name

Long-chain alkyl trimethylammonium Hexadecyltrimethylammonium Hexadecyltrimethylammonium bromide

CnTMA CTA, C16TMA CTAB, C16TMABr CTAC, C16TMACl Cn-s-n

Tetrapropyl orthosilicate TPOS Tetrabutyl orthosilicate TBOS OCH2CH2 EO OCH2CH2CH2 EO20PO70EO20

PO P123

Cn-s-1

EO26PO39EO26

P85

CnTEA CnEOx

EO17PO59EO17 EO132PO50EO132

P103 P108

TEOS TMOS

EO106PO70EO106 F127 1,3,5-Trimethylbenzene TMB

Hexadecyltrimethylammonium chloride Gemini surfactant CnH2nþ1N(CH3)2(CH2)sN(CH3)2CnH2nþ1 Divalent surfactant CnH2nþ1N(CH3)2(CH2)sN(CH3)3 Long-chain alkyl triethylammonium Nonionic surfactant CnH2nþ1N(OCH2CH2)mOH Tetraethyl orthosilicate Tetramethyl orthosilicate

Abbreviation

We will introduce mesoporous silica-based material syntheses, structures, compositions, properties, and applications in detail in the following sections of this chapter.

8.2 Synthesis Characteristics and Formation Mechanism of Ordered Mesoporous Materials For convenience, the most mentioned chemicals in this chapter are list in Table 8.1 (based on the most-used abbreviation in the literatures). 8.2.1

Mesostructure Assembly System: Interaction Mechanisms between Organics and Inorganics

History and Discovery of Hexagonal Mesoporous Silica Materials According to the open publications and patent literature, the earliest synthesis of the hexagonal mesoporous silica material was reported in US Patent 3,556,725 by Chiola, Ritsko, and Vanderpool.[52] This patent was applied for in February 1969 and was approved in January 1971. The goal of this patent was to prepare low-bulk-density silica that would be one component of luminous powder (coating fluorescent-lamp tube’s). This patent claimed that high-purity silica, having controlled bulk density within the range of about 6 to about 23 pounds/cubic foot, can be prepared by conducting the ammoniacal hydrolysis of tetraalkyl orthosilicates in the presence of a cationic surfactant. TEOS as silica resource and CTAC as surfactant were used in an example. Iler described this patent in the classical book on silica and silicates ‘Silica Chemistry’: ‘Cationic surfactants are adsorbed on the silica surface and promote coagulation. This was used by Chiola, Ritsko, and Vanderpool to recover fine, pure silica from the aqueous hydrolysis of ethyl silicate with ammonia. The product bulk density was about 0.1 g/ cm3’ on page 562.[53] In 1997, Di-Renzo[54] repeated the synthesis shown in the patent and found the product to be a highly ordered hexagonal mesoporous silica material.

Synthesis, Structure, and Characterization of Mesoporous Materials

473

In 1990, Kuroda in Waseda University, Japan reported a 3-D ‘microporous’ SiO2 material.[55] This silica material was called FSM-16[4] or KSW-1 mesoporous material after the report of MCM-41 in 1992. Their initial goal was to synthesize alkyltrimethylammonium–kanemite complexes, a possible pillared clay. Because deformation of the layers in kanemite can be expected on organic intercalation, a specific silica network would form if organic guest substances were selected appropriately. During the ionexchange (typically with alkyltrimethylammonium ions), the silicate layers of the singlelayered polysilicate (kanemite) are condensed to form 3-D silicate networks, which is confirmed by 29Si magic-angle spinning (MAS) NMR (ration of Q3 to Q4) and transmission electron microscopy X-ray diffraction (TEM). The (XRD) patterns of the dodecyltrimethylammonium reaction products show two peaks at 3.7 nm and 3.1 nm. The peak at 3.1 nm was ascribable to dodecyltrimethylammonium-intercalated kanemite, which has no interlayer condensation. The peak at 3.7 nm was possibly attributable to the interlayer condensed product. Two small peaks at 4–6 (2u, Cu-Ka) were not mentioned in the paper by the authors. These two peaks should be the 2nd and 3rd peaks for the hexagonal phase, and the peak at 3.7 nm should be the 1st peak. By calcinations, the silicate–organic complexes were converted into ‘microporous’ materials with uniform pore-size distributions. The specific surface areas of the calcined products are 900 m2/g and the pore size (2–4 nm) can be altered by variation of the alkyl-chain length (from C12 to C18) of the alkyltrimethylammonium ions used. The precise observation of the TEM image suggested that the lamellar lines indicated the formation of uneven layers. The image probably exhibited just the overall arrangement of the condensed silica skeleton at its macromolecular level. The typical hexagonal pore array of TEM image was not shown in the paper. Later, they reported the applicability of trimethylsilylation for the kanemite– organic complexes to control the pore-size distribution more precisely.[56] Soon after MCM-41 was reported, they reported[4] the successful formation of highly ordered mesoporous materials derived from layered polysilicate kanemite by optimizing the reaction conditions of the ion-exchange reaction described previously.[55] The product was highly ordered hexagonal silica and there is no significant difference between the product and the MCM-41 from their TEM, XRD, and 29Si MAS NMR analyses. At the end of 1980s, scientists at Mobil discovered the ordered mesoporous material and applied a series of patents at the beginning of 1990. After their first US patents were approved,[57,58] they published their well known paper ‘Ordered Mesoporous MolecularSieves Synthesized by a Liquid-Crystal Template Mechanism’[3] in Nature in 1992. This short letter showed extremely persuasive powerful experimental evidence: SEM image of near-hexagonal-shaped MCM-41; TEM image showing the MCM-41 structure in a large area; electron diffraction confirmed the hexagonal structures; XRD patterns for MCM-41 and MCM-48 with index results; nitrogen adsorption isotherm with mesoporous characteristics; and the concept of the liquid-crystal template. Soon afterwards, their full paper[2] discussed the detailed synthesis and characterizations. Synthesis System The synthesis mixture for mesoporous materials contains four major components: inorganic precursors, organic template molecules, solvent, and acid or base catalyst. The formation of a material with a desired structure and morphology depends on a delicate interplay between several basic processes, whose relative rates determine the

474

Chemistry of Zeolites and Related Porous Materials

structure and properties of the final structure. These are the self-assembly of the organic template molecules to form organized structures that serve as templates, the sol–gel chemistry that generates the inorganic network, and the specific interaction at the interface between the organic assemblies and the forming inorganic oligomers. There are different models to describe how the inorganic phase interacts with the organic surfactant molecules (see the mechanism discussion section). The typical synthesis of mesoporous material can be divided into two main stages: (i) Formation of the organic–inorganic liquid-crystal phase (mesophases or mesostructure) results from the self-assembly of surfactant molecules and inorganic species which are polymerizable (or condensable) under synthesis conditions. Moreover, this mesostructure has a crystal lattice with the cell length in the nanometer range. (ii) Removal of surfactant from the mesostructure by calcination at high temperatures or other physical or chemical treatments results in the formation of mesopores (the space occupied by surfactant molecules) in the mesostructure. The surfactant may be cationic, or anionic, even nonionic. The type most used for MCM-41 and MCM-48 is cationic quaternary ammonium surfactant. The inorganic resources are those monomers or oligomers that are polymerizable or condensable under certain conditions (concentration, temperature, pressure, pH, solvent, etc.) and form inorganic solid matter such as gel, ceramic, and glass. The depolymerization or hydrolysis is necessary sometimes, for example with TEOS, Ti(OC2H5)4, silicon sol, sodium silicate, and amorphous silica. The synthesis procedure for mesoporous material is simple, and synthesis parameters can be controlled easily. The simple procedure does not mean that the reactions or interactions among reactants in the synthesis system are simple. Many complicated reactions, interactions, and assemblies occur in the mesoporous material synthesis system. The synthesis involves three main components: Inorganic species for the formation of the inorganic wall; template (surfactant in most cases) whose assembly will guide the formation of mesophase; the reaction media (solvent). Figure 8.3 shows the interactions between the three main components.[59] These interactions play the key roles during synthesis. The surfactant molecules in the solution will self-assemble into a micelle or liquid-crystal phase; of course, various factors can affect the assembly process,

Figure 8.3 The relationships between various components in synthesis system. Reproduced with permission from [59]. Copyright (2003) Elsevier

Synthesis, Structure, and Characterization of Mesoporous Materials

475

such as concentration, temperature, chemical additives, and inorganic precursor. The structure of the micelle or liquid crystal will determine the structure of the mesoporous product. The micelle or liquid-crystal phase possesses very strong template effects, although the formation mechanism for most syntheses is not a simple replication of water molecules in liquid-crystal phase by inorganic species. The interaction between inorganic species and solvent (water) is common for most inorganic reactions and solid-material synthesis, which involve dissolution, hydrolysis, and condensation, and so on. A sol–gel process is a typical example. In general, this is a kinetically controlled process. The synthesis parameters including pH, catalyst, organic or inorganic additives, and reaction conditions (temperature, time, and so on) will have effects on the sol–gel process. This process and the inorganic composition will determine the structure and the stability of the final product. The interaction between template and inorganic species is the real key to the formation of the mesostructure. The formation of stable and ordered, structured mesophase requires interactions of suitable type and strength. In the real synthesis system, these interactions affect each other, either causing enhancement or restriction. For example, template assemble will result in an increase in concentration of inorganic species in the organic–inorganic interface. The high concentration of inorganic species will help the condensation of inorganics. In another way, the condensation of inorganics will stabilize the template assembly. Mesostructure Assembly System: Interaction Mechanisms between Organics and Inorganics The interaction between the organic template and the inorganic species (e.g., charge matching) is the key to mesoporous material synthesis. It guides the synthesis process. Any kind of interaction between inorganic and organic species is feasible. Based on this concept, Stucky explored different inorganic–organic combinations and proposed a universal synthesis principle[9] (see below for details). Various new mesoporous materials have been synthesized and new combinations of inorganic–organic species have been found based on this synthesis concept. In order to form a mesophase, it is necessary to adjust the properties (e.g., charge) of the surfactant polar head. Figure 8.4 shows the various types of silica–surfactant interface (synthesis pathways). These interfaces exist in the whole synthesis process from the initial interaction between surfactant and inorganic precursor to the interaction between surfactant and inorganic framework (wall) in the solid product. In aqueous solution, some inorganic cations or anions are condensable under certain conditions (e.g., pH value and concentration). If these species are ordered at the beginning and early stages of their condensation, the product gel or solid will be structurally highly ordered. The synthesis pathways could be classified according to the type of interaction between surfactants and the inorganic precursors (see Table 8.2). Even for the same interaction (synthesis pathway), the strength of the interaction is adjustable. Here, I represents for inorganic species and S represents the surfactant molecule (for example, Sþ for cationic surfactant, S for anionic surfactant, and S0 for nonionic surfactant). In the simplest case, the charges of the S and the I are opposite, under the synthesis conditions (pH). Two main direct synthesis pathways have been identified: Sþ I  and S I þ . The organization (self-assembly) can occur among the same charged surfactant and inorganic species, but the counterion is needed to balance the charge. These two

476

Chemistry of Zeolites and Related Porous Materials

Figure 8.4 Schematic representation of the different types of silica–surfactant interface. Solvent molecules are not shown, except for the I 0S 0 case (triangles); dashed lines correspond to H-bonding interactions. Reproduced with permission from [51]. Copyright (2002) American Chemical Society

pathways, considered to be indirect, also yield mesophases. The Sþ X  I þ path takes place under acidic conditions, in the presence of halogenide anions ðX ¼ Cl ; Br Þ; the S M þ I  route is characteristic of basic media, in the presence of alkaline cations ðMþ ¼ Naþ ; Kþ Þ. The synthesis pathway can be a nonelectrostatic force. The typical example is by use of a nonionic surfactant (e.g., Brij surfactant or long-chain amine) as template (I 0 S0 pathway). The different possible inorganic–organic interfaces are schematized in Figure 8.4. The interface between organic and inorganic species might vary during the synthesis process.[40] For example, the initial interface is Sþ X Iþ such as þ CTMAþ Cl SiOþ and CTMAþ NO for the synthesis of mesoporous silica in 3 SiO extremely acidic media, while the interface changes gradually into near-I X Sþ in the solid product. In basic solution ðpH > 10Þ, the silicate ion is anionic. The simplest method is the use of a cation template (e.g., cationic surfactant) to organize these silicate ions. This interaction (synthesis pathway), Sþ I  , was applied in the classical synthesis of mesoporous silica in basic media. In the acidic route (with pH < 2), both kinetic and thermodynamic controlling factors need to be considered. First, the acid catalysis speeds up the hydrolysis of silicon alkoxides. Second, the silica species in solution are positively charged as  SiOHþ 2 þ (denoted as I ). Finally, the siloxane bond condensation rate is kinetically promoted near the micelle surface. The surfactant ðSþ Þ–silica interaction in Sþ X Iþ is mediated by the counterion X . The micelle–counterion interaction is in thermodynamic equilibrium. Thus the factors involved in determining the total rate of nanostructure formation are the counterion adsorption equilibrium of X on the micellar surface, surface-enhanced concentration of Iþ , and proton-catalysed silica condensation near the micellar surface. From consideration of the surfactant, the surfactants first form micelles as a combination of the Sþ X assemblies, which then form a liquid crystal with molecular silicate species, and finally the mesoporous material is formed through inorganic polymerization and condensation of the silicate species. In the Sþ X Iþ model, the surfactant-to-counteranion

Synthesis, Structure, and Characterization of Mesoporous Materials

477

Table 8.2 Different interactions between the surfactant and the inorganic framework Surfactant: S or N Inorganic precursor: I

Interaction

Examples (structure) (hex ¼ hexagonal and lam ¼ lamellar)

Cationic surfactant Sþ Anionic inorganic I

Sþ I electrostatic force

MCM-41 and FSM-16 (hex), MCM-48 (cubic), SBA-2 (hex-cubic) tungsten oxide (lam, hex), antimony (V) oxide (lam, hex, cubic), tin sulfide (lam), aluminophosphate

Cationic surfactant Sþ Counterion X Cationic inorganic Iþ

Sþ X Iþ electrostatic force

SBA-1 (cubic Pm3n), SBA-2 (hex-cubic), SBA-3 (hex), zirconia (lam, hex), titanium dioxide (hex), zinc phosphate

Cationic surfactant Sþ Counterion F Neutral inorganic I0

Sþ F I0

Silica (hex)

Anionic surfactant S Cationic inorganic Iþ

S Iþ electrostatic force

Mg, Al, Ga, Mn, etc. oxides (lam), alumina (hex), gallium oxide (hex), titanium dioxide (hex), tin oxide (hex)

Anionic surfactant S Counterion Mþ Anionic inorganic I

S Mþ I electrostatic force

AMS-n, zinc oxide(lam), alumina

Nonionic surfactant S0 Neutral inorganic I0

S0I0 hydrogen bond

HMS (near hex)

Nonionic surfactant (amine) N0 Neutral inorganic I0

N0I0 hydrogen bond

MSU-X (near hex), oxides (Ti, Al, Zr, Sn hex)

Nonionic surfactant S0 Counterion X Cationic inorganic Iþ

(S0Hþ )X Iþ

SBA-15 (hex)

Nonionic surfactant N0 Counterion F Cationic inorganic Iþ

N0F Iþ

silica (hex)

Nonionic surfactant S0 Counterion Mþ Neutral inorganic I0

ðS0 Mnþ ÞI0

Metal-containing silica (hex, cubic)

Surfactant S Inorganic I

SI covalent (complex)

silica, Nb, Ta oxides Ta (hex)

ratio is 1:1 and the Hþ concentration in the solution does not change during the synthesis. In the acidic synthesis route, the surfactant/silicate interaction in Sþ X Iþ is weaker; thus, it can usually lead to many topological constructions. The use of the weaker interface interaction in acid synthesis permits a more subtle competition of the kinetics of silica condensation and surfactant/silicate assembly. The advances in block copolymer

478

Chemistry of Zeolites and Related Porous Materials

templating and the formation of mesoporous silica films, gyroids, fibers, and ropes are based on the acid synthesis route. The physical chemistry in the acidic route is quite different from that of the alkaline route. For the I0S0 pathway, the hydrogen bond between template and silica species plays the key role. The electrostatic interaction also exists, but it is not the main interaction, because the nonionic surfactant is partially positively charged (even in neutral pH) and the silica species bear a partially negative charge (when pH >
3). Owing to the weak interaction, the surfactants in the solid product prepared by this synthesis pathway can be removed easily (e.g., extraction with organic solvent at a mild temperature). In general, the solid product from the synthesis pathway with weak interaction between template and inorganic precursor is of low order. As a new example of SI covalent synthesis pathway, siloxane oligomers with alkoxy functionality and covalently attached alkyl chains were used as both silica source and ‘surfactant’ for the synthesis of ordered silica–organic hybrids without the use of an additional template.[60] 8.2.2

Formation Mechanism of Mesostructure: Liquid-crystal Template and Cooperative Self-assembly

Establishing a mechanistic understanding is needed for better control of the synthesis process. A better understanding of the formation mechanism via combined characterization techniques and modeling may lead to a more rational approach for tuning the pore structure of mesoporous materials. In principle, the reaction mechanism can be viewed at three length scales: (i) the molecular one, which concentrates on the interaction between the organic and inorganic precursors and on the silica-polymerization process; (ii) the mesoscopic scale, which involves the development of the micellar structures and the onset of the long-range order; and finally (iii) the macroscale, which is related to the shape/morphology of the final product. It is clear that the changes that occur at the molecular level are the driving force for the mesoscale structure, but the question is how the two scales are correlated. To date, most of the in situ mechanistic studies have concentrated on molecular-level observations using spectroscopic techniques such as electron paramagnetic resonance (EPR), nuclear magnetic resonance (NMR), and infrared (IR). In addition, a number of in situ smallangle X-ray scattering (SAXS) studies targeting the mesoscale have been reported. It is difficult to get clear correlation between them, because most of the studies were carried out on different systems and under different conditions. Since MCM-41 was reported, the formation mechanism of MCM-41 and other M41S members have been an important study topic. Even now, we do not have a complete understanding of formation mechanisms. There have been a number of mechanisms proposed to explain the formation of mesoporous materials and to provide a rational basis for the various synthesis pathways. The main formation mechanisms include: liquidcrystal templating (LCT) by Beck[2], charge-density matching by Monnier,[5] generalized liquid-crystal template mechanism based on the interaction between surfactant and inorganic species by Stucky,[8,9] silicate-rod assembly mechanism by Davis,[61] folding sheets by Inagaki,[4] real liquid-crystal templating by Attard and Antonietti,[13,62] and more. All these mechanisms are derived from the initial LCT mechanism (in two possible

Synthesis, Structure, and Characterization of Mesoporous Materials

479

Figure 8.5 The initial LTC mechanism for the formation of MCM-41: (1) liquid crystal template; (2) co-assembly template. Reproduced with permission from [2]. Copyright (1992) American Chemical Society

ways: liquid-crystal templating or cooperative interaction, see Figure 8.5).[2] The first way assumes that the primary structure-templating element is the lyotropic liquid-crystal mesophase. The second way suggests that the addition of the silicate orders the subsequent silica-encased micelles. The initial LCT mechanism was based on the formation of MCM-41, MCM-48, and MCM-50 in basic synthesis media. Other mechanisms can be considered as the supplement, improvement, consummation, and detailed description of the initial LCT mechanism. Synthesis Characteristics of Mesoporous Material: Supramolecular Assembly as Template The main difference between the synthesis of MCM-41 mesoporous material and traditional synthesis of zeolite or silica molecular sieve is the use of different templates. An individual organic molecule or metal cation is used for the traditional synthesis of silica microporous molecular sieve. For example, the typical template for ZSM-5 synthesis is tetrapropylammonium ion; the crystal is formed through the condensation of silicate species around the template molecule, while for the formation of MCM-41, the typical template is the assembly of large molecules containing one hydrophobic chain with more than 10 carbons. Surfactant, Micelle, and Lyotropic Liquid Crystal In the typical surfactant–water mixture at a given composition and temperature, from a molecular point of view the assembly of the surfactant is determined by a balance between three general types of free-energy contributions. One is associated with the tendency of the alkyl chains to minimize their water contact and maximize their interorganic interactions. The second involves the columbic and dipolar interactions among the charged headgroups and their associated anions. This contribution determines the mean area-per-headgroup, ao , that is available to each surfactant headgroup in an aggregate. In most classical discussions of liquid-crystal aggregates, the counterion of the surfactant is implicitly include in ao . The third type of free-energy contribution includes solvation energies that arise from the presence of water, alcohol, or organic molecules in the hydrophilic, intermediary hydrophobic–hydrophilic ‘palisade’, and hydrophobic regions.

480

Chemistry of Zeolites and Related Porous Materials

Figure 8.6 C12EO8–water lyotropic liquid-crystal phase diagram. Reproduced with permission from [63]. Copyright (1983) Royal Society of Chemistry

C12EO8 is a typical surfactant and can form micelles and lyotropic liquid crystals in water. Its phase diagram[63] based mainly on optical microscope observations is shown in Figure 8.6. The content of water or other solvent molecules changes the self-assembled structures. The structures of the liquid-crystal phases depend on the surfactant concentration. At very low surfactant concentration, the molecules will be dispersed randomly without any ordering: an ideal solution, W. At slightly higher (but still low) concentration [above critical micelle concentration (CMC)], amphiphilic molecules will spontaneously assemble into micelles, L1. These micelles do not order themselves in solution. At higher concentration, the assemblies will become ordered. As the surfactant concentration increases, first a discontinuous cubic phase I1 (a cubic array of spherical micelles) forms, then 2-D hexagonal columnar phase H1, and bicontinuous cubic phase V1, form. At still higher concentration, a lamellar phase, La, may form, wherein extended sheets of surfactants are separated by thin layers of water. S is the undissolved surfactant solid. The subscript 1 means the normal micelle (the headgroup forward outward, water-in-oil). The reversed micelle (oil-in-water) is described with the subscript 2. Other surfactants show the similar lyotropic liquid-crystal phase behavior and follow the same succession of phases, but not all of the phases are always present. Figure 8.7 shows a phase diagram for the CTAB–water binary system. CMC can also be classified CMC1 (spherical micelle) and CMC2 (rod-shaped micelle). The relatively small group of phase symmetries (space group) is observed for typical amphiphilic, hydrophilic, and polymer-based surfactant system. In region I1, more liquidcrystal-phases are possible for certain systems; they include cubic (Pm3n, Fm3m, or Im3m) and hexagonal (P63/mmc). Fd3m is common in region I2. For simple surfactants (e.g., quaternary ammonium, neutral CmEOn), I1 means the Pm3n mesophase for charged surfactants in most cases, while for block copolymer surfactant, I1 is Im3m in

Synthesis, Structure, and Characterization of Mesoporous Materials

481

Figure 8.7 The lyotropic liquid crystal phase diagram for CTAB–water system

most cases. 2-D hexagonal mesophase is the most common phase, exists in most surfactant system, and occupies a large region in the phase diagram. In general, V1 is cubic (Ia3d), but co-exists with other cubics such as Pn3m and Im3m in rare cases. Between 2-D hexagonal (H1) and lamellar mesophases, several intermediate phases can be formed; for example, R3m, c2mm, p2g, etc. Their structures can be considered as modifications of hexagonal or lamellar mesophase. In addition to the highly ordered mesophases, lower-ordered mesophase exists in certain systems under certain conditions. For example, an array of worm-like micelles was found in mixed C12EO6 and C16EO6 systems. The formation of lyotropic liquid-crystal mesophase depends on the structure and properties of surfactant, solvent, and reaction conditions. Although studies on lyotropical liquid crystals have been carried out for many years, the structure and properties of some mesophases are still not very clear. Since lyotropic liquid crystals rely on a subtle balance of intermolecular interactions, it is difficult to analyse their structures and properties, the boundary in the phase diagram may be not accurate and the minor phase may be missed. However, the formation of additional, new lyotropic mesophases is also possible. Even within the same phases, their self-assembled structures are tunable by concentration: for example, in lamellar phases, the inter-layer distances increase with the solvent volume. For lyotropic mesophases of surfactants and lipids, it is very difficult to obtain clear conclusions about the exact nature of structures since the quality of the X-ray diffraction data is inadequate, and high-quality electron microscopy data are difficult to obtain. In reality, the pore or cage shapes, wall thicknesses, and geometries of mesophases are highly tunable with almost an infinite variability. Inorganic mesophases with better longrange ordering quality and excellent stability are helpful in characterizing both new liquid crystal-like mesostructures and structural modulations.

482

Chemistry of Zeolites and Related Porous Materials

It does not seem surprising that mesostructured silicas display topologies closely related to those observed for the liquid-crystal-like phases typical of surfactant–water systems. The structural evolution observed in the case of the silica mesophases is clearly reminiscent of that characteristic of the liquid-crystal phases as a function of the surfactant concentration. In short, such a characteristic evolution with the surfactant concentration can be expressed as follows: cubic (discontinuous phases), hexagonal, cubic (bicontinuous phases), random bicontinuous (sponge-like), and lamellar. In addition, other phases, intermediate between the hexagonal and lamellar phases, are sometimes detected. Hence, accurate knowledge of the mesophase diagram of a given pure surfactant–water system may be a priori a useful guide to the synthesis of mesoporous solids, but such phase diagrams can only be considered as rough guides. Original Liquid-crystal Templating (LCT) Mechanism In order to explain the formation of MCM-41, Beck[2,3] suggested the ‘liquid-crystal templating’ (LCT) mechanism based on the similarity between the mesostructure of lyotropic liquid crystals from surfactant assemblies and the structures of M41S materials. This mechanism includes two main general pathways, in which either (i) the liquidcrystal phase pre-exists before the silicate species are added or (ii) the addition of the silicate anions promotes the long-range ordering of the surfactant to form the hexagonal arrangement. In either case, the inorganic precursors that bear a negative charge under the synthesis conditions preferentially interacted with the positively charged ammonium headgroups of the surfactants and condensed into a continuous solid framework. Later, based on experimental evidence gained through 2H and 29Si NMR spectroscopy, as well as neutron scattering, Firouzi[15] showed that a micellar solution of CTAB transformed into a hexagonal lyotropic phase in the presence of silicate anions [pathway (ii)]. In the literature, pathway (ii) has been referred to as the cooperative self-assembly mechanism. As the accumulation of mesoporous-silica synthesis knowledge increases, researchers found the LCT mechanism to be too simple to explain some new experimental phenomena. For example, the small-angle neutron-scattering experiment indicated the long-range ordered organic–inorganic assembly exists at room temperature and at low concentration of CTA (for example 2% CTAB). Such low concentration of CTAB give only spherical micelles without any liquid-crystal mesophase. Much higher CTAB concentrations are needed for the formation of liquid-crystal mesophase (e.g., above 28 wt% for hexagonal phase, and higher than
60 wt% for bicontinuous cubic phase). In fact, M41S materials can be obtained at much lower concentration (e.g.,
2 wt% for MCM-41 and < 10 wt% for MCM-48 in the original synthesis by Mobil). Moreover, silicate species cannot polymerize and form solid products in the absence of surfactant under the typical synthesis conditions (pH, temperature, and time). Another extreme example is the cage-structured SBA-2[64] synthesis using Gemini surfactant Cn-s-m. SBA-2 contains mainly a 3-D hexagonal mesostructure (P63/mmc) domain, while the P63/mmc mesophase has not been observed in quaternary ammonium surfactant systems to date. The above examples suggested that no preformed LC phase was necessary for the formation of MCM-41 or other silica mesoporous materials. Pathway (1) of the LCT mechanism did not take place (this pathway was shown to be possible under different synthesis conditions – see the discussions for real LCT mechanism). The second mechanistic pathway of LCT was vaguely postulated as a cooperative self-assembly of

Synthesis, Structure, and Characterization of Mesoporous Materials

483

the ammonium surfactant and the silicate species at a low concentration of surfactant. To date, the actual details of MCM-41 or other mesoporous-silica formation have not yet been fully agreed upon. Several mechanistic models have been advanced which share basic ideas about the detailed process of the silicate species-promoted LC-phase formation and a common description of different synthesis processes. Cooperative Formation Mechanism The second mechanistic pathway in the original Mobil mechanism involves the liquidcrystal mesophase as the template for the formation of MCM-41, but that the liquid-crystal mesophase is formed after the addition of inorganic precursor. The formation of organic– inorganic mesostructure is a cooperative self-assembly of the ammonium surfactant and the silicate species. The cooperative self-assembly can be observed in two main aspects: (1) the acceleration of inorganic-species polymerization by surfactant (or micelle), and (2) the enhanced assembly of surfactant (or micelle) by inorganic species and its polymerization. The interactions between organic–inorganic substrates (e.g., static electric attraction, hydrogen bonding) concentrate the inorganic species in the interface and accelerate the polymerization. This mechanism is helpful in our understanding of certain experimental results such as the synthesis at low surfactant concentration,[65] discovery of new mesophases,[64] and phase transformation during synthesis and treatment.[1] However, there are different ways to describe the details of the cooperative selfassembly of the surfactant and the silicate species. The two mechanisms proposed by Davis[61,66] and Stucky[8] are most noteworthy. Davis and coworkers found that the hexagonal mesophase did not develop during MCM-41 synthesis, based on in situ 14N NMR spectroscopy. They proposed a silicate rod assembly mechanism (see Figure 8.8). Under the basic synthesis conditions, the formation of MCM-41 began with the deposition of two to three monolayers of silicate precursor onto isolated surfactant micellar rods driving by the Coulombic force between the micelle surface and silicate species. The silicate-encapsulated rods were randomly ordered, eventually packing into a hexagonal mesostructure (the low-energy structure). Heating and aging then completed the condensation of the silicates into the assynthesized MCM-41 mesostructure. This possible model suggested for mesostructured-materials synthesis using surfactants was that of coating preassembled organic surfactant arrays with the inorganic phase and then assembling these coated organic arrays into a 3-D periodic structure. These are features of this model that make it attractive, giving a direct explanation for the analogous symmetries of the silicate structures to those of liquid-crystal chemistry, and it is consistent with the paradigm for biomimetic synthesis: First create an organized organic array, and then condense an inorganic phase on the preorganized organic surface.

Figure 8.8 Davis’s MCM-41 formation mechanism. Reproduced with permission from [66]. Copyright (1993) Elsevier

484

Chemistry of Zeolites and Related Porous Materials

Image Not Available In The Electronic Edition

Stucky and coworkers[8,15,65] carried out systematic syntheses under different conditions, as well as NMR studies. Based on previous work, they proposed an extensive mechanism to describe the detailed cooperative self-assembly (Silicatropic Liquid Crystals) process (see Figure 8.9). In its simplest kinetic form, at low temperatures the synthesis is a three-stage process in which the first step can be viewed as the net ionic exchange of a polycharged anionic species. Single-tail surfactant molecules react preferentially with silicate polyanions (e.g. dimers, double three- and four-rings) which displace the original surfactants’ counteranions to form a ‘silicatropic liquid crystal’ phase. This was consistent with the effect of electrolytes on micellar-phase transformations. Micelles serve as a surfactant molecule source or are rearranged according to the anion charge density and shape requirements. The charge density of the inorganic species determines how many surfactant molecules are associated with a given inorganic molecular unit as well as the preferred orientation of the surfactant headgroup relative to the molecular inorganic species. This ‘charge-density matching’ determines the average intermolecular spacing between surfactant headgroups and, within that constraint and the energetics of the intermolecular interactions of the inorganic species, the molecular ion-pairs adapt the preferred liquid-crystal-array morphology. These ionpairs can then organize into a new liquid-crystal-like array. Nucleation and rapid precipitation of organized arrays takes place with configurations determined by the cooperative interactions of ion-pair charges, geometries, and organic van der Waals’ forces. The silicate counterions were reactive, but the silicate condensation at this stage at low temperatures is minimal. Under synthesis conditions that prevented condensation of the silicate species, such as low temperatures and high pH, a true cooperative selfassembly of the silicates and surfactants was found to be possible. The silicate phase condenses further with increasing time and temperature. The final step is condensation of

Synthesis, Structure, and Characterization of Mesoporous Materials

485

the inorganic phase. With time and/or temperature silicate inorganic units condense and, not surprisingly, one can observe phase transformation. The silicate-framework charge decreases during this process and may lead to liquid-crystal-like phase transformations as the surfactant phase tries to reorganize the changing interface charge density. Of course, Davis’s mechanism is correct for certain cases under certain conditions, but it could not be generalized and cannot explain some experimental facts. For example, the channel of typical MCM-41 is much longer than the length of rod micelles. The spherical micelle exists with a rod micelle under synthesis conditions. The spherical and rod micelles should have very similar surfaces. Why did the spherical micelle not involve the templating process in the case of synthesis of MCM-41, MCM-48, and MCM-50? MCM41 can be synthesized by using a very low concentration of surfactant (below CMC2). The formation of MCM-48 and MCM-50 cannot be explained by this mechanism. Stucky’s mechanism can explain more experimental results in a wide range. After improvements,[15] it guides the synthesis of mesoporous inorganic materials. In fact, the initial basis of Stucky’s mechanism is the charge-density-matching model based on the fact of phase transformation from lamellar phase to hexagonal phase. Charge-density Matching The charge-density matching mechanistic model is the basis of Stucky’s cooperative formation mechanism. This charge-density matching model was proposed by Monnier[5] in 1993 and suggested that MCM-41 could be derived from a lamellar phase. The initial solid phase of the synthesis mixture was a layered mesostructure (based on the XRD result), and was formed from the electrostatic attraction between the anionic silicates and the cationic surfactant headgroups (see Figure 8.10). This mechanism indicated that the three closely coupled phenomena are identified as being crucial to the formation of surfactant–silicate mesophase. These include: (1) multidentate binding of silicate oligomers; (2) preferred polymerization of silicates at the surfactant–silicate interface; and (3) charge-density matching across the interface. Figure 8.10 shows a mechanism proposed for the transformation of a surfactant– silicate system from the lamellar to the hexagonal mesophase. On the left, small silica oligomers act as mutildentate ligands, which have sufficiently high charge density to

Image Not Available In The Electronic Edition

486

Chemistry of Zeolites and Related Porous Materials

permit a lamellar surfactant configuration. As the silicate species began to condense, the charge density was reduced. Accompanying this process, curvature was introduced into the layers to maintain the charge-density balance with the surfactant headgroups, which drive the transformation of lamellar mesostructure into the hexagonal mesostructure. The core of this mechanism is that the interaction of organic and inorganic materials determines the product structure. The cooperative formation of the inorganic–organic interface can be expected to create new organic array configurations. Mesostructure syntheses can be carried out under conditions in which the silicate alone would not condense to solid (at pH 12
14 and low silicate concentration) and the surfactant CTAB (concentration < 2%) alone would not form a lyotropic liquid-crystal phase. The rapid formation of MCM-41 when surfactant solution and silicate solution are combined indicates that there is strong interaction between the cationic surfactant and anionic silicate species in the formation of mesophases. Mesophase formation and associated silica polymerization are intimately tied to Coulombic interactions between surfactant and silicate species at the micelle interfaces. The oligomeric silica polyanions can easily act as multidentate ligands for the cationic headgroups of the surfactants, leading to a strongly interacting surfactant–silicate interface. This was confirmed by a later NMR study.[65] Preferential multidentate binding of the silicate polyanions causes the interface to quickly become populated by tightly held silicate oligomers, which can subsequently polymerize further. Silicate polymerization within the interface region is favorable for a main reason: the concentration of silicate species near the interface is high. Furthermore, as polymerization proceeds, the formation of highly connected silicate polyanions further enhances the cooperative interaction between the surfactant and silicate species. This process leads to precipitation of a given mesophase from solution. Now, we know that the phase transformation from lamellar phase to hexagonal phase is not the sole pathway to form MCM-41. In other words, not all MCM-41 species are formed from the lamellar precursors. Folding-sheets Mechanism Inagaki[4] proposed a mechanism (see Figure 8.11) to explain the formation of FSM-16 using kanemite as starting precursor material. Kanemite (NaHSi2O5.3H2O) is a type of hydrated sodium silicate composed of single-layered silica sheets. After the surfactants were ion-exchanged into the layered structure, the silicate layers of kanemite can wind

Figure 8.11 Folding of silicate sheets around intercalated surfactant molecules. (a) Ionexchange, (b) calcination. (a) Reproduced with permission from [4]. Copyright (1993) Royal Society of Chemistry and (b) Reproduced from [67]. Copyright (1998) Elsevier

Synthesis, Structure, and Characterization of Mesoporous Materials

487

around the exchanged alkyltrimethylammonium ions. This causes the condensation of silanol groups on the adjacent silicate layers of kanemite, since the silicate layers have the flexibility to wind due to its single layered structure. The final product was claimed to be very similar to MCM-41, with no resemblance to the original kanemite structure. However, the formation mechanism of the mesostructured precursors for FSM-16 has not been confirmed explicitly. It has been proposed that 2-D hexagonal mesostructured precursors are formed by the transformation of layered intermediates composed of CnTMA ions and fragmented silicate sheets. It is difficult to say if FSM-16 and MCM-41 are the same material.[67] Generalized Liquid-crystal-templating Mechanism A generalized mechanism of formation based on the specific type of electrostatic interaction between a given inorganic precursor I and surfactant headgroup S was proposed by Stucky.[9] The basic synthetic route involves the direct co-condensation of anionic inorganic species with a cationic surfactant ðSþ I Þ; the MCM-41 and MCM-48 are prototypic examples. By operating below the isoelectric point of silica (pH
2) under extreme acidic conditions, the silica species become cationic ðIþ Þ; the silicate mesostructure is formed through the Sþ X Iþ pathway. In the energetic self-organization it is thought that the packing of the organic surfactant and the charge-density matching between the surfactant and the inorganic precursor are essential for the formation of the ordered mesostructure. The surfactant packing depends on the molecular geometry of the surfactant species, such as the number of carbon atoms in the hydrophobic chain and the size or charge of the polar headgroup. In addition, the formation of mesostructures was affected by the solution conditions, including the surfactant concentration, pH, the presence of co-surfactant, and its concentration and temperature. The generalized liquid-crystal-template mechanism focuses on the interaction between surfactant molecules and inorganic species. Surfactant molecules assemble with inorganic species to form the liquid-crystal-like mesophase. The three main interactions between surfactant and inorganic species are: (i) electrostatic interaction (the chargedensity match plays the key role), (2) hydrogen bond, in particularly for those neutral templates, and (iii) covalent bond. This generalization is useful, especially when other types of inorganic–organic interactions are considered. This mechanism is also suitable for the formation of nonsilicon mesoporous materials. The success of the cooperative templating model was illustrated by the diverse compositions of organic–inorganic mesostructures found to be possible. True Liquid-crystal Templating Mechanism (LCT) Attard and Antonietti[13,62] showed that monolithic mesoporous silicates could be prepared using a true liquid crystal (very high surfactant concentration) as template. The inorganic precursors replace some of water and polymerize in situ around the micelle (liquid crystal) structure and provide a negative replica of liquid crystal mesophase. This method is expensive due to the high concentrations of surfactant required, and the synthesis solutions are highly viscous and difficult to manipulate. The fact that the final structure is usually the same as the initial liquid-crystal phase does, however, give this method the benefit of some predictability. This synthetic approach could be considered a true LCT route, which supports the viability of pathway 1 of the originally proposed LCT mechanism for MCM-41 (see Figure 8.5). In this method of preparation, the

488

Chemistry of Zeolites and Related Porous Materials

organic–inorganic interaction was less important than the actual presence of a liquidcrystal phase. The true liquid-crystal-templating approach is more successful for the syntheses with the nonionic surfactants such Brij-n[68] and copolymer[69] as template. The nonionic surfactant C16EO8 has been successfully used to produce three liquid-crystalline phases: hexagonal (H1), cubic (Ia3d), and lamellar silicas. The true liquid-crystal-templating approach shows promise in fabricating silica monoliths of the desired size and shape and in significantly enhancing the control over surfactant-phase domains. El-Safty[70] used microemulsion lyotropic liquid-crystal phases of Brij 56 as templates to synthesize mesoporous silica monoliths (HOM-n) with large-caged cubic structures, uniform pore size up to 8 nm, and walls of 10 nm thickness. Other Possible Mechanisms Beside mechanisms mentioned above, other possible mechanisms have been proposed for the explanation of certain experimental details. For example, the silicates were organized into layers, with rows of the cylindrical rods intercalated between the layers. Aging the mixture caused the layers to pucker and collapse around the rods, which then transformed into MCM-41 mesostructure.[43,44] MCM-41 intermediate structures were found in the form of clusters of rod-like micelles ‘wrapped’ by a coating of silicate through lowtemperature TEM and small-angle X-ray scattering.[43,44] The clusters of elongated micelles were found before precipitation occurred. As the reaction progressed, the silicate species diffused to and deposited onto the individual surfaces of the micelles within the cluster; the clusters of elongated micelles eventually became clusters of silicate-covered micelles. Thus, the clusters of micelles served as nucleation sites for MCM-41 formation. New Progress in Mechanistic Study The recent mechanism studies focus on the formation of mesophase with block copolymer surfactant as template. Synthesis conditions such as low temperature, low acidity, and low ionic strength that increase the induction time give rise to the morphologies of mesoporous silica with increased curvatures. The particle-growth process of rodlike SBA-15 materials from solutions has been examined by directly observing the morphologies of particles as a function of time. A colloidal phase-separation mechanism for the formation of mesoporous materials[71] suggested that the formation process of mesoporous materials involves three stages: (1) cooperative self-assembly of inorganic/organic composites; (2) formation of a new crystal-like phase rich in aggregates of block copolymer/silica species; and (3) phase separation of this liquid-crystal-like phase from the solution and further growth of solid mesostructures driven by further condensation of silica species. The morphologies of mesoporous materials are developed after the phase-separation stage and are influenced by the competition mainly between the free energy of mesostructure self-assembly and the colloidal surface free energy, as well as other interactions. The evolution of the solution microstructures during the formation of the hexagonal mesoporous material SBA-15 was studied by direct imaging and freeze-fracture replication cryo-TEM.[72] A continuous transformation from spherical micelles, into threadlike micelles, which become longer and stiffer, followed by the formation of bundles with dimensions similar to those found in the final material, was observed. The direct imaging

Synthesis, Structure, and Characterization of Mesoporous Materials

489

cryo-TEM and FFR results show that the transition to the hexagonal order takes place after precipitation. The newer inorganic-driven phase-separation mechanism[71,73] is also gaining experimental support. In this mechanism, interacting inorganic–surfactant species (that can be but are not necessarily coated micelles) aggregate into larger liquid-like particles where microphase separation of the inorganic species and surfactant under high-concentration conditions results in formation of the final mesostructure. 8.2.3

Surfactant Effective Packing Parameter: g and Physical Chemistry of Assembly and Interface Considerations

Effective Packing Parameter: g The main controlling forces of micelle structures involve hydrophobic affinity on the hydrocarbon–water interface and hydrophilic repulsion, ionic repulsion, and steric repulsion of charged headgroups. The surfactant must pack to fill space and, thus, maximize favorable van der Waals’ interactions between the hydrophobic tails while avoiding high-energy repulsive interactions between the charged or polar headgroups. A key question for a researcher in mesoporous materials is how to relate the continuum surface in the mesostructure model to surfactant molecular structure. Fortunately several investigators have shown that it is possible to mathematically relate molecular size, charge, and shape to the more global surface curvature, bending energies, and morphology. The classical and contemporary molecular description of surfactant organization in liquid-crystal arrays has been described in terms of the local effective surfactant packing parameter,[74] g ¼ V=ao l, where V is the total volume of surfactant chains plus any co-solvent organic molecules between the chains, ao is the effective headgroup area at the micelle surface per hydrophilic headgroup, and is related to both the size and the charge on the surfactant headgroup and is affected by the electrostatic environment around the surfactant headgroup, and l is the kinetic surfactant tail length. The interface surface-bending energy can be written in terms of g, the actual surfactant-packing parameter adopted by the aggregating chains in the phase.[75] The counterion in this classical model is not explicitly included. It is not immediately clear that this relatively simple molecular model can be used as a first approximation to explain and predict product structure and phase transformations for the inorganic mesostructures. Stucky determined that the molecular-packing-parameter model used in liquid-crystal chemistry is useful in designing inorganic-organic composite mesostructures. In classical micelle chemistry, as the g value increases above critical values, mesophase transformations occur. The expected mesophase sequence as a function of the packing parameter is shown in Table 8.3. These transitions reflect a decrease in surface curvature from cubic (Pm3n) through lamellar. For surfactants to associate in a spherical structure, the surface area occupied by the surfactants’ polar headgroup should be large. If the headgroups are permitted to pack tightly, on the other hand, the aggregation number will increase, and rod or lamellar packing will be favored. The values of g (between ˆ
¯ and ˆ
˜) for p6mm (2-D hex) and (between ˆ
˜ and ˜
¯) for Ia3d (cubic) depend upon the volume fraction of surfactant chains.[75] An examination of a large number of surfactants, coupled with a study of the effects of co-solvents, has confirmed that to a first and relatively good approximation

490

Chemistry of Zeolites and Related Porous Materials

Table 8.3 Relationship of g and mesostructures Packing parameter g ¼ lcVa0

Micellar shape

g < 13

Surfactant

Mesophase example

spherical

single tail and large headgroup

g ¼ 13
12

Cylindrical

g ¼ 12
23

3-D cylindrical

g¼1

layer

g>1

(reversed micelle)

single tail and small headgroup single tail and small headgroup double tail and small headgroup double tail and small headgroup

SBA-1 (Pm3n) and SBA-2 (P63/mmc) MCM-41 (p6mm) MCM-48 (Ia3d) MCM-50 (lamellar)

the molecular-packing-parameter model can be used in a predictive way to generate structures analogous to those found in conventional liquid-crystal chemistry (see Table 8.3). In the real synthesis systems, the surfactant effective packing parameter, g, are mainly affected by the following factors: (1) charge, composition, molecular shape, and structure of surfactant, (2) the interactions between surfactant and inorganic species (e.g., chargedensity matching), (3) reaction parameters and conditions: concentration, pH, ion strength, temperature, etc. Block copolymer surfactants cannot be simply described with surfactant effective packing parameter, but the ratio of hydrophilic part/hydrophobic part or the fractions of hydrophilic part and hydrophobic part may be used to give a similar description. For example, the x parameter phase space of block copolymers[76] as a function of solvent and temperature can be used to help define the packing parameter and hydrophobic and hydrophilic volume fractions of the block copolymer. The increase in the number of EO units in the EOx-POy-EOx triblock copolymer corresponds to an increase in the curvature of the surfactant layer toward water. While the hydrophilic group tends to dissolve in water, the hydrophobic moieties gather together to form a hydrophobic core. The shapes of the mesophase thus obtained may be determined by the balance of the attractive and the repulsive forces acting at the hydrophobic interfaces of the aggregates. Similar to mesophases found in surfactants and amphiphilic block copolymers, the formation of mesoporous silica materials with different mesostructures can be explained by the hydrophilic–hydrophobic balance of the structure-directing agents, as suggested for cationic surfactants[14] and for amphiphilic block copolymers such as the triblock copolymer (EOx-POy-EOx). The PEO in the polymer acts as the hydrophilic headgroup. The EO/PO ratio, i.e., hydrophilic–hydrophobic balance, has a large effect on the formation of the mesostructured silica. By varying the PEO fraction (x from 50 to 61) and keeping a similar PPO fraction of copolymer surfactants (EOx-POy-EOx), different mesostructures were obtained.[77] As the number of EO units increases, the headgroup area, a0, increases and consequently the packing parameter is diminished. The number of EO units is essential in determining which silica mesophase is obtained. Lamellar structure is obtained with short EO chains (4 units); hexagonal structure SBA-15, with

Synthesis, Structure, and Characterization of Mesoporous Materials

491

Figure 8.12 The synthesis results with nonionic surfactants. Reproduced with permission from [78]. Copyright (2002) American Chemical Society

medium-length EO chains (17–37 units); and cubic structure SBA-16, with long EO chains (132 units). When blends of nonionic amphiphilic block copolymers were used as templates, the mesostructures are transformed from lamellar to 2-D hexagonal, 3-D hexagonal, and cubic symmetries, as the size of hydrophilic EO headgroup increases. Some successful synthesis examples[78] are shown in Figure 8.12. Physical Chemistry of Mesostructure Assembly and Interface Considerations The formation of surfactant–silica mesostructure can be considered as a result of the assembly of surfactant and silica species. This assembly involves many physical and chemical processes, such as the assembly of surfactant molecules, and hydrolysis and polymerization of inorganic species. The balance of all processes and all interactions between surfactant and inorganic species determines the final mesostructure. The synthesis involves the organization of hydrophobic and hydrophilic entities into a homogeneous biphase composite. Three types of nearest-neighbor assembly are evident: inorganic–organic (IO), organic–organic (OO), and inorganic–inorganic (II). We therefore consider the free energy and kinetics of formation of the mesostructure in terms of the relative contributions of (i) formation of the interface between the inorganic/organic phases, (ii) organization of the organic array, (iii) interactions between inorganic units, including polymerization and hydrogen bonding, and (iv) the chemical potential of the surrounding solution phase during synthesis. In this context the free-energy terms[5] can be summarized as Equation (8.1), where A is headgroup area and P is a variable, defined below. G ¼ Ginter ðA; PÞ þ Gorg ðA; PÞ þ Ginorg ðA; PÞ þ Gsol ðPÞ ð8:1Þ We now discuss the different synthesis systems. For a typical ionic surfactant, A is the effective area occupied by the ionic surfactant headgroup in the organic phase. The optimal headgroup area, Ao, is obtained by minimizing the free-energy change. P is a generic variable representing the state of phase by specifying the organization and charge distribution of the various species within it. 1/A is a measure of the average organic surface charge density. Experimentally

492

Chemistry of Zeolites and Related Porous Materials

it is observed and readily modeled that, for a given surfactant, a curved organic array surface has a substantially lower charge density at equilibrium than does a flat surface ˚ 2, 46.7 A ˚ 2, and 23.3 A ˚ 2 for spherical, rod-like, and lamellar C12TMAB, (Amin ¼ 70 A respectively, at room temperature). The critical role of the effective headgroup area in determining the energetics of surface curvature and the type of biphase surfactant structure formed is extensively documented and can be expected to be as important in matching inorganic molecular species such as silicate anions as it is with simpler inorganic ions such as Cl or OH . A is therefore an obviously important parameter when electrostatic ion-pairing between surfactant and counterion is energetically dominating. The interface-wall charge and thickness are defined by P. For pure aqueous surfactant phases the interface-wall thickness is also a function of the cation–cation repulsion due to the cationic charge densities and the cation–anion/ solvent interactions. Gorg(A, P) results from the van der Waals forces and conformational energy of the hydrocarbon chains and the van der Waals and electrostatic interactions of the headgroups within the organic subphase. It is related, but different, to the free energy of micellization for ionic surfactants defined by Evans et al.[79] Ginorg(A, P) reflects the contributions arising from the inorganic subphase. If the inorganic phase consists of unpolymerized inorganic molecular species, Ginorg describes the energetics of the charge and inorganic intermolecular interactions. This is also a measure of the polymerized inorganic phase (e.g., polysilicate) structural free energy, including the solvent, van der Waals, and electrostatic interactions within the inorganic framework (or wall) if they are present. Ginter(A, P) accounts for the van der Waals and electrostatic effects associated with inorganic–organic array interactions. Gsol(P) represents the contribution of the mother liquor solution. This contribution sets the chemical potential of the various species during nucleation and within the precipitate. From the viewpoint of interactions among reactants (organic, O, and inorganic, I), the main interaction could determine both synthesis progress and direction. Because assembly of the composite is determined by the relative strengths of the thermodynamic driving forces and the relative rates of the kinetic processes, stabilized organic arrays do not have be used if the OI interaction is weak. An example is the Sþ X Iþ that is obtained by combining cationic surfactants with cationic silica species at acidic media (pH value below the aqueous silica isoelectric point).[9] (1) For OO > OI, II, the assembly of surfactant is stable (no change) during synthesis. This is a real liquid-crystal mechanism. The final product is a replica of the surfactant assembly. This situation means that an organized organic array controls the assembly and also defines the ultimate configuration of the composite phase. In fact, this is the basis for what has been described as a central tenet of biomineralization that states that nucleation, growth, and the final morphology of biominerals are determined by the existence of a preorganized assembly of organic molecules. Biomimetic approaches and modeling of biomineralization have relied on this paradigm for experimental design and have accordingly focused on the use of known stable organic arrays or stabilization. Nevertheless, using preorganized organic assembly to control morphology and nucleation is a potentially powerful approach to mesoporous materials synthesis, particularly

Synthesis, Structure, and Characterization of Mesoporous Materials

493

in terms of macroscale shaping, as a template that can be created with the required acid– base or molecular structure characteristics, and as a liquid support phase for bulk processing or synthesis. The concept of using an organized organic array as a template is a statement that the most important free energy or kinetic contribution to biphase composite formation is the organization of the organic array. Inorganic deposition and subsequent polymerization do not significantly perturb that array morphology. Several possible ways to approach that goal are (1) to strengthen the organic intra-array coupling by cross-linking; (2) to stabilize the organic array by interfacing it to an inorganic substrate; and (3) to decrease the organic–inorganic (OI) interface interactions relative to the organic–organic (OO) interactions. (2) For OI > OO, the introduction of inorganic species will affect the assembly of surfactant and the interactions between surfactant and inorganic species are more important. (2a) For OO > II, this is a cooperative mechanism. This situation suggests that one can expect the original organic array, if present, to be disrupted upon interaction with the inorganic species; and then to reorganize to a new configuration determined by kinetic and thermodynamic assembly parameters associated with the newly formed molecular or oligomeric inorganic–organic unit. The formation constant for OI is large in this case with strong OI interaction due to interactions such as multidentate binding and chelation, or the forces between highly charged inorganic–organic species. The same result is obtained by creating the organic–inorganic molecular or oligomeric unit by using covalent bonding between the desired organic and inorganic moieties. This situation has been experimentally confirmed by in situ studies of the basic ðpH > 9Þ synthesis of mesoporous silicate phase using colloidal silica and cetyltrimethylammonium cations as the template; and by the use of covalently bound organosilicon or transition metal organometallic precursor in mesoporous-phase synthesis. The cooperative templating perspective for the mechanism of formation of the MCM-41 and MCM-48 silicate phases is that the electrostatic interface interaction between soluble anionic inorganic molecular species and cationic single-tailed surfactant molecules is responsible for initiating the composite synthesis and that the preorganized organic array is not necessary, or if present not necessarily related to the morphology of the final composite product. This point of view argues that at the very least the kinetics of low-temperature syntheses of periodic mesocomposite are better described by kinter ðA; PÞ > korg ðA; PÞ > kinorg ðA; PÞ, where kinter ; korg , and, kinorg are, respectively, the relative rates for cation–anion formation, the organic array assembly as it exists in the final mesostructure configuration, and silica polymerization. This model also states that, for surfactant–polycharged inorganic ion combinations, Ginter ðA; PÞ < Gorg ðA; PÞ. All free-energy terms are important so that the particular approach to biomimetics and nanocomposite-structure synthesis can be varied by changing organic and inorganic interactive forces, temperature, and concentration. For example, van der Waals forces and organic-array stability are greater for bitailed surfactants than for monotailed surfactants so that Gorg ðA; PÞ becomes a more important contribution. (2b) for II > OO, in these microporous ð<1:5 nmÞ phases, organic–inorganic (OI) interactions dominate during synthesis while what is generally the kinetically slower inorganic condensation (II) progresses so that each template is encapsulated within a cage or pore opening with only secondary van der Waals’ organic–organic (OO)

494

Chemistry of Zeolites and Related Porous Materials

interactions found in the final product (OI > OO and II > OO). Charge matching of the available charge/unit area of the organic species with that of the encapsulating inorganic species is required. Water molecules of salvation play a key role in this assembly process, both in terms of solubilizing the template and as an entropic thermodynamic driving force provided upon release of the water molecules as inorganic–organic assembly takes place. Taking into account these various factors, rigid, bulky, and relatively short (< 1 nm for the longest axis) molecules with moderate hydrophobicity are the best templates for high-silica microporous molecular-sieve synthesis. The biomimetic approaches start with an organized organic array and carry out inorganic growth on its surface, and so the most important free energy or kinetic contribution to biphase composite formation is the organization of the organic array. The free-energy changes associated with the interface between the inorganic and organic phases, and the organization/condensation of the inorganic phase, are regarded as perturbations on the organized array, i.e. Gorg ðA; PÞ < Ginter ðA; PÞ < Ginorg ðA; PÞ. This is the description of the mechanism proposed by Davis for the formation of the MCM-41. (2c) If OI > II, solvation and co-solvent effects are for the moment implicitly included in the OO, OI, and II assembly processes, an interesting variation in materials synthesis takes place if OO, II are both much less kinetically or thermodynamically favorable than OI (OI  OO, II). This permits the very clever sequential growth process in which monolayers of organic, then inorganic, etc., composition can be deposited to give lamellar phases that can be structurally designed over more than 100 repeats. This spatial control of composite ordering makes it possible to alter, for example, every nth repeating unit or to create compositionally laddered arrays. It is important to emphasize that the situations described above are different synthesis strategies that expectedly lead to composite and porous materials that have distinctly different properties. In the preparation of mesoporous materials, procedural variables define a very complex system in which kinetic parameters (time, basic operations sequence) may play a determinant role.

8.3 Mesoporous Silica: Structure and Synthesis Silica materials have been studied extensively because of the structural flexibility of silica (through SiO4 tetrahedral connections), easy control of hydrolysis and polymerization of silica species, high thermal stability of silica framework, easy modification of the silica surface, and well known silica and zeolite chemistry. Amorphous silica is also the main inorganic component for certain natural materials obtained from bioassembly, such as diatoms. Various mesoporous silica materials have been reported, which are very important for both fundamental research and applications. 8.3.1

Structural Characteristics and Characterization Techniques for Mesoporous Silica

In general, there are two main methods to control the mesostructures: (1) change various reactant concentrations, reaction temperature, and reaction-mixture composition;

Synthesis, Structure, and Characterization of Mesoporous Materials

495

(2) adjust the surfactant effective-packing parameter, g,[1] by using different structured surfactants, adding additives, or changing synthesis conditions. Until now, many well ordered mesoporous materials have been successfully synthesized and it has been claimed that these have different mesostructures. Most structures have been proposed on the basis of a combination of powder XRD studies and an accumulated knowledge of liquid-crystalline phases, whilst only a few have been precisely determined by electron crystallography. The typical mesostructures include the M41S family, SBA family, FDU family, etc. Table 8.4 lists various mesostructures and material examples. The cooperative assembly of surfactant–inorganic precursor composites is driven by weak Coulombic or van der Waals’ forces, or hydrogen bonding; therefore, the high symmetries, e.g., close-packing mesostructures of cubic or hexagonal symmetries, can be easily formed. If the driving force between silicates and surfactant assembly is stronger, the low-symmetry mesostructure could be formed. According to the structure and composition of materials and analysis requirements of the researcher, the following analysis techniques can be selected for the characterization of mesoporous materials: XRD, TEM, adsorption–desorption (N2 or other gas), solid MAS NMR (29Si, 27Al, 13C, etc.), scanning electron microscopy (SEM), catalysis test, Fourier Transform infra-red (FT-IR), thermal analysis, UV-visible, and chemical analysis. IR, X-ray photoelectron spectroscopy (XPS), X-ray absorption near-edge structure XANES, extended X-ray absorption fine structure EXAFS and other spectral methods are commonly used to analyse metal elements such as Ti in the mesoporous material frameworks. It is particularly important to note that the symmetries determined by XRD or even electron diffraction are best viewed as ‘average’ symmetries, and a careful analysis of the diffraction intensities, high-resolution TEM, solid-state NMR, and adsorption–desorption data is required in order to characterize the details of the cage, wall, and pore structures. Because of the limited amount of scattering data available for materials with high void densities and large unit cells, diffraction modeling must be done on the continuumsurface basis and not at the discrete atom level used for microporous crystal-structure determination. The continuum-surface model has been particularly useful in understanding mesostructures and mesophase transformations of materials in which surfactants play a dominating role in determining the overall structural symmetry, and it applies equally well to inorganic mesostructured composites. Because of the large unit cell, the XRD diffraction peaks appear at very low angles of 2u. Some problems for structural analysis of mesoporous materials by powder XRD are the lack of enough diffraction peaks, overlapping of the peaks, and loss of structurefactor phase information. Since most samples have the low ‘crystalline’ quality, the calibration of equipment is very important. The normal standards for calibrating a diffractometer (e.g., Si, quartz) are not very efficient, so standards for low-angle calibration should be used (e.g., lead stearate). The defects (faults) and intergrowth can be often found in zeolite or microporous molecular sieve crystals. Mesoporous material synthesis is controlled kinetically and there is no defect-free mesoporous material. Various defects can be found easily in mesoporous materials, even for so-called high-quality materials.[106] The XRD patterns appearing in this chapter were collected by using Cu-Ka, unless specified otherwise.

Table 8.4 Structure characteristics of various mesoporous materials Pore system

Crystal system

(Low-ordered, channel)

(near hexagonal)

Symmetry

Lamellar (no channel) Channel

hexagonal

hexagonal cubic

p6mm (17) p6m (old name) c2mm (9) cmm (old name) P63/mmc (194) Pm3n (223) Im3m (229) Fd3m (227)[85] Fm3m(225)[87]

3-D Channels

MSU-n[80] HMS KIT-1[81]

long-chain primary amine, nonionic poly(ethylene oxide) surfactant, CTA long-chain alkyl trimethylammonium (CTA), long-chain alkyl trimethylammonium (CTA), block copolymer (P123) Bola surfactant (quaternary ammonium)

MCM-41 SBA-3 SBA-15 FSM-16 TMS-1 SBA-8[82] KSW-2 SBA-2[64] SBA-1[9,83] SBA-6[84] HOM-C9 SBA-16[27] HOMC-1 FDU-2[85] AMS-8[86]

cubic-hexagonal intergrowth orthorhombic

Pm3m (221) Ia3d (230) Fm3m (225), P63/mmc (194) Pmmm (47)

KIT-5[88] FDU-12[89] HOM-C10 SBA-11[27] HOM-C5[90] SBA-2, SBA-7, SBA-12[89]FDU-1 FDU-13[91]

tetragonal

Fmmm (69) P4/mmm (123)

silica membrane FDU-11[91]

rhombohedral cubic

P42/mnm (136) R3m (166) Im3m (229)

AMS-9[93] silica thin film[94] ‘Plumber’s Nightmare’[95,96] MCM-48[2,3,97] FDU-5 HOM-5[98] KIT-6[99]

Ia3d (230)

2-D Channels

Typical template used

MCM-50

tetragonal 3-D Cage

Typical materials

tetragonal trigonal

Pn3m (224)[103,104] I41/a (88) R3m (166)

HOM-7[103] AMS-10[105] CMK-1 HUM-1

C16-3-1 Long-chain alkyl triethylammonium block copolymer (F127) tri-head group quaternary ammonium, anionic surfactant F127 C16-3-1 C16EO10 F68 C16-3-1, Brij76 tetra-headgroup rigid bolaform quaternary ammonium C3-12-12-3 PE6800[92] tetra-headgroup rigid bolaform quaternary ammonium C3-12-12-3 N-lauroylglutamic acid P123 poly(isoprene-b-ethylene oxide) (PI-b-PEO) CTA, Gemini quaternary ammonium, P123, Brij76[100], EO(17)MA(23) diblock copolymers;[101] P103 with sodium iodide,[102] P123-BuOH[99] Brij56, anionic surfactant

Synthesis, Structure, and Characterization of Mesoporous Materials

497

In most cases, TEM is an indispensable characterization tool for the investigation of nanostructures and can be used to analyse mesostructures. TEM enables researchers to probe the structure of materials at the nanometer scale, providing information on the shapes, sizes, crystallinity, and organization of the observed objects. Conventionally, a 2-D image is obtained from an inherently 3-D sample. However, an unambiguous interpretation of the 2-D images may be difficult as the structural features of the 3-D sample are projected and may overlap in the 2-D image. In order to obtain correct TEM images, preparation of the sample is critical, in particular for noncubic samples. Su[107] found the regular striped patterns (160–200 nm), which were believed to indicate an ordered macrolamellar structure, is actually an artifact arising from the ultramicrotoming of the sample embedding in epoxy resin. The gas-adsorption technique is a very convenient and effective method[108] used to analyse the pore system of mesoporous materials. Commonly used gases include nitrogen and argon. The characteristic mechanism of adsorption on mesoporous material is capillary condensation, while the adsorption on microporous material is controlled by the strong interaction between the adsorbate molecule and the inorganic wall. The adsorption–desorption isotherm for typical mesoporous material including M41S and SBA family belongs to type-IV adsorption isotherms, according to the classification of IUPAC. The isotherm can be type I, when the pore size is small (e.g., < 3 nm). As the pressure is increased, adsorption in mesopores proceeds via monolayer–multilayer adsorption on the pore walls, followed by capillary condensation; that is, filling of the pore core with condensed gas. The subsequent pressure decrease results in desorption via capillary evaporation, which is the emptying of the pore core, followed by the desorption from the multilayer on the pore walls. Gas adsorption is a prominent method for determining the specific surface area, pore volume, and pore-size distribution of mesoporous materials. It is also possible in favorable cases to extract information about the pore connectivity. It is important to couple adsorption isotherm data with electron crystallography for the complete structural solution of mesoporous structures. Such methods when applied together offer accurate and reliable data that can be further used for the design and application of these complex structures. 29 Si MAS NMR is a powerful technique for analysing the local environment of silicon in the mesoporous wall, and indicates the polymerization level of silica. 29Si MAS NMR spectra for MCM-41 are similar with those for amorphous silica. There are two main resonance peaks: 100 ppm for Q3, i.e., Si(OSi)3OH, and 110 ppm for Q4, i.e., Si(OSi)4. In some cases, a peak at 90 ppm for Q2 can be observed. According to the different silicon species distribution, one may calculate the quantity of silanol. 27 Al MAS NMR can distinguish Al atoms with different coordination: the chemical shift for AlO4 is at about 53 ppm, and that for AlO6 is about 0 ppm. In some cases, the impurities (amorphous alumina, dense phase of alumina, or aluminosilicate) add their contribution to the NMR spectra. The analysis should combine with other techniques. 8.3.2

2-D Hexagonal Structure: MCM-41, SBA-15, and SBA-3

The 2-D hexagonal materials with honeycomb arrays of nonintersecting primary channels, its symmetry (space group) is 2-D hexagonal p6mm. Typical materials include:

498

Chemistry of Zeolites and Related Porous Materials

the well known MCM-41 and SBA-15, SBA-3 made from acidic media, and FSM-16 made from kanemite. MCM-41 The earliest MCM-41 synthesis is that which involved addition of surfactant C16H33(CH3)3OH/Cl solution to a solution of sodium silicate, and the resulting mixture was heated at 100  C for 6 d. Of course, an Al resource was added to the reaction mixture when the synthesis target was Al-MCM-41 (Al-containing MCM-41). The pore size for MCM-41 is adjustable in the range of 1.5–10 nm and
4.0 nm is for the classical MCM-41 made by using CTAC as template. The channel may go through a whole ‘crystal’ (or particle). To date, the structure for MCM-41 pore system (a 2-D array of channels) is clear, but the shape of the channel is not clear. Is it either hexagonal or cylindrical? Current analysis results cannot answer this question yet. The most probable is somewhere between the two configurations. In fact, an experiment indicated the crosssection of hexagonal liquid-crystal phase change from hexagonal to cylindrical as the surfactant concentration increased.[109] The synthesis of MCM-41 was extended into very wide reaction conditions and various reactants. The silicon resource may be either organic silicon compounds (e.g., TEOS, TMOS, TBOS) or inorganic compounds (such as amorphous silica, soluble silicate). Synthesis temperature can be from lower than room temperature to high temperature (
150  C). Reaction time may vary from several minutes to a few weeks. The synthesis media can be from very basic to near neutral. The long-chain quaternary ammonium (CnTMA) surfactant is the best template. For the high-quality MCM-41 sample, XRD can give more than four hk0 diffraction peaks. Figure 8.13 shows the XRD pattern of high-quality MCM-41 made from an extremely low surfactant-concentration system.[110]. These peaks, except the first one, are very weak. The position of diffraction peaks can be indexed with 2-D hexagonal p6mm symmetry. Figure 8.14 shows high-resolution TEM images[111] of a calcined MCM-41 sample. Figure 8.14(a) shows a uniform hexagonal arrangement of bright dots corresponding to the straight channels of MCM-41. The diffraction pattern in the inset of Figure 8.14(a) clearly shows that the incident beam is along the [001] direction. The brightness and the shape of the channels seem to differ slightly from place to place. This is a result of slight changes in diffraction conditions, which show a domain character. Note that all the TEM images are projected along the incident-beam direction; therefore, the value of pore diameter from Figure 8.14(a) is the projection of the pores along the [001] direction and may be somewhat less than the actual value. Nevertheless, in principle, the pore diameter which is calculated from Figure 8.14(b) may give more reliable value than that obtained from other methods, on the assumption that the channels are cylindrical. The TEM method requires appropriate methods of data analysis to avoid artifacts that can lead to an incorrect result. Gas adsorption at low temperature (e.g., in liquid nitrogen) is a prominent method for determining the specific surface area, pore volume, and pore-size distribution for mesoporous materials. The typical N2-adsorption isotherm for MCM-41 belongs to a Type IV isotherm. A high-quality MCM-41 sample has a narrow pore-size distribution, high surface area (> 1000 m2/g), and large pore volume (> 0:7 cm3/g). Shown in Figure 8.15 are argon- and nitrogen-adsorption isotherms[112] acquired at 77 K for MCM-41 silicas

Synthesis, Structure, and Characterization of Mesoporous Materials

499

Figure 8.13 XRD pattern for high-quality MCM-41. Reproduced with permission from [110]. Copyright (1999) Elsevier

Figure 8.14 TEM images of calcined mesoporous MCM-41 along the channel direction (c axis; a) and perpendicular to it (b). Reproduced with permission from [111]. Copyright (2000) Wiley-VCH

500

Chemistry of Zeolites and Related Porous Materials

Figure 8.15 Argon- and nitrogen-adsorption–desorption isotherms measured at 77 K for a series of MCM-41 silicas with pores of diameter 2.4–6.5 nm. Reproduced with permission from [112]. Copyright (2003) American Chemical Society

with pore diameters from 2.4 to 6.5 nm. In both cases of adsorbate, the same pattern of adsorption–desorption behavior is observed. Multilayer adsorption that takes place at lower pressures is followed by the capillary condensation at pressures gradually and systematically increasing with the pore diameter. Adsorption–desorption isotherms for smaller pores are reversible, but adsorption–desorption hysteresis is observed for larger pore diameters.[1] It is notable that, in the case of argon, hysteresis disappears for pores of diameter between 3.2 and 3.5 nm, which is appreciably smaller than the lowest diameter for which hysteresis is observed for nitrogen (somewhere between 3.9 and 4.2 nm; the hysteresis loop for 4.2 nm MCM-41 is too narrow to be seen clearly in Figure 8.15). Al-MCM-41 Like aluminosilicate zeolite, the acid sites of Al-MCM-41, which come from tetrahedral Al in the inorganic wall, are active sites for most catalysis reactions. Many efforts have been made to introduce tetrahedral Al into the silicate wall of MCM-41. The Al resources for the synthesis of Al-MCM-41 can be sodium aluminate, aluminum sulfate, or other Alcontaining compounds. The Si/Al of MCM-41 can be lowered to
1. The introduction of Al would decrease the long-range order of MCM-41 structure, and lower and broaden the XRD peaks. Sometimes, the broad peaks result from the particle size of Al-MCM-41 (nanoparticles). SBA-15 One great breakthrough in mesoporous material synthesis was the discovery of SBA15.[113] The typical synthesis of SBA-15 is carried out by using poly(ethylene oxide)– poly(propylene oxide)–poly(ethylene oxide) (PEO–PPO–PEO) triblock-copolymer as

Synthesis, Structure, and Characterization of Mesoporous Materials

501

template in an acidic synthesis system. The formation of SBA-15 is believed to be a ðS0 Hþ ÞðX Iþ Þ process. SBA-15 may exhibit a large variety of morphologies depending on the synthesis conditions. The template can be removed by calcination at high temperature (e.g., 500  C) or extraction with a solvent (e.g., EtOH). Pore size can be from
4 to 30 nm by varying the synthesis composition and conditions (e.g., addition of swelling agent TMB). The silica wall (
2 to 6 nm) is much thicker than that of MCM-41. Owing to the large pore size ð> 4 nmÞ of SBA-15, the nitrogen adsorption–desorption isotherm for SBA-15 sample shows clearly an H1 hysteresis loop.[1,114] SBA-15 has a high thermal stability (> 900  C) and hydrothermal stability (to hot water). With regard to pore structure of SBA-15, it was initially thought of as an array of hexagonally packed cylindrical channels similar to MCM-41. However, it was soon realized that the typical SBA-15 contains a significant number of micropores in its framework (the wall for the mesopore). The micropore volume is about 0.1 cm3/g. This microporosity is the result of silica templating by PEO fingers forming a corona around each micelle. In other words, these micropores result from the insertation of hydrophilic PEO of the surfactant into the silica wall. The templating polymer within both mesopores and micropores can be removed stepwise.[115] Ryoo confirmed strongly the existence of micropores by the successful synthesis of mesoporous carbon by using SBA-15 as template[34] This difference between MCM-41 and SBA-15 can be observed clearly in their adsorption results (as-plot).[116,117] Figure 8.16 shows as-plots of MCM-41 (C22TEA as template) and SBA-15. The adsorption on MCM-41 usually exhibits excellent linearity in the low-pressure range (see Figure 8.16 left), which indicates the absence of micropores. The initial parts of the as-plots for the calcined SBA-15 samples were significantly nonlinear, which indicates the presence of micropores, whose volume was estimated as about 0.12 and 0.06 cm3 g1 , respectively, for samples A and B (see Figure 8.16 right). The initial parts of the as-plots did not exhibit any steps, which in this case is indicative of broad pore-size distributions in the micropore/small-mesopore range.

Figure 8.16 as-Plot for MCM-41 (left, Reproduced with permission from [118]. Copyright (2000) American Chemical Society) and SBA-15 (right, Reproduced with permission from [117]. Copyright (2000) American Chemical Society)

502

Chemistry of Zeolites and Related Porous Materials

The mesopore of SBA-15 is adjustable; the micropores in the wall are also adjustable by varying the synthesis conditions or post-synthesis treatment.[27,113,119] When the synthesis temperature is low (30–40  C), the pore walls of the material are too thick ð
4 nmÞ for the microporous channels to actually connect adjacent cylindrical mesopores. As the synthesis temperature increases to 80–100  C, the precursor mesophase is gradually modified by a decrease in the strength of the interaction between the surfactant and its environment and by a densification of silica inside the walls. The mesopore size increases, the wall thickness decreases, and the collapse of the ultramicroporosity brings about the formation of secondary porosity. The mesoporous channels become connected by bridges. At even higher temperature, i.e., 130  C, the micropores tend to vanish, mesopores of more than 9 nm become separated by 2 nm-thick silica walls which present no ultramicroporosity, whereas the secondary mesopores (1.5–5 nm) bridging the main channels still occur. These changes may be explained by the effect of temperature on PEO hydrophobicity. The post-synthesis hydrothermal treatment can attain similar results (decrease micropores) (see Figure 8.17). In addition to the synthesis temperature, some other means to control the occurrence of micropores were reported. In particular, it was found that inorganic salts such as KCl, NaF, and NaCl have the ability to inhibit the formation of micropores during the synthesis of SBA-15.[120] The information on SBA-15 pore structure can be obtained indirectly from the synthesis of mesoporous carbon by using SBA-15 as template. When SBA-15 silicas prepared at above 80  C are used as template, ordered mesoporous carbon rods (CMK-3) and pipes (CMK-5) with 2-D hexagonal symmetry were synthesized successfully. The

Figure 8.17 The effect of hydrothermal treatment on SBA-15 pore system. Reproduced with permission from [59]. Copyright (2003) Elsevier

Synthesis, Structure, and Characterization of Mesoporous Materials

503

carbon particles were stabilized by carbon bridges formed within micropores and/or secondary mesopores connecting the primary mesoporous channels of the SBA-15 templates. Similarly, SBA-15 was used as template to generate a platinum replica, containing bridges between parallel rods, which were observed directly by TEM.[27,113,119] Sayari[121] used a combination of designed synthesis conditions for SBA-15 to gain new insights into the genesis of micropores and secondary mesopores bridging the primary mesoporous channels. A modified mesoporous SBA-15 (3-D SBA-15) with interconnecting 3-D large-pore networks has been purposely synthesized by high-temperature (130  C) hydrothermal treatment, even introducing TMB into embryo mesostructures.[122] The 3-D SBA-15 has the average mesostructure of hexagonal plane group symmetry p6mm. The four diffraction peaks can be clearly observed and assigned to 100, 110, 200, and 210 reflections similar to those for conventional SBA-15. The 3-D SBA-15 has many nanosized (2–8 nm) connections/tunnels, which are randomly distributed between the 1-D channels. The hydrothermal stability of SBA-15 materials with different structural parameters, such as pore size, pore volume, and wall thickness has been tested.[123] The materials with thicker walls and more micropores show relatively better hydrothermal stability. To the SBA-15 materials with more micropores, the recombination of Si O Si bonds during high-temperature steam treatment may not cause direct destruction to the wall structure. As a result, SBA-15 materials with more micropores show better stability in pure steam at 600  C. Thermal treatment at 900  C can enhance the polymerization  Si   bonds and effectively improve the hydrothermal stability of degree of   O Si  SBA-15. However, this approach will cause very serious shrinkage of the mesopores, resulting in smaller pore diameter and low surface area. A carbon-propping thermal treating method was employed to enhance the polymerization of  Si O Si  bonds and minimize the serious shrinkage of mesopores at the same time. In general, the extremely acidic medium for SBA-15 synthesis does not favor the introduction of Al into the ‘framework’ or wall of SBA-15, since Al species are very soluble under there synthesis condition. Post-synthesis treatment may introduce Al into SBA-15. The synthesis of zeolite seeds as precursor can introduce Al into SBA-15, and result in the highly acidic material. A simple and effective ‘pH-adjusting’ method has been used to graft a large number of heteroatoms such as Al and Ti to mesoporous silica material SBA-15.[124] The heteroatom source is added to the initial reaction mixture in strongly acidic media ðpH < 0Þ, just as in the case of direct synthesis; when the mesostructure is basically formed, the pH value of the system is adjusted from a strong acid ðpH < 0Þ to neutral (pH 7.5), followed by hydrothermal treatment for another period of time, during which a large number of heteroatoms can be introduced into the mesophase. The products prepared by this method show highly ordered mesostructures with large surface areas and uniform mesopore size distribution. SBA-3 SBA-3 is the acidic version of MCM-41. A typical synthesis of SBA-3 was carried out in extreme acidic media (1–7 M acid, 2 M HCl is preferred). A typical example is that of Sþ X Iþ synthesis pathway.[8,9] The CnTMA cationic surfactants are the best template

504

Chemistry of Zeolites and Related Porous Materials

for SBA-3. Owing to the strongly acidic synthesis media, a long-chain amine [e.g., C16H33N(CH3)2] will bear a positive charge and become a good template for SBA-3. The hexagonal silica mesophase can be formed at room temperature within a few minutes. The sample with short synthesis time has poor stability, even a wash with water (before being completely dry) would destroy its structure. Longer reaction time and high synthesis temperature will improve the stability of SBA-3. To avoid destroying the SBA-3 mesostructure, the wash step (after filtration) may be omitted because the acid will escape from the solid during drying, and extra surfactant (outside the particles) will be removed with the template (inside the pore) during calcination or solvent extraction. This mediated ðSþ X Iþ Þ-synthesis model is supported by the following evidence: (i) the cationic silica species which are present at pH < 2; (ii) the near 1:1 surfactant-tochlorine ratio in the hexagonal and lamellar products; (iii) the easy removal of the surfactant with ethanol; (iv) the observation that TEOS and SiCl4 hydrolyse and form mesophase products, while Cab-O-Sil, which does not readily hydrolyse in acidic conditions, forms no mesophase products; and (v) the anion dependence of the synthesis, e.g., the different products obtained with Cl , Br, and oxyanions. It should be emphasized that the acid- and base-synthesized silica mesophases have little in common other than sometimes having the same space-group symmetry. They do not have the same composition since mesophase samples synthesized below the silica isoelectric point require a counteranion, generally a halide anion, for each surfactant molecule that is present. Terminal Si O groups are protonated so that the bulk compositions of M41S (e.g., MCM-41) and acid-prepared materials (e.g., SBA-3) made with the same surfactants are completely different in hydrogen and halide-ion content. The ultimate periodic symmetry is determined in both cases by the nanophase surfactant-packing requirements, so that similar space group and lattice symmetries may be observed by XRD and TEM. However, the XRD peaks of the two phases for a given surfactant have clearly different diffraction intensities, indicating different pore and wall structures. SBA-3 (see Figure 8.18) and other mesoporous silica from acidic synthesis systems have regular ‘crystal’ morphology, even curved shapes.

Figure 8.18 SEM of SBA-3 (scale bar 10 mm). Reproduced with permission from [9]. Copyright (1994) Nature Publishing

Synthesis, Structure, and Characterization of Mesoporous Materials

505

Since in the acid mesophase product the surfactant cationic charge is exactly balanced by a chloride anion, the ion-pair surfactants are removable by extraction without providing an exchangeable cation as required for the anionic framework of MCM-41. It was found that most of the surfactant could be removed from the as-synthesized material by stirring it in pure ethanol at room temperature or by overnight reflux. Removal of the surfactant was also possible by calcination. Calcined SBA-3 has similar stability and absorption–desorption properties as has MCM-41. 8.3.3

Cubic Channel Mesostructures: MCM-48, FDU-5, and Im3m Materials

This family of mesostructures has the 3-D interconnected channels. To date, there are a few of members in the structure family: Ia3d, Im3m and Pn3m. The corresponding lyotropic liquid-crystal phase is V1 and its cubic neighbors. The V1 region is much smaller than that for the H1 (2-D hexagonal) phase in phase diagrams of most surfactant– water binary systems. The similar reasons result in the relative difficulty of synthesis of materials with these mesostructures. From a view of material applications, a 3-D channel is more favorable for mass transfer than is a 1-D channel. The materials (e.g., MCM-48) with 3-D channels seem to be more interesting candidates as adsorbents, separation media, or catalyst supports than do MCM-41. MCM-48 The structure of MCM-48 is based on a bicontinuous cubic surfactant phase with symmetry Ia3d shown in Figure 8.2. MCM-48 structure may be represented by an enantiometic pair of three-dimensional channel systems (Q230), which are wrapped by the silica wall corresponding to the continuous gyroid minimal surface, periodic Gsurface, i.e., Equation (8.2) holds. ðsin px  cos pyÞ þ ðsin pz  cos pxÞ þ ðsin py  cos pzÞ ¼ 0

ð8:2Þ

MCM-48 has two independent chiral channel systems with opposite handedness (see black and white channels in Figure 8.2). The structure has been inferred from modeling studies and fitting of powder X-ray profiles and high-resolution electron microscopy (HREM) images. The initial structure model proposed by Monnier[5] based on a lyotropic liquid-crystal model gave a good agreement with XRD and TEM experimental results. Figure 8.19 shows the TEM image for MCM-48. Figure 8.20 shows the high-resolution XRD pattern of MCM-48 (synchrotron X-ray resource). All diffraction peaks can be indexed with Ia3d, and all peaks (in the low-angle region) are visible, none is missing. The adsorption experimental results at different temperatures[125] indicate that MCM-48 possesses a uniquely sized channel. As one member of the M41S mesoporous silica materials, MCM-48 is the target for many research projects. MCM-41 can be made easily under different synthesis conditions. The synthesis of MCM-48 is believed to be difficult. Low-quality MCM-48 samples were obtained in most cases, even for repeated application of well known recipes. To date, there are many synthesis recipes for MCM-48 available. The basic concept for these syntheses is to control the effective surfactant-packing parameter g between the limits 1/2 and 2/3; in other words, to increase the palisade-layer volume of micelles. The following are some successful synthesis strategies.

506

Chemistry of Zeolites and Related Porous Materials

Image Not Available In The Electronic Edition

(A) Use CnTMA as template with necessary organic additives. Polar organic additives are important in the formation of MCM-48 when CnTMAþ is used as template. For normal surfactant–water systems, it is widely believed that the role of the alcohol is to prevent the growth of the aggregates into infinite rods (hexagonal)

Figure 8.20 XRD pattern of MCM-48 (synchrotron X-ray resource, wavelength 0.17 nm). Reproduced with permission from [1]. Copyright (1996) American Chemical Society

Synthesis, Structure, and Characterization of Mesoporous Materials

(B)

(C)

(D) (E)

(F)

507

or bilayers. Polar organic additives [such as EtOH, (CH3)2NCH2CH2OH, N(CH2CH2OH)3] are solubilized by the micelle and are located in the palisade layer (the region between the headgroup and the hydrophobic core) because of their polarity and hydrogen-bonding to water. The addition of EtOH will favor the formation of MCM-48.[1] When TEOS is used as silica source, the EtOH coming from the hydrolysis of TEOS is considered as the organic additive. The presence of polar organic additives [such as (CH3)2NCH2CH2OH, N(CH2CH2OH)3] favor the formation of MCM-48; even inorganic silica sources (e.g., Cab-o-sil or sodium silicate) are used. Use special templates.[1] Gemini surfactant Cm-12-m favors the formation of MCM48, even without organic additives using Cab-o-sil (SiO2) or sodium silicate solution as the silica source. Its micellar structure is similar to the C16TMAB–polar additive– water mixture in certain respects. The spacer (C12H24) is long enough to penetrate the hydrophobic core of the micelle, but the spacer remains in the outer portion of the micelle since it is bonded to the headgroups. Long-chain Gemini surfactants favor the MCM-48 mesophase, even at extremely low ratios of surfactant/silica and low surfactant concentration. For example, C22-12-22 (only 0.4 wt % in water) gave an MCM-48 sample at room temperature after aging for 1 d. Another good template for MCM-48 is cetyldimethylbenzylammonium (CDBA) which favors the formation of MCM-48 but not MCM-41. CDBA could be considered as a C16TMAþ surfactant with a solubilized benzene molecule anchored on the trimethylammonium headgroup. The benzene group, like the spacer C12H24 of Cm-12-m, can penetrate to a certain degree the hydrophobic core of the micelle, thus decreasing the headgroup size. Use a mixture of cationic and anionic surfactants. Li synthesized MCM-48 using a mixture of CTAB and C12H25COONa as template.[126] The pair of cationic and anionic surfactants has higher hydrophobicity compared with a pure cationic surfactant. The pair prefer to penetrate to the core of the micelle to a certain degree, increase the packing parameter g. Use a mixture of cationic and nonionic surfactants. High-quality MCM-48 can be obtained by using a mixture of CTAB and nonionic surfactants.[127] Control synthesis conditions (pH, temperature, time, pressure, etc.). The pH value of solution can affect the charge density of silica species, and affect the charge matching between surfactant–inorganic interface.[15] Higher synthesis temperatures preferentially lead to the formation of MCM-48.[128] For a synthesis system of Cab-O-Sil fumed silica, trimethylammonium hydroxide (TMAOH), CTAB, and water at 80–120  C, samples exhibited MCM-41. At 130  C, there was a dramatic change in the morphology of the material. This sample comprised a mixture of MCM-41 and MCM-48 mesophases. By careful control of the synthesis temperature at 132  C, a sample containing only the faceted particles MCM-48 was obtained. For synthesis temperatures of 140 and 150  C, a lamellar phase formed. MCM-48 can also be obtained by a phase transformation (from as-made MCM-41) process at 150  C.[129] Control the effect of counteranion. MCM-48-like mesoporous silica[130] has been synthesized by using cetyltriethylammonium bromide as template and HNO3 as acid source at 0  C for 1 d. In this synthesis, HNO3 plays a key role and favors the formation of the Ia3d mesophase with a larger g parameter.

508

Chemistry of Zeolites and Related Porous Materials

FDU-5 and other Ia3d Mesophase Synthesis from Acidic Media with Polymer as Template Zhao has reported the synthesis of FDU-5 (Ia3d)[131] at room temperature in ethanol solution by a solvent-evaporation method. In the synthesis of FDU-5, triblock polymer P123 was used as template, TEOS as silica source, and a small amount of 3-mercaptopropyltrimethoxysilane (MPTS), benzene, or a benzene derivative (methyl-, ethyl-, dimethyl-, or trimethylbenzene) as additive. FDU-5 has uniform large pores (4.5–9.5 nm). The organosiloxane or organic additive plays an essential role in the formation of Ia3d mesostructures under acidic conditions at room temperature. Without organosiloxane, a 2-D hexagonal (p6mm) mesostructure is obtained from TEOS as sole silica source. The addition of organosiloxane or small organic molecules may increase the hydrophobic/hydrophilic ratio and result in a transition of the cooperatively assembled inorganic–organic mesostructure from high-curvature hexagonal p6mm to low-curvature bicontinuous cubic Ia3d. FDU-5 has also been successfully prepared by using mixed surfactants of P123 and anionic sodium dodecyl sulfate (SDS) as templates through an acid-catalysed silica sol–gel process. Zhao[132] demonstrated a simple solvothermal post-treatment method to prepare ordered large-pore ð> 7 nmÞ FDU-15 (Ia3d phase) via phase transformation from 2-D hexagonal mesophases (an SBA-15 prepared by the evaporation-induced self-assembly process). This synthetic strategy of solvothermal post-treatment can be simply performed in many organic solvents such as n-hexane at 60–100  C, and be extended to the syntheses of other silica-based mesoporous materials. KIT-6[133] is another high-quality large mesoporous cubic Ia3d silica. The synthesis of KIT-6 was done by utilizing a mixture of P123 and butanol as the template in aqueous solution. The cubic phase domain is remarkably extended by controlling the amounts of butanol and silica source correspondingly. The different synthesis temperatures resulted in the formation of different-pore-sized KIT-6 materials. Figure 8.21 shows the N2 adsorption–desorption isotherms for KIT-6 silica samples[133] synthesized at different hydrothermal treatment temperatures. AMS-10 (Pn3m Mesophase) AMS-10 with bicontinuous double-diamond cubic Pn3m symmetry[105] was prepared with anionic surfactant N-myristoyl-L-glutamic acid as template and N-trimethoxysilylpropyl-N,N,N-trimethylammonium chloride as co-structure-directing agent (CSDA). The silica wall of AMS-10 follows a typical diamond minimal surface (D surface) in analogy to the gyroid minimal surface (G surface) as observed for the silica wall of MCM-48. AMS-10 has a bicontinuous structure composed of an enantiomeric pair of 3-D mesoporous networks that are interwoven. Each network, which is divided by a D surface, consists of a tetrahedral connection of pores, although in the case of a bicontinuous cubic Ia3d structure each network that is divided by a G-surface consists of three connected pores. 8.3.4

Caged Mesostructures

The similar organization energies of the hcp and ccp lattices make their intergrowth easy in mesoporous materials synthesized from both simple surfactant and copolymer surfactant. Intergrowth with cubic close-packed (ccp) structure was typically observed

Synthesis, Structure, and Characterization of Mesoporous Materials

509

Figure 8.21 N2 adsorption–desorption isotherms for KIT-6 silica samples synthesized at different hydrothermal treatment temperatures. Reproduced with permission from [133]. Copyright (2005) American Chemical Society

for silica-based ordered mesoporous materials of hexagonal close-packed structure (hcp). Typical 3-D hexagonal (P63/mmc) and cubic (Fm3m) intergrowth mesostructures include SBA-2, SBA-12, and FDU-1. In their TEM images, the cubic and hexagonal structures can be observed in different ratios. They have the same packing method: the close packing of cages (spheres, sphere-like polyhedrons) in which each cage has 12 cages as next neighbors. The cages are interconnected via a window (or short channel). SBA-2 SBA-2 was first made as the product of a synthesis using divalent quaternary ammonium surfactants, Cn-s-1 in both basic and acidic media in 1995.[64] Based on the powder XRD patterns (see Figure 8.22) and TEM images, Huo, Stucky and coworkers proposed that SBA-2 has a 3-D hexagonal cage mesostructure with a space group P63/mmc (No. 194),[64] and the mesostructure is derived from a hexagonal close-packing of globular surfactant–silica micellar arrays. P63/mmc symmetry had not been previously reported for conventional liquid-crystal phases before SBA-2 was reported.[134] SBA-2 can be obtained with varying unit-cell and cage sizes by using different chain lengths, n, of Cns-1 surfactant over a wide synthesis range (from cell parameter c ¼ 7:7 nm for C12-3-1 to c ¼ 10:8 nm for C20-3-1). The SBA-2 crystal growth is plate-like and excellent for making thin films and membranes with the six-fold axis normal to the sheet direction. As expected for this geometry, the unit-cell parameter c/a ratio is about 1.62. After calcination, the large

510

Chemistry of Zeolites and Related Porous Materials

Figure 8.22 XRD patterns for SBA-2 (see reference [64] for details). Reproduced with permission from [64]. Copyright (1995) American Association for the Advancement of Science

cage-structured mesoporous silica framework remains. The template in SBA-2 can be removed by calcination at high temperature (500–600  C). The calcined SBA-2 has an N2 Brunaver, Emmett, and Teller (BET) surface area of 500–800 m2g1 . The N2 adsorption– desorption isotherms is type IV with an H2 hysteresis for even small-pore SBA-2 samples ð< 2:5 nmÞ. Later, more extensive TEM studies[135] suggested that the SBA-2 is an intergrowth of hcp structure and ccp structure. This phenomenon is similar with the case of intergrowth of EMT/FAU in zeolites. The intergrowth phases such as zeolite ZSM-3 and ZSM-20 were believed to be pure hexagonal phase initially. Even so, the index of XRD pattern with P63/mmc symmetry is still the most convenient method to identify SBA-2-type materials. The window size of SBA-2 is strongly dependent on the conditions of its synthesis.[136] For the low-temperature (4  C) synthesis in the presence of high concentrations of the

Synthesis, Structure, and Characterization of Mesoporous Materials

511

dicationic surfactant C16-3-1, basic conditions yield materials having smaller pores than the materials prepared in acidic media. The adsorption of nitrogen and the hydrocarbons cyclopentane and mesitylene reveal that SBA-2 prepared in basic media has a cage structure where the cages are linked through small ð< 0:4 nmÞ micropores, whereas the silicas (SBA-2) prepared in acidic media have larger pores after calcination, and are able to adsorb cyclopentane (diameter
0:5 nm) but not mesitylene ð
0:8 nmÞ. FDU-1 FDU-1 is an ordered silica with large cage-like mesopores of diameter about 8.15 nm, which can be synthesized under acidic aqueous conditions using the poly(ethylene oxide)–poly(butylene oxide)–poly(ethylene oxide) triblock copolymer (EO39BO47EO39) as a template.[137] FDU-1, first reported to have body-centered-cubic Im3m symmetry, was shown to exhibit face-centered-cubic Fm3m (ccp) mesostructure with, however, clear evidence of intergrowths with 3-D hexagonal P63/mmc (hcp).[138] Figure 8.23 shows the XRD pattern of FDU-1. FDU-1 shows an excellent hydrothermal stability, it retains its ordered structure on the basis of XRD even after 9 d of heating in water at 100  C. The feasibility of tailoring the pore-cage diameter (from
9:5 to 14.5 nm) and pore-entrance diameter (from below 4 to
8 nm) was done simply by adjusting the hydrothermal treatment temperature and time.[139] Doubling the amount of block copolymer, adding sodium chloride, and lowering the acid concentration will give FDU-1 samples with a high pore volume and a

Figure 8.23 Low-angle X-ray scattering pattern of FDU-1 (indexed with Fm3m) and its TEM image. Reproduced with permission from [138]. Copyright (2003) American Chemical Society

512

Chemistry of Zeolites and Related Porous Materials

narrower pore-size distribution. The addition of NaCl was found to significantly narrow the pore-size distributions and to improve the uniformity of entrances to the cage-like mesopores, whereas the pore diameter, specific surface area, and pore volume were similar to those for FDU-1 silicas obtained in the absence of NaCl. Although typical FDU-1 silicas and other polymer-templated silicas with large cagelike mesopores tend to be significantly microporous, the FDU-1 sample calcined at 1000  C appeared to be essentially free from microporosity, as can be inferred from the relation between its surface area, pore volume, and pore diameter. SBA-1 SBA-1 and SBA-6 were synthesized by using different surfactants and from acidic and basic synthesis media, respectively. They have the same structure and show the similar XRD patterns. Their structure is similar with cubic I1 phase, spherical micelles packed in Pm3n symmetry, in lyotropic liquid-crystal phase diagram for surfactant–water systems. SBA-1 was first prepared at room temperature under acidic conditions using C16TEAB, a surfactant with large headgroup, as template and TEOS as the silica source.[1,8,9] XRD pattern of SBA-1 (see Figure 8.24) is indexable with Pm3n symmetry. The use of large headgroup of the template supports further that it is the structural analogue of lyotropic liquid crystal I1 (Pm3n) mesophases, which needs small surfactantpacking parameter g and large headgroup area ao. Hartmann[140] investigated the influence of HCl/surfactant ratio (nHCl/nS) and synthesis time on the synthesis of SBA-1. The samples prepared at an nHCl/nS ratio of 280 were found to exhibit a higher degree of structural ordering, a higher specific surface

Figure 8.24 XRD pattern of SBA-1 with a synchrotron X-ray source (wavelength 0.17 nm). Reproduced with permission from [8]. Copyright (1994) American Chemical Society

Synthesis, Structure, and Characterization of Mesoporous Materials

513

Figure 8.25 SEM of material with Pm3n structure (a) HMM-3, (b) SBA-1. Reproduced with permission from [141]. Copyright (2000) American Chemical Society, and with permission from [142]. Copyright (2001) American Chemical Society

area, and a higher specific pore volume as compared with materials prepared at a lower nHCl/nS ratio. The high-quality ‘crystal’ of SBA-1 and other materials with Pm3n structure (e.g., hybrid silica HMM-3[141]) can be obtained by varying synthesis conditions. Figure 8.25 shows two examples: HMM-3 (BTME as silica source and C16TMA as template) and SBA-1 (synthesized at 5  C).[142] The stability of SBA-1[140] toward washing with water was improved by increasing the crystallization time. Based on the results of n-heptane, cyclohexane, and nitrogen adsorption. SBA-1 is mechanically more stable as compared with hexagonal mesoporous materials such as MCM-41 and SBA-15, but exhibits similar mechanical stability as compared with the cubic MCM-48 material. The pore size of SBA-1 can be expanded by increasing the amount of water in synthesis gel without any addition of organic cosolvent.[143] The more water for the same acidity (H2O/HCl molar ratio) and the same amount of siliceous species, the more the amount of Hþ surrounding the siliceous species. Thus the synthesis-process behavior at higher H2O/ TEOS molar ratios was similar to that at higher acidity, which leads to highly promoted silica condensation. However, the concentration of siliceous species tends to decrease with increasing amount of water, which could prevent the condensation of siliceous species. On the other hand, a possible reason is that water gives the hydrophilic solvation interaction with the charged headgroups in concentrated liquid-crystal arrays. The more water, the larger the volume of micelle, which leads to the expanded pore size of SBA-1. The window size of caged mesoporous silicas has been studied by various methods. For example, treatment of dehydrated SBA-1 and SBA-2 with two differently sized silylamines,[144] tetramethyldisilazane [HN(SiHMe2)2] and tetramethyldiphenyldisilazane [HN(SiMe2Ph)2], at ambient temperature. The intraporous silylation is controlled by the different sizes of the pore entrances (cage windows) of these mesoporous silica materials. The vast majority of the channels and pore openings, interconnecting the supercages of SBA-2, display a diameter larger than the size of an HN(SiHMe2)2 molecule; however, this is smaller than that of an HN(SiMe2Ph)2 molecule. The inner surface of SBA-1 is fully accessible to both disilazane reagents.

514

Chemistry of Zeolites and Related Porous Materials

SBA-6 The cubic Pm3n structure of SBA-6 silica templated by a extra-long divalent alkylammonium surfactant C18H37OC6H4O(CH2)4N(CH3)2(CH2)3N(CH3)3Br2 (18B4-3-1) in basic media.[84] Therefore, its unit cell is very big (a ¼ 18 nm for as-made sample). This mesostructure features two kinds of mesopore cages of diameters
7:3 and
8:5 nm. The structure of SBA-6 has been solved with electronic microscopy techniques.[84] 3-D HREM analysis gives details on asymmetric pore sizes and pore configuration that cannot be obtained by other techniques. The diffraction data for SBA-1 and SBA-6 clearly show that either Pm3n (no. 223) or P43n (no. 218) is an allowed space group. However, the space group Pm3n was choosen,[84] because the crystal morphologies of SBA-1 show point-group symmetries of m3m, whereas P43n should have a point symmetry of 43m. The crystal structure factors are obtained from a set of Fourier transforms of the HREM images (see Figure 8.26) by correcting for the effect of an objective lens contrasttransfer function (CTF). 42 Independent reflections out of 44 reflections with a larger spacing than 2.0 nm were obtained. A threshold in the potential density is determined for differentiating between the amorphous wall and enclosed cavities, giving the 3-D structure shown in Figure 8.27. A-cages and B-cages, differing in size, are arranged as in an MEP clathrate type (A3B) structure [see Figures 9.22(A) and 9.24(B)]. HREM analysis reveals an unusual aspect of the structures of SBA-1 and SBA-6, namely

Figure 8.26 TEM images of SBA-6. Reproduced with permission from [84]. Copyright (2000) Nature Publishing

Synthesis, Structure, and Characterization of Mesoporous Materials

515

Figure 8.27 The 3-D structure for SBA-6. Reproduced with permission from [84]. Copyright (2000) Nature Publishing

bimodal (micro–meso) porosity. In the structure of SBA-6, a B-cage is surrounded by 12 A-cages that are connected through mesoporous openings of 2.0 nm, while the openings between A-cages are about 3:3  4:1 nm. For SBA-6, the diameters of cage A and cage B are 8.5 nm and 7.3 nm, respectively. For SBA-1, the diameters of cage A and cage B are 4.0 nm and 3.3 nm, respectively. Cubic Structure (Im3m): SBA-16 The common structure models for Im3m symmetry are shown in Figure 8.28. Among the well known lyotropic liquid-crystal mesophases, these are at least two mesostructures with Im3m symmetry; one locates near the I1 region in the phase diagram, with a possible spherical micelle packed structure. Another one is close to the V1 region, and its most probable structure can be described by a P surface.

Figure 8.28

The periodic minimum surfaces for description of Im3m

516

Chemistry of Zeolites and Related Porous Materials

SBA-16[27] of cubic (Im3m) symmetry (body-centered cubic structure, bcc) synthesized using F127 was the first well documented example of ordered mesoporous silica with cage-like pores of diameter well above 5 nm. The mesostructure of SBA-16 can be described as a body-centered arrangement of cages with diameter 9.5 nm connected through large windows with diameter 2.3 nm along the [111] directions. The surface of SBA-16 is related to the minimal surface I-WP and can be described by the simple analytical function shown in Equation (8.3). ðcos px  cos pyÞ þ ðcos px  cos pzÞ þ ðcos py  cos pzÞ ¼ 0

ð8:3Þ

SEM and TEM studies of SBA-16[84] confirmed that SBA-16 has an Im3m symmetry. Figures 9.22(B) and 9.24(A) show the schematic structure for SBA-16. Each mesopore in the bcc structure has eight nearest-neighbor mesopores, and electron crystallography study[84] suggests that each mesopore is actually connected with its eight adjacent pores, thus forming a multidirectional pore-network system. The cage diameter is
9.5 nm and the entrance of the cage is
2.3 nm in diameter. Owing the cage structure of SBA-16, the preferred template for SBA-16 is a polymer surfactant with high PEO fraction such as F127 and F108. The formation of the SBA-16 mesophase was also achieved for somewhat lower EO/PO ratios by using blends of Pluronic F127 with Pluronic P123 copolymer.[78,145] The pore-cage diameter (from
4.5 to 9 nm) and entrance size in SBA-16 can be enlarged over a wide range not only by increasing the synthesis temperature and time, but also by increasing the content of P123 copolymer in the polymer mixture. This synthesis strategy for the pore structure design is expected to be generally applicable for other polymer- or oligomer-templated silicas with cage-like mesopores arranged in various structures. The pore-entrance size in SBA-16 can be tailored from the sub-nanometer range to at least 6 nm, making this silica a true mesoporous molecular sieve. Figure 8.29 shows the nitrogen-adsorption isotherm on SBA-16 at liquid-nitrogen temperature.[125] Its typical type-H2 hysteresis loop indicates that SBA-16 has a cage

Figure 8.29 Nitrogen-adsorption isotherm on SBA-16. Reproduced with permission from [149]. Copyright (2002) Chinese Chemical Society

Synthesis, Structure, and Characterization of Mesoporous Materials

517

structure. A simulation[146] suggested that the I-WP (see Figure 8.28) is closer to the experimental result than is P (see Figure 8.28). The structure of SBA-16 has been solved by electron microscopy techniques. 3-D HREM analysis gives details on asymmetric pore sizes and pore configuration that cannot be obtained by other techniques. The as-made SBA-16 single crystals synthesized with F108 and K2SO4 at 38  C are composed of
100% single-crystal particles uniform in morphology (
1 mm in size, Figure 8.30).[147] All of these particles have the shape of a rhombdodecahedron, consisting of 12 well defined crystal faces. These 12 faces can be indexed to 110 planes. The crystal has four three-fold axes and three four-fold axes and exhibits cubic symmetry which belongs to the m3m point-group class. Figure 8.31 shows the XRD pattern[149] of SBA-16 and the index result with Im3m. Small-pore (
2.6 nm) thick-walled (7.7 nm) SBA-16 (ST-SBA-16) can be synthesized[150] by using an oligomeric surfactant with ultra-long hydrophilic chains, Brij700.

Figure 8.30 SEM images of mesoporous SBA-16 single crystals. Reproduced with permission from [147]. Copyright (2002) American Chemical Society

518

Chemistry of Zeolites and Related Porous Materials

Figure 8.31 XRD patterns for as-made and calcined SBA-16. Reproduced with permission from [149]. Copyright (2002) Chinese Chemical Society

NMR and IR results show that ST-SBA-16 products have large numbers of silanol groups. The mesoscopic ordering of ST-SBA-16 can be greatly improved by the addition of a suitable amount of TMB. The pore size and surface area can be effectively tailored by changing the hydrothermal treatment time and the calcination temperature. Fm3m (FDU-12 and Others) The use of Gemini surfactant generally affords the SBA-2 type of structure as a mixture of 3-D hexagonal and Fm3m symmetry.[135] However, the pure Fm3m mesostructure was achieved under extremely low reactant concentrations by starting with a clear solution. Synthesis under low-concentration conditions would make micelles in the solution more diffuse, and crystallization more kinetically controlled. As suggested from the reconstructed 3-D structure, the shape of the micelles is highly spherical. Although the formation energy of ccp stacking would be expected to be close to that of hcp stacking, the micellar constituents should prefer to stack not by hexagonal close packing but by cubic close packing. This evokes the notion that the particle morphologies result in an icosahedral or decahedral shape with multiple twinning, and a triangular plate with single twinning.[151] Large-pore face-centered cubic (Fm3m) mesoporous silica FDU-12[87] without intergrowth with other phases has been synthesized by using F127 as template. Figure 8.32 shows the XRD pattern and nitrogen-adsorption isotherm of FDU-12.

Synthesis, Structure, and Characterization of Mesoporous Materials

519

Figure 8.32 XRD patterns (left) and adsorption isotherms (right) of FDU-12. Reproduced with permission from [87]. Copyright (2003) Wiley-VCH

The approximate pore structures of FDU-12s can be regarded as a face-centered cubic (fcc) close-packing of spherical cages, each connected to 12 nearest neigboring cages. However, the ideal spherical model is only valid for the FDU-12s prepared at low temperature. The 3-D structure models show that the synthesis of FDU-12 at a low temperature results in not only very large pore cages but also a very spherical-like cage shape. The cage shape of the FDU-12s synthesized at high temperature deviates from perfect spheres and is accompanied by an entrance enlargement. High-temperature hydrothermal treatment causes much more efficient enlargement of the cage entrances than that of the cages themselves, partially due to a release of expansion strains. The temperature-dependent behavior of the PEO block affected the micelles and hence the cage configuration. The cage and entrance size could be controlled by tailoring the amount of swelling agent, synthesis, and hydrothermal-treatment temperature. Normally, a conventional large-pore cagelike mesoporous silica, such as SBA-16, FDU-12, or FDU-1, shows a large H2 hysteresis loop. On the other hand, no hysteresis loop is observed in SBA-1, SBA-2, and SBA-12 in which the sizes of cavities and entrances are relatively small. In the case of FDU-12 materials, the type-H2 hysteresis loop for the sample synthesized at low temperature became a type-H1 hysteresis loop for the sample synthesized at high temperature. This result indicates that the entrance of the cage is enlarged and the resulting structure looks like a channel pore system. Accordingly, the shift of desorption branches to high relative pressure indicates that the entrance sizes of FDU-12 have been enlarged. When the hydrothermal-treatment temperature is increased from 373 to 413 K, the cavity size is increased to a small extent from 10 to 12.3 nm (increasing by 23%), while the entrance size is increased much more significantly from
4 to 8.9 nm (increasing by 123%).

520

Chemistry of Zeolites and Related Porous Materials

By employing a low-temperature synthetic pathway, highly ordered FDU-12[152] with the largest pores (up to 27 nm) and unit cells (up to 44 nm) have been successfully obtained. Meanwhile, the entrance dimension can be adjusted to between <4 and 16.7 nm. The unit-cell parameter does not change very much in the temperature range 23–60  C (29.5–26.0 nm), but swells sharply from 29.5 to 44.5 nm when the synthesis temperature, T, decreases to 15  C. Disordered mesostructured products were obtained at T < 14  C, and no solid precipitate was observed when T was lower than 10  C. 8.3.5

Deformed Mesophases, Low-order Mesostructures, and Other Possible Mesophases

MCM-41, SBA-3, and SBA-15 have perfect 2-D hexagonal mesostructures, whereas SBA-16 and FDU-12 have perfect cubic mesostructures. The mesostructures for their forms as a film on substrate can be deformed during drying or calcination, because their shrinkage is not isotropic. For example, the initial as-made 2-D hexagonal films on substrate were composed of cylindrical micelles, a highly organized structure with the c axis preferentially aligned parallel to the surface plane. During drying and calcination, unidirectional contraction of the initial mesostructure leads to 2-D-centered rectangular (c2mm) mesoporous structures. This phenomena have already been observed in 2-D hexagonal mesoporous films of both silica and TiO2. The bulk materials with deformed mesostructure have also been synthesized via a judicious choice of template or synthesis conditions. SBA-8 (c2mm) In surfactant lyotropic liquid-crystal systems,[153] the term ‘intermediate phase’ has been used for those noncubic phases occurring at compositions between hexagonal and lamellar phases. Ribbon phases are the intermediate phases that appear closest to the normal hexagonal phase. These phases are most easily described as a 2-D array of elongated noncircular tubes that are packed on 2-D lattices. Experimentally, intermediate phases in charged surfactant systems are observed whenever the hydrophobic chain is rather long or restricted in flexibility.[153] However, it is not clear why increasing the chain length or rigidity should favor the formation of these phases over that of the bicontinuous cubic structure. Stucky described how pore-structure modification can be achieved in a highly ordered fashion through the use of bolaform surfactants containing a rigid unit in the hydrophobic chain. The deformed hexagonal mesostructured silica SBA-8,[82] synthesized using bolaform surfactants at room temperature, has a 2-D pffiffiffi pore structure with a 2-D centered rectangular lattice (space group cmm, 1 < a=b < 3), which has no reported lyotropic liquid-crystal analogue. Bolaform amphiphiles, for example [(CH3)3Nþ(CH2)12OC6H4C6H4O(CH2)12Nþ (CH3)3](Br)2, R12, are closely related to Gemini surfactants CmH2mþ1N(CH3)2 CsH2sN(CH3)2CpH2pþ1Br2, Cm-s-p, with two hydrophilic moieties connected by a hydrophobic chain (of length s), but without the hydrophobic tails (m ¼ p ¼ 1). Doubling the chain length seems to have a larger influence on the aggregation behavior than doubling the number of headgroups. It has been predicted by theory and shown by experiments that surfactants containing rigid units have more specific aggregation

Synthesis, Structure, and Characterization of Mesoporous Materials

521

Figure 8.33 Lattice image and electron diffraction patterns (inset) of transmission electron micrographs for SBA-8 (a) perpendicular to channels and (b) parallel to channels. Reproduced with permission from [82]. Copyright (1999) American Chemical Society

behavior than do conventional surfactants with a flexible hydrophobic chain. In addition, introducing oxygen atoms into the bridging chain increases the solubility and makes it possible to substantially increase the bridging chain length. The unusual properties that result from these substitutions can be expected to introduce pore-channel size and shape anisotropy to the inorganic mesophase. For these reasons, the bolaform surfactants with rigid chain, Rn ðn ¼ 4; 6; 8; 10; 12Þ, were used as template. TEM images and electron-diffraction patterns (Figure 8.33) of calcined SBA-8 show a cylinder array along the zone plane[6] and a distorted hexagonal array along the zone plane,[5] confirming that calcined SBA-8 has a highly ordered 2-D mesostructure with centered rectangular symmetry. XRD patterns of as-synthesized and calcined silicate mesostructure SBA-8 prepared using a bolaform surfactant R12 at room temperature are shown in Figure 8.34. The XRD pattern of as-made SBA-8 shows two strong reflection peaks with d spacing of ˚ , in the 2u range of 2–2.5 , and several well resolved diffraction peaks 41.3 and 37.9 A in the 2u range of 3–10 . The XRD pattern can be indexed as a 2-D centered ˚ rectangular (space group c2mm) lattice with cell parameters a ¼ 75:7; b ¼ 49:2 A, a=b ¼ 1:53. pffiffiffi In the SBA-8 c2mm structure (a=b < 3), the elongation of the hexagonal channels takes place along the b direction or shrinkage takes place along the a direction, while in the Ma structure the elongation is along the a direction or shrinkage is along the b direction (see Figure 8.35), e.g., a=b ¼ 2:2 for SDS and CTAB.[154] Under basic synthesis conditions, the same type of surfactant Rn ðn ¼ 10; 8; 6; 4Þ led to the formation of MCM-41, and only the long-spacer surfactant R12 yields SBA-8. This result is in agreement with the lyotropic behavior of long, rigid-chain surfactants. The fact of transformation of SBA-8 synthesized at room temperature readily into MCM-41 through post-synthesis treatment confirms the intermediate nature of this phase. SBA-8 (c2mm) can be induced to transform into MCM-41 (p6mm) in the mother synthesis solution at 70–100  C. When the as-made SBA-8 is treated in water at 100  C (pH 7–10), the phase transformation to MCM-41 is completed within 2 h.

522

Chemistry of Zeolites and Related Porous Materials

Figure 8.34 XRD patterns of (a) as-made SBA-8 and (b) calcined SBA-8. Reproduced with permission from [82]. Copyright (1999) American Chemical Society

Surfactant R12 gives SBA-3 (p6mm) in acidic (HCl) synthesis media, which also confirms the close relationship of c2mm and p6mm mesophases. A simple method for obtaining SBA-8 mesoporous silica (with a=b ratio of 1.44) is the use of alkyltrimethylammonium (such as CTAB) as template, and atrane complexes (a silatrane made from TEOS and triethanolamine) as silicon precursors.[155] Dealing with CTAB, it is well known that polar organic species (such as ethanol, triethanolamine, and (CH3)2NCH2CH2OH which was mentioned for MCM-48 synthesis) penetrate the hydrophobic and palisade regions of micelles and thereby induce structural rearrangements of the surfactant phase. This penetration should affect both the micelle geometry and its charge density at the interface. In synthesis, that resulted in the formation of elliptically shaped micelles and led to the anisotropic rectangular c2mm symmetry. The rapid silica polymerization quickly locks the distorted mesostructure in place. KSW-2 (C2mm) Almost all mesostructures reported far have been governed by the geometrical packing of surfactants. The mesopores are near-cylindrical channels or near-spherical cages. Here is an exception. KSW-2[156] has a mesostructure of rectangular arrangements of square or lozenge-shaped 1-D channels. KSW-2 was synthesized by mild acid treatment (pH value

Synthesis, Structure, and Characterization of Mesoporous Materials

523

Figure 8.35 A scheme for deformation of 2-D hexagonal mesophase into 2-D centered rectangular mesophase. Reproduced with permission from [82]. Copyright (1999) American Chemical Society

524

Chemistry of Zeolites and Related Porous Materials

Figure 8.36 (a) Typical TEM scan of calcined KSW-2 (pH 4.0) and the corresponding electron diffraction ED pattern indexed as hk0 projection. (b) Typical TEM and the corresponding ED pattern of as-made KSW-2 (pH 6.0). (c) Another TEM scan of the as-synthesized KSW-2 (pH 6.0). Arrows imply the observed place of the bending of silicate sheets derived from kanemite. Reproduced with permission from [156]. Copyright (2000) Wiley-VCH.

below 6.0) of a layered alkyltrimethylammonium (CnTMA)-kanemite complex. The preparation of the layered complex without altering the structure of the silicate framework is the most important step in obtaining the mesostructured precursor. A typical TEM image and the corresponding electron-diffraction (ED) pattern of the calcined KSW-2 obtained under the adjusted pH value of 4.0 are shown in Figure 8.36. The TEM image of the calcined KSW-2 exhibits relatively ordered square arrangements that display a periodic distance of adjacent pores of about 3.3 nm. Striped patterns with the same periodic distance were also observed, supporting the presence of 1-D mesopores. The ED pattern showed the angle of diagonal lines connecting the spots was close to 90 .[105] All the powder XRD peaks of the as-synthesized and calcined KSW-2 are assigned to an orthorhombic structure (C2mm) (as-synthesized: a ¼ 5:34 nm, b ¼ 5:05 nm, c ¼ 1; calcined: a ¼ 4:84 nm, b ¼ 4:26 nm, c ¼ 1). Authors observed that the structural units of kanemite are partly retained and therefore chose a 3-D symmetry (C2mm) rather than a 2-D one (c2mm).

Synthesis, Structure, and Characterization of Mesoporous Materials

525

The layered silicate network originating from the structure of kanemite connects twodimensionally and is not destroyed under the synthesis conditions. Thus, the bending of individual silicate sheets may be attributed to the limited structural changes that result from partial intralayer condensation and the accompanying structural change in C16TMA assemblies during their gradual leaching. The resulting square pore system is not defined by the geometrical packing of surfactant molecules. Low-order Mesostructures: KIT, MSU, etc. Compared with MCM-41, typical KIT[81], MSU-n[80], and HMS[157] generally possess disordered structure, thicker walls, high surface area, and superior thermal stability upon calcination. Their structures are close to hexagonal phase, but their channels are not straight, and interconnect each other. Their 3-D-like structures, although not ordered, result in the good mass transportation property. 8.3.6

Phase Transformation and Control

From a view point of reaction time, the typical preparation of mesoporous material can be divided three main steps: (1) interaction between surfactant and silica (or other inorganic) species in solution and the formation of ordered mesostructure; (2) the further reaction (polymerization or condensation for silica) at a certain temperature for a time period. A possible phase transformation may occur; (3) recovery of solid product by filtration, washing, and drying. The phase transformation may also occur in this step; (4) removal of template from the solid product by calcination or extraction with solvent. The phase transformation is also possible even in this step. Phase transformations are associated with changes in the curvature of organic– inorganic interface and may be understood phenomenologically as a competition between the elastic energy of bending of the interfaces and energies resulting from the constraints of interfacial and charge separation. The different entropic and interaction energies in the nanoscale organic, inorganic, and interface regions result in structural frustration with incompatible local packing constraints that forbid an optimal geometry where the energy is everywhere minimized. The inorganic–organic structures therefore readily undergo structural changes or transformations to relieve this stress. In fact, since the mesophase synthesis or reaction is a kinetically control process and the solid formed is not a thermally stable phase, the phase transformation is very common during the synthesis of mesoporous materials. The phase transformations include the transition from one structure or symmetry into another structure or symmetry, or the transition from a disordered phase to an ordered phase, or from an ordered phase to a disordered phase. The intermediate phase can be isolated as a product or be observed by analysis techniques. The phase transformation can occur during the synthesis process or in a post-synthesis treatment. The early famous example is the transition of lamellar mesophase of silicate into hexagonal mesophase in basic or near neutral media.[5] Once those parameters which have more influence on mesophase stability are known, the new synthetic pathways utilizing structural rearrangements to create new materials will become a reality. An investigation of phase transformation[158] under hydrothermal conditions indicated the most mildly basic conditions utilized (pH 9), which favor silica condensation, best inhibit the phase transformation, and thus produce the most kinetically

526

Chemistry of Zeolites and Related Porous Materials

stable composites. High-pH treatment allows for the most simple rearrangements. Condensation occurring during composite synthesis rather than during hydrothermal treatment has a much smaller effect on phase stability, probably because much of the condensation that occurs during synthesis is random and not optimally coupled to the nanoscale architecture. Materials that start out poorly condensed, by contrast, can be extensively hydrothermally modified so that the final material has an inorganic framework with a highly uniform silica density; this provides the maximum resistance to transformation and the highest kinetic stability. Changes in the packing of template in the mesophase drives the transformation, indicating that surfactant packing is a dominant factor in determining the overall structure of products.[159] Matching between the interfacial charge density of the inorganic silica framework and the charge density of the surfactant headgroups is also found to affect the kinetics of transformation.[160] There are many phase transitions observed, for example, from lamellar to MCM-41,[5] cubic Pm3n mesophase to P63/mmc mesophase gradually.[161] A phase transformation from cubic SBA-1 to hexagonal SBA-3 was observed upon drying the precipitates.[162] Solvent evaporation upon drying of the precipitate was found to be the key factor for the phase transformation. This phase transformation occurred only when the wet precipitate, crystallized for a short period (less than 2 h), was dried in open air. After a few min, the 100 peak of the hexagonal phase has grown relatively strong and the 110 and 200 peaks can be clearly seen. The phase transformation apparently is somehow related to the solvent evaporation. However, for the precipitate crystallized for a long period of time such as 4 h, or with higher acid concentration, the cubic phase was retained even after the precipitates were dried. The differences between the stable and unstable cubic SBA-1 crystallites should be the extent of silica condensation.

8.4 Pore Control The biggest advantage of ordered mesoporous materials is their uniform mesopores; pore control is very important for theses mesoporous materials. The mesopore system (pore shape and array of pores) can be controlled by varying different mesostructures. In this section, the general methods to control pore size will be discussed. 8.4.1

Pore-size and Window-size Control

Pore Size The pore sizes of M41S family mesoporous materials are about 3–4 nm.[2,3] The pore sizes of mesoporous silica are extended to
30 nm by SBA-15 and FDU-12, etc. material. The classical method to adjust pore size is to vary the chain length of the surfactant or to add swelling agent. Now, there are many strategies available for changing pore size.[163] The basic principle is the same: change the size and volume of micelles. Most methods are listed as follows: (1)

Choice of the tail length of surfactant (e.g., alkyltrimethylammonium and amine surfactants).[2,8,17] In general, the surfactants with a chain length of C16 give the

Synthesis, Structure, and Characterization of Mesoporous Materials

(2)

(3) (4) (5) (6)

(7) (8) (9)

(10)

(11) (12) (13)

(14) (15)

527

pore size of
3.8 nm. The materials with pore size of 2–5 nm can be obtained with a judicious choice of surfactant and synthesis conditions. Use of cationic surfactant and low-polar or nonpolar organic swelling agents (e.g., aromatic hydrocarbons, alkanes, or long-chain amines) in the direct synthesis. The swelling agents can enlarge the pore size up to
10 nm. Use of the larger surfactant such as block copolymer. For example, surfactant P123 gives SBA-15 with the pore of 2–30 nm without or with swelling asgents.[27,113] Synthesis at high temperatures. The pore size can be controlled to between 2 and 7 nm by using the same surfactant. Post-synthesis hydrothermal treatment in mother liquor or water. The pore size can be enlarged up to
7 nm. Post-synthesis hydrothermal treatment in the presence of amine. The treatment of as-made 3.5 nm pore MCM-41 with long-chain trialkylamines or N; N-dimethylalkylamines at 100–130  C for 2–3 d increased the pore size to
11 nm. The pore volume also increased from 0.8 to 2.4 cm3/g. Post-synthesis solvent-thermal treatment. Alcohol-thermal treatment can enlarge the pore size of SBA-15 and improve its ‘crystallinity’. Modification of pore surface with chemicals such as silane. This will reduce the pore size and improve the stability of material. Chemical vapor deposition (CVD) modification after partial removal of surfactant. After removal of surfactant located in the entrance of pores with organic solvent, modify the material with CVD of TEOS, then remove all surfactant in the channel. This results in reducing the pore-entrance size only. Use of mixture of surfactant (e.g., C18-3-1 and CTAB for MCM-41, Gemini surfactant mixture for MCM-48) as template and post-synthesis hydrothermal treatment.[1] This method can give high-quality and large-pore mesoporous silica materials. Use of a mixture of surfactants with different chain lengths to tune finely pore size. For example, C12TMA and C16TMA mixture. Control of pH of synthesis system to tune finely pore size of MCM-41. Chang of synthesis temperature. For example, SBA-15 with different pore size can be obtained by changing the synthesis temperature.[27] Low temperature helps TMB to swell the micelle of F127 and to synthesize highly ordered mesoporous silicas with very large pores.[152] Vary the synthesis parameters and conditions such as concentration of surfactant and reaction time. Use of nanometer-sized liquid or solid (e.g., emulsion or colloid) as template. The pore size is larger than 50 nm. Details will be discussed in the Macroporous Material Section.

Window Size for Caged Mesostructure For the caged structured materials, the window size (or entrance size) is very important for their adsorption and other properties. The low surface area measured for calcined mesocaged AMS-8,[86] 408 m2/g, in comparison with other cubic caged mesostructures such as SBA-1 (>900 m2/g) suggests that smaller cages are not fully accessible to the external surface. There are 16 small

528

Chemistry of Zeolites and Related Porous Materials

˚ ) and 8 large (76.0 A ˚ ) cages in the unit cell interconnected via cage windows of (56.0 A ˚ ˚ (for large–small 14–25 A (for large–large connectivity) and smaller windows of <5 A and small–small connections). For the small window (pore entrance)-sized materials, the classical molecular probe adsorption or reaction can be used to analyse window size of materials. For example, dehydrated SBA-1 and SBA-2 were treated with two differently sized silylamines, tetramethyldisilazane [HN(SiHMe2)2] and tetramethyldiphenyldisilazane [HN(SiMe2 Ph)2], at ambient temperature.[144] The intraporous silylation is controlled by the different sizes of the cage windows of these materials. SBA-2 displays a diameter larger than the size of an HN(SiHMe2)2 molecule; however, this is smaller than that of an HN(SiMe2Ph)2 molecule. The inner surface of SBA-1 is fully accessible to both disilazane reagents. The more general technique for window-size analysis is based on the study of adsorption isotherms. The conventional methods of pore-size analysis assuming cylindrical pore channels, such as the Barret–Joyner–Halenda (BJH) method, were shown not to be appropriate for cage-type structures. The nonlocal density functional theory (NLDFT) has been used to characterize mesocaged structures.[164] NLDFT analysis gives accurate information about the cage size, the total meso- and micropore volumes and surface area, and the pore-wall thickness in combination with XRD measurements. Argon- and nitrogen-desorption data on FDU-1 provided evidence that there are two major populations of pore entrances. Argon desorption was superior in providing information about pore connectivity in FDU-1 samples. Kim[145] discussed the relation between the lower limit of adsorption–desorption hysteresis and the pore diameter, the latter being reflected by the capillary condensation pressure. It is often assumed that the lower limit of hysteresis, that is the lower closure point of the hysteresis loop, is a function of the adsorbate and temperature. It was also suggested that the lower limit of hysteresis depends on the pore shape and the pore diameter.[165] The window size of caged mesostructure may affect the adsorption behavior. Thus the adsorption behavior of cage-like pores is likely to be a function not only of pore diameter, but also of entrance size (and number of entrances), especially when the latter size is relatively large. It seems that the larger pore-entrance size shifts the pressure of capillary evaporation at the lower limit of hysteresis to higher relative pressures. This is consistent with an observation that the lower limit of hysteresis in caged mesopores appears to be shifted to somewhat lower pressures in comparison to the limit for cylindrical mesopores. The most successful example is the control of window size of caged mesoporous silica materials synthesized with block copolymer as templates. The pore-entrance diameter of SBA-16 increases as a function of synthesis or hydrothermal-treatment temperature. This is likely to be related to the known phenomenon of the decrease in hydrophilicity of PEO blocks as the temperature increases. In the mesostructure of the F127–silica composite, the cores of the (spherical) micelles are constituted by PPO blocks, whereas the micelle corona, which consists of PEO blocks, interacts with the silica framework. At lower temperature, PEO blocks are expected to favorably interact with hydrophilic silica species and thus to have a tendency to be intimately mixed with the silica framework. When the F127-silica composite is subjected to the treatment at higher temperatures,

Synthesis, Structure, and Characterization of Mesoporous Materials

529

these interactions become less favorable, which is expected to lead to a higher degree of aggregation of PEO blocks in the silica wall. Post modification can also be used to control the window size of mesoporous materials. The modifications include the methods mentioned above for window size-analysis. An efficient method is CVD with TEOS on as-made, surfactant partially removed, or calcined sample. 8.4.2

Macroporous Material Templating Synthesis

Although macroporous materials are beyond the scope of this chapter, we believe it is necessary to briefly introduce some macroporous silica materials and their syntheses, because they are very close to the mesoporous silica materials we described in this chapter. Ordered macroporous materials with pore sizes in the sub-micrometer range have applications in low-dielectric-constant materials and lightweight structural materials. Macroporous oxides such as silica, titania, and zirconia as well as polymers with well defined pore sizes in the sub-micrometer regime have been successfully synthesized.[166,167] Several synthesis methods have been used to prepare 3-D macroporous inorganic materials. These include sol–gel, salt precipitation, nanocrystal infiltration, and polymerization. All the synthesis methods rely on the use of a polymeric or inorganic template, usually in a form of spheres packed in a periodic fashion. For sol–gel synthesis of ordered macroporous materials in which metal alkoxides dissolved in alcohol are impregnated into the voids of the template, monodisperse particles such as emulsion droplets or latex spheres, which self-assemble to form an ordered 3-D array, can be employed as templates. Hydrolysis and condensation take place at the sphere surface, leading to the formation of the inorganic framework. The templates are then removed either by heat treatment or dissolution with a solvent leaving a periodic array of holes in the inorganic matrix. The final macropore dimensions are
15–30% smaller than the original size of the spheres due to the shrinkage of the inorganic framework. This is caused by a large volume loss during sol–gel process as an alcohol is evaporated. Significant shrinkage of the inorganic framework during template removal by heat treatment may result in severe cracking and loss of long-range order. Several single, binary, and tertiary oxides have been prepared using the sol–gel chemistry. Colloidal Crystals Template A macroporous silica can be synthesized through a slow sedimentation of colloidal particles onto a template. In a fast, single-step reaction (see Figure 8.37), the monomeric alkoxide precursors permeate the array of bulk polystyrene spheres and condense in air at room temperature. Close-packed, open-pore structures with 320–360 nm voids are obtained after calcination of the organic component at high temperatures. Polymer latex particles[169] in the range from 20 to 400 nm (or larger) with different surface functionalities can be employed as templates for the synthesis of macroporous materials. The route of templating polymer dispersions is complementary to the synthesis in lyotropic liquid-crystalline phases, leading to a bimodal size distribution of the pores. Dual template synthesis combined with a modified bulk sol–gel process can be used to prepare the 3-D bimodal ordered porous silica, in which the macropore wall is mesoporous and both the pores are interconnected. The macropores were replicated

530

Chemistry of Zeolites and Related Porous Materials

Image Not Available In The Electronic Edition

from the template of the dried polystyrene colloidal crystal after being removed by calcination at high temperature, whilst the mesopores were achieved by burning off the surfactant in the gel. Titania, zirconia, and alumina samples[168] with periodic 3-D arrays of macropores were synthesized also from the corresponding metal alkoxides, using latex spheres as templates. Microemulsion Imhof [166] reported a method for producing highly monodisperse macroporous materials with pore sizes ranging from 50 nm to several micrometers. Macroporous titania, silica, and zirconia were formed by using the nonaqueous monodisperse emulsion droplets of isooctane (2,2,4-trimethylpentane) in formamide as templates around which material was deposited through a sol–gel process. Monodisperse emulsions assemble spontaneously into an ordered lattice at sufficiently high volume fraction (>50%). Gelation and removal of the droplets led to a macroporous amorphous titania gel. The pore size can be accurately controlled, and the technique should be applicable to a wide variety of metal oxides and even organic polymer gels. Bacterial Template Mann[170] showed how a bacterial superstructure, consisting of a thread of coaligned multicellular filaments of Bacillus subtilis, can be used as template to make macroporous

Synthesis, Structure, and Characterization of Mesoporous Materials

531

Figure 8.38 Scheme of two routes to the formation of ordered macroporous inorganic frameworks using bacterial superstructural templates. Reproduced with permission from [170]. Copyright (1997) Nature Publishing

silica materials (Figure 8.38). They produced ordered macroporous fibers of either amorphous silica or MCM-41 by template-directed mineralization of the interfilament spaces followed by removal of organic material by heating to 600  C. The inorganic macrostructures consist of a macroporous framework of 0.5 mm-wide channels with curved walls of either silica or mesoporous silica, 50 to 200 nm in thickness. Other textures from animal or plant can be used as template. Siliceous Mesostructured Cellular Foams Siliceous mesostructured cellular foams (MCFs) have welldefined ultra-large mesopores and hydrothermally robust frameworks. MCFs[171] can be obtained by adding a sufficiently large amount of an organic co-solvent (e.g., TMB) in the SBA-15 synthesis system. MCFs composed of uniformly sized, large spherical cells that are interconnected by uniform windows to create a continuous 3-D pore system. The interconnected nature of the large uniform pores makes these new mesostructured silicas promising candidates for supports for catalysts and in separations involving large molecules. The MCFs consist of uniform spherical cells measuring 24–42 nm in diameter,[171] possess BET surface areas up to 1000 m2/g and porosities of 80–84%, and give higherorder scattering peaks even in the absence of long-range order. Windows with diameters of 8–22 nm and narrow size distribution interconnect the cells. The pore size can be controlled by adjusting the amount of the organic swelling agent that is added and by varying the aging temperature. Adding ammonium fluoride selectively enlarges the windows by 50–80%. In addition, the windows can be enlarged by post-synthesis treatment in hot water. The MCF materials resemble aerogels, but offer the benefits of a facilitated synthesis in combination with well defined pore and wall structure, thick walls, and high hydrothermal stability. 8.4.3

The Synthesis of Hierarchical Porous Silica Materials

Recent interest is being devoted to the development of hierarchically ordered porous structures (ordered on multiple lengthscales with controlled multiscale porosity), because

532

Chemistry of Zeolites and Related Porous Materials

hierarchical materials with different pore sizes integrated in one body can be expected to combine reduced resistance to diffusion and high surface areas for yielding improved overall reaction and adsorption–separation performances and can be extended to biological applications. There are many studies reported on the bimodal micro-mesoporous, double mesoporous, macro-mesoporous, and trimodal porous materials. The syntheses can be done in various ways. The most common involves secondary templates such as colloidal crystals, emulsion, or vesicle droplets for the formation of macropores. The use of cationic surfactant and neutral homopolymer [e.g., poly(ethylene glycol)] mixtures can lead to the formation of hierarchical pore structures through a multiscale selfassembly process. Micro-mesoporous Bimodal Materials SBA-15 made at low temperatures contains micropores and mesopores. The micropore volume of SBA-15 could be systematically controlled by varying synthesis parameters (some were mentioned above); for example, the synthesis temperature and the TEOS/ surfactant ratio.[172] Plugged hexagonal templated silicas[173] (see Figure 8.39) have been reported. They are hexagonally ordered materials, with internal microporous silica nanocapsules; they

Figure 8.39 Nitrogen isotherm at 77 K and a model of a typical plugged mesoporous silica material. Reproduced with permission from [173]. Copyright (2002) American Chemical Society

Synthesis, Structure, and Characterization of Mesoporous Materials

533

have a combined micro- and mesoporosity and a tunable amount of both open and encapsulated mesopores. These plugged hexagonal templated silicas have two types of micropore (originating from the walls and the nanocapsules, respectively) and a tunable number of both open and encapsulated mesopores. The micropore volumes have a high value (up to 0.3 cm3/g) and the total pore volume exceeds 1 cm3/g. The obtained materials are much more stable than the conventional micellar templated mesostructures, and can easily withstand severe hydrothermal treatments and mechanical pressures. An ordered mesoporous aluminosilicate with completely crystalline zeolite ZSM-5 wall structure has been synthesized with mesoporous carbon CMK-3 as template.[174] Meso-macroporous Bimodal Materials Stucky[175] developed a procedure for the synthesis of sponge-like silica membranes with 3-D meso-macrostructures. The process utilizes multiphase media composed of a mesoscopically ordered block copolymer–silica phase that macroscopically separates from an electrolyte phase. The different characteristic lengthscales in hierarchically organized composite structures can be independently adjusted. Micro-macroporous Materials Mixed solutions of cationic surfactants and nonionic poly(ethylene glycol) or block copolymers were employed for the synthesis of monolithic trimodal porous silica.[176] Lyotropic mixtures of block copolymers of different lengths with hydrophilic linear PEO chains were also applied to their nanocasting into bimodal micro-mesoporous silica to formulate the dependence of the mesopore sizes and the microporosity on the lengths or sizes of the hydrophobic and the hydrophilic blocks, but the mesostructures were wormtype in morphology and several hundred nanometers or more in size. Trimodal Porous Materials Zeolite monlith with macropores can be considered as a micro-macroporous material. Mechanically stable zeolite monoliths[177] containing 3-D, ordered, closed macropores have been fabricated by hydrothermal treatment of nano-zeolite seeded mesoporous silica spheres. The easy speed of sedimentation and digestion renders the whole process suitable for large-scale production of macroporous zeolite materials.

8.5 Synthesis Strategies 8.5.1

Synthesis Methods

Heating Methods The most common synthesis method for mesoporous materials is through a reaction in aqueous solution at different temperatures (from 4 to
150  C). The heating methods include the normal oven or microwave oven.[178,179] Microwave energy has been employed in many recent chemical-reaction studies and has been found to change the kinetics and selectivity, often in favorable ways.[179] The synthesis of many mesoporous materials could be shortened by over an order of magnitude if microwave energy was employed. MCM-41 was first synthesized[178] using microwave heating in 1996. Later, many mesoporous silica materials, including MCM-48, SBA-15, FDU-1, and PSU-1, were prepared by microwave hydrothermal methods. All report an

534

Chemistry of Zeolites and Related Porous Materials

enhanced crystallization compared with conventional hydrothermal synthesis. Typically, the crystals produced are smaller, uniform particles at higher yield, indicating rapid homogeneous nucleation. Evaporation-induced Self-assembly (EISA) Process The EISA process[180] enables the rapid production of mesophases or mesoporous materials in the form of films, fibers, or powders. Beginning with a homogeneous solution of soluble silica and surfactant prepared in ethanol–water solvent with co  cmc, preferential evaporation of ethanol concentrates the nonvolatile surfactant and silica species (Figure 8.40). The progressively increasing surfactant concentration drives self-assembly of silica–surfactant micelles and their further organization into liquid-crystal mesophases. The result is rapid formation of thin-film mesophases that are highly oriented with respect to the substrate surface. Through variation of the initial reaction-mixture composition it is possible to form different final mesostructures. The EISA process is perfect for the synthesis of films and monoliths[181] and can be used for transition metal-containing silica materials and nonsilica mesoporous materials.[33] The organization through EISA of mesostructured silica materials is given by a delicate balance of different processes in competition with each other: silica polycondensation and phase separation or organization of the template. Speeding up of the silica reactivity gives, in general, a lower degree of organization, in accordance with the ideal order for self-assembly of the kinetic constants involved in the process: kinter > korg > kinorg , where kinter , korg , and kinorg are, respectively, the relative rates for interface formation, organic-array assembly (as it exists in the final mesostructure), and silica polycondensation. On the other hand, the inorganic or hybrid bricks used to build up the pore walls of mesoporous materials are synthesized using sol–gel chemistry which has to be combined with the micelle organization and the formation of an interface between the bricks and the surfactant. At acidic conditions (pH
2), the condensation rates of silica species are slow enough to allow the formation of ordered mesophases during self-assembly.

Figure 8.40 Approximate trajectory taken in ethanol–water–CTAB phase space during the EISA process. Point A corresponds to the initial composition of entrained solution, Point B is near the drying line, and Point C corresponds to the dried product. Reproduced with permission from [180]. Copyright (1999) Wiley-VCH

Synthesis, Structure, and Characterization of Mesoporous Materials

535

Some synthesis methods can be considered to be modifications of the general EISA synthesis. For example, the direct calcination of wet surfactant–silicate gel and some post-synthesis treatments (phase transformation). 8.5.2

Surfactant, its Effect on Product Structure and Removal from Solid Product, and Nonsurfactants Template

The importance of surfactant in the synthesis of mesoporous materials has already been discussed in this chapter. A great deal of effort has been devoted to evaluating the effects of the chemical nature, shape, and steric hindrance of organic surfactants on the mesophase configuration of silica-based materials. Surfactant control of the pore system and mesophase topology has been a fruitful approach to the design of mesoporous materials. Nevertheless, the need for special surfactants constitutes a serious breakdown of such an approach. The surfactant plays the key role in formation of the mesophase and determines the synthesis pathway at certain levels. According to their charge properties, we can divide them into three types: cationic [such as CnH2nþ1N(CH3)3þBr], anionic(such as CnH2nþ1OSO3-Naþ and CnH2nþ1OPO3H2), and nonionic (neutral) (such as long-chain amine, and block copolymer P123 and F127). Cationic Surfactants For the systematic investigation of the formation of mesoporous silica materials, a series of cationic surfactants was selected with and without organic additives, which favor a range of g values when used as templates to synthesize silica mesophases under different reaction conditions.[1] (1) Long-chain quaternary ammonium surfactants CnTMAX (n ¼ 10–18, X ¼ Br , Cl, or OH) are the most used templates for M41S materials including MCM-41 and MCM-48. (2) Gemini surfactants (Cn-s-n): A more subtle and variable way to fine tune the surfactant molecular shape and charge is by using multicharged oligomeric units. This has been done with Gemini surfactants with considerable success. Gemini surfactant, Cn-s-n, is a name assigned to a family of synthetic amphiphiles consisting of, in sequence, a long hydrocarbon chain, an ionic group, a spacer, a second ionic group, and another hydrophobic tail.[182] These surfactants are particularly interesting from a fundamental point of view: their structure can be considerably modified by acting independently on the length and the nature of either side chain or the spacer group. The relative positions and distances of headgroups of conventional monoquaternary ammonium surfactants are determined primarily by electrostatic interactions and also by the packing requirements of disordered alkyl chains. Formally, the Cn-s-n surfactants may be considered as dimers of the monosurfactants. The two headgroups in Cn-s-n are chemically linked through an adjustable polymethylene spacer (CsH2s). The presence of the spacer makes it possible to fine tune the distance between the headgroups and thereby control the effective headgroup area, ao. By this means we can change the packing parameter, g, of a surfactant by adjusting its spacer length. The mesostructure-templating behavior of Cn-s-n [e.g., C16-s-16, C16H33Nþ(CH3)2(CH2)sNþ(CH3)2C16H33] is similar to that in surfactant–water binary systems[182] and gives the mesostructures expected for charge-density matching. Small-s (2 or 3) surfactants favor MCM-50, medium-s (5 or 6) surfactants

536

(3)

(4)

(5) (6)

(7)

(8) (9)

Chemistry of Zeolites and Related Porous Materials

favor MCM-41. Cn-12-n (n ¼ 16–22) gives MCM-48 at both room temperature and high temperature (100  C), while C12-12-12 gives MCM-41 at room temperature. The latter observation illustrates the ability to fine tune with the individual tail lengths. Note that an aqueous solution of C12-12-12 remains micellar over the entire range of composition and does not form lyotropic liquid-crystal phases. This synthesis result demonstrates the importance of the inorganic species in the cooperative templating mechanism for the concentration region in which the syntheses are carried out. Divalent and multiply charged surfactants: The divalent quaternary ammonium surfactant Cn-s-1 may be considered as an end member of Gemini surfactant or a highly charged large-headgroup surfactant. The effective headgroup area relative to total hydrophobic tail volume and length is effectively almost doubled, greatly decreasing the packing parameter, g. This puts us back in the cage side of the structural phases. SBA-2 (P63/mmc) is the typical product when Cn-s-1 is the template. Large-headgroup surfactants: The effective headgroup area can be modified by simple headgroup substitution (see Table 8.5). Surfactants having large headgroups favor globular micellar aggregates and result in cage-structured mesoporous materials. Bitail surfactant with one charged headgroup such as (C12H25)2N(CH3)2Br gave generally lamellar phase due to its large packing parameter, g. Bola surfactants such as [(CH3)3Nþ(CH2)12OC6H4C6H4O(CH2)12Nþ(CH3)3](Br)2 (for synthesis of SBA-8) and [(CH3)3NCHCH2CH2N(CH3)2CH2(CH2)11OC6H4 C6H4O(CH2)11CH2N(CH3)2CH2CH2CH2N(CH3)3](Br)4 (for FDU-11 and FDU13)[91] favor lower-symmetry mesophases. These lower symmetries are not stable for the normal surfactant under common conditions. The molecular configuration and the rigid property of these surfactants made these low-symmetry mesophases possible. Extra-long rigid-chain surfactant such as C18H37OC6H4O(CH2)4N(CH3)2(CH2)3N(CH3)3Br2 (for the synthesis of SBA-6) is useful to make large pore material because its long chain cannot coil, while a normal long chain would coil (for example C22TMA).[1] Positively charged surfactants in acidic media such as C16H33N(CH3)2(CH2)3SO3 and C16H33N(CH3)2 act as templates, similarly to quaternary ammonium surfactants.[1] Hydroxy-functionalized surfactant: the hydroxy group in the surfactant CnH2nþ1Nþ(CH3)2(CH2)mOH decreases the hydrophobicity of the headgroup and the headgroup charge is more shielded by water[183] of solvation or silicate or their anions in solution, thus decreasing the effective cationic headgroup area, ao. Table 8.5 Effect of headgroup of surfactant C16NR1R2R3 on product structure[1] R1

R2

R3

Product

H CH3 CH3 CH3 C2H5

CH3 CH3 CH3 C2H5 C2H5

CH3 CH3 C2H5 C2H5 C2H5

SBA-3 SBA-3 SBA-3 SBA-1 SBA-1

Synthesis, Structure, and Characterization of Mesoporous Materials

537

It therefore plays an important role in the entropic and enthalpic contributions of water organization to structure direction. The hydroxy-functionalized headgroup surfactants favor formation of mesophases with low interface curvatures such as 2D hexagonal or lamellar mesophases[8] due to the small effect area, ao. The product is a lamellar mesophase when C16H33Nþ(CH3)2(CH2)2OH is used, while a similarly structured surfactant with a smaller headgroup, C16H33Nþ(CH3)2C2H5, gave only a relatively high-interface-curvature mesophase, MCM-41. A hydroxy group in the hydrocarbon chain of the surfactant, e.g., b-substituted, C14H29CH(OH) CH2Nþ(CH3)3, also has a small effective headgroup area, and templates the formation of highly ordered lamellar silicate (up to sixth-order in Bragg reflections in XRD pattern). In addition, high-angle diffraction peaks are observed that are characteristic of the hydrocarbon-chain packing within organized surfactant layer structures. Anionic Surfactants In the S Iþ direct-synthesis pathway, anionic surfactants are used as templates for mesoporous metal oxides. For example, C12H25OSO3 Naþ (SDS) as template for Al2O3 and Ga2O3 mesoporous materials with Al13[AlO4Al12(OH)24(H2O)127þ] and Ga13-Keggin [GaO4Al12(OH)24(H2O)127þ] as Al and Ga source, respectively.[184] A novel route for the synthesis of high-quality mesoporous silicas using anionic surfactants and organosilane groups as co-templates has been recently reported by Che.[185] This synthesis pathway involves the use of the co-structure-directing agents (CSDA): aminosilane or quaternized aminosilane, e.g., N-trimethoxysilylpropyl-N,N,Ntrimethylammonium chloride and (3-aminopropyl)trimethoxysilane under alkaline conditions. The alkoxysilane site of CSDA is co-condensed with inorganic precursors. The introduction of CSDAs into the reaction system makes it feasible to produce electrostatic interactions between anionic surfactants and inorganic species and to control the packing parameter. These are key factors in the formation of highly ordered mesophases and to select a certain structure. Various mesophases AMS-n (anionic-templated mesoporous silicas) (see Table 8.4), including lamellar, 2-D hexagonal p6mm, tetragonal P42/mnm, 3-D hexagonal P63/mmc, cubic Pm3n, cubic Fd3m, cubic Ia3d, modulated structure, and a chiral mesostructure with helical arrangement of the pores, were successfully prepared based on this synthesis route, though some of them have not been observed even in liquid-crystal systems. The mechanism has been generalized as S Nþ I, where S is the surfactant, Nþ represents the positively charged amine moiety in the aminosilane group, and I the condensing silica framework. Nonionic (Neutral) Surfactants The nonionic surfactants used for the synthesis of mesoporous materials include longchain amines and poly(ethylene oxide). Most of the nonionic surfactants may be charged in extreme acidic conditions. The common surfactants used in synthesis of mesoporous silica materials include: (1) long-chain amine surfactants such as C12NH2[14,17] give loworder mesoporous materials with thicker walls, (2) long-chain poly(ethylene oxide) surfactants,[14,17] (3) the molecule containing more than amine groups such as NH2(CH2)nNH2 (n ¼ 10–22) and CnH2nþ1NH(CH2)2NH2 (n ¼ 10; 12; 14) can also be used as templates for synthesis of mesoporous silica.[186] The templated silica materials have good thermal and hydrothermal stabilities.

538

Chemistry of Zeolites and Related Porous Materials

Polymeric Surfactant Block copolymers are valuable supramolecular templates for the synthesis of ordered large-pore mesoporous silica and other metal oxides because of their simple templating ability, low-cost commercial availability, and biodegradability. Copolymer surfactants such as PEO-PPO-PEO and PI-b-PEO[187] consist two or three blocks: one hydrophobic and one or two hydrophilic blocks. Various self-organizing structures such as micelles, vesicles, microemulsions, and liquid crystals can be formed in amphiphilic block copolymer–water and amphiphilic block copolymer–water–oil systems. The ratio of hydrophilic block to hydrophobic block determines the surfactant-packing property in the micelle. The biggest difference between long-tail quaternary ammonium surfactants and copolymer surfactants is their hydrophilic part. The micelle formed with quaternary ammonium surfactants is like a hard ball, while the micelle formed with block copolymer surfactants is like a soft ball with fluffy PEO shell. The flexibility of the micelle of polymer surfactant also enhances the possibilities of formation of more mesostructures. The differences result in the material that is templated by polymer surfactant having a thicker wall. The wall may contain micropores, because a certain fraction of the PEO chains is tightly embedded inside the silica walls as a consequence of the molecular templating of single PEO chains.[77] The main characteristics for block copolymer surfactant templating include: (1) the product has larger pores because of the large size of the micelle, (2) the synthesis media must be acidic because the PEO will be partially charged in acidic media. This will increase the interactions between surfactant and inorganic species, (3) mesoporous product has a thicker wall which results in the high stability of silica framework, (4) the surfactant is easy to remove due to the weak interactions between surfactant and inorganic wall (compared with the strong interaction between positively charged quaternary ammonium surfactant and negative charged inorganic wall), (5) the micropores in the inorganic wall can be controlled by choosing suitable synthesis conditions, (6) more mesostructures, including low-symmetry structures, can be synthesized, (7) most copolymer surfactants are commercially available at low cost, and have low toxicity, (8) easy to extend to nonsilica-based mesoporous materials,[33] (9) structure can be controlled easily in multiple lengthscales to make the mesomacroporous, meso-microporous, and meso-mesoporous (two different pore sizes or shapes) materials, (10) pore size and shape, and window size for caged mesostructures, can be adjusted easily. Compared with ionic surfactants, block copolymers have become more and more popular in the synthesis of mesoporous inorganic solids, because of their diverse structural characteristics and rich phase behavior. Different synthesis methodologies have been developed, carefully manipulating reaction parameters such as temperature, pH, ionic strength, reaction time, and solution composition. Mixed Surfactants It is of practical interest to find a way to change the effective head- and tailgroup sizes systematically and continuously in order for their more efficient use. The use of mixed surfactants is the easiest way. Such control of the hydrophilic–hydrophobic balance is often impractical because surfactant molecules with various chain sizes that cover the range to be studied are not easily available.

Synthesis, Structure, and Characterization of Mesoporous Materials

539

The cooperative assembly of the composite mesostructures is both a kinetically and thermodynamically determined process, a blend of two surfactants with different headgroup sizes or chain lengths might be a good template for the synthesis. In this way, the surfactant’s effective headgroup size could be controlled by varying the molar ratio between the surfactants with small and large headgroups. The effect of mixing unlike surfactants can be thought of as a simple averaging of two surfactant-packing parameters. For example, a mixture of C16-12-16 and C16-3-1 is used in silicate mesophase synthesis.[1] The products vary from MCM-48 to SBA-2 through MCM-41 as the fraction of C16-3-1 increases in the mixture. The mixture of Cn-3-1 and CmTMAþ can result in the formation of good quality MCM-41 which easily gives five or more well defined XRD peaks. It is worth noting that high-quality MCM-41 can still be obtained when m ¼ 22, while the single surfactant CmTMAþ (m ¼ 20 or 22) favors the lamellar phases and does not give MCM-41 at 100  C. The high-quality MCM-41 obtained using a mixture of surfactants is thermally stable and the calcined sample has still at least five XRD peaks. This synthesis result indicates that CmTMAþ is a good, but not ideal, template for the formation of MCM-41. CmTMAþ has only one charge per hydrophobic chain. More charges (from Cn-3-1) in the headgroups in the mixed surfactants are more favorable for the formation of high-quality MCM-41. There are many combinations of surfactants reported. For example, cationic CTA surfactant and cationic Gemini surfactant, CTA and long-chain pyridine quaternary ammonium surfactant,[1] anionic and cationic surfactant,[126] CTA and amine (C12H25NH2), neutral and cationic surfactant,[176] and neutral–neutral surfactants have been used to prepare silica with interesting mesostructures (e.g., Ia3d cubic) or morphologies (e.g., nanoparticles). Effect of Surfactant Chain (Tail) Length The effect of chain (tail) length of quaternary surfactants is often ignored in most cases. Ottaviani[188] reported that the kinetics of formation of mesostructured aluminosilicate is strongly dependent on the surfactant (CnTMAB, n ¼ 8–18) chain length, larger micelles inducing a faster formation. Ryoo[189] found that the structural order of MCM-41 can be improved remarkably if mixed surfactants of n-alkyltriethylammonium bromides (CnTEAB, n ¼ 12; 14; 16; 20 and 22), and n-alkyltrimethylammonium bromides (CnTMAB) are used. The optimum mixing ratio can be tuned according to the length of the alkyl groups. Under the same conditions, the length of the surfactant micelles increases with the chain length. In general, the change of chain length of alkyltrimethylammonium (CnTMA) surfactant does not affect the surfactant-packing parameter, g, because the ratio of V/l is almost unchanged when n ¼ 10–18, while when n > 20, the long chain is easily coiled, V is increased and l decreased, and g is increased. Thus, CnTMA (n ¼ 20; 22) give generally the lamellar mesophases, not the 2-D hexagonal phase.[1] Surfactants with longer chain lengths, corresponding to higher hydrophobicity, have the stronger tendency to form elongated micelles. Removal of Surfactant from Solid Product For making the porous materials, the surfactant must be removed from the solid product. The main methods include calcination at high temperature and extraction with solvent. Heat treatment has a great influence on product structure. Shrinkage of pores or channels is a common phenomenon for calcined samples. Two-step calcination can reduce the

540

Chemistry of Zeolites and Related Porous Materials

influence. The surfactant is decomposed at low temperature (e.g., 150  C for CTAB in MCM-41), then the organics are removed at high temperatures such as 500  C. Extraction with solvent has less influence on product pore-structure. The common solvents used include ethanol, methanol, and water. To remove cationic surfactant more efficiently, HCl is added in the solvent. The surfactant recovered can be reused. Nonionic surfactant can be extracted easily, even without addition of HCl. The removal of surfactant with solvent extraction can be combined with modification of the product. Heating with microwaves was used to remove surfactant in SBA-15, SBA-16, MCM41, etc. materials. All surfactants can be removed within a short time period (2–5 min) and more Si OH groups left. Supercritical-liquid (e.g., CO2) extraction in the presence and absence of co-solvents was also used to remove surfactant from as-made mesoporous materials.[190] Use of oxidizing agents (e.g., O3, N2O, and NO2) can decrease the calcination temperatures to 423–623 K (150–350  C). Nonsurfactant Templates Wei[191] developed a novel, low-cost, biocompatible nonsurfactant pathway for preparing mesoporous materials. With use of small nonsurfactant molecules such as glucose, fructose, dibenzoyltartaric acid, and urea as a pore-forming agent, silica mesoporous materials have been prepared with high surface area (
1000 m2/g) and pore volume (
0.5–1.0 cm3/g). The pore size is in the range of 2–6 nm with narrow distributions. No evidence of the existence of self-assembly in these syntheses and highly ordered mesostructure products was found. Other templates include triphenylene-based charge-transfer (CT) complexes, inorganic tubules, ionic liquids, and so on.[192] Mesoporous Carbon as Template Highly ordered mesoporous silica can be regenerated from a mesoporous carbon CMK-3 that is a negative replica of mesoporous silica SBA-15, indicating reversible replication between carbon and inorganic materials.[193] The advantage of this synthesis method is that it does not need to make the same silica material (template). This method is likely to be a valuable complement to the existing methods for the fabrication of new mesoporous silicas and other composition materials. The regenerated HUM-1 (I41/a),[194] a silica replica of mesoporous carbon, is a cubic mesoporous silica that is distinctly different from the original MCM-48 silica and represents a previously unreported new mesoporous silica material. HUM-1 does not possess the two noninterconnecting channel systems found in the starting MCM-48 framework. This new mesoporous silicate was prepared by a cyclic serial replication process and it may not be possible to make this material using the current conventional surfactant assembly methods employed for the synthesis of mesoporous silicas. Monodisperse and high-surface-area mesoporous inorganic spheres of various compositions including metal oxides, mixed oxides, and metal phosphates[195] have been prepared by templating mesoporous carbon spheres replicated from spherical mesoporous silica. Owing to the rigid and thermally stable framework of the carbon template, the crystalline phases of the obtained metal oxide spheres can be readily tailored by controlling crystallization temperatures.

Synthesis, Structure, and Characterization of Mesoporous Materials

8.5.3

541

Stabilization of Silica Mesophases and Post-synthesis Hydrothermal Treatment

Zeolites such as Y, Beta, and ZSM-5 are widely used commercial catalysts, but their applications are strongly limited by their small pore sizes. One solution is to use ordered mesoporous materials such as MCM-41 and SBA-15. These materials exhibit good catalytic properties for the catalytic conversion of bulky reactants. Although the mesoporous silica materials made from normal synthesis processes, such as MCM-41 and MCM-48, have high thermal stability, their hydrothermal stability is poor. Calcined samples can be destroyed with moisture or water, even at room temperature. Most calcined samples became amorphous in cold water within a few minutes. The main reason is the hydrolysis reaction of the amorphous silica wall (Si O Si bonds broken). The relatively low catalytical activities of mesoporous materials such as MCM-41 and MCM-48, as compared with zeolites, can be typically attributed to the low acidity or low oxidation ability of catalytically active species, which is strongly related to the amorphous nature of the mesoporous walls. Therefore, increasing acidity, oxidation ability, and hydrothermal stability are great goals for rational syntheses of ordered mesoporous materials. Through studies over the last few decades, several methods have been developed for improvement of the stabilities of mesoporous silica materials. These strategies were shown to increase either the wall thickness or the degree of polymerization of the silica wall. Incorporation of metal oxides, such as zirconia, alumina, and vanadium oxides, was also demonstrated to be an effective method to improve the hydrothermal stability. The main routes for improving hydrothermal stability of mesoporous silica materials will be discussed in this section. Stabilization of Silica Mesophases 1. Increase the polymerization level of silica wall Most methods described here are confirmed by 29Si MAS NMR results: the silica wall was further polymerized. (a) Ryoo[196] improved markedly the stability of MCM-41 by a shift in the reaction equilibrium when acetic acid was repeatedly added during the synthesis. Adjustment of pH with acetic acid to pH
11 shifted the reaction equilibrium toward the formation of MCM-41. This synthesis method gives much higher quality MCM-41 than do procedures using a pH adjustment at the beginning of the reaction. The materials were stable up to 973 K (700  C) under hydrothermal heating in humid air. (b) The addition of inorganic salts (such as KCl, NaCl, and NaF, etc.) during the crystallization process or post-synthesis hydrothermal process has been demonstrated to be an effective approach to improve the hydrothermal stability of such materials in boiling water.[197] (c) Increases of synthesis temperature and reaction time can improve the stability of mesoporous silica. Extreme high-temperature (>150  C) synthesis of ultra-stable ordered mesoporous silica-based materials by using fluorocarbon–hydrocarbon surfactant mixtures or mesoporous carbons as templates will be discussed later.

542

Chemistry of Zeolites and Related Porous Materials

(d) The addition of an organic amine to the synthesis system can increase the hydrothermal stability of products. (e) Use of an amine (e.g., methylamine, dimethylamine, ethylamine) to replace NaOH as base source improved the stability of MCM-41. (f) Amorphous silica as silica source in better than TEOS to make hydrothermal stable materials. (g) High-temperature calcination can increase the hydrothermal stability of mesoporous silica. For example, a calcined sample at 800  C shows a clear MCM-41 XRD pattern after 2 h of heating in water at 100  C, while a calcined sample at 500  C loses its structure under the same conditions. 2. Chemical modification to build a protective layer (a) The post-synthesis modification can be alumination, grafting, or silylation which leads to lower surface polarity and increased hydrothermal stability. (b) The introduction of F into the synthesis system favors the formation of stable mesoporous silica such as MCM-48.[198] (c) The wall surface of SBA-15 or MCF materials was coated with ZSM-5 or NaY zeolite by using a dilute, clear solution containing primary zeolite units.[199] The resulting materials have high acidity and improved steam stability as compared with the corresponding mesoporous materials. The method is quite simple, general, and applies to various kinds of zeolite guests and mesoporous materials hosts. 3. Increase the wall thickness (a) Use of nonionic surfactants as templates to synthesize materials with thicker walls; for example, block copolymer P123 templating SBA-15. (b) Synthesis with mixed surfactants as template. (c) Post-synthesis modifications such as grafting or silylation by using CVD or reaction in solution. (d) Low-order materials such as KIT-1 and MSU materials have thicker walls. 4. Change wall composition and structures (a) Use zeolite building units as the building blocks for the wall of mesoporous silica materials (details will be described later in this chapter). (b) Use calcined MCM-41 as silica source to make stable MCM-41. (c) Introduction of Al or certain other metals (such Zr or V) into the mesoporous wall. Post-synthesis Hydrothermal Treatment Maximizing the ordering influence of the organic surfactants can be done by (1) using low temperatures to minimize organic lower-order and short reaction times to kinetically create only partially condensed silica frameworks that can structurally follow the organic organization and minimize interphase frustration, (2) annealing the as-made solid product at room temperature to further optimize long-range order, and then (3) carrying out the silica polymerization in water. The latter greatly reduces the silica charge relative to what

Synthesis, Structure, and Characterization of Mesoporous Materials

543

it would have in the mother liquor so that the organic–inorganic interactions are correspondingly reduced. This, and the reduced solubility at the pH used, makes it possible to retain the templating introduced at low temperatures by the more organized organic to the partially polymerized silica. Post-synthesis hydrothermal treatment[1,18] is a simple and powerful technique to optimize or change mesostructure, or improve certain properties of the product. The socalled treatment is a heating process of as-made material in water with or without additives. Phase transformations are frequently observed as a consequence of mild treatment of as-made samples. The strength and rigidity of the material is enhanced and, in some cases, its unit cell and pore size enlarge. Post-synthesis hydrothermal treatment can improve the properties (XRD quality, silica polymerization, stability, etc.) of most asmade mesostructured silica materials, including the products from both of acidic and basic media. The main reason is that the as-made mesophases are still flexible after the low-temperature synthesis step. In almost all cases, the treatment improves the structural ordering,[1] however, in some cases there is no significant expansion of the unit cell. If the treatment is for enlargment of pore size, the poorly polymerized silica samples should be chosen because they have more space to be adjusted. In general, low-temperature and short-time synthesis leads to poorly polymerized materials. When TEOS is used as starting reagent at pH
12, mixed surfactants (C22-3-1 and C18TMAþ) as template, at room temperature or lower, polymerization of the silica begins and a precipitate rapidly forms.[1] Silica polymerization of this partly condensed phase is interrupted by using short reaction times (0.5–2 h) and then ‘ripening’ the filtered, air-dried, solid product at room temperature. The solubility[53] of amorphous silica is a minimum in water at neutral media (pH 7
8), and is more than an order of magnitude less than that at the normal mother liquor pH used in MCM-41 synthesis. Two-week treatment at 100  C in water (pH
7) enlarged the cell of hexagonal silica mesostructure from 5.45 to 7.82 nm, and two additional weeks’ of treatment gave a cell of 7.96 nm. The XRD patterns for large-unit-cell mesophases show seven to eight XRD peaks and retain their structure on calcination, with only small cell shrinkage. Nitrogen-adsorption measurements reveal that this material has a pore size of 6.0 nm, a pore volume of 1.6 cm3/g, and a surface area of 1086 m2/g. An important feature is that the apparent wall thickness based on the adsorption measurement and XRD data is 1.7 nm, which is substantially greater than that obtained from conventional MCM-41 (
0.8–1.0 nm). This is not surprising in view of the reduced charge associated with the silica phase at the lower pH. Three well distinguished regions of the adsorption branch of the adsorption–desorption isotherm were observed: monolayer–multilayer adsorption, capillary condensation, and multilayer adsorption on the outer surface. In contrast to nitrogen-adsorption results of MCM-41 with pore size less than
4.0 nm, a clear type-H1 hysteresis loop in the adsorption–desorption isotherm is observed and the capillary condensation occurs at high relative pressure, consistent with the large pore size. However, no significant expansion of the unit cell is observed for MCM-41 and MCM48 containing single CnTMAþ template, although seven to eight XRD peaks are obtained for the MCM-41 and about 20 peaks are generated with MCM-48.[1] The kinetic matching of organic and inorganic ordering during assembly and silica polymerization is critical to the structural and property design of mesophase materials.

544

Chemistry of Zeolites and Related Porous Materials

Figure 8.41 XRD patterns for the calcined SBA-3 obtained from the CnTMAX–HNO3–TEOS– H2O systems with (A,B) and without (C,D) hydrothermal treatment: (A) C18TMAC; (B) C16TMAB; (C) C18TMAC; (D) C16TMAB. Reproduced, with amendment, from [200]. Copyright (2000) Wiley-VCH

The post-synthesis ammonia hydrothermal treatment can simultaneously restructure pore size, nanochannel regularity, and morphology of SBA-3.[200] The restructured products become highly ordered and more stable while giving the nanosized tubular form of the silica. After ammonia hydrothermal treatment at 150  C for 2 d, the products exhibit 5 to 7 sharp XRD peaks (see Figure 8.41) while the products before hydrothermal treatment only have 2 to 4, broader peaks (see Figure 8.41). The existence of the higherorder reflection peaks such as 300, 220, and 310 suggests that well ordered hexagonal structures are formed. There is also a lattice expansion 1.4 and 1.3 nm after the hydrothermal restructuring. The shrinkage of the d100 value, after calcination, of these treated samples is only 0.1 nm. From the 29Si MAS NMR result of the as-synthesized materials before and after ammonia hydrothermal treatment, the Q3/Q4 ratio changes from 1.0 to 0.4. This indicates that the silica structure further condenses during ammonia hydrothermal treatment and makes it more thermally stable. The effect of ammonia hydrothermal treatment is to shift the equilibrium to stronger surfactant/silicate binding and thus more pore expansion. Figure 8.42 shows the N2-adsorption isotherms of the mesoporous materials before and after ammonia hydrothermal treatment. The sample without hydrothermal treatment has broader pore-size distribution. After ammonia hydrothermal treatment, the samples posses a sharp pore-size distribution. Notably, increasing the hydrothermal time and temperature to 150  C can expand the pore size to about 5.0 nm.

Synthesis, Structure, and Characterization of Mesoporous Materials

545

Figure 8.42 Nitrogen-adsorption isotherms for samples obtained from the C18TMAC–HNO3– TEOS–H2O system before and after ammonia hydrothermal treatment. (A) Before treatment. (B) Hydrothermal treatment at 100  C for 4 d. (C) Hydrothermal treatment at 150  C for 1 d. (D) Hydrothermal treatment at 150  C for 2 d. Reproduced from [200]. Copyright (2001) Wiley-VCH

These results show that the ammonia hydrothermal process would not only restructure the nanostructure of the silica into a highly ordered nanostructure, but would also increase the thermal and hydrothermal stability of the nanostructure. Sayari[163,20] found the unprecedented expansion of the pore size (from
3.15 nm to 25 nm) and pore volume (from 0.85 to 3.31 cm3/g) of MCM-41 after hydrothermal treatment in the presence of N,N-dimethyldecylamine (DMDA). The pore wall thickness did not undergo any significant changes and shows the extreme structural flexibility of the MCM-41. However, the pore structure was no longer ordered as inferred from the XRD and TEM results. The effect of long-chain amines can be observed also in the direct synthesis at
70  C: upon adding a long-chain amine directly to the synthesis system or to a synthesis system, it was shown that the amine was not added as a reactant, but was the resulting decomposition product of the surfactant under the synthesis or treatment conditions. N,N-Dimethylhexadecylamine (DMHA) is another convenient expander; other amines, such as trioctylamine and tridodecylamine, were also found to be suitable for the preparation of large-pore MCM-41 but were not as efficient as DMHA. Addition of DMDA, dimethyloctylamine (DMOA), and small trialkylamines to the synthesis gel did

546

Chemistry of Zeolites and Related Porous Materials

not lead to pore-size enlargement, but decreased the structural uniformity and in some cases resulted in development of microporosity. The pore size increased and the structural ordering decreased as the amount of the amine in the synthesis gel increased. The post-synthesis treatment can give a better result. Mesopores (3.5 nm) of silica material were enlarged up to 25 nm without losing surface area. Figure 8.43 shows a

Figure 8.43 Schematic representation of MCM-41 pore-expansion, selective extraction, and calcination. Reproduced with permission from [201]. Copyright (2005) American Chemical Society

Synthesis, Structure, and Characterization of Mesoporous Materials

547

schematic representation of MCM-41 pore-expansion, selective extraction, and calcination.[201] Depending on the conditions, materials with controlled pore sizes from 3.5 up to 25 nm were obtained. The pore volume varied accordingly from, typically, 0.8 up to 3.6 cm3/g, whereas the specific surface areas hardly changed. In the presence of water, the alkylamine self-assembles into an inverted cylindrical micelle inside the alkyltrimethylammonium micelle and is held via attractive hydrophobic forces. The expander can be removed selectively in the presence of ethanol even at room temperature. The material A has an open-pore structure and readily accessible amine groups and material B has a highly porous, hydrophobic material with large surface area and pore volume. Materials A and B were found to be fast, sensitive, high-capacity, recyclable adsorbents for metallic cations and organic pollutants, respectively. Post-synthesis hydrothermal treatment in salt solution[202] could be a convenient method for pore expansion and silica-wall thickening for improvement of its stability. The pore size and wall thickness vary with the type of anion in the salt and its concentration. The salt effect follows the well known binding strength of the Hofmeister series of anions for the cationic surfactant, NO3 > Br > Cl > SO42
F. The anion binds with cationic surfactant molecules in solution to shift the equilibrium of surfactant/silicate binding, leading to less surfactant and water in the pore, and hence less pore expansion. 8.5.4

Zeolite Seed as Precursor and Nanocasting with Mesoporous Inorganic Solids

Zeolite Seed as Precursor[203,204] There are several successful examples of the self-assembly of preformed zeolite primary and secondary structural building units (nanoclusters) with surfactant micelles. This selfassembly will form ordered mesoporous aluminosilicates with strong acidity and high hydrothermal stability. Pinnavaia[205,206] reported that steaming-stable mesoporous aluminosilicates with strongly acidic sites are assembled from a zeolite seed solution that normally nucleates the crystallization of microporous zeolites. Assembling these zeolite seeds into a mesostructure enhances acidity and hydrothermal stability that begins to approach that of zeolites, even though the framework walls remain atomically disordered. The mesoporous aluminosilicates (e.g., MSU-S) had been initially synthesized from zeolite seeds of the faujasitic type, followed by synthesis from zeolite seed solutions of ZSM-5 and Beta in alkaline media. Both XRD and nitrogen-adsorption data showed that these mesoporous materials were hydrothermally stable. Catalytic tests in cumene (isopropylbenzene) cracking show these mesoporous materials are much more active than conventional mesoporous aluminosilicates, indicating that the acidic strength is greatly enhanced, as compared with that of MCM-41. Interestingly, the mesoporous walls of ˚ , but MSU-S is hydrothermally stable. MSU-S are very thin, only at 9–12 A [207] Xiao reported the preparation of ordered hexagonal mesoporous aluminosilicates (MAS-5) with uniform pore sizes from self-assembly of preformed aluminosilicate precursors with CTAB surfactant. The XRD pattern for a typical as-made MAS-5 sample shows four well resolved peaks that can be indexed as (100), (110), (200), and (210) reflections associated with the hexagonal symmetry. No diffraction peak was observed in the region of higher angles 10–40 , which indicates the absence of large

548

Chemistry of Zeolites and Related Porous Materials

microporous crystals in the sample. A similarly high degree of mesoscopic order is observed for hexagonal MAS-5 even after calcination at 900  C. After treatment of the calcined sample in boiling water for more than 300 h or at 800  C for 2 h in a flowing steam, the XRD patterns still show those peaks assigned to hexagonal symmetry, suggesting that the MAS-5 sample has a remarkable hydrothermal stability even at high temperatures. The acidity of the MAS-5 sample was characterized by temperature-programmed desorption of ammonia. Apparently, the desorption temperature of ammonia on MAS-5 is about 410  C, which is much higher than that on MCM-41 (320  C) with similar wall composition. These results indicate that the acidic strength of MAS-5 is much higher than that of MCM-41 and is comparable to that of microporous Beta zeolite. In catalytic cracking of 1,3,5-triisopropylbenzene and catalytic alkylation of isobutane with butene, MAS-5 exhibit is greater catalytic activities. The IR, UV-Raman, and NMR results suggest that the pore walls of MAS-5 contain primary and secondary structural building units of zeolites. Such unique structural features might be responsible for the observed strong acidity and high thermal stability of the mesoporous aluminosilicates with ordered hexagonal symmetry. Furthermore, the preformed zeolite precursors are extended to zeolite MFI and L nanoclusters. All of these mesoporous aluminosilicates prepared from preformed zeolite nanoclusters with primary and secondary building units exhibit much higher hydrothermal stability and acidity than those of conventional MCM-41. The first example for the preparation of ordered mesoporous aluminosilicates assembled from preformed zeolite Beta precursors in strongly acidic media has been reported by Xiao.[203] He reported that Al is effectively introduced into ordered mesoporous silica materials (MAS-7) by assembly of preformed zeolite Beta nanoclusters with copolymer surfactant P123 in strongly acidic media (pH < 0) by a two-step procedure. First, preformed zeolite precursors containing zeolite primary and secondary building units were prepared. Secondly, the zeolite precursors were assembled with a polymer surfactant in strongly acidic media. In this way, aluminum species were fixed in the framework of the preformed zeolite nanoclusters in the first step and then directly introduced into the mesoporous walls along with the preformed zeolite nanoclusters in the second step. The most significant advantage of this method is that the Al species in the mesostructure are mostly located at zeolite-like sites, which give the products very high catalytic activity. Characterization of nitrogen-adsorption isotherms and TEM shows that the mesoporous wall (4–5 nm) of MAS-7 is much thicker than that of SBA-15, which is enough for assembly of preformed zeolite precursors (2–3 nm). MAS-7 has much higher hydrothermal stability than do SBA-15 and Al-SBA-15. The micropore volume of MAS-7 sample (0.15 cm3/g) is much more than that of the SBA-15 sample prepared under the same conditions (0.05 cm3/g). Even the fact that the MAS-7 sample has thicker walls than SBA-15 was considered. The larger micropore volume in MAS-7 may be attributed to the existence of zeolite primary units in the ˚ mesoporous walls. The TEM image[203] of MAS-7 shows obvious white dots of 7 A diameter except for the mesopores (pore size of 7.4 nm), which could possibly be assigned to micropores in the mesoporous walls of MAS-7. This confirmed that the mesoporous walls are partially polycrystallized although the size is relatively small (around 2–3 nm).

Synthesis, Structure, and Characterization of Mesoporous Materials

549

Table 8.6 Synthesis of mesoporous materials with zeolite precursors as starting reactants Material

Zeolite seeds

MAS-3 MAS-5 MAS-7 MAS-9 MPS-9 MTS-9 MTS-10 MSU-S MSU-S/H MSU-S/F

Zeolite L Zeolite b Zeolite b ZSM-5 Silicalite-1 TS-1 TS-1 Zeolite Y, b, ZSM-5 Zeolite Y, b, ZSM-5 Zeolite Y, b, ZSM-5

Template CTAB CTAB P123 CTAB P123 P123 F127 CTAB Block copolymer (MCF structure)

Reference [203] [203, 207] [203] [203] [203] [203] [203] [205] [205] [206]

The synthesis has also been extended into other preformed zeolite precursors. Table 8.6 summarized the results for syntheses of hydrothermally stable and highly catalytically active mesoporous materials assembled from various preformed zeolite precursors. The above approach can be considered as the ‘bottom-up’ methodology. The ‘topdown’ approach[208] has also been reported. Hexagonal mesostructured MSU-Z was made through the base hydrolysis of ZSM-5 in the presence of cetyltrimethylammonium ions. The five-membered ring subunits can be readily incorporated into the framework walls of MSU-Z. One-step direct synthesis of the highly stable mesoporous silica-based material is also possible. MMS-H[209] has a structure analogous to that of MCM-48 but which contains zeolite building units. A mixture of CTAB and Brij30 was used as template for the mesopores. The use of TPAOH without the assistance of NaOH helps to introduce zeolite secondary building units, as well as the direct formation of acidity after removal of the template. This material was also found to possess superior thermal, hydrothermal, steam, and mechanical stabilities. Nanocasting with Mesoporous Inorganic Solids[210] Nanocasting allows us to create mesoporous materials with new compositions, controllable structure, and specific functionality. Figure 8.44 illustrates the nanocasting pathway. For practical applications, the nanocasting pathway provides an opportunity to create the

Figure 8.44 Illustration of the nanocasting pathway. Reproduced from [210]. Copyright (2005) Acade´mie des sciences, Elsevier Paris

550

Chemistry of Zeolites and Related Porous Materials

materials with desired macroscopic shape directly from the template with the macroscopic shape. Several novel strategies have been developed in the past decade. The replication process can be practically perfect; the template is actually used as a true mold to produce the mesoporous materials with controllable pore size, pore shape, and distribution. Two kinds of template, viz. hard template and soft template, are usually available for nanocasting processes. The true liquid crystal templating synthesis can be considered a soft-template process. In general, the hard template means an inorganic solid. For example, mesoporous silica as a template to replicate other materials, such as carbon or metal oxides, by which the pore structure of the parent can be transferred to the generated porous materials. A 3-D pore network in the template is necessary to create a stable replica. Mesoporous silica and carbon are commonly used templates for nanocasting synthesis. Ordered mesoporous silica seems to be an ideal hard template, which can be used as a mold for other mesostructures with various compositions, such as ordered mesoporous carbon and metal oxides. Mesoporous silicas with various different structures are available, and silica is relatively easily dissolved in HF or NaOH. Alternatively, mesoporous carbons with a solid skeleton structure are also suitable choices as hard templates due to their excellent structural stability on thermal or hydrothermal and chemical treatment. A pronounced advantage of carbon is the fact that it is much easier to remove than silica by simple combustion. The nanocasting synthesis of mesoporous carbon by using mesoporous silica as template will be discussed in detail in the section on mesoporous carbon. In many cases, silica is unsuitable for synthesizing framework compositions other than carbon, since the leaching of the silica typically affects the material which is filled into the silica pore system. To obtain well ordered porous products by nanocasting, two requirements should be satisfied by the precursor of the expected composition. (i) The precursor should have a high solubility in a suitable solvent or should exist in the liquid state to maximize the loading amount and produce a sufficiently rigid skeleton in the calcined solid to avoid collapse. Basically, a very high concentration of the precursor solution, if at all possible, the neat precursor, is highly recommended. (ii) The resulting material should have a melting point higher than the temperature at which the carbon templates are combusted if the mesoporous carbon is used as template. Silica has an excellent thermal stability, and TEOS as silica precursor allows a high filling degree of the pore system of the porous carbon. In addition, the moderate hydrolysis rate helps to avoid a too rapid and vigorous reaction. Alternatively, an aqueous Na2SiO3 solution can be used as silica precursor.[211] Zhao has demonstrated that 3-D mesoporous silica can be used as hard template and hydrated metal nitrates as metal source to fabricate various mesostructured crystalline metal oxides (Co3O4, Mn2O3, CeO2, and In2O3).[132,212] The nanocasting pathway can be extended to other nonsilica compositions which are not accessible by solution-based methods. Roggenbuck and Tiemann[213] have succeeded in the synthesis of hexagonal magnesium oxide via a nanocasting pathway using CMK-3 carbon as template. 8.5.5

Synthesis Parameters and Extreme Synthesis Conditions

The synthesis parameters in determining the degree of mesostructure, mesophase composition, and morphologies of product have been emphasized and discussed on the

Synthesis, Structure, and Characterization of Mesoporous Materials

551

basis of synthesis results. There are many synthesis parameters in mesoporous-material synthesis. The main parameters include: (1) inorganic wall components and their potential of crystallization; (2) inorganic precursors and their hydrolysis and condensation kinetics; (3) surfactant type (cationic, anionic, and nonionic) and structure (surfactant-packing parameter, g); (4) concentration of surfactant (molecular, micellar, liquidcrystal templating); (5) reactant composition and their ratios; (6) pH value of media; (7) reaction temperatures; (8) synthesis time; (9) additives (salts, organic molecules); (10) the procedure of adding the reactants; (11) solvent and co-solvent; (12) synthesis methods; (13) requirement of product macroscopic shape (monolith, film, fiber, nanoparticles); (14) post-synthesis treatment; (15) the methods for removing the templates. Inorganic Species The sol–gel process, coordinate chemistry, and condensation reactions of inorganic species will affect the formation of the mesostructure. The physical and chemical properties of inorganic species should be considered when choosing inorganic starting materials. Interactions such as static electric interaction and hydrogen bonding should exist between the inorganic species and template. The processes that drive the co-assembly of organic and inorganic units into a 3-D continuous composite with spatially distinct organic and inorganic regions of the nanostructure are strongly correlated, which a priori makes the separation of the various contributing factors difficult to resolve. Certainly, one would expect that the process of mesophase organization would be strongly coupled to the time-dependent polymerization kinetics of silica species at the inorganic–organic interface.[65] In order to separate the effects of silica polymerization from the thermodynamics of mesophase self-assembly, Stucky[65] used low temperatures and careful pH control to control silica polymerization relative to the overall mesophase assembly. This approach has been used to show that, in the absence of inorganic polymerization, these mesophases have liquid-crystalline properties similar to those of conventional lyotropic liquid-crystal systems. In order to maintain liquid-crystal-like properties and optimize long-range composite ordering during polymerization of the inorganic species, the inorganic and organic domains must be able to reorganize on the same kinetic timescale into mutually compatible configurations. Temperature The effect of temperature on the condensation of silica has been mentioned before (e.g., stabilization of mesoporous silica). The effect of the temperature on the formation of the mesostructure can be understood by considering the g parameter of quaternary ammonium surfactants. As the mixture gel is heated, the conformational disorder of the surfactant tail increases, increasing the surfactant molecular volume and as a result the g value. At the same time, the repulsion of the charged headgroups is also increased with heating, which leads to an increase in the headgroup area ao value. Temperature may affect the structure or stability of products, which can be seen from the early discussions in this chapter. Surfactant Concentration In mesophase synthesis, both the silica and surfactant show similar effects on the formation of mesophases. The effects of the surfactant concentration can be explained in two ways. One is packing of the surfactant, and the other is charge density matching

552

Chemistry of Zeolites and Related Porous Materials

between the surfactant and silica. The g value of the surfactant is affected by the electronic environment around the surfactant. The lower the surfactant concentration, the smaller the amount of micelle that is formed, suggesting that the micelles are surrounded by more inorganic species or protons than in the systems with higher surfactant concentration. Thus the synthesis behavior at lower surfactant concentration was similar to that at high inorganic concentration or higher acidity. Organic Additives Organic additives are particularly effective in controlling phase and interface geometry during the synthesis of both mesoporous inorganic solids[1,2,64] and lyotropic mesophases of surfactant–water binary system. We have already mentioned the use of organic additives in some syntheses. The control comes from being able to ‘solvate’ the interface headgroup, palisade, and hydrophobic regions associated with the organic surfactant arrays (see Figure 8.45). For example, when a hydrophobic, non polar organic additive such as TMB is added, it seeks the most hydrophobic region that is at the tail end of the surfactant array and swells the micelle size. Both v and l are affected and the net result can be either a phase change or an increase in the effective pore or cage size. Thus, when TMB is added as a swelling agent, relatively large pore-size changes are observed. This approach has been used in large-pore MCM-41 synthesis and frequently, but not always, works for other mesoporous silica syntheses.[2,8,9] C16-3-1 gives SBA-2 with a ¼ 6:2 nm and c ¼ 10 nm (based on P63 =mmc symmetry) when TMB/TEOS ¼ 1.1, while a ¼ 5:4 nm and c ¼ 8:7 nm without TMB. In these examples, the result is as if a longer-chain surfactant (increased l) has been used to increase the pore size of the product. However, phase changes in some instances are also induced. C16TMAþ favors

Image Not Available In The Electronic Edition

Synthesis, Structure, and Characterization of Mesoporous Materials

553

MCM-41 over a wide range of reaction compositions. At moderately high pH values, if TMB is added, the MCM-41 is replaced by a lamellar mesophase, suggesting that the increased surfactant tail volume v is more important. If this lamellar-phase sample is heated before silica condensation, it reverts back to the hexagonal mesostructure.[65] A suitable polar additive is able to enter the hydrophilic–hydrophobic palisade region (first few carbon atoms) of the micelle, with a relative increase in the volume of the hydrophobic core to form surfactant molecular aggregates with lower-curvature intersurfaces, e.g., from sphere to rod. Thus when t-amyl alcohol, a polar additive, is added to the synthesis mixture in basic synthesis media, the SBA-2 product is replaced by MCM-41 when C16-3-1 as the template (see Figure 8.45). Under acidic synthesis conditions, one can make SBA-1 using C16TEA as template without t-amyl alcohol, but SBA-3 if t-amyl alcohol is used. Another example is the effect of EtOH on the formation of MCM-48 as we mentioned before. The effect of addition of polar additives is quite predictable and one can very effectively use this to generate desired mesophases. TMB isomers have a different effect[161] on the synthesis of mesoporous silica with CTAB as template in acid system. 1,3,5-TMB tends to be associated with the hydrophobic part of the surfactant micelle. One-side-substituted 1,2,3-TMB can enter the hydrophobic– hydrophilic palisade region of the micelle, with a relatively large increase in the volume of the hydrophobic core to form surfactant molecule aggregates with a lower-curvature surface. pH The pH value of solution affects the hydrolysis and condensation of silica species, which will determine the synthesis pathway and product mesostructure. The pH of solution may affect other reactions and interactions in the solution or synthesis mixture and hence result in the formation of different materials. For the acidic synthesis system, silica condensation causes the positive charge density of the silicate network to decrease. It can be considered that the organic surfactants pack to form a high surface curvature to adjust the effective headgroup area, maintaining charge matching in the interface so that the higher-curvature mesophases are formed under higher-acidity conditions. The mesophases are always transformed from the lower-curvature one into the higher-curvature one in the acidic synthesis gel. When the classical surfactant CnTMA is used as template, the acidic medium is advantageous in producing various regular morphologies and macroscopic shapes (e.g., film, sphere, fiber), while basic media are advantageous in the product’s structural quality (e.g., high-quality XRD pattern). In the synthesis system with anionic surfactant N-myristoyl-L-glutamic acid as template and N-trimethoxysilylpropyl-N,N,N-trimethylammonium chloride as CSDA, the packing of the micelle was controlled by simply adjusting the neutralization degree of the C14GluA surfactant. Different mesophases, ranging from tetragonal P42/mnm (cage type, AMS-9), cubic Fd3m (cage type, AMS-8), to 2-D hexagonal p6mm (cylindrical, AMS-3), and an bicontinuous double-diamond cubic Pn3m mesophase (AMS-10), were obtained by decreasing the amount of NaOH that was added to the reaction system (Figure 8.46).[105] The neutralization degree of the surfactant increased with the amount of NaOH. AMS-10 may exhibit a lower curvature close to bicontinuous cubic Ia3d from the sequence of the mesophases. The effect of the neutralization degree

554

Chemistry of Zeolites and Related Porous Materials

Figure 8.46 XRD patterns of calcined mesoporous silica prepared with C14GluA and TMAPS. The compositions of the gels are C14GluA/TMAPS/TEOS/H2O/NaOH ¼ 1 : 1.5 : 15 : 1983 : x; a) x ¼ 2 (P42 =mnm), b) x ¼ 1:5 (Fd3m), c) x ¼ 1 (p6mm), and d) x ¼ 0:75 (Pn3m; AMS-10). Reproduced with permission from [105]. Copyright (2006) Wiley-VCH

on the formation of mesostructures can be explained in terms of the surface-charge density of the surfactant micelles, which increases with the amount of NaOH added to the carboxylic acid surfactant solution. It is useful to introduce the surfactant-packing parameter, g ¼ v=al. The lower charge density contributes to a partial decrease in the electrostatic repulsion between the charged surfactant headgroups and a decrease in the effective headgroup area of surfactant, a, therefore resulting in an increase in the g value. It is well known that the g parameter of lyotropic liquid-crystal phases increases in the order: micellar tetragonal P42 =mnm, micellar cubic Fd3m < cylindrical 2-D hexagonal p6mm < bicontinuous cubic Pn3m. Thus, it is reasonable that a lower charge density on the micelle surface facilitates the formation of the cubic Pn3m mesophase with a larger g parameter. Effects of Anions and Salts Inorganic anions are the most common component in the synthesis which includes the counterion of inorganic precursor and surfactant, acid, base or salt additives. The acidic route is interesting in that it offers versatile structures and morphologies due to its weaker surfactant–silicate interaction in S þ X  Iþ , which suggests that the presence of the counteranion is important for the acidic synthesis systems.

Synthesis, Structure, and Characterization of Mesoporous Materials

555

The presence of various counteranions in the interfacial region of the silicate– surfactant mesophase introduces opportunities for manipulation of the phase structure.[161] Well-ordered P63 =mmc, Pm3n, p6mm, and Ia3d mesoporous silica materials can be synthesized with the same surfactant (cetyltriethylammonium bromide), depending on the kind of acid used. The formation of various mesophases is strongly affected not only by the anions but also by the reactant composition such as the acidity, the molar ratio of TEOS to surfactant, and the molar ratio of anion to surfactant in the synthesis mixture. For the synthesis system with CTEABr as template at 0  C for 1 d, the P63/mmc mesostructures are obtained by using H2SO4 and HCl as the acid; H2SO4 gives this mesophase over a wider composition range than does HCl. The Pm3n mesostructures are obtained with three acids: H2SO4, HCl, and HBr; HCl gives this mesophase over the widest composition range. The p6mm mesostructure is synthesized over a wide range of reactant compositions when HBr or HNO3 is used. Only HNO3 was able to produce the Ia3d mesophase, although the range was narrow. The XRD pattern of the material synthesized with H3PO4 shows one broad peak, which cannot be indexed to any ordered mesophase. The other acids HF, HI, and CH3COOH produced an amorphous solid, an opaque gel, and a transparent gel, respectively. Too large and too small counteranions are not able to form the ordered mesophase, and instead form disordered or amorphous solid products. The weak acids did not have the ability to protonate the silanol group, and the formation of mesoporous materials was not observed even at a high acid concentration. The effect of counteranions on the formation of mesostructures can be explained in terms of the adsorption strength on the headgroups of the micelle. The X ions are more or less hydrated in the surfactant solution. Less strongly hydrated ions have, in general, smaller ionic radii and bind more strongly on the surface of the surfactant micelles. The adsorption ability or the aggregation number is reported to decrease in the following    order: ½SO2 4 > Cl > Br > NO3 . The well known binding-strength Hofmeister series of anions for the cationic surfactant is increased in the following order: SO42 < Cl < Br < NO3. It is also parallel to the effectiveness of the counterions in decreasing the CMC of the cationic surfactant. For the synthesis with nonionic block polymer as template, inorganic salts can dramatically widen the syntheses domain (in temperature, surfactant concentration, etc.) and broaden the range of surfactants that can be utilized to produce highly ordered mesostructures.[214] Synthesis through Acid–Base Pair Zhao[215] described how the self-adjusted inorganic–inorganic (II) interplay between two or more inorganic precursors is guided by acid–base chemistry considerations. A wide variety of highly ordered, large-pore, homogeneous, stable, and multicomponent mesostructured minerals, including metal phosphates and metal borates, as well as various metal oxides and mixed metal oxides, have been obtained based on this route. It takes a different perspective to address the influence of inorganic–inorganic (I–I) interplay on the synthesis of mesoporous materials (see Table 8.7). This ‘acid–base pair’ concept is particularly important for the preparation of multicomponent mesoporous materials. The routes are especially useful for metal phosphates, mixed metal phosphates, metal borates, metal oxides, and mixed metal oxides.

556

Chemistry of Zeolites and Related Porous Materials

Table 8.7 Syntheses of various mesoporous metal phosphates through ‘acid–base pair’ principles Material TiPO TiPO AlPO ZrPO NbPO

Acid–base pair A: Chloride–alkoxides A: Chloride–alkoxides D: Acid–salt A: Chloride–alkoxides B: Ester–salt

Inorganic 2 (I2)

Inorganic 1 (I1)

Surfactant Structure

PCl3 PCl3 H3PO4 PCl3 OP(OCH3)3

Ti(OC3H7)4 Ti(OC3H7)4 AlCl3 Zr(OC3H7)4 NbCl5

P123 F108 P123 F108 F108

p6mm, Ia3d Im3m p6mm Im3m Im3m

In the selection of possible pair combinations, normally the larger the acidity or alkalinity difference between the metallic or nonmetallic sources, the more effectively the pairs will form and function. No extra reagents (such as HCl or NH3) are added to adjust the pH of the desired sol–gel reactions during the entire preparation. Instead, two or more inorganic species were used to self-generate a reaction medium with the correct acidity to favor the sol–gel process of the solvated products, and give a self-adjusted sol– gel synthesis. The solvents used must be polar organic solvents. Amphoteric solvents, such as C2H5OH or CH3OH, are recommended because they can easily act as oxygen donors and aid proton transfer within the synthetic system. The multicomponent assembly can be guided by the relative acid–base character of all the species present, so to assemble a multicomponent (I1, I2) composition, it is necessary to match acid–base interactions of the various species present during nucleation of the mesostructured phases in the order I1 I2  I1 I1 ; I2 I2 ; and OðI1 I2 Þ  OI1 ; OI2. Synthesis at High Temperatures It can be expected that the silica-condensation level will be enhanced by increasing the crystallization temperature. However, mesostructured materials prepared at relatively low temperatures (in general, < 150  C) generally exhibit imperfectly condensed walls with large numbers of terminal hydroxy groups, resulting in low hydrothermal stability of the product. Surfactant molecules will not be able to template mesostructure formation due to unfavorable conditions for micelle formation at the higher temperatures. Fluorocarbon surfactants are a kind of stable surfactant which are widely used at high temperatures (> 200  C). However, owing to the rigidity and strong hydrophobicity of the fluorocarbon chains, fluorocarbon surfactants cannot form an ordered surfactant micelle. Xiao[216] reported syntheses of the stable mesoporous silica-based material, JLU-20, over the temperature range 160–220  C by using high-temperature stable surfactants of fluorocarbon–hydrocarbon surfactant mixtures and cationic modified ionic liquids as templates. The XRD pattern of calcined JLU-20 showed four clear peaks that can be indexed as p6mm hexagonal symmetry. In contrast, ordered mesostructured silica cannot be formed under the same synthesis conditions in the absence of fluorocarbon surfactant. Notably, JLU-20 is much more hydrothermally stable than is SBA-15. The unit cell of JLU-20 does not contract during calcination at 650  C for 2 h and demonstrates its excellent thermal stability. The 29Si MAS NMR

Synthesis, Structure, and Characterization of Mesoporous Materials

557

spectrum of the as-synthesized JLU-20 provides direct evidence of the extent of silica condensation. JLU-20 gave a very high Q4/Q3 ratio (6.5). In contrast, SBA-15 has typical peaks corresponding to Q2, Q3, and Q4 silica species respectively, and the ratio of Q4/Q3 þ Q2 is 1.9. Later, another high-temperature synthesis was reported: mesoporous carbon (CMK-3) used as the template to synthesize mesoporous silica materials (RSC-3) in strongly acidic media at high temperatures (160–240  C).[217] RSC-3 samples exhibit 2-D hexagonal (p6mm) mesostructure and much higher hydrothermal stability than do conventional silica materials synthesized at relatively low temperatures (< 150  C). 29Si MAS NMR spectra indicate that as-synthesized RSC-3 samples are primarily made up of fully condensed Q4 silica units (d  112 ppm) with a small contribution from incompletely cross-linked Q3 (d  102 ppm) as deduced from the high Q4/Q3 ratio of 5.5–9.5, implying the fully condensed walls of RSC-3. Such unique structural features should be attributed directly to the high-temperature synthesis, which could be responsible for the observed high hydrothermal stability of RSC-3. Syntheses in Nonaqueous Solvents and Supercritical Fluids Nonaqueous solvent synthesis systems were used for silica or other composition (e.g., transition metal oxides,[26] sulfide mesophase[26]). Most syntheses through EISA and acid–base-pair pathways were done in nonaqueous synthesis media. The solvents used are generally polar organic solvents. Amphoteric solvents, such as glycol, C2H5OH, or CH3OH, are recommended because they can easily act as oxygen donors and aid proton transfer within the synthetic system. Non aqueous synthetic media are used to maximize the utility of these synthesis methods; a solvent that is compatible with the reactants, solvolysis, and polymerization reactions of the inorganic precursor reactants must be chosen. Well ordered mesoporous silicate films were prepared in supercritical carbon dioxide.[218] In the synthesis in aqueous or alcoholic solution, film morphology of preorganized surfactants on substrate cannot be fully prescribed before silicaframework formation, because structure evolution is coincident with precursor condensation. The rapid and efficient preparation of mesostructured metal oxides by the in situ condensation of metal oxides within preformed nonionic surfactants can be done in supercritical CO2. The synthesis procedure is as follows. A copolymer template is prepared by spin-coating from a solution containing a suitable acid catalyst. Upon drying and annealing to induce microphase separation and enhance order, the acid partitions into the hydrophilic domain of the template. The template is then exposed to a solution of metal alkoxide in humidified supercritical CO2. The precursor diffuses into the template and condenses selectively within the acidic hydrophilic domain of the copolymer to form the incipient metal oxide network. The templates did not go into the CO2 phase because their solubility is very low. The alcohol by-product of alkoxide condensation is extracted rapidly from the film into the CO2 phase, which promotes rapid and extensive network condensation. Because the template and the metal oxide network form in discrete steps, it is possible to pattern the template via lithography or to orient the copolymer domains before the formation of the metal oxide network.

558

Chemistry of Zeolites and Related Porous Materials

8.6 Composition Extension of Mesoporous Materials The change of composition of mesoporous materials can be done by direct synthesis and post-synthesis modification. Now, the composition of mesoporous materials can be extended to nonsilica oxides, phosphates, sulfides, even metals. The study of nonsilica mesoporous materials started much later than that for silica-based materials. The main reasons include: the hydrolysis and condensation reactions of transition metal precursors is difficult to control; the inorganic wall easily crystallizes and results in the loss of mesostructures; the synthetic procedure is difficult to repeat. 8.6.1

Chemical Modification

The one advantage of mesoporous materials is their huge inner surface which can be easily modified. In order to achieve desired applications such as catalysis, adsorption, chromatography, and chemical sensing, the immobilization of functional groups on mesoporous materials is indispensable, although the mesoporous silica materials directly synthesized can be used in some cases. It is possible to modify the mesoporous material frameworks with a wide variety of organic groups. For modifying the mesoporous materials through a covalent (or other kind of) linkage between functional groups and silica framework, the major pathways (see Figure 8.47) include grafting

Figure 8.47 Incorporation of organic functions in mesoporous silica: (a) Surface grafting of organic functions on the mesopore walls by post-synthesis; direct incorporation of organic functions by co-condensation of organosilanes (b) or bridging silasesquioxanes (c). Reproduced with permission from [51]. Copyright (2002) American Chemical Society

Synthesis, Structure, and Characterization of Mesoporous Materials

559

approach (the organic functionalities are bound on the silica surface by a post-synthesis treatment), co-condensation of tetraalkoxysilanes and terminal organotrialkoxysilanes (direct synthesis), substitution of inorganic wall with nonsilicon atoms, ion-exchange, or by loading active species (e.g., Cu, Ni), and techniques related with periodic mesoporous organosilicates. Here the grafting and co-condensation methods will be discussed in detail. Grafting Grafting is one of the modification methods for pre-synthesized mesoporous silica (see Figure 8.48[219]), which makes use of the silanol groups present on the silica surface of the ordered mesoporous silica, because they are the anchor sites for metal species or silane-coupling agents.[220] This method can be carried out by one or more procedures. Various functional groups including amino, thiol, and alkyl groups can be introduced by direct reaction of organosilanes onto the silica surface. Other functionalities can be fixed to the previously introduced functional groups through covalent bonding or molecular recognition. Similarly the pre-introduced functionalities are often converted into the other groups by chemical reaction. The grafting of a specific organosilane into the preformed mesoporous silica is widely used for the introduction of organic functions. The distribution and concentration of functional groups are influenced by the reactivity of the organosilane and their accessibility to surface silanols, which are limited by diffusion and steric factors. In post-synthesis functionalization, it is often desirable to start with a large number of surface silanol groups (Si OH) on the mesoporous silica wall after the removal of the surfactants when a high surface coverage of functional groups is preferred. However, high-temperature calcination usually precedes functionalization in standard methods. This would lead to more surface silanol condensation and reduce the Si OH density for the subsequent surface modification. The surface silanols can be recovered by adding a trace amount of water. Moreover, excess of water will also cause self-condensation of the silanes among themselves. Several laboratories have developed the methods of mesoporous silica surface functionalization with simultaneous removal of the surfactant and grafting of organic functionalities without prior calcination. A convenient and highly controllable approach to the surface functionalization of mesoporous silica[221] employs an alcoholic solution of

Figure 8.48 Schematic of reactions between MCM-41 and TiO2 precursors. Reproduced with permission from [219]. Copyright (2000) Royal Society of Chemistry

560

Chemistry of Zeolites and Related Porous Materials

silanes to allow the simultaneous surfactant–silyl exchange process, which results in a more uniform monolayer coverage of the surface and a higher number of surface attachments of silane. The loading of surface silane, depending on the solution concentration of the silane, can be sufficiently controlled to be by monolayer. The surfactant–silyl exchange process is greatly favored when applied to mesoporous silica prepared under acidic conditions because of the weaker interactions between the silica walls and the surfactants compared with those in materials prepared under basic conditions. The mesoporous silica from acidic systems is believed to possess more abundant silanol groups. The grafting on both outer surface and inner surface of mesoporous silica can be done separately. First, by modifying the outer surface with as-made material, then removing surfactant and getting the inner surfactant available. The second modification is carried out on the outer surface-modified mesoporous material. This method can result in two different functional-group-modified materials.[222] Mesoporous silicas with various pore sizes are hydrophobic by silylation with silanes. Changes in the pore structure as a result of the silylation reactions are monitored in order to assess the distribution of the hydrophobic groups. Extensive polymerization of dimethyldichlorosilane causes blocking of the micropore fraction. For silica with pore sizes in the supermicroporous range (2 nm), this leads to hydrophobization of almost exclusively the outer surface. While for trimethylchlorosilane a smaller number of molecules react with the surface, modification is more homogeneous and an open structure is optimally preserved. Both silanes lead to lower surface polarity and increased hydrothermal stability, i.e., preservation of the porous structure during exposure to water.[223] Grafting can result in useful materials; for example, introduction of Ti into mesoporous silica is showed in Figure 8.48. TiO2 was dispersed in the channels of mesoporous silica and formed the bonds of Si O Ti species. The TiO2 is a singlelayer dispersion without crystal TiO2 phase. The Ti-MCM-41 showed good catalysis performance.[16] Another good example is for wastewater treatment. Feng[21] functionalized mesoporous silica with  SH groups. The resulting material is a very good adsorbent for heavy metals. Alkylphenols and alkylanilines can be asorbed by alkylgrafted MCM-41. The rate of adsorption could also be optimized based on the functional group. Co-condensation Another major approach for functionalization of mesoporous silica is the cocondensation method where an organosilane is condensed or polymerized together with conventional silica sources such as TMOS and TEOS. The one-pot pathway of the co-condensation method provides several advantages such as homogeneous distribution of the functional groups, unique pore size, and short preparation time. The organic functionalities have to be selected in consideration of experimental conditions in the synthetic process, hydrothermal treatment, and solvent extraction. The introduced functionalities are also converted into another group. For example, thiol-bearing hybrid mesoporous silica is first synthesized and then an appropriate oxidation procedure converts the thiol into a sulfonic acid. Treatment of SBA-15 with  CN groups with H2SO4 can remove surfactant P123 and convert  CN groups into  COOH groups.

Synthesis, Structure, and Characterization of Mesoporous Materials

561

Co-condensation vs. Grafting The post-synthesis grafting procedure has often been the method of choice since the early development of mesoporous silica. Co-condensation procedures are also popular because they are simple and often give higher loading. However, all methods can lead to an inhomogeneous distribution of the organic groups in the pores. Different pathways have their limitations and inherent drawbacks: When using the grafting approach, the diffusion limitations of the rather large grafting species within the pores may lead to an inhomogeneous distribution, meaning that they are often located in the vicinity of the pore openings. Therefore the negative effects of pore blocking accompanied by a limited loading with the organic groups and reduced free-channel volume may occur. Using cocondensation methods, silasesquioxane precursors with terminal organic groups must be co-assembled to some extent with TEOS to form an ordered periodic mesoporous structure for most cases. The negative effect of pore blocking is circumvented and a more homogeneous distribution of the organic functionalities can be realized, because the organic units are direct parts of the mesoporous framework. The inhomogeneity is mainly due to the different reaction rates (hydrolysis and condensation) of both components, which can lead more likely to homo- than to co-condensations and the tendency of the (large) organic groups to protrude into the pores. The second disadvantage is that the organic content of the material is usually limited to around 25%, otherwise a structural collapse of the mesophase may occur, because there may be phase separation at high loadings. This disadvantage can be avoided by using a single silicon precursor in some cases. 8.6.2

Synthesis Challenges for Nonsilica Mesoporous Materials[224]

There are enormous difficulties in synthesizing ordered mesoporous crystalline transition metal oxides. One of the major problems in texturing transition metals is the high reactivity of metal alkoxide precursors towards hydrolysis–condensation reactions, which usually leads to a very fast nucleation, promoting an uncontrolled segregation of a dense oxidic phase. It must be kept in mind that hydrolysis–polymerization processes are orders of magnitude faster for transition metals than for silica. In principle, there are two ways to control the fast processes of hydrolysis and condensation. The first approach consists of working in dilute alcoholic media or with very low water-to-metal ratios. The second approach consists of the addition of condensation inhibitors such as chelating agents. Two general aspects for adjusting self-assembly and inorganic wall composition are: reaction rate and neutral electric point of inorganic precursor, and the creation of an inorganic–organic interface for both silica-based materials and nonsilica materials preparation. For nonsilica synthesis, the following should be considered: (1) Control condensation process, avoiding the growth of an inorganic network (too fast and random growth). (2) Enhance nano- or sub-nanophase separation (e.g., increase the inorganic– organic interaction). (3) Control the curvature of product structure, avoiding the formation of lamellar phase. Special Synthesis Techniques for Nonsilica-based Materials According to our understanding of the formation of mesoporous materials and knowledge of silica-based materials, the following methods have been developed and used in nonsilica mesoporous materials. (1) Select a suitable inorganic precursor, adjust the

562

Chemistry of Zeolites and Related Porous Materials

hydrolysis rate or precursor, pH of solution, and reactant concentration. (2) Reduce condensation of inorganics by adding acid, complex agent, etc. (3) Utilize nonaqueous media (reduce the concentration of water). (4) Use solvent evaporation to achieve selfassembly. (5) Change the oxidation state of the main framework element. (6) Use preformed nanocluster, particles, seeds, or sol. (7) Use block copolymer as template. It is believed that the interactions taking place between PEO-based surfactants and the metallic centers are a key feature in obtaining mesostructured transition metal oxides. Mesostructural transition metal oxides have such poor thermal stability that the ordered mesostructure is easily destroyed during their synthesis by heating.[26] The main reason for the collapse is believed to be crystallization of the wall during heating. One strategy[225] is to design a mesoporous nanocomposite which is similar to the ‘bricks and mortar’ configuration in architectural engineering. The designed structure is made up of a large number of functional nanocrystals and a smaller quantity of the multicomponent glass phase. The nanocrystals that govern the properties of the nanocomposites were designed as basic building blocks of the mesopores. The glass phase was designed to do the following: (i) Provide predictable and controllable properties, because glass is a good host for most oxides; (ii) help in forming and maintaining a threedimensional network composed of oxygen polyhedral units during molecular selfassembly; (iii) control in situ crystallization of materials on a nanometer scale; (iv) form a ‘glue’ between the nanocrystals. The compositions for glass phase and crystalline phase can be the same or different. Stability of Non-silica-based Materials Crystallization of the amorphous wall of transition metal oxides materials during heat treatment results in collapse of the uniform mesostructure. The following methods can help to increase the stability of nonsilica mesostructured solids. (1) Increasing inorganic wall thickness. (2) Wild hydrothermal treatment before calcination. (3) Heating slowly during calcination, or multiple-step calcination under inert atmosphere. (4) Extraction of all or part of the surfactant before calcination. (5) Reducing the crystallization of inorganic wall by adding cations or anions (sulfate, phosphate, etc.) or another phase. (6) Use of preordered nanobuilding blocks. 8.6.3

Metal-containing Mesoporous Silica-based Materials

The introduction of heteroatoms into the mesostructured silica wall will change the chemical properties of mesoporous silica materials. Many examples have been reported. Here, only a few of examples are introduced. Ti-containing Silica Mesoporous Materials We choose Ti-containing materials as an example to discuss heteratom-containing silica material and nonsilica mesoporous materials, because Ti-containing materials are useful catalysts and were studied extensively. Microporous titanium silicate (e.g., TS-1, Ti-b, Ti-ZSM-12, Ti-mordenite) is an effective molecular-sieve catalyst for the selective oxidation of alkanes, the hydroxylation of phenol, and the epoxidation of alkenes with aqueous H2O2. The range of organic compounds that can be oxidized is greatly limited, however, by the relatively small pore size (about 0.6 nm) of the host framework.

Synthesis, Structure, and Characterization of Mesoporous Materials

563

Since the first synthesis of Ti-MCM-41, several Ti-containing mesoporous molecular sieves, such as Ti-HMS, Ti-MCM-48, Ti-MSU-1, and Ti-SBA-15 have been synthesized and characterized. In most instances the syntheses were accomplished using organic precursors, namely, TEOS and titanium tetraethoxide (TEOT) or titanium tetrabutoxide (TBOT) as the Si and Ti sources, respectively. It is easy to form a homogeneous gel using organic precursors. Ti-MCM-41 was prepared by either grafting titanium precursor onto surface silanols via a post-synthetic procedure or depositing titanium precursor on MCM-41 from the sol obtained by controlled hydrolysis of a titanium alkoxide precursor followed by calcination. Corma[7] has prepared Ti-MCM-41 by direct hydrothermal synthesis. Ti-MCM-41 is the first example of ordered mesoporous titanosilicates, and catalytic results exhibited good properties for the oxidation of bulky reactants under mild conditions. Thomas[16] reported direct grafting of an organometallic complex onto the inner walls of mesoporous silica MCM-41 to generate a shape-selective catalyst with a large concentration of accessible, well spaced, and structurally well defined active sites. Attachment of a titanocene-derived catalyst precursor to the pore walls of MCM-41 produces a catalyst for the epoxidation of cyclohexene and more bulky cyclic alkenes. Through the grafting route, titanocene complex on mesoporous silica exhibits high activity for catalytic conversion of bulky molecules The following methods have been used for the synthesis of Ti-SBA-15: (1) Direct synthesis, (2) treatment of SBA-15 with a Ti-containing compound, (3) synthesis of Ti-SBA-15 with microwave heating, (4) Ti-containing zeolite seeds as precursor, e.g., MTS-9.[204] Under the extreme acidic conditions for SBA-15 synthesis, most Ti-containing compounds hydrolyse much faster than do silicon compounds. It is difficult to synthesize high-quality Ti-SBA-15 with Ti highly dispersed in the wall. Zhao[226] successfully synthesized Ti-SBA-15 by a direct approach under conventional hydrothermal conditions. The products obtained by treating titanium butoxide with acetylacetone were used as the titanium precursor. The substitution of Ti for Si does not change the highly ordered mesostructure of SBA-15. The titanium is totally tetrahedral and highly dispersed. The Ti-SBA-15 demonstrates relatively high catalytic ability on the oxidation of cyclohexene. Other Metal-substituted Silica Mesoporous Materials Owing to the amorphous wall of mesoporous material, the heteratom substitution is much easy than that for zeolite. There are many elements that have been introduced into mesoporous silica materials (MCM-41, MCM-48, SBA-15, etc.); for example, B, Ga, Cu, Fe, Co, V, Mn, Sn, Cr, Mo, W, Zr, Nb, etc. 8.6.4

Inorganic–Organic Hybrid Materials

In the inorganic–organic hybrid materials, the inorganic component provides the mechanically, thermally, structurally stable framework (wall), and the organic component enhances the structure and provides more possibility for modification of properties. Hybrid organic–inorganic materials can be divided into two distinct classes. In class I, only weak bonds (e.g., hydrogen bond, van der Waals’) give cohesion to the whole

564

Chemistry of Zeolites and Related Porous Materials

structure. In class II the organic and inorganic components are linked together by strong chemical bonds (e.g., covalent bond). In the construction of the mesostructured hybrid network, the interactions between the surface of the inorganic precursors and the organic texturing agent are of paramount importance. Traditional mesostructure synthesis routes resort to class I hybrid intermediates that associate organic and inorganic components though weak interactions. The ligand-assisted templating approach is based on the chemical bonding (class II hybrids) between molecular precursors and the complexing headgroups of the surfactants. Both types of hybrids could be useful in the synthesis of hierarchically ordered materials. Class I hybrids generally allow more versatility in the tuning of the geometric and charge-balance constraints needed at the interface to build highly ordered assemblies. Ordered inorganic–organic hybrid silica-based composites include the following:[48] (1) as-made (uncalcined or unextracted) surfactant–silica mesophase, (2) surface-modified mesoporous silica, (3) periodic mesoporous organosilica (PMO), (4) organiccontaining mesoporous SiO2. Periodic Mesoporous Organosilicas (PMOs)[28,227] Many of the problems of a conventional mesoporous silica material can be avoided while ensuring efficient use of the bridging organic functional groups of the silasesquioxane. PMOs represent an exciting new class of organic–inorganic nanocomposites targeted for a broad range of applications such as catalysis, sensing, adsorption, chromatography, nanowires, low-k materials, and microelectronics. In PMOs, organic and inorganic components are integrated by means of chemical, bottom-up, and self-assembly approaches. The organic functionalities are not only completely homogeneously distributed but possess additionally ordered pore arrangements accompanied with sharp pore-size distributions. Use of bridged silasesquioxanes [(OR)3Si-R-Si(OR)3] as silica precursors for mesoporous silica material permits addition of organic functional groups in the framework without pore blocking. Wide varieties of organic functionalities, even large heterocyclic bridging groups, can be incorporated into the silica framework. Typical surfactant templates for mesoporous silica (e.g., ionic surfactants, nonionic block copolymers) can be used for the synthesis of PMOs. PMOs can have different framework compositions depended on the precursor used. Organosilicas with terminal organic groups can be prepared by grafting organic groups onto the channel walls using the reactivity of the silanol groups in the material or by cocondensing TEOS with an organosiloxane of the type RSi(OEt)3. These modifications result in materials with useful properties, such as alkanethiol groups in the channels, which can bind toxic heavy metals. Organosilicas with bridging organic groups and high organic-group content can be obtained when (EtO)3Si-R-Si(OEt)3 is used as the sole precursor, providing that the group R is sufficiently short and rigid and has favorable condensation kinetics. Because co-assembly with another precursor like TEOS is often not necessary, a homogeneous distribution of the organic groups in the pore walls can be ensured. Thus, higher organic loading and greater avoidance of pore blockage can be achieved than with terminal organic groups because the material consists entirely of SiO3-R-SiO3 building blocks. In PMOs assembled from (EtO)3Si-R-Si(OEt)3, the number of organic groups is restricted

Synthesis, Structure, and Characterization of Mesoporous Materials

565

to one organic group attached to each Si atom according to the exclusive presence of SiO3R building units. Since the organic groups in PMOs are the carriers of the desired properties, it is imperative to find pathways that would allow a higher replacement of bridging O atoms by bridging organic groups within the channel walls. This could lead not only to improved physical properties but also to order in the channel walls and new mesostructured topologies or morphologies. Such high organic-group-content PMOs would consist of SiO2R2, SiOR3, or SiR4 building units. This requires a redesign of the precursor to include these building blocks pre-organized in a molecule that is polymerizable in all three spatial dimensions. The ordered benzene–silica hybrid material[228] has a hexagonal array of mesopores ˚ , and crystal-like pore walls that exhibit structural with a lattice constant of 52.5 A ˚ along the channel direction. The periodic wall periodicity with a spacing of 7.6 A structure results from alternating hydrophilic and hydrophobic layers, composed of silica and benzene, respectively. 8.6.5

Metal Oxides, Phosphates, Semiconductors, Carbons, and Metallic Mesoporous Materials

Many mesoporous oxide materials have been synthesized in the past; for example, Al2O3, TiO2, ZrO2, MnO2, Ga2O3, Nb2O5, Ta2O5, HfO2, Fe2O3 and SnO2. A few examples will be given in this section. Alumina Aluminas are extensively used in catalyst, catalyst support, and adsorption applications, thus control of their surface area and pore-size distributions has received a great deal of attention. The advent of the self-assembly method for synthesizing mesoporous materials from metal oxides gives promise for the production of high surface-area aluminas with narrow pore-size distributions in the mesoporous range. Unfortunately, the synthesis of such materials has proven to be more problematic than that of their silica-based analogs. Nonetheless, a wide range of synthesis procedures[14,229,230] has been reported for templating mesoporous aluminas, including both nonionic and electrostatic assembly mechanisms. The mesoporous aluminas synthesized using a nonionic templating method are thermally stable not only to template removal, but also to prolonged heating at elevated temperature. Therefore, these aluminas would be able to maintain their unique structural features in fairly demanding catalyst preparations and catalytic applications. Unlike sol– gel-derived aluminas, the synthesis temperature used for the hydrolysis and condensation of the aluminum alkoxide did not affect the resulting thermal evolution from the aluminum hydroxide to transitional alumina and the subsequent thermal stability of the transitional alumina. The only observed effect of synthesis temperature was the impact on median pore diameter and pore volume.[231] Titania If mesoporous TiO2 could be prepared with an anatase crystalline wall, it would be a useful material applicable to high-performance photocatalyst and wet-type solar cell applications. Ionic surfactants were used initially in the synthesis of mesoporous titania because their amphiphilic nature provides well organized micelles around which the titania

566

Chemistry of Zeolites and Related Porous Materials

framework can be assembled by electrostatic interactions. The mesoporous hexagonal TiO2 with high surface area and narrow pore-size distribution was first synthesized by a modified sol–gel method with phosphorus surfactants as templates.[12] It is difficult to make mesoporous TiO2 materials with a crystalline wall. One major reason is that phase transformation from amorphous to crystalline by heat treatment or UV light irradiation usually induces the collapse of mesopores because the wall is too thin to retain the 3-D mesoporous structure during crystallization. Shibata[232] demonstrated a direct method for preparing crystallized mesoporous TiO2 by using a sol–gel reaction of TiOSO4 in the presence of CTAB at low temperature for 24 h. Wang[233] reported preparation of mesoporous TiO2 with wormhole-like framework structures by using long-chain amines as the template and titanium isopropoxide as the precursor by using a sonochemical synthesis method. The selective synthesis of anatase and rutile structure of TiO2 could be achieved by using titanium isopropoxide and titanium tetrachloride as the precursors, respectively. Use of a high-intensity ultrasound probe can result in mesoporous TiO2 with a bicrystalline framework of anatase and brookite.[234] The mesoporous TiO2 prepared using traditional phosphate surfactants had low photocatalytic activity, because the strong interactions between surfactant and titania walls resulted in that the surfactant could not be removed completely, by either calcination or solvent extraction. Another problem was that the integrity of the crystalline framework was difficult to maintain under the typical high-temperature thermal or hydrothermal treatment conditions. Thus, nonionic surfactants appeared to be a potential alternative, given that hydrogen bonding mediates the formation of the metal oxide– surfactant composites involved in the inorganic framework organization. The weak surfactant–precursor interactions involved in the synthesis reactions made possible the effective elimination of the block copolymer by solvent extraction. Moreover, the use of acidic synthesis conditions provided titania pore walls composed of nanocrystalline anatase. Yang[26] prepared mesoporous TiO2 by using a block copolymer as the template and TiCl4 as the Ti precursor. Mesoporous TiO2 synthesized using a block copolymer as a template have thick inorganic walls, in which small crystallites of anatase TiO2 can be formed during calcination.[235] The incorporation of Ce3+ ions into the channel wall dramatically improves the stability of the mesoporous structure. Mesoporous TiO2 films have been made by different methods. For example, dipcoating for evaporation-induced self-assembly followed by a straight thermal treatment was employed to make highly organized mesoporous anatase films exhibiting favorable properties for photocatalysis and photovoltaic applications by the hydrolysis– condensation of TiCl4 in the presence of poly(hydroxybutyrate)–poly(ethylene oxide) (PHB–PEO) block copolymer templates.[236] High-optical-quality TiO2 films[237] with cubic or 2-D-hexagonal symmetry were fabricated by combining trifluoroacetate (TFA)modified titanium precursors with PEO-PPO-PEO copolymers. TFA coordinates the titanium center and forms a stable complex that is subsequently organized by the block copolymer species into ordered mesostructures. PEO is predominantly incorporated within the TFA-modified titania, and the PPO environments encompass both microphase separated regions and interfacial regions composed of mixed PPO and TFA-modified titania.

Synthesis, Structure, and Characterization of Mesoporous Materials

567

Other Oxides Various mesoporous oxides have been reported; here only a few examples are listed. Ga2O3-based dodecyl sulfate mesophase[184] with a hexagonal structure was synthesized by the homogeneous precipitation method using urea. On removal of the incorporated surfactants by chemical or thermal treatment the mesophase structure collapsed. Antonelli and Ying synthesized stable hexagonal mesoporous Ta2O5[238] and Nb2O5 through a ligand-assisted templating mechanism.[239] In this approach, the inorganic precursor is covalently bonded to the template throughout the synthesis. Tian[240] prepared hexagonal and cubic phases of manganese oxide mesoporous structures by means of the oxidation of Mn(OH)2. The hexagonal materials form thick walls (1.7 nm) and exhibit exceptional thermal stability (1000  C). The walls of the mesopores are composed of microcrystallites of dense phases of Mn2O3 and Mn3O4, with MnO6 octahedra as the primary building blocks. Aluminophosphates (AlPO) Early efforts towards these resulted only in the production of lamellar aluminophosphates. Later, mesoporous aluminophosphates were synthesized successfully by using CTAB or CTAC as template and Al(iPO)3 or Al(OH)3 as aluminum source.[241] F127 can give ordered large-pore (up to 12 nm) and stable mesoporous AlPO under nonaqueous conditions starting from inorganic precursors based on an acid–base-pair route.[242] Three acid–base pairs, including AlCl3/H3PO4, AlCl3/OP(OCH3)3, and Al(OC4H9)3/PCl3 are confirmed to be efficient for the assembly of periodic mesoporous frameworks. This method can also be applied to synthesize other mesoporous MeAlPOs (Me ¼ Co, Ni, Cu, Zn, Mn, Sn, Sb, B, Cr, Nd, Y, and Ce). Other Phosphates Many mesoporous phosphates including Ti, V, Zr, Fe, and Sn phosphates have been reported. Mesoporous titanium(IV) phosphates with cationic framework topologies obtained by using both cationic and anionic surfactants are reported.[243] They show anion-exchange capacity due to the framework phosphonium cation, and cation-exchange capacity due to defective P-OH groups. Hexagonal, cubic, and lamellar vanadium phosphates were hydrothermally synthesized by using V metal as a reducing agent and a V source.[244] The respective phases were tunably synthesized over the pH ranges 2.63–2.95, 3.00–3.36, and 3.45–4.45. It is suggested that the formation of these mesostructured materials depends on the solution species of vanadium and phosphorus. These mesostructured materials possess an amorphous wall of vanadium–phosphorus oxides. Highly ordered mesostructured zirconium oxophosphate[245] with Pm3n symmetry was obtained by using tri-headgroup quaternary ammonium surfactant CH3(CH2)16CH2N (CH3)2CH2CH2N(CH3)2CH2CH2CH2N(CH3)3Br3 (C18-2-3-1) as a template and zirconium sulfate as an inorganic precursor under hydrothermal conditions. The cubic mesostructured zirconium oxophosphates are thermally stable up to 500  C. In addition, a phase transformation from Pm3n through mixed mesophase to p6mm symmetry was observed through fine tuning of the spacer methylene chain length among the hydrophilic headgroups in the multi-charged amphiphile (from C18-2-3-1 through C18-3-3-1 to C18-3-4-1).

568

Chemistry of Zeolites and Related Porous Materials

Mesoporous Semiconductor Materials Kanatzidis[246] reported the synthesis of large single-crystal particles (
2 micrometer) of chalcogenido mesostructured materials with a highly ordered cubic structure, accessible pore structure, and semiconducting properties. Building blocks with square-planar bonding topology, Pt2þ and [Sn2Se6]4, in combination with long-chain pyridinium surfactants (CnPyBr, n ¼ 18, 20) favor formation of faceted single-crystal particles with the highest possible space-group symmetry Ia3d. The surfactant molecules can be ionexchanged reversibly and without loss of the cubic mesostructure and particle morphology. Mesoporous Carbons Porous carbons are commonly used as adsorbents and catalyst supports. Many porous carbons are known to exhibit periodic structures resulting from the uniform stacking of graphene sheets and periodic arrangement of atoms within these sheets. Carbons with periodic microporous or mesoporous structures have been reported only recently. Zeolites were already employed as templates in the synthesis of microporous carbon with ordered structures.[247] The discovery of ordered mesoporous silica materials opened new opportunities in the synthesis of periodic carbon structures using the templating approach. By employing mesoporous silica structures as hard templates, ordered mesoporous carbon replicas have been synthesized from a nanocasting strategy. The synthesis is quite tedious and involves two main steps: (i) Preparation and calcination of the silica mesophase, and (ii) filling the silica pore system by a carbon precursor, followed by the carbonization and selective removal of the silica framework. Various carbon precursors such as sucrose,[32] furfuryl alcohol,[248] acetylene gas, and phenol–formaldehyde resin[249] can be used for the synthesis of ordered mesoporous carbons. These organic substances, after carbonization, formed rigid carbon frameworks in the mesopores of the silica template. Following formation of the carbon replica the silica substrate can be removed by dissolution in sodium hydroxide or HF solution and the structural order of the carbon frameworks can be retained. These ordered mesoporous carbons exhibit high specific surface areas (typically 1300–2000 m2 g1), uniform pore diameters (2–6 nm), large adsorption capacities (1–2 cm3 g1), and high thermal, chemical, and mechanical stability. For the low-temperature (700–1000  C) synthesis, the resulting carbons are usually amorphous because no graphitization (heating in argon at a temperature between 2000 and 3000  C) was performed. The use of carbon precursors that contain graphitic building blocks (e.g., pitch, acenaphthene) afforded carbons with some degree of graphitization even without an additional graphitization process. The first mesoporous carbon examples are CMK-1[32] with symmetry I4132 from sucrose, and CMK-4 from acetylene with symmetry Ia3d, by using MCM-48 as template. For the synthesis of CMK-1, MCM-48 was impregnated with sucrose and sulfuric acid, both as an aqueous solution. The impregnated MCM-48 was heated to a desired temperature, in the range of 1073–1373 K (800–1100  C), under vacuum or in an inert atmosphere. The sucrose was converted into carbon by such a process using sulfuric acid as the catalyst. Finally, the silica framework was removed by dissolution in aqueous solution containing NaOH and ethanol. Both enantiomeric channel systems of MCM-48 separated by the silica walls corresponding to periodic G-surface were statistically equally filled with carbon,

Synthesis, Structure, and Characterization of Mesoporous Materials

569

maintaining the cubic Ia3d space group with inversion centers on the original G-surface. For CMK-1, removal of the silica wall caused a new XRD peak to appear. This change indicated that the structure of the resultant carbon transforms[32] to a new ordered structure (cubic I4132 or I41/a).[250] The space-group symmetry of CMK-4 was reported to be Ia3d.[250] Later, other mesoporous silicas were used as templates for periodic mesoporous carbons. It is noteworthy that two types of carbon with SBA-15 as template but different framework configurations, rod- and tube-type, respectively, were obtained. The rod-type p6mm mesoporous carbon CMK-3 was synthesized by a complete filling of mesoporous cylindrical channels of SBA-15 silica. The CMK-5 carbon resulted from an incomplete filling of the SBA-15 channels which, after silica dissolution, gave interconnected nanopipes. Ordered mesoporous carbon can also be synthesized by a catalytic CVD method.[251] The ordered carbons possess bimodal pores: the pores arise from the replica frameworks of the template and the pores correspond to carbon nanotubes formed in the channels of the template. The silica materials with various pore-wall thicknesses are suitable as templates for mesoporous carbons with controlled pore diameters.[252] The carbons exhibit wide varieties of pore shape, connectivity, and pore-wall thickness, depending on the silica templates that are synthesized with various structures and pore diameters. The syntheses of mesoporous carbons are summarized in Table 8.8. Figure 8.49 show the XRD pattern of CMK-3 (keeps original symmetry) [34] and Figure 8.50 shows the changes in powder XRD patterns during synthesis of the CMK-1 with its silica template MCM-48 (creates new symmetry).[32] Table 8.8 Synthesis and characterization of mesoporous carbons Symmetry, pore system (pore size, surface area, pore volume)

Material

Template

CMK-1

I41/a, 3 nm, 1500–1800 m2/g, 0.9–1.2 cm3/g Al-MCM-48 I41/a SBA-1 cubic SBA-15 hex, 4.5 nm, 1500 m2/g, 1.3 cm3/g MCM-48 cubic Ia3d SBA-15 hexagonal packing of carbon tubes, 1500–2200 m2/g, 1.5 cm3/g HMS low order MSU-H (SBA-15) low order, 3.9 nm, 1230 m2/g, 1.26 cm3/g MCF-Si carbon sphere, 7–9 nm, 290 m2/g, 0.39 cm3/g MCM-41 random carbon rods, <2 nm, 1170 m2/g SBA-16 Im3m KIT-6 rod-type KIT-6 (tube-type) (FDU-14 as precursor) Cubic Ia3d

SNU-1 CMK-2 CMK-3 CMK-4 CMK-5 SNU-2 C-MSU-H MCF-C C-41 CSU CFA CMK-8 CMK-9 C-FDU-14

MCM-48

Reference [32] [249] [253] [34] [250] [35, 165, 251] [254] [255] [256] [248] [257] [99, 258] [99] [259]

570

Chemistry of Zeolites and Related Porous Materials

Figure 8.49 Powder XRD patterns of CMK-3 carbon and SBA-15 silica used as template for the CMK-3 synthesis. Reproduced with permission from [34]. Copyright (2000) American Chemical Society

Figure 8.50 XRD patterns of template MCM-48 and CMK-1. Reproduced with permission from [32]. Copyright (1999) American Chemical Society

Synthesis, Structure, and Characterization of Mesoporous Materials

571

Figure 8.51 CMK-5 (a) TEM image viewed along the direction of the ordered nanoporous carbon and the corresponding Fourier diffractogram. (b) Schematic model for the carbon structure. (c) XRD pattern indicating the hexagonal order between carbon cylinders. Reproduced with permission from [35]. Copyright (2001) Nature Publishing

Figure 8.51 show the XRD pattern of CMK-5, TEM image, and schematic structure.[35] The structural model is provided to indicate that the carbon nanopores are rigidly interconnected into a highly ordered hexagonal array by carbon spacers. The outside diameter of the carbon structures is controllable by the choice of a template SBA-15 aluminosilicate with suitable diameter; the inside diameter is controllable by the amount of the carbon source. The (10) diffraction peak is lower than (11) in intensity, owing to the diffraction interference between the walls and the spacers interconnecting adjacent cylinders. Nanoporous carbon with narrow pore-size distribution can be prepared directly via a one-step nanocasting technique by carbonization of cyclodextrin–silica organic–inorganic hybrid composite.[260] The method consists of preparing a cyclodextrin-templated silica mesophase via soft chemistry, followed by direct carbonization of the occluded cyclodextrins. The nanocasting procedure provides granules (in millimeter scale) or monoliths (in centimeter scale). Cyclodextrins can be employed not only as structuredirecting templates but also as carriers for metal nanoparticle precursors. Zhao[259,261] used resol (phenol/formaldehyde) as a precursor, which can be a threeconnected benzene-ring framework, and independently demonstrated a reproducible synthesis of highly ordered mesoporous polymers and carbon frameworks with p6mm, Im3m and Ia3d symmetries via an EISA process of triblock copolymers. Unlike the nanorod or nanopipe array structures of the CMK carbon materials derived from a hardtemplate method, zeolite-like pore structures are observed in these novel mesoporous polymers and carbons. Since the mesostructures are formed on the surface by the EISA strategy, it meets the demands of preparing mesostructured films. The synthesis approach is quite similar to that of the mesoporous silica. The obtained mesoporous polymer and carbon have a large surface area (up to 1150 m2/g) and uniform pore size (
3 nm). Mesoporous Metals The electrochemical reduction of platinum sails confined to the aqueous environments of lyotropic liquid-crystalline phases leads to the deposition of platinum films[262] that have a well defined long-ranged porous nanostructure and high specific surface areas. These results suggest that the use of liquid-crystalline plating solutions could be a versatile way to create mesoporous electrodes for batteries, fuel cells, electrochemical capacitors, and sensors.

572

Chemistry of Zeolites and Related Porous Materials

2-D and 3-D metal nanowire thin films[263] with tunable 3–10 nm wire diameters have been obtained by electrodeposition into mesoporous silica thin-film templates, resulting in nanowire arrays that reflect the pore structure of the template. Removal of silica is achieved via annealing followed by etching to leave mechanically strong freestanding metal nanowire films. Nanoporous platinum sponges,[264] which exhibit characteristic X-ray diffraction patterns due to structural order on the mesoscale, have been obtained after removing the silica template with HF following synthesis within the pores of MCM-48 and SBA-15 with Pt(NH3)4(NO3)2. A 3-D networked osmium nanomaterial[265] was prepared by thermal decomposition of Os3(CO)12 within mesopores of MCM-48. The novel osmium nanomaterial shows high catalytic activity and excellent reusability in the oxidation reactions of unsaturated compounds under mild conditions. Mesoporous Carbides and Nitrides Nonoxide ceramics, such as silicon carbides, silicon nitrides, and boron nitrides, have unique mechanical and functional characteristics. Silicon carbides with high thermal conductivity, high thermal stability, excellent mechanical strength, and chemical inertness are especially considered as effective catalyst supports. However, there have been few reports on the successful fabrication of ordered mesoporous carbides and nitrides with high surface areas and large adjustable pores. The difficulties are mainly the extremely high synthesis temperatures required and the lack of proper precursors. A few of examples are shown here for the successful synthesis of mesoporous carbides and nitrides. Zhao[266] demonstrated the successful synthesis of highly ordered mesoporous silicon carbides with unusually high surface areas (430–720 m2/g), uniform pore sizes (<3.5 nm), and extremely high thermal stabilities (up to 1400  C) replicated by mesoporous silica hard templates via a one-step nanocasting process. Highly ordered 2-D hexagonal (p6m) and bicontinuous cubic (Ia3d) SiC nanowire arrays have been cast from the hard templates, mesoporous silica SBA-15 and KIT-6, respectively. Highly ordered hexagonal mesoporous carbon nitride material[267] can be synthesized by the following procedure. Calcined SBA-15 is added to a mixture of ethylenediamine and carbon tetrachloride. The resultant mixture is refluxed at 90  C for 6 h. The template carbon nitride polymer composite is then heat-treated in a nitrogen flow at 600  C. The mesoporous carbon nitride is recovered after dissolution of the silica framework in HF solution.

8.7 Morphology and Macroscopic Form of Mesoporous Material For industrial applications, the particle size, morphology, and texture of the mesoporous material are important, which include several critical points such as mechanical stability and macroscopic shapes with well defined properties. Morphology control is one of the most interesting issues in the research field of mesoporous materials. It plays a very important role in understanding the basic synthesis mechanism. The control of morphology in mesoporous materials is thought to be governed by kinetic effects as the self-assembly of surfactant molecules and nucleation processes

Synthesis, Structure, and Characterization of Mesoporous Materials

573

occur simultaneously as a result of the hydrolysis of the silica source, usually TEOS, into small silicate oligomers. The particle-growth process is dominated by the condensation of these oligomers. Co-condensation with organic functional groups in the form of organoalkoxysilanes has been identified as an important parameter in the formation of different particle morphologies. Synthetic conditions such as temperature, stirring rate, ionic strength, acidity, and reaction-mixture composition that affect the mesostructures and macrostructures of materials have been extensively studied in the surfactanttemplating system. 8.7.1

‘Single Crystal’ and Morphologies of Mesoporous Silicas

‘Single Crystal’ The mesoscaled structural regularity causes various features that are observed for mesoporous materials, such as crystal-like particle morphologies and the existence of sharp XRD peaks. Although no atomic-scaled structural ordering is observed in the mesoporous ‘crystal’, these particles can be described as being artificial single-crystalline with a mesoscopic periodic structure. Highly ordered pure-silica MCM-41 materials possessing well defined morphology (up to 2 mm) have been successfully prepared with extremely low surfactant concentration at room temperature. Experimental results suggested that highly ordered MCM-41 material was formed by a deposition mechanism with silicate rod-like micelles.[110] Large-pore cubic (Im3m) mesoporous silica single crystals[147] can be synthesized by using nonionic block copolymers F108 as templates, and inorganic salts (e.g., K2SO4) as additives (see Figure 8.30). These single crystals possess exclusively uniform rhombdodecahedral shapes (
1 mm) with
100% yield. The mesopore lattice array in each crystal face is solved by TEM, further confirming that these crystals are perfect single crystals. Kim and Ryoo[268] synthesized truncated rhombic dodecahedral crystals of MCM-48 by a hydrothermal procedure using sodium silicate, CTAB, and various kinds of alcohol. Guan[141] reported cubic (Pm3n) hybrid organic–inorganic mesoporous crystal with a decaoctahedral shape (
5 mm in diameter). Che[142] developed a method for highly ordered SBA-1(Pm3n) with well resolved XRD patterns and particular morphology having 54 or more crystal faces by adjusting temperature, synthesis time, acidity, and surfactant concentration. Trikalitis[246] synthesized single-crystal mesostructured semiconductors with cubic Ia3d symmetry. The SEM images of the product reveal that the particles are in fact single crystals with rhombic dodecahedral morphology, as large as 2 mm, shown in Figure 8.52. Morphologies of Mesoporous Silicas The integration of hydrogen-bonding interactions at the organic–inorganic interface, and the use of sol–gel and emulsion chemistry in acidic media have proven to be a general route for the easy processing of ordered mesoporous materials into desired morphologies. Yang[24] described a remarkable array of shapes and surface patterns of SBA-3. The acidic reaction conditions favor curved including toroidal, disk-like, spiral and spherical, shapes. Some morphologies of SBA-3 made Yang are shown in Figure 8.53. Zhao[269] reported a morphological control approach using block copolymers, co-surfactants, co-solvents, or the additive of strong electrolytes to selectively form

574

Chemistry of Zeolites and Related Porous Materials

Figure 8.52 SEM of mesostructured platinum chalcogenides. Reproduced with permission from [246]. Copyright (2002) American Chemical Society

Figure 8.53

Morphologies of SBA-3. Reproduced [32]

Synthesis, Structure, and Characterization of Mesoporous Materials

575

micrometer-sized hard sphere, fiber, doughnut, rope, egg-sausage, gyroid, and discoidlike mesoporous silica SBA-15. The shape of mesoporous silica SBA-15 powder is controlled by several complex factors including (1) the condensation rate of silica species, (2) shape of surfactant micelles, (3) the concentration of inorganic salts, and (4) stirring rate. Some mesoporous crystals show characteristic morphology, such as MCM-48 crystals with truncated rhombic dodecahedra;[268] HMM-3/SBA-1 with well defined decaoctahedra,[141,270] and 74 faces as well as 54 faces with crystal-like shape;[161,271] SBA-7 (P63/ mmc) with 20 distinct crystal faces with one six-fold axis;[130] and FDU-1 with rhombododecahedra.[147] Furthermore, morphological evolution of SBA-1 was investigated with respect to various synthetic parameters such as temperature, acidity, time, concentration, and so on.[143] All morphologies reported in the above references were commensurate with crystallographic point-group symmetries, and were observed from the crystals synthesized under well controlled conditions for crystal growth processes. The morphology of a well ordered mesoporous silica crystal of the cage type with cubic Fm3m symmetry, that shows a well resolved powder XRD profile, does not fit any crystallographic point-group symmetry. Morphologies such as an icosahedron or a decahedron[151] of cubic Fm3m structure seem to be inconsistent with the corresponding point-group symmetry, m3m. 8.7.2

Macroscopic Forms

Nanoparticles Minimization of particle size to the nanometer range for intracellular drug or agent delivery is critical in the biological usage of mesoporous silica because most cell uptake occurs in this size range. Several synthetic strategies to control the sizes of mesoporous nanoparticles have been reported. Lu[272] reported a rapid, aerosol-based process for synthesizing solid, well ordered spherical particles with stable pore mesostructures of hexagonal and cubic topology, as well as layered (vesicular) structures. This method relies on evaporationinduced interfacial self-assembly confined to spherical aerosol droplets. This simple, generalizable process can be modified for the formation of ordered mesostructured thin films. Single-domain mesoporous silica nanoparticles as small as 20–100 nm in particle size have been synthesized by modifying the conventional method based on the hydrolysis of TEOS mixed with an aqueous solution of surfactant,[176,272] by self-assembly of dilute TEOS and low surfactant concentration with NH4OH as a catalyst,[273] or from a highly dilute aqueous solution of surfactant–sodium silicate.[274] The mesoporous silica nanoparticles may be a new class of potential magnetic resonance imaging contrast agent, ideal for a completely novel biomedical application of mesoporous silica materials. Recently, mesoporous silica MCM-41 nanoparticles with a diameter of
20 nm and without deformation of the hexagonal mesostructure have been successfully synthesized.[176] In this synthesis, a nonionic surfactant (F127) was used as a suppressant of grain growth to achieve a balance between the ordered mesostructure and the nanoparticle.

576

Chemistry of Zeolites and Related Porous Materials

Thin Films Mesoporous thin films[275] have potential applications as low-dielectric-constant films, low-refractive-index films, hydrogen sensors, biomolecular sensors, nanostructured magnetic materials, photomodulated mass-transport layers, nanostructured solar cells, and nanostructured thermoelectrics. For most of these applications, pore connectivity, mass-transport, ion-transport, or electron-transport are key factors to control in the engineering of new devices. In addition, for applications such as thermoelectrics, the small pores size afforded by this synthesis route (2–10 nm) make these films better candidates for generating nanostructured materials that exhibit quantum confinement when compared to other film synthesis methods that typically yield larger pore sizes (>10 nm). Thus, one of the ongoing and grand challenges in the field of self-assembled nanostructured thin films is to control phase topology, order, orientation, and composition. Many different phases (symmetry, space group) and orientations of self-assembled nanostructured thin films have been synthesized by dip-coating or spin-coating, including (1) a two-dimensional hexagonal phase with hexagonal plane group p6mm (which distorts to a rectangular plane group c2mm upon drying), (2) a three-dimensional hexagonal phase with space group P63/mmc, (3) a body-centered cubic (bcc) phase with space group Im3m (which distorts to an orthorhombic Fmmm space group upon drying), (4) a primitive cubic phase with space group Pm3n, (5) body-centered tetragonal phase with space group I4/mmm, (6) a distorted cubic phase based on space group Ia3d, and (7) a rhombohedral structure with space group R3m. The early preparations of mesoporous silica film were conducted by growth from solution.[20,276]. The basic principle for the synthesis of ordered mesoporous films by growth from solution is to bring the synthesis solution (including a solvent, surfactant, and inorganic precursor) into contact with a second phase, e.g. solid (ceramic), gas (air), or another liquid (oil). The two-phase system is kept under specific conditions and the ordered film is formed at the interface. When the second phase is solid, it is the support on which the ordered film or membrane is grown. When the second phase is air or oil, the solid films are self-standing. The solvent-evaporation technique[23] is a very effective method to make thin films. This method involves formation of a liquid film containing the solvent, surfactant, and silica precursor followed by evaporation of the solvent. Several methods can be used to form liquid films. These include dip-, spin-coating, and film casting. Solvent evaporation has been suggested as the driving force for the organization of surfactant species into mesophases around which condensation of silicate species takes place. In the preparation of mesostructured films, the evaporation rate of the solvent must be controlled for selforganization to occur. For 2-D hexagonal thin films, substrate–micelle interactions cause the long axes of the micelle domains to align parallel to the plane of the substrate. Upon drying and calcination, the original circular micelle cross-sections exhibit elliptical shapes in thinner films.[277] In thicker films, the elliptical shape can evolve into a more-or-less rectangular profile with rounded ends when viewed in cross-section. Because of the hexagonal stacking, the overall result is a self-assembled brick-wall morphology that provides a uniform and well defined porosity. Control of mesopore shapes may be useful for controlling the sizes or orientations of reactant molecules entering the mesopores.

Synthesis, Structure, and Characterization of Mesoporous Materials

577

The uniformity of this morphology makes it desirable for many applications, including investigations of the thin-film stress system that forms the mesopore shape. The film is more useful when it may be able to yield good accessibility to the pore system and the substrate. However, hexagonal film has 1-D channel structures with the pore channels oriented parallel to the substrate surface. Ordered disclinations may yield domains of pores that are directed perpendicular to the substrate over limited regions. Another promising approach to forming nanoporous thin films with good accessibility to the pore structure and underlying substrate is the use of 3-D mesostructures to yield films with an interconnected pore network. The benefit with this approach is that orientation of the film is not an issue. An alternative approach to synthesize films with good pore accessibility is to form cage-like structures that are interconnected by windows or microporosity. Zhao[278] reported continuous mesoporous silica thin films with 3-D accessible pore structures (Pm3n, P63/mmc space groups) by a dip-coating technique using cationic surfactants as the template in nonaqueous media under acidic conditions. Now, it is possible to control mesoporous film structure on macroscopic scales. The excellent example is the synthesis of mesoporous silica films with a single-crystalline 3-D hexagonal (P63/mmc) porous structure,[279] in which a long-range order of cage-like pores is maintained over centimeter scales. The films are prepared through epitaxial growth of a self-organized mixture of hydrolysed TEOS and template on the rubbingtreated polyimide coating. The syntheses of nonsilica-based mesoporous films have been studied too. Stucky and Chmelka[280] prepared well ordered mesostructured titania films using triblock copolymer P123 as template. By varying the volume ratio between the copolymer and inorganic components of the precursor solution, silica and titania thin films with cubic, 2-D hexagonal, and lamellar mesostructures were prepared. The regions over which the three phases were obtained correspond well with those of the water-block copolymer binary phase diagram when considered in terms of the volume fraction of copolymer incorporated. In particular, a cubic mesostructure with crystalline TiO2 (anatase) in the walls, stable to 400  C, was synthesized. Recently, a simple preparation of homogeneous mesoporous silica films (coating) using basic conditions has been reported,[281] which involves the substrates (both plate and powder) being put into a homogeneous solution containing TEOS, CTAC, methanol, water, and ammonia and allowed to react at room temperature or 3  C for several h. This synthesis is a versatile method to prepare nanoporous silica thin layers on a solid substrate, especially when the reported procedure is not applicable: a substrate with complex morphology and/or being unstable in acidic solutions such as hydrotalcite and ZnO. Spheres and Balls Mesoporous materials with spherical morphology are quite attractive due to the potential applications in macromolecular separation, drug delivery, catalysis support, and template agents for photonic crystals. A general approach for preparing hollow spheres of mesoporous materials was based on sol–gel/emulsion technologies or the use of organic polymer beads as the templates that control the void formation and its volume. If an oil-in-water interface is used as an inorganic growth medium with the growth direction into the aqueous phase, morphological control of the resulting inorganic–organic

578

Chemistry of Zeolites and Related Porous Materials

composite assembly can be achieved at micron and longer length scales. The morphology of the preorganized organic liquid phase can be preserved during the mesophase synthesis. Stucky and coworkers[18] have used the combination of long-range oil-in-water emulsion and oil–water interface physics with shorter-range molecular assembly of silica and surfactants at the emulsion interface to create ordered composite mesostructured phases that are also macroscopically structured and shaped. Hydrodynamic, long-range forces can therefore be used to define emulsion morphology and the configuration of the emulsion oil–water interface. Hollow spheres (1–100 mm) of mesostructures of silica (SBA-1, SBA2, and SBA-3) were made (see Figure 8.54). They were created by control of the interface on two different length scales simultaneously. Micellar arrays controlled the nanometerscale assembly, and at the static boundary between an aqueous phase and an organic phase, control was achieved on the micrometer to centimeter scale.

Image Not Available In The Electronic Edition

Synthesis, Structure, and Characterization of Mesoporous Materials

579

Figure 8.55 Hard spheres of mesoporous silica. Reproduced with permission from [282]. Copyright (1997) American Chemical Society

Large, hard, transparent, mesoporous silica spheres[282] (see Figure 8.55) were synthesized in one step by using oil-in-water emulsion chemistry under basic conditions with cationic surfactants and TBOS. The pores are shown by TEM and nitrogen-absorption studies to be monodispersed in size, with total surface area over 1000 m2/g. Pang[283] successfully synthesized millimeter-sized mesoporous Co-SiO2 spheres through the introduction of CoIII-(3-aminopropyl)trimethoxysilane (APTMOS) complex into the reaction mixture with CTAB as template in basic media. The mesoporous Co-SiO2 spheres have a high surface area and disordered pores. Microspherical particles of MCM-41 have been synthesized.[284] All the properties of the spherical MCM-41 particles are essentially the same as those of the MCM-41 materials synthesized by well established methods. Highly ordered SBA-16 spheres (2–4 mm in diameter)[149] have been synthesized using triblock copolymer as templates in the presence of inorganic salts. The use of inorganic salts greatly improves the quality of the resultant mesostructure by increasing the self-assembly ability of organic species and results in the formation of millimeter spheres by enhancing the interaction at the inorganic/organic interface. The sphere synthesis has been extended into other materials. For example, mesoporous carbon capsules[285] with hollow core and mesoporous shell structures were synthesized using submicrometer-sized solid core and mesoporous-shell silica spheres as templates. Fibers High-quality SBA-3 fibers[287] can be grown by the general procedure for one-step, twophase, room-temperature synthesis. The silica fibers, which are 1–5 mm diameter and up to 5 cm long, have excellent long-range order, a narrow pore-size distribution, and a pore volume of up to 0.78 cm3/g. The birefringent fibers can be used as high-surface-area optical fibers. The mesopores in silica fibers run in a circular direction around the

580

Chemistry of Zeolites and Related Porous Materials

Figure 8.56 (a) SEM image of as-made silica nanofibers. (b) TEM images of as-made nanofibers. Reproduced with permission from [290]. Copyright (2004) American Chemical Society

axis.[288] These fibers are a unique example of a material that is highly ordered on the nanometer scale, but where the ordering does not lead to translational invariance in all three dimensions. The kinetics of fiber formation and the product distribution vary, being strongly dependent on the silica source and the amount of additional oil.[288] Mesoporous silica fibers[289] were also directly drawn from a highly viscous triblock copolymer–silicate sol. These fibers have uniform exterior diameters of submicron to several hundred micron size and no apparent limit on aspect ratios. Figure 8.56 shows SEM and TEM images of nanofibers. Rods A reverse amphiphilic mesophases approach to create hierarchically arrayed silica nanorods[291] has been demonstrated. In nonpolar solvent systems, the cylinder nanorods with a uniform diameter (10 nm) generated from highly ordered 2-D reverse hexagonal mesophases are aligned within lamellar macrostructures (
150 nm). The diameter size of silica nanorods can be adjusted from 9 to 15 nm by varying the chain length of the EO segment and the concentration of the surfactants. Low surfactant concentration and long hydrophilic EO segment chain yield silica nanorods with a large diameter. The calcined material has a relatively small surface area of 275 m2/g. Monodispersed SBA-15 materials with
100% rodlike morphologies (
1–2 mm long, see Figure 8.57) have been obtained by using P123 as template under static conditions without using salts.[120] This method afforded rodlike particles with temperaturedependent pore sizes ranging from 5.8 to 12.5 nm but with similar external dimensions.

Synthesis, Structure, and Characterization of Mesoporous Materials

581

Figure 8.57 SEM image of SBA-15 nanorod. Reproduced with permission from [120]. Copyright (2004) American Chemical Society

In order to prepare short monodispersed SBA-15 rods, the absence of stirring is essential, and the temperature of the first stage should not exceed 60  C. Tubes Careful control of the surfactant–water content and the rate of condensation of silica at high alkalinity resulted in hollow tubules 0.3 to 3 mm in diameter.[292] The wall of the tubules consisted of coaxial cylindrical pores, nanometers in size, that are characteristic of those of MCM-41. The formation of this higher-order structure may take place through a liquid-crystal-phase transformation mechanism involving an anisotropic membrane-totubule phase change. Chiral Mesoporous Silica Chirality is widely expressed in organic materials, perhaps most notably in biological molecules such as DNA, and in proteins, owing to the (homo)chirality of their components (d-sugars and l-amino acids). However, the occurrence of large-scale chiral pores in inorganic materials is rare. The assembly of chiral anionic surfactants and inorganic precursors in the presence of aminosilane or quaternized aminosilane[185] provides a potential method to synthesize mesoporous materials with inherent chirality Che et al.[148] have reported the surfactant-templated synthesis of ordered chiral mesoporous silica, together with a general approach for the structural analysis of chiral mesoporous crystals by electron microscopy. SEM indicates that the material has a twisted hexagonal rodlike morphology (with a hexagonal cross-section), with outer diameter 130–180 nm and length 1–6 mm (see Figure 8.58). From six distinct surfaces, the helical pitch along the rod axis was estimated to be
1.5 mm. TEM combined with

582

Chemistry of Zeolites and Related Porous Materials

Figure 8.58 (a) SEM image. (b) Schematic drawing of a structural model. (c) Cross-section. (d) One of the chiral channels in the material. Reproduced with permission from [148]. Copyright (2004) Nature Publishing

computer simulations confirms the presence of hexagonally ordered chiral channels of 2.2 nm diameter winding around the central axis of the rods. Monoliths The direct liquid-crystal templating pathway is effective in making mesostructured silica monoliths. The advantage of this method is the feasibility of fabricating large monoliths of periodic mesoporous materials in combination with a high degree of control over the phase structure and pore size. For example, Feng[69] synthesized monolithic mesoporous silica in ternary and microemulsion-type quaternary systems while preserving long-range ordering at the mesoscopic level. Large-sized, crack-free silica monoliths with highly ordered mesostructure are prepared in a fast and easy way via liquid–paraffin–medium-protected solvent evaporation.[293] By employing the inert liquid paraffin as the morphology ‘protector’, cracks in the materials can be successfully avoided. The block copolymer–silica composite monoliths are transparent and crack-free with a large size. The mesoporous silica monoliths have a highly ordered hexagonal mesostructure of space group p6mm and narrow pore-size distribution, with a mean pore diameter of 5.65 nm. In addition, metal ions can be easily doped into the monoliths. Ribbons Single-crystal mesoporous silica ribbons[294] that are 50–250 nm thick, 0.4–1.5 mm wide, and hundreds of micrometers long have been synthesized by using a one-phase route in dilute aqueous solutions of surfactants and silica species under acidic conditions. The mesoporous ribbons have excellent long-range order with tracklike pore channels oriented perpendicularly to the length of the ribbon and hexagonally organized. Confined Mesostructures In a physically confined environment, interfacial interactions, symmetry breaking, structural frustration, and confinement-induced entropy loss can play dominant roles in determining molecular organization. Wu[295] studied the confined assembly of silica– copolymer composite mesostructures within cylindrical nanochannels of porous anodic

Synthesis, Structure, and Characterization of Mesoporous Materials

583

alumina with varying diameters from 18 nm to 80 nm. Using exactly the same precursors and reaction conditions that form the two-dimensional hexagonal SBA-15 mesostructured thin film, unprecedented silica mesostructures with chiral mesopores such as singleand double-helical geometries spontaneously form inside individual alumina nanochannels. On tightening the degree of confinement, a transition is observed in the mesopore morphology from a coiled cylindrical to a spherical cage-like geometry. Self-consistent field calculations carried out to account for the observed mesostructures accord well with experiment. The mesostructures produced by confined syntheses are useful as templates for fabricating highly ordered mesostructured nanowires and nanowire arrays.

8.8 Possible Applications, Challenges, and Outlook 8.8.1

Possible Applications

Mesoporous materials have many present and potential future applications for catalysis, life science, energy, environmental, functional materials, and so on. In comparison to zeolites, ordered mesoporous materials overcome the pore-size constraint of zeolites and allow easier diffusion of bulky substrates. Unrestricted diffusion of reactants and products for mesoporous materials was observed even after the incorporation of large catalytically active sites in the mesopore system. Ordered mesoporous oxides have been used as supports for metals and metal oxides, and as a host material for anchoring stereo- and enantio-selective species. In reactions that require milder acidity and also involve bulky reactants and products, such as mild hydrocracking reactions, mesoporous materials exhibit great potential.[40,296] Mesoporous yttrium zirconium oxide or mesoporous cerium oxide anodes and a mesoporous lanthanum strontium manganate cathode constructed from nanocrystallites of these materials organized by a templating mesophase can be used in solid-oxide fuel cells. Mesoporous TiO2 has a potential application in dye-sensitized solar cells. The field of mesoporous silicas and carbons doped with biologically interesting molecules has already exhibited its diversity and potential applications in many frontiers of modern materials science including biocatalysis, biosensing, drug delivery and control release,[297] photocontrolled reversible release of guest molecules,[298] immobilization of biomolecules,[299] and separation of biological molecules.[300] The ability of mesoporous adsorbents to separate molecules by size exclusion has been demonstrated. The use of ordered mesoporous materials in medical and life science is a young and growing but still challenging field with still high demand for new and improved materials. In an effort to develop selective solid adsorbents for acidic gas removal from natural gas mixtures, amine-modified mesoporous silica materials have been applied.[301] With large numbers of basic amine groups on the surface, the adsorbents are able to selectively bind the acidic gases CO2 and H2S. High adsorption capacities and adsorption rates were observed, compared with the other adsorbent supports including activated carbon and silica gel. These adsorbents can be regenerated under mild conditions such as those used in pressure-swing or temperature-swing adsorption processes. Repeated adsorption– desorption cycles revealed that the adsorbent exhibited good cyclic stability. Modified mesoporous silica can be applied to toxic-ion (e.g., Hg2+) adsorptions.[302]

584

Chemistry of Zeolites and Related Porous Materials

Ordered mesoporous materials, due to their periodic and size-controllable pore channels and high surface areas, have been regarded as ‘a nano-reactor’ to construct novel ordered and well dispersed nanostructured composites with controlled size and size distribution.[303] A number of studies have reported on the encapsulation of guest materials, such as metal oxides,[304] semiconductors, metal sulfides,[305] carbon, metals,[306] and polymers into mesoporous silica hosts. Polyaniline filaments within the mesoporous channel host (aluminosilicate) have significant conductivity, and this demonstration of conjugated polymer with mobile charge carriers in nanometer channels represents a step toward the design of nanometer electronic devices. These composites have potential as stable molecular wires, which can be applied in the design of batteries and systems to accumulate electric charge. SBA-15 with polyaniline inside the pore channels was used as a dispersed phase in electrorheological (ER) fluids. Ordered mesoporous materials are highly interesting hosts for optically functional materials.[307] The mesopores have specific effects on guests. For example, silver nanoparticles were separated and highly uniformly dispersed within the pores of the mesoporous silica;[308] the resultant material displays optical switching and memory effects under different ambient conditions. Low-k dielectrics are important as packaging materials for future developments of microelectronics. Thin-film dense silica has always been the material of choice for this application. Their low dielectric constants, plus their good thermal and mechanical stability and hydrophobicity, suggest the potential utility of mesoporous organosilica films as low-k layers in microelectronics.[272,309] Metal- or metal-oxide-based gas sensors are the most widely used solid-state devices for detecting gases in the environment and atmosphere. The high specific surface areas and uniform mesopores of mesoporous materials will result in a higher probability for a gas to interact with the sensing compounds or sites, which is likely to increase the sensitivity of the material.[310] 8.8.2

Challenges and Outlook

Template-directed synthesis is a convenient and versatile method for generating porous materials. It is also a cost-effective and high-throughput procedure that allows the complex topology of a template to be duplicated in a single step or a few of steps. With the use of mesoscale objects as templates, the dimensions of these pores can be significantly extended to cover a wide range of
2 nm to 10 mm. Ordered mesoporous materials exhibit tunable pore size, high surface area and pore volume, ease of surface functionalization, and controllable morphology and macroscopic shape, all these being highly promising properties for numerous applications. Considerable scientific effort has been focused on the preparation, characterization, and application of ordered mesoporous silicas. The followings are still challenges for mesoporous material researchers. (1) The utility of mesoporous solids is still rather lacking. Make the useful devices and products available for targeting special applications (able to compete with current technologies), although some achievements have been reached in functional materials, physical properties, and catalyst-application fields.

Synthesis, Structure, and Characterization of Mesoporous Materials

585

(2) The formation mechanisms are clearly not settled and are still a matter for active investigation.[311] Most mechanisms described a certain stage of a synthesis under special synthesis conditions. (3) Detailed control of the structural and properties such as pore topology, pore diameter, pore connectivity, controlled multiscale porosity, surface properties, reactivity, functionalization, morphology, and macroscopic shape are desirable to reach the ultimate goals of industrial and commercial applications. Make mesoporous materials able to compete with other current using materials (e.g., zeolites). (4) The synthesis of mesoporous material should be an ecologically friendly, low toxicity, low-cost, simple procedure. A possible reuse of these materials (as adsorbent or catalyst) has to be investigated. Commercial issues such as cost and scale up of the mesoporous material have to be assessed in competition with existing materials. (5) The mesoporous materials should be mechanically, thermally, and hydrothermally stable and have a long lifetime. One of the major problems to be solved is the stability of mesoporous silicas upon prolonged exposure to aqueous solutions. (6) The pore system and wall composition (in particular, nonsilica-based materials) are controllable and available. (7) Find new properties for new potential applications. Use really the unique structure and composition of mesoporous materials. In conclusion, ordered mesoporous materials are unique materials due to their specific structure and property. They have attracted more and more attention owing to their potential applications in many fields. The applications are the biggest driving force for mesoporous material research. How to utilize and enhance the characteristics of mesoporous materials should be the direction of the main effort in this field. Many investigations have explored the suitability of mesoporous materials for potential practical applications, for example, catalysis, optically active materials, polymerization science, separation technology, optical waveguides, low-k dielectric layers for microelectronics, photoelectrochemical applications in solar cells, thin-film sensors, encapsulation of drugs and biomolecules for targeted or controlled-release applications. With successes in the synthesis of various mesoporous materials, we have reason to expect that further application possibilities coming soon. Mesoporous materials will play an important role in science and technology of this century.

References [1] Q.S. Huo, D.I. Margolese, and G.D. Stucky, Surfactant Control of Phases in the Synthesis of Mesoporous Silica-based Materials, Chem. Mater., 1996, 8, 1147–1160. [2] J.S. Beck, J.C. Vartuli, W.J. Roth, M.E. Leonowicz, C.T. Kresge, K.D. Schmitt, C.T.W. Chu, D.H. Olson, E.W. Sheppard, S.B. McCullen, J.B. Higgins, and J.L. Schlenker, A New Family of Mesoporous Molecular-Sieves Prepared with Liquid-Crystal Templates. J. Am. Chem. Soc., 1992, 114, 10834–10843. [3] C.T. Kresge, M.E. Leonowicz, W.J. Roth, J.C. Vartuli, and J.S. Beck, Ordered Mesoporous Molecular-Sieves Synthesized by a Liquid-Crystal Template Mechanism. Nature (London), 1992, 359, 710–712.

586

Chemistry of Zeolites and Related Porous Materials

[4] S. Inagaki, Y. Fukushima, and K. Kuroda, Synthesis of Highly Ordered Mesoporous Material from a Layered Polysilicate. J. Chem. Soc., Chem. Commun., 1993, 680–682. [5] A. Monnier, F. Schuth, Q. Huo, D. Kumar, D. Margolese, R.S. Maxwell, G.D. Stucky, M. Krishnamurty, P. Petroff, A. Firouzi, M. Janicke, and B.F. Chmelka, Cooperative Formation of Inorganic-Organic Interfaces in the Synthesis of Silicate Mesostructures. Science, 1993, 261, 1299–1303. [6] A. Corma, V. Fornes, M.T. Navarro, and J. Perezpariente, Acidity and Stability of Mcm-41 Crystalline Aluminosilicates. J. Catal., 1994, 148, 569–574. [7] A. Corma, M.T. Navarro, and J.P. Pariente, Synthesis of an Ultralarge Pore Titanium Silicate Isomorphous to Mcm-41 and Its Application as a Catalyst for Selective Oxidation of Hydrocarbons. J. Chem. Soc. Chem. Commun., 1994, 147–148. [8] Q.S. Huo, D.I. Margolese, U. Ciesla, D.G. Demuth, P.Y. Feng, T.E. Gier, P. Sieger, A. Firouzi, B.F. Chmelka, F. Schuth, and G.D. Stucky, Organization of Organic Molecules with Inorganic Molecular Species into Nanocomposite Biphase Arrays. Chem. Mater., 1994, 6, 1176–1191. [9] Q.S. Huo, D.I. Margolese, U. Ciesla, P.Y. Feng, T.E. Gier, P. Sieger, R. Leon, P.M. Petroff, F. Schuth, and G.D. Stucky, Generalized Synthesis of Periodic Surfactant Inorganic Composite Materials. Nature (London), 1994, 368, 317–321. [10] P.T. Tanev, M. Chibwe, and T.J. Pinnavaia, Titanium-Containing Mesoporous Molecular Sieves for Catalytic Oxidation of Aromatic Compounds. Nature (London), 1994, 368, 321– 323. [11] C.G. Wu and T. Bein, Conducting Polyaniline Filaments in a Mesoporous Channel Host. Science, 1994, 264, 1757–1759. [12] D.M. Antonelli and J.Y. Ying, Synthesis of Hexagonally Packed Mesoporous TiO2 by a Modified Sol-Gel Method, Angew. Chem., Int. Ed. Engl., 1995, 34, 2014–2017. [13] G.S. Attard, J.C. Glyde, and C.G. Goltner, Liquid-Crystalline Phases as Templates for the Synthesis of Mesoporous Silica. Nature (London), 1995, 378, 366–368. [14] S.A. Bagshaw, E. Prouzet, and T.J. Pinnavaia, Templating of Mesoporous Molecular-Sieves by Nonionic Polyethylene Oxide Surfactants. Science, 1995, 269, 1242–1244. [15] A. Firouzi, D. Kumar, L.M. Bull, T. Besier, P. Sieger, Q. Huo, S.A. Walker, J.A. Zasadzinski, C. Glinka, J. Nicol, D. Margolese, G.D. Stucky, and B.F. Chmelka, Cooperative Organization of Inorganic-Surfactant and Biomimetic Assemblies. Science, 1995, 267, 1138–1143. [16] T. Maschmeyer, F. Rey, G. Sankar, and J.M. Thomas, Heterogeneous Catalysts Obtained by Grafting Metallocene Complexes onto Mesoporous Silica. Nature (London), 1995, 378, 159– 162. [17] P.T. Tanev and T.J. Pinnavaia, A Neutral Templating Route to Mesoporous Molecular Sieves. Science, 1995, 267, 865–867. [18] S. Schacht, Q. Huo, I.G. VoigtMartin, G.D. Stucky, and F. Schuth, Oil-Water Interface Templating of Mesoporous Macroscale Structures. Science, 1996, 273, 768–771. [19] H. Yang, N. Coombs, I. Sokolov, and G.A. Ozin, Free-standing and Oriented Mesoporous Silica Films Grown at the Air-Water Interface. Nature (London), 1996, 381, 589–592. [20] H. Yang, A. Kuperman, N. Coombs, S. MamicheAfara, and G.A. Ozin, Synthesis of Oriented Films of Mesoporous Silica on Mica. Nature (London), 1996, 379, 703–705. [21] X. Feng, G.E. Fryxell, L.Q. Wang, A.Y. Kim, J. Liu, and K.M. Kemner, Functionalized Monolayers on Ordered Mesoporous Supports. Science, 1997, 276, 923–926. [22] M. Kruk, M. Jaroniec, and A. Sayari, Application of Large Pore MCM-41 Molecular Sieves to Improve Pore Size Analysis using Nitrogen Adsorption Measurements. Langmuir., 1997, 13, 6267–6273. [23] Y.F. Lu, R. Ganguli, C.A. Drewien, M.T. Anderson, C.J. Brinker, W.L. Gong, Y.X. Guo, H. Soyez, B. Dunn, M.H. Huang, and J.I. Zink, Continuous Formation of Supported Cubic and

Synthesis, Structure, and Characterization of Mesoporous Materials

[24] [25] [26]

[27]

[28] [29]

[30] [31] [32] [33]

[34]

[35]

[36] [37] [38] [39] [40] [41] [42]

[43]

587

Hexagonal mesoporous Films by Sol Gel Dip-coating, Nature (London), 1997, 389, 364–368. H. Yang, N. Coombs, and G.A. Ozin, Morphogenesis of Shapes and Surface Patterns in Mesoporous Silica, Nature (London), 1997, 386, 692–695. P.D. Yang, T. Deng, D.Y. Zhao, P.Y. Feng, D. Pine, B.F. Chmelka, G.M. Whitesides, and G.D. Stucky, Hierarchically Ordered Oxides, Science, 1998, 282, 2244–2246. P.D. Yang, D.Y. Zhao, D.I. Margolese, B.F. Chmelka, and G.D. Stucky, Generalized Syntheses of Large-pore Mesoporous Metal Oxides with Semicrystalline Frameworks, Nature (London), 1998, 396, 152–155. D.Y. Zhao, Q.S. Huo, J.L. Feng, B.F. Chmelka, and G.D. Stucky, Nonionic Triblock and Star Diblock Copolymer and Oligomeric Surfactant Syntheses of Highly Ordered, Hydrothermally Stable, Mesoporous Silica Structures. J. Am. Chem. Soc., 1998, 120, 6024–6036. T. Asefa, M.J. MacLachan, N. Coombs, and G.A. Ozin, Periodic Mesoporous Organosilicas with Organic Groups Inside the Channel Walls. Nature (London), 1999, 402, 867–871. S. Inagaki, S. Guan, Y. Fukushima, T. Ohsuna, and O. Terasaki, Novel Mesoporous Materials with a Uniform Distribution of Organic Groups and Inorganic Oxide in their Frameworks. J. Am. Chem. Soc., 1999, 121, 9611–9614. S.A. Jenekhe and X.L. Chen, Self-assembly of Ordered Microporous Materials Rod-coil Block Copolymers. Science, 1999, 283, 372–375. B.J. Melde, B.T. Holland, C.F. Blanford, and A. Stein, Mesoporous Sieves with Unified Hybrid Inorganic/Organic Frameworks. Chem. Mater., 1999, 11, 3302–3308. R. Ryoo, S.H. Joo, and S. Jun, Synthesis of Highly Ordered Carbon Molecular Sieves via Template-mediated Structural Transformation, J. Phys. Chem. B, 1999, 103, 7743–7746. P.D. Yang, D.Y. Zhao, D.I. Margolese, B.F. Chmelka, and G.D. Stucky, Block Copolymer Templating Syntheses of Mesoporous Metal Oxides with Large Ordering Lengths and Semicrystalline Framework. Chem. Mater., 1999, 11, 2813–2826. S. Jun, S.H. Joo, R. Ryoo, M. Kruk, M. Jaroniec, Z. Liu, T. Ohsuna, and O. Terasaki, Synthesis of New, Nanoporous Carbon with Hexagonally Ordered Mesostructure, J. Am. Chem. Soc., 2000, 122, 10712–10713. S.H. Joo, S.J. Choi, I. Oh, J. Kwak, Z. Liu, O. Terasaki, and R. Ryoo, Ordered Nanoporous Arrays of Carbon Supporting High Dispersions of Platinum Nanoparticles, Nature (London), 2001, 412, 169–172. T. Bein, Synthesis and Application of Molecular Sieve Layers and Membranes. Chem. Mater., 1996, 8, 1636–1653. N.K. Raman, M.T. Anderson, and C.J. Brinker, Template-based Approaches to the Preparation of Amorphous, Nanoporous Silicas. Chem. Mater., 1996, 8, 1682–1701. A. Sayari, Catalysis by Crystalline Mesoporous Molecular Sieves. Chem. Mater., 1996, 8, 1840–1852. X.S. Zhao, G.Q.M. Lu, and G.J. Millar, Advances in Mesoporous Molecular Sieve MCM-41. Ind. Eng. Chem. Res., 1996, 35, 2075–2090. A. Corma, From Microporous to Mesoporous Molecular Sieve Materials and their Use in Catalysis. Chem. Rev., 1997, 97, 2373–2419. K. Moller and T. Bein, Inclusion Chemistry in Periodic Mesoporous Hosts. Chem. Mater., 1998, 10, 2950–2963. T.J. Barton, L.M. Bull, W.G. Klemperer, D.A. Loy, B. McEnaney, M. Misono, P.A. Monson, G. Pez, G.W. Scherer, J.C. Vartuli, and O.M. Yaghi, Tailored Porous Materials, Chem. Mater., 1999, 11, 2633–2656. U. Ciesla and F. Schuth, Ordered Mesoporous Materials. Microporous Mesoporous Mater., 1999, 27, 131–149.

588

Chemistry of Zeolites and Related Porous Materials

[44] J.Y. Ying, C.P. Mehnert, and M.S. Wong, Synthesis and Applications of Supramoleculartemplated Mesoporous Materials. Angew. Chem., Int. Ed., 1999, 38, 56–77. [45] A. Stein, B.J. Melde, and R.C. Schroden, Hybrid Inorganic-Organic Mesoporous Silicates – Nanoscopic Reactors Coming of Age. Adv. Mater., 2000, 12, 1403–1419. [46] G. Ferey, Microporous Solids: From Organically Templated Inorganic Skeletons to Hybrid Frameworks . . . Ecumenism in Chemistry. Chem. Mater., 2001, 13, 3084–3098. [47] C. Sanchez, G. Soler-Illia, F. Ribot, T. Lalot, C.R. Mayer, and V. Cabuil, Designed Hybrid Organic-inorganic Nanocomposites from Functional Nanobuilding Blocks. Chem. Mater., 2001, 13, 3061–3083. [48] A. Sayari and S. Hamoudi, Periodic Mesoporous Silica-based Organic–Inorganic Nanocomposite Materials. Chem. Mater., 2001, 13, 3151–3168. [49] F. Schuth, Non-siliceous Mesostructured and Mesoporous Materials. Chem. Mater., 2001, 13, 3184–3195. [50] M.E. Davis, Ordered Porous Materials for Emerging Applications. Nature (London), 2002, 417, 813–821. [51] G.J.D. Soler-Illia, C. Sanchez, B. Lebeau, and J. Patarin, Chemical Strategies to Design Textured Materials: From Microporous and Mesoporous Oxides to Nanonetworks and Hierarchical Structures. Chem. Rev., 2002, 102, 4093–4138. [52] V. Chiola, J.E. Ritsko, and C.D. Vanderpool, US Patent Vol. 3,556,725 (1971). [53] R.K. Iler, The Chemistry of Silica, John Wiley & Sons, Inc., New York, 1971. [54] F. DiRenzo, H. Cambon, and R. Dutartre, A 28-Year-Old Synthesis of Micelle-Templated Mesoporous Silica, Microporous Mesoporous Mater., 1997, 10, 283–286. [55] T. Yanagisawa, T. Shimizu, K. Kuroda, and C. Kato, The Preparation of Alkyltrimethylammonium-Kanemite Complexes and Their Conversion to Microporous Materials. Bull. Chem. Soc. Jpn., 1990, 63, 988–992. [56] T. Yanagisawa, T. Shimizu, K. Kuroda, and C. Kato, Trimethylsilyl Derivatives of Alkyltrimethylammonium-Kanemite Complexes and Their Conversion to Microporous SiO2 Materials. Bull. Chem. Soc. Jpn., 1990, 63, 1535–1537. [57] J.S. Beck, US Patent Vol. 5,057,296 (1991). [58] C.T. Kresge, M.E. Leonowicz, W. Roth, and J.C. Vartuli, US Patent Vol. 5,102,643 (1992). [59] G. Soler-Illia, E.L. Crepaldi, D. Grosso, and C. Sanchez, Block Copolymer-templated Mesoporous Oxides. Curr. Opin. Colloid Interface Sci., 2003, 8, 109–126. [60] A. Shimojima, Z. Liu, T. Ohsuna, O. Terasaki, and K. Kuroda, Self-assembly of Designed Oligomeric Siloxanes with Alkyl Chains into Silica-based Hybrid Mesostructures. J. Am. Chem. Soc., 2005, 127, 14108–14116. [61] C. Chen, H.X. Li, and M.E. Davis, Studies on Mesoporous Materials. I. Synthesis and Characterization of MCM-41. Microporous Mater., 1993, 2, 17–26. [62] C.G. Goltner, S. Henke, M.C. Weissenberger, and M. Antonietti, Mesoporous Silica from Lyotropic Liquid Crystal Polymer Templates. Angew. Chem., Int. Ed., 1998, 37, 613–616. [63] D.J. Mitchell, G.J.T. Tiddy, L. Waring, T. Bostock, and M.P. McDonald, PhaseBehavior of Polyoxyethylene Surfactants with Water – Mesophase Structures and Partial Miscibiliity (Cloud Points). J. Chem. Soc., Faraday Trans. I, 1983, 79, 975–1000. [64] Q.S. Huo, R. Leon, P.M. Petroff, and G.D. Stucky, Mesostructure Design with Gemini Surfactants. – Supercage Formation in a 3-Dimensional Hexagonal Array. Science, 1995, 268, 1324–1327. [65] A. Firouzi, F. Atef, A.G. Oertli, G.D. Stucky, and B.F. Chmelka, Alkaline Lyotropic Silicatesurfactant Liquid Crystals. J. Am. Chem. Soc., 1997, 119, 3596–3610.

Synthesis, Structure, and Characterization of Mesoporous Materials

589

[66] C.Y. Chen, S.L. Burkette, H.X. Li, and M.E. Davis, Studies on Mesoporous Materials. II. Synthesis Mechanism of MCM -41. Microporous Mater., 1993 2, 27–34. [67] Y. Sakamoto, S. Inagaki, T. Ohsuna, N. Ohnishi, Y. Fukushima, Y. Nozue, and O. Terasaki, Structure Analysis of Mesoporous Material ‘FSM-16’ – Studies by Electron Microscopy and X-ray Diffraction. Microporous Mesoporous Mater., 1998, 21, 589–596. [68] N.R.B. Coleman and G.S. Attard, Ordered Mesoporous Silicas Prepared from both Micellar Solutions and Liquid Crystal Phases. Microporous Mesoporous Mater., 2001, 44, 73–80. [69] P.Y. Feng, X.H. Bu, G.D. Stucky, and D.J. Pine, Monolithic Mesoporous Silica Templated by Microemulsion Liquid Crystals. J. Am. Chem. Soc., 2000, 122, 994–995. [70] S.A. El-Safty and T. Hanaoka, Microemulsion Liquid Crystal Templates for Highly Ordered Three-dimensional Mesoporous Silica Monoliths with Controllable Mesopore Structures. Chem. Mater., 2004, 16, 384–400. [71] C.Z. Yu, J. Fan, B.Z. Tian, and D.Y. Zhao, Morphology Development of Mesoporous Materials: a Colloidal Phase Separation Mechanism. Chem. Mater., 2004, 16, 889–898. [72] S. Ruthstein, J. Schmidt, E. Kesselman, Y. Talmon, and D. Goldfarb, Resolving Intermediate Solution Structures during the Formation of Mesoporous SBA-15. J. Am. Chem. Soc., 2006, 128, 3366–3374. [73] C.C. Egger, M.W. Anderson, G.J.T. Tiddy, and J.L. Casci, In situ NMR and XRD Studies of the Growth Mechanism of SBA-1. Phys. Chem. Chem. Phys., 2005, 7, 1845–14855. [74] J.N. Israelachvili, D.J. Mitchell, and B.W. Ninham, Theory of Self-Assembly of Hydrocarbon Amphiphiles into Micelles and Bilayers. J. Chem. Soc., Faraday Trans., II, 1976, 72, 1525–1568. [75] S.T. Hyde, Interfacial Architecture in Surfactant Water Mixtures – Beyond Sphers, Cylinders and Planes. Pure Appl. Chem., 1992, 64, 1617–1622. [76] F.S. Bates and G.H. Fredrickson, Block copolymers – Designer Soft Materials. Phys. Today, 1999, 52, 32–38. [77] P. Kipkemboi, A. Fogden, V. Alfredsson, and K. Flodstrom, Triblock Copolymers as Templates in Mesoporous Silica Formation: Structural Dependence on Polymer Chain Length and Synthesis Temperature. Langmuir., 2001, 17, 5398–5402. [78] J.M. Kim, Y. Sakamoto, Y.K. Hwang, Y.U. Kwon, O. Terasaki, S.E. Park, and G.D. Stucky, Structural Design of Mesoporous Silica by Micelle-packing Control using Blends of Amphiphilic Block Copolymers, J. Phys. Chem. B, 2002, 106, 2552–2558. [79] D.F. Evans, D.J. Mitchell, and B.W. Ninham, Ion Binding and Dressed Micelles. J. Phys. Chem., 1984, 88, 6344–6348. [80] S.S. Kim, T.R. Pauly, and T.J. Pinnavaia, Non-ionic Surfactant Assembly of Wormhold Silica Molecular Sieves from Water Soluble Silicates. Chem. Commun., 2000, 835–836. [81] R. Ryoo, J.M. Kim, C.H. Ko, and C.H. Shin, Disordered Molecular Sieve with Branched Mesoporous Channel Network. J. Phys. Chem., 1996, 100, 17718–17721. [82] D.Y. Zhao, Q.S. Huo, J.L. Feng, J.M. Kim, Y.J. Han, and G.D. Stucky, Novel Mesoporous Silicates with Two-dimensional Mesostructure Direction using Rigid Bolaform Surfactants. Chem. Mater., 1999, 11, 2668–2672. [83] H.M. Kao, Y.W. Liao, C.C. Ting, Synthesis of cubic mesoporous silica SBA-1 with bulky headgroup surfactant cetyltripropylammonium bromide. Microporous and Mesoporous Materials, 2007, 98, 80–88. [84] Y. Sakamoto, M. Kaneda, O. Terasaki, D.Y. Zhao, J.M. Kim, G. Stucky, H.J. Shim, and R. Ryoo, Direct Imaging of the Pores and Cages of Three-dimensional Mesoporous Materials, Nature (London), 2000, 408, 449–453. [85] S.D. Shen, Y.Q. Li, Z.D. Zhang, J. Fan, B. Tu, W.Z. Zhou, and D.Y. Zhao, A Novel Ordered Cubic Mesoporous Silica Templated with Tri-head Group Quaternary Ammonium Surfactant. Chem. Commun., 2002, 2212–2213.

590

Chemistry of Zeolites and Related Porous Materials

[86] A.E. Garcia-Bennett, K. Miyasaka, O. Terasaki, and S.N. Che, Structural Solution of Mesocaged Material AMS-8. Chem. Mater., 2004, 16, 3597–3605. [87] J. Fan, C.Z. Yu, T. Gao, J. Lei, B.Z. Tian, L.M. Wang, Q. Luo, B. Tu, W.Z. Zhou, and D.Y. Zhao, Cubic Mesoporous Silica with Large Controllable Entrance Sizes and Advanced Adsorption Properties. Angew. Chem. Int. Ed., 2003, 42, 3146–3150. [88] F. Kleitz, D.N. Liu, G.M. Anilkumar, I.S. Park, L.A. Solovyov, A.N. Shmakov, and R. Ryoo, Large Cage Face-centered-cubic Fm3m Mesoporous Silica: Synthesis and Structure. J. Phys. Chem. B, 2003, 107, 14296–14300. [89] Y. Sakamoto, I. Diaz, O. Terasaki, D.Y. Zhao, J. Perez-Pariente, J.M. Kim, and G.D. Stucky, Three-dimensional Cubic Mesoporous Structures of SBA-12 and Related Materials by Electron Crystallography, J. Phys. Chem. B, 2002, 106, 3118–3123. [90] S.A. El-Safty, T. Hanaoka, and F. Mizukami, Design of Highly Stable, Ordered Cage Mesostructured Monoliths with Controllable Pore Geometric and Sizes. Chem. Mater., 2005, 17, 3137–3145. [91] S.D. Shen, A.E. Garcia-Bennett, Z. Liu, Q.Y. Lu, Y.F. Shi, Y. Yan, C.Z. Yu, W.C. Liu, Y. Cai, O. Terasaki, and D.Y. Zhao, Three-dimensional Low Symmetry Mesoporous Silica Structures Templated from Tetra-headgroup Rigid Bolaform Quaternary Ammonium Surfactant. J. Am. Chem. Soc., 2005, 127, 6780–6787. [92] S. Besson, C. Ricolleau, T. Gacoin, C. Jacquiod, and J.P. Boilot, Highly Ordered Orthorhombic Mesoporous Silica Films. Microporous Mesoporous Mater., 2003, 60, 43–49. [93] A.E. Garcia-Bennett, N. Kupferschmidt, Y. Sakamoto, S. Che, and O. Terasaki, Synthesis of Mesocage Structures by Kinetic Control of Self-assembly in Anionic Surfactants. Angew. Chem. Int. Ed., 2005, 44, 5317–5322. [94] B.W. Eggiman, M.P. Tate, and H.W. Hillhouse, Rhombohedral Structure of Highly Ordered and Oriented Self-assembled Nanoporous Silica Thin Films. Chem. Mater., 2006, 18, 723– 730. [95] A.C. Finnefrock, R. Ulrich, A. Du Chesne, C.C. Honeker, K. Schumacher, K.K. Unger, S.M. Gruner, and U. Wiesner, Metal Oxide Containing Mesoporous Silica with Bicontinuous ‘‘Plumber’s Nightmare’’ Morphology from a Block Copolymer-Hybird Mesophase. Angew. Chem. Int. Ed., 2001, 40, 1207–1211. [96] A.C. Finnefrock, R. Ulrich, G.E.S. Toombes, S.M. Gruner, and U. Wiesner, The Plumber’s Nightmare: A New Morphology in Block Copolymer-Ceramic Nanocomposites and Mesoporous Aluminosilicates. J. Am. Chem. Soc., 2003, 125, 13084–13093. [97] A. Carlsson, M. Kaneda, Y. Sakamoto, O. Terasaki, R. Ryoo, and S.H. Joo. The Structure of MCM-48 Determined by Electron Crystallography. J. Electron Microsc., 1999, 48, 795–798. [98] S.A. El-Safty, T. Hanaoka, and F. Mizukami, Large-sclae Design of Cubic Ia3d Mesoporous Silica Monoliths with High Order, Controlled Pores, and Hydrothermal Stability. Adv. Mater., 2005, 17, 47–53. [99] F. Kleitz, S.H. Choi, and R. Ryoo, Cubic Ia3d Large Mesoporous Silica: Synthesis and Replication to Platinum Nanowires, Carbon Nanorods and Carbon Nanotubes. Chem. Commun., 2003, 2136–2137. [100] S.A. El-Safty and J. Evans, Formation of Highly Ordered Mesoporous Silica Materials adopting Lyotropic Liquid Crystal Mesophases. J. Mater. Chem. 2002, 12, 117–123. [101] Y.T. Chan, H.P. Lin, C.Y. Mou, and S.T. Liu, Ia3d Cubic Mesoporous Silicas using EO(17)MA(23) Diblock Copolymers made from ATRP. Chem. Commun., 2002, 2878–2879. [102] K. Flodstrom, V. Alfredsson, and N. Kallrot, Formation of a New Ia3d Cubic Mesostructured Silica via Triblock Copolymer-assisted Synthesis. J. Am. Chem. Soc. 2003, 125, 4402–4403. [103] S.A. El-Safty and T. Hanaoka, Monolithic Nanostructured Silicate Family Templated by Lyotropic Liquid-crystalline Nonionic Surfactant Mesophases. Chem. Mater., 2003, 15, 2892–2902.

Synthesis, Structure, and Characterization of Mesoporous Materials

591

[104] I. Honma, H.S. Zhou, D. Kundu, and A. Endo, Structural Control of Surfactant-templated Hexagonal Cubic, and Lamellar Mesoporous Silicate Thin Films Prepared by Spin-casting. Adv. Mater., 2000, 12, 1529–1533. [105] C.B. Gao, Y. Sakamoto, K. Sakamoto, O. Terasaki, and S.N. Che, Synthesis and Characterization of Mesoporous Silica AMS-10 with Bicontinuous Cubic Pn3m Symmetry. Angew. Chem., Int. Ed., 2006, 45, 4295–4298. [106] J. Feng, Q. Huo, P.M. Petroff, and G.D. Stucky, Morphology Definition by Disclinations and Dislocations in a Mesostructured Silicate Crystal. Appl. Phys. Lett., 1997, 71, 620–622. [107] Z.Y. Yuan, M.F. Six-Boulanger, and B.L. Su, Do Mesostructured Materials have an Unusual Macrolamellar Structure? Angew. Chem. Int. Ed., 2003, 42, 1572–1573. [108] M. Kruk and M. Jaroniec, Gas Adsorption Characterization of Ordered Organic-inorganic Nanocomposite Materials. Chem. Mater., 2001, 13, 3169–3183. [109] R. Vargas, P. Mariani, A. Gulik, and V. Luzzati, Cubic Phases of Lipid-Containing Systems – the Structure of Phase-Q223 (Space Group Pm3n) – an X-Ray-Scattering Study. J. Mol. Biol., 1992, 225, 137–145. [110] Q. Cai, W.Y. Lin, F.S. Xiao, W.Q. Pang, X.H. Chen, and B.S. Zou, The Preparation of Highly Ordered MCM-41 with Extremely Low Surfactant Concentration, Microporous Mesoporous Mater., 1999, 32, 1–15. [111] Z. Liu, Y. Sakamoto, T. Ohsuna, K. Hiraga, O. Terasaki, C.H. Ko, H.J. Shin, and R. Ryoo, TEM Studies of Platinum Nanowires Fabricated in Mesoporous Silica MCM-41. Angew. Chem., Int. Ed., 2000, 39, 3107–3110. [112] M. Kruk and M. Jaroniec, Argon Adsorption at 77 K as a Useful Tool for the Elucidation of Pore Connectivity in Ordered Materials with Large Cagelike Mesopores. Chem. Mater., 2003, 15, 2942–2949. [113] D.Y. Zhao, J.L. Feng, Q.S. Huo, N. Melosh, G.H. Fredrickson, B.F. Chmelka, and G.D. Stucky, Triblock Copolymer Syntheses of Mesoporous Silica with Periodic 50 to 300 angstrom Pores. Science, 1998, 279, 548–552. [114] A. Sayari, P. Liu, M. Kruk, and M. Jaroniec, Characterization of Large-pore MCM-41 Molecular Sieves Obtained via Hydrothermal Restructuring. Chem. Mater., 1997, 9, 24992506. [115] A.H. Lu, W. Schmidt, B. Spliethoff, and F. Schuth, Synthesis and Characterization of Nanocast Silica NCS-1 with CMK-3 as a Template. Chem. Eur. J., 2004, 10, 6085–6092. [116] M. Kruk, M. Jaroniec, C.H. Ko, and R. Ryoo, Characterization of the Porous Structure of SBA-15. Chem. Mater., 2000, 12, 1961–1968. [117] R. Ryoo, C.H. Ko, M. Kruk, V. Antochshuk, and M. Jaroniec, Block-copolymer-templated Ordered Mesoporous Silica: Array of Uniform Mesopores or Mesopore-Micropore Network? J. Phys. Chem. B, 2000, 104, 11465–11471. [118] M. Kruk, M. Jaroniec, Y. Sakamoto, O. Terasaki, R. Ryoo, and C.H. Ko, Determination of Pore Size and Pore Wall Structure of MCM-41 by Using Nitrogen Adsorption, Transmission Electron Microscopy, and X-ray Diffraction. J. Phys. Chem. B, 2000, 104, 292–301. [119] A. Galarneau, N. Cambon, F. Di Renzo, R. Ryoo, M. Choi, and F. Fajula, Microporosity and Connections between Pores in SBA-15 Mesostructured Silicas as a Function of the Temperature of Synthesis. New J. Chem., 2003, 27, 73–79. [120] A. Sayari, B.H. Han, and Y. Yang, Simple Synthesis Route to Monodispersed SBA-15 Silica Rods. J. Am. Chem. Soc., 2004, 126, 14348–14349. [121] A. Sayari and Y. Yang, SBA-15 Templated Mesoporous Carbon: New Insights into the SBA15 Pore Structure. Chem. Mater., 2005, 17, 6108–6113. [122] J. Fan, C.Z. Yu, L.M. Wang, B. Tu, D.Y. Zhao, Y. Sakamoto, and O. Terasaki, Mesotunnels on the Silica Wall of Ordered SBA-15 to Generate Three-dimensional Large-pore Mesoporous Networks. J. Am. Chem. Soc., 2001, 123, 12113–12114.

592

Chemistry of Zeolites and Related Porous Materials

[123] F.Q. Zhang, Y. Yan, H.F. Yang, Y. Meng, C.Z. Yu, B. Tu, and D.Y. Zhao, Understanding Effect of Wall Structure on the Hydrothermal Stability of Mesostructured Silicas SBA-15. J. Phys. Chem. B, 2005, 109, 8723–8732. [124] S. Wu, Y. Han, Y.C. Zou, J.W. Song, L. Zhao, Y. Di, S.Z. Liu, and F.S. Xiao, Synthesis of Heteroatom Substituted SBA-15 by the ‘‘pH-adjusting’’ Method. Chem. Mater., 2004, 16, 486–492. [125] K. Morishige, N. Tateishi, and S. Fukuma, Capillary Condensation of Nitrogen in MCM-48 and SBA-16. J. Phys. Chem. B, 2003, 107, 5177–5181. [126] F.X. Chen, L.M. Huang, and Q.Z. Li, Synthesis of MCM-48 using Mixed Cationic-Anionic Surfactants as Templates. Chem. Mater., 1997, 9, 2685–2686. [127] R. Ryoo, S.H. Joo, and J.M. Kim, Energetically Favored Formation of MCM-48 from Cationic-Neutral Surfactant Mixtures. J. Phys. Chem. B, 1999, 103, 7435–7440. [128] A. Sayari, Novel Synthesis of High-quality MCM-48 Silica. J. Am. Chem. Soc., 2000, 122, 6504–6505. [129] K.W. Gallis and C.C. Landry, Synthesis of MCM-48 by a Phase Transformation Process. Chem. Mater., 1997, 9, 2035–2038. [130] S.N. Che, S.Y. Lim, M. Kaneda, H. Yoshitake, O. Terasaki, and T. Tatsumi, The Effect of the Counteranion on the Formation of Mesoporous Materials under the Acidic Synthesis Process. J. Am. Chem. Soc., 2002, 124, 13962–13963. [131] X.Y. Liu, B.Z. Tian, C.Z. Yu, F. Gao, S.H. Xie, B. Tu, R.C. Che, L.M. Peng, and D.Y. Zhao, Room-temperature Synthesis in Acidic Media of Large-pore Three-dimensional Bicontinuous Mesoporous Silica with Ia3d Symmetry. Angew. Chem. Int. Ed., 2002, 41, 3876–3878. [132] B.Z. Tian, X.Y. Liu, L.A. Solovyov, Z. Liu, H.F. Yang, Z.D. Zhang, S.H. Xie, F.Q. Zhang, B. Tu, C.Z. Yu, O. Terasaki, and D.Y. Zhao, Facile Synthesis and Characterization of Novel Mesoporous and Mesorelief Oxides with Gyroidal Structures. J. Am. Chem. Soc., 2004, 126, 865–875. [133] T.W. Kim, F. Kleitz, B. Paul, and R. Ryoo, MCM-48-like Large Mesoporous Silicas with Tailored Pore Structure: Facile Synthesis Domain in a Ternary Triblock Copolymer-ButanolWater System. J. Am. Chem. Soc., 2005, 127, 7601–7610. [134] M. Clerc, A New Symmetry for the Packing of Amphiphilic Direct Micelles, J. Phys. II, Fr., 1996, 6, 961–968. [135] W.Z. Zhou, H.M.A. Hunter, P.A. Wright, Q.F. Ge, and J.M. Thomas. Imaging the Pore Structure and Polytypic Intergrowths in Mesoporous Silica. J. Phys. Chem. B, 1998, 102, 6933–6936. [136] A.E. Garcia-Bennett, S. Willliamson, P.A. Wright, and I.J. Shannon, Control of Structure, Pore Size and Morphology of Three-dimensionally Ordered Mesoporous Silicas Prepared using the Dicationic Surfactant [CH3(CH2)(15)N(CH3)(2)(CH2)(3)N(CH3)(3)]Br-2. J. Mater. Chem., 2002, 12, 3533–3540. [137] C.Z. Yu, Y.H. Yu, L. Miao, and D.Y. Zhao, Highly Ordered Mesoporous Silica Structures Templated by Poly(butylene oxide) Segment Di- and Tri-block Copolymers. Microporous Mesoporous Mater., 2001, 44, 65–72. [138] J.R. Matos, M. Kruk, L.P. Mercuri, M. Jaroniec, L. Zhao, T. Kamiyama, O. Terasaki, T.J. Pinnavaia, and Y. Liu, Ordered Mesoporous Silica with Large Cage-like Pores: Structural Identification and Pore Connectivity Design by Controlling the Synthesis Temperature and Time. J. Am. Chem. Soc., 2003, 125, 821–829. [139] M. Kruk, V. Antochshuk, J.R. Matos, L.P. Mercuri, and M. Jaroniec, Determination and Tailoring the Pore Entrance Size in Ordered Silicas with Cage-like Mesoporous Structures. J. Am. Chem. Soc., 2002, 124, 768–769. [140] A. Vinu, V. Murugesan, and M. Hartmann, Pore Size Engineering and Mechanical Stability of the Cubic Mesoporous Molecular Sieve SBA-1. Chem. Mater., 2003, 15, 1385–1393.

Synthesis, Structure, and Characterization of Mesoporous Materials

593

[141] S. Guan, S. Inagaki, T. Ohsuna, and O. Terasaki, Cubic Hybrid Organic-inorganic Mesoporous Crystal with a Decaoctahedral Shape. J. Am. Chem. Soc., 2000, 122, 5660–5661. [142] S. Che, Y. Sakamoto, O. Terasaki, and T. Tatsumi, Control of Crystal Morphology of SBA-1 Mesoporous Silica. Chem. Mater., 2001, 13, 2237–2239. [143] S.N. Che, Y. Sakamoto, O. Terasaki, and T. Tatsumi, The Structure and Morphology Control of Mesoporous Silica under Acidic Conditions, Microporous Mesoporous Mater., 2005, 85, 207–218. [144] C. Zapilko and R. Anwander, Size-selective Surface Silylation of Cagelike Mesoporous Silica SBA-2 with Disilazane Reagents. Chem. Mater., 2006, 18, 1479–1482. [145] T.W. Kim, R. Ryoo, M. Kruk, K.P. Gierszal, M. Jaroniec, S. Kamiya, and O. Terasaki, Tailoring the Pore Structure of SBA-16 Silica Molecular Sieve through the Use of Copolymer Blends and Control of Synthesis Temperature and Time. J. Phys. Chem. B, 2004, 108, 11480– 11489. [146] P.I. Ravikovitch and A.V. Neimark, Density Functional Theory of Adsorption in Spherical Cavities and Pore Size Characterization of Templated Nanoporous Silicas with Cubic and Three-dimensional Hexagonal Structures, Langmuir, 2002, 18, 1550–1560. [147] C.Z. Yu, B.Z. Tian, J. Fan, G.D. Stucky, and D.Y. Zhao, Nonionic Block Copolymer Synthesis of Large-pore Cubic Mesoporous Single Crystals by Use of Inorganic Salts. J. Am. Chem. Soc., 2002, 124, 4556–4557. [148] S. Che, Z. Liu, T. Ohsuna, K. Sakamoto, O. Terasaki, and T. Tatsumi, Synthesis and Characterization of Chiral Mesoporous Silica, Nature (London), 2004, 429, 281–284. [149] C.Z. Yu, J. Fan, and D.Y. Zhao, Block Copolymer Synthesis of High-quality Cubic, Large Pore Mesoporous Millimeter Spheres in the Presence of Inorganic Salts, Acta Chim. Sin., 2002, 60, 1357–1360. [150] L.M. Wang, H. Fan, B.Z. Tian, H.F. Yang, C.Z. Yu, B. Tu, and D.Y. Zhao, Synthesis and Characterization of Small Pore Thick-walled SBA-16 Templated by Oligomeric Surfactant with Ultra-long Hydrophilic Chains. Microporous Mesoporous Mater., 2004, 67, 135–141. [151] K. Miyasaka, L. Han, S.N. Che, and O. Terasaki, A Lesson from the Unusual Morphology of Silica Mesoporous Crystals: Growth and Close Packing of Spherical Micelles with Multiple Twinning. Angew. Chem. Int. Ed., 2006, 45, 6516–6519. [152] J. Fan, C.Z. Yu, J. Lei, Q. Zhang, T.C. Li, B. Tu, W.Z. Zhou, and D.Y. Zhao, Low-temperature Strategy to Synthesize Highly Ordered Mesoporous Silicas with Very Large Pores. J. Am. Chem. Soc., 2005, 127, 10794–10795. [153] U. Henriksson, E.S. Blackmore, G.J.T. Tiddy, and O. Soderman, Intermediate Liquid Crystalline Phases in the Binary System C16TACL-H2O: – An NMR and Low-Angle X-ray Diffraction Study. J. Phys. Chem., 1992, 96, 3894–3902. [154] X. Auvray, C. Petipas, I. Rico, and A. Lattes, X-Ray-Diffraction Study of the Intermediate Phases Formed by Sodium Dodecyl-Sulfate in Aqueous Mixtures of N-Methylformamide, Liq. Cryst., 1994, 17, 109–126. [155] J. El Haskouri, S. Cabrera, M. Caldes, C. Guillem, J. Latorre, A. Beltran, D. Beltran, M.D. Marcos, and P. Amoros, Surfactant-assisted Synthesis of the SBA-8 Mesoporous Silica by Using Nonrigid Commercial Alkyltrimethyl Ammonium Surfactants. Chem. Mater., 2002, 14, 2637–2643. [156] T. Kimura, T. Kamata, M. Fuziwara, Y. Takano, M. Kaneda, Y. Sakamoto, O. Terasaki, Y. Sugahara, and K. Kuroda, Formation of Novel Ordered Mesoporous Silicas with Square Channels and their Direct Observation by Transmission Electron Microscopy. Angew. Chem., Int. Ed., 2000, 39, 3855–3859. [157] P.T. Tanev and T.J. Pinnavaia, Mesoporous Silica Molecular Sieves Prepared by Ionic and Neutral Surfactant Templating: A Comparison of Physical Properties. Chem. Mater., 1996, 8, 2068–2079.

594

Chemistry of Zeolites and Related Porous Materials

[158] A.F. Gross, V.H. Le, B.L. Kirsch, and S.H. Tolbert, Chemical Control of Phase Transformation Kinetics in Periodic Silica/Surfactant Composites. J. Am. Chem. Soc., 2002, 124, 3713–3724. [159] S.H. Tolbert, C.C. Landry, G.D. Stucky, B.F. Chmelka, P. Norby, J.C. Hanson, and A. Monnier, Phase Transitions in Mesostructured Silica/Surfactant Composites: Surfactant Packing and the Role of Charge Density Matching. Chem. Mater., 2001, 13, 2247–2256. [160] C.C. Landry, S.H. Tolbert, K.W. Gallis, A. Monnier, G.D. Stucky, F. Norby, and J.C. Hanson, Phase Transformations in Mesostructured Silica/Surfactant Composites. Mechanisms for Change and Applications to Materials Synthesis. Chem. Mater., 2001, 13, 1600–1608. [161] S.N. Che, H.C. Li, S. Lim, Y. Sakamoto, O. Terasaki, and T. Tatsumi, Synthesis Mechanism of Cationic Surfactant Templating Mesoporous Silica under an Acidic Synthesis Process. Chem. Mater., 2005, 17, 4103–4113. [162] M.C. Liu, H.S. Sheu, and S.F. Cheng, Drying Induced Phase Transformation of Mesoporous Silica. Chem. Commun., 2002, 2854–2855. [163] A. Sayari, Y. Yang, M. Kruk, and M. Jaroniec, Expanding the Pore Size of MCM-41 Silicas: Use of Amines as Expanders in Direct Synthesis and Postsynthesis Procedures. J. Phys. Chem. B, 1999, 103, 3651–3658. [164] P.I. Ravikovitch and A.V. Neimark, Relations between Structural Parameters and Adsorption Characterization of Templated Nanoporous Materials with Cubic Symmetry. Langmuir, 2000, 16, 2419–2423. [165] M. Kruk, M. Jaroniec, T.W. Kim, and R. Ryoo, Synthesis and Characterization of Hexagonally Ordered Carbon Nanopipes. Chem. Mater., 2003, 15, 2815–2823. [166] A. Imhof and D.J. Pine, Ordered Macroporous Materials by Emulsion Templating. Nature (London), 1997, 389, 948–951. [167] O.D. Velev, T.A. Jede, R.F. Lobo, and A.M. Lenhoff, Porous Silica via Colloidal Crystallization, Nature (London), 1997, 389, 447–448. [168] B.T. Holland, C.F. Blanford, and A. Stein, Synthesis of Macroporous Minerals with Highly Ordered Three-dimensional Arrays of Spheroidal Voids, Science, 1998, 281, 538–540. [169] M. Antonietti, B. Berton, C. Goltner, and H.P. Hentze, Synthesis of Mesoporous Silica with Large Pores and Bimodal Pore Size Distribution by Templating of Polymer Latices. Adv. Mater., 1998, 10, 154–159. [170] S.A. Davis, S.L. Burkett, N.H. Mendelson, and S. Mann, Bacterial Templating of Ordered Macrostructures in Silica and Silica-Surfactant Mesophases. Nature (London), 1997, 385, 420–423. [171] P. Schmidt-Winkel, W.W. Lukens, P.D. Yang, D.I. Margolese, J.S. Lettow, J.Y. Ying, and G.D. Stucky, Microemulsion Templating of Siliceous Mesostructured Cellular Foams with Well-defined Ultralarge Mesopores. Chem. Mater., 2000, 12, 686–696. [172] K. Miyazawa and S. Inagaki, Control of the Microporosity within the Pore Walls of Ordered Mesoporous Silica SBA-15. Chem. Commun., 2000, 2121–2122. [173] P. Van Der Voort, P.I. Ravikovitch, K.P. De Jong, M. Benjelloun, E. Van Bavel, A.H. Janssen, A.V. Neimark, B.M. Weckhuysen, and E.F. Vansant, A New Templated Ordered Structure with Combined Micro- and Mesopores and Internal Silica Nanocapsules. J. Phys. Chem. B, 2002, 106, 5873–5877. [174] Y.M. Fang and H.Q. Hu, An Ordered Mesoporous Aluminosilicate with Completely Crystalline Zeolite Wall Structure. J. Am. Chem. Soc., 2006, 128, 10636–10637. [175] D.Y. Zhao, P.D. Yang, B.F. Chmelka, and G.D. Stucky, Multiphase Assembly of Mesoporous-Macroporous Membranes. Chem. Mater., 1999, 11, 1174–1178. [176] K. Suzuki, K. Ikari, and H. Imai, Synthesis of Silica Nanoparticles having a Wellordered Mesostructure using a Double Surfactant System. J. Am. Chem. Soc., 2004, 126, 462–463.

Synthesis, Structure, and Characterization of Mesoporous Materials

595

[177] A.A. Dong, Y.J. Wang, Y. Tang, Y.H. Zhang, N. Ren, and Z. Gao, Mechanically Stable Zeolite Monoliths with Three-dimensional Ordered Macropores by the Transformation of Mesoporous Silica Spheres. Adv. Mater., 2002, 14, 1506–1510. [178] C.G. Wu and T. Bein, Microwave Synthesis of Molecular Sieve MCM-41. Chem. Commun., 1996, 925–926. [179] G.A. Tompsett, W.C. Conner, and K.S. Yngvesson, Microwave Synthesis of Nanoporous Materials. ChemPhysChem, 2006, 7, 296–319. [180] C.J. Brinker, Y.F. Lu, A. Sellinger, and H.Y. Fan, Evaporation-induced Self-assembly: Nanostructures made easy. Adv. Mater., 1999, 11, 579–585. [181] N.A. Melosh, P. Davidson, and B.F. Chmelka, Monolithic Mesophase Silica with Large Ordering Domains. J. Am. Chem. Soc., 2000, 122, 823–829. [182] D. Danino, Y. Talmon, and R. Zana, Alkanediyl-Alpha, Omega-Bis(Dimethylalkylammonium Bromide) Surfactants (Dimeric Surfactants). 5. Aggregation and Microstructure in Aqueous-Solutions. Langmuir, 1995, 11, 1448–1456. [183] R. Zana and H. Levy, Micellar Properties of Omega-Hydroxyalkyldodecyldimethylammonium Bromides in Aqueous-Solutions. J. Colloid Interface Sci., 1995, 170, 128–133. [184] M. Yada, H. Takenaka, M. Machida, and T. Kijima, Mesostructured Gallium Oxides Templated by Dodecyl Sulfate Assemblies. J. Chem. Soc., Dalton Trans., 1998, 1547–1550. [185] S. Che, A.E. Garcia-Bennett, T. Yokoi, K. Sakamoto, H. Kunieda, O. Terasaki, and T. Tatsumi, A Novel Anionic Surfactant Templating Route for Synthesizing Mesoporous Silica with Unique Structure. Nature Mater. (London), 2003, 2, 801–805. [186] S.S. Kim, W.Z. Zhang, and T.J. Pinnavaia, Ultrastable Mesostructured Silica Vesicles. Science, 1998, 282, 1302–1305. [187] M. Templin, A. Franck, A. DuChesne, H. Leist, Y.M. Zhang, R. Ulrich, V. Schadler, and U. Wiesner, Organically Modified Aluminosilicate Mesostructures from Block Copolymer Phases. Science, 1997, 278, 1795–1798. [188] M.F. Ottaviani, A. Galarneau, D. Desplantier-Giscard, F. Di Renzo, and F. Fajula, EPR Investigations on the Formation of Micelle-templated Silica. Microporous Mesoporous Mater., 2001, 44, 1–8. [189] R. Ryoo, C.H. Ko, and I.S. Park, Synthesis of Highly Ordered MCM-41 by Micelle-packing Control with Mixed Surfactants. Chem. Commun., 1999, 1413–1414. [190] S. Kawi and M.W. Lai, Supercritical Fluid Extraction of Surfactant Template from MCM-41. Chem. Commun., 1998, 1407–1408. [191] J.B. Pang, K.Y. Qiu, Y. Wei, X.J. Lei, and Z.F. Liu, A Facile Preparation of Transparent and Monolithic Mesoporous Silica Materials. Chem. Commun., 2000, 477–478. [192] A. Stein, Advances in Microporous and Mesoporous Solids – Highlights of Recent Progress. Adv. Mater., 2003, 15, 763–775. [193] A.H. Lu, W. Schmidt, A. Taguchi, B. Spliethoff, B. Tesche, and F. Schuth, Taking Nanocasting One Step Further: Replicating CMK-3 as a Silica Material. Angew. Chem., Int. Ed., 2002, 41, 3489–3492. [194] J.Y. Kim, S.B. Yoon, and J.S. Yu, Template Synthesis of a New Mesostructured Silica from Highly Ordered Mesoporous Carbon Molecular Sieves. Chem. Mater., 2003, 15, 1932–1934. [195] A.G. Dong, N. Ren, Y. Tang, Y.J. Wang, Y.H. Zhang, W.M. Hua, and Z. Gao, General Synthesis of Mesoporous Spheres of Metal Oxides and Phosphates. J. Am. Chem. Soc., 2003, 125, 4976–4977. [196] R. Ryoo and J.M. Kim, Structural Order in Mcm-41 Controlled by Shifting Silicate Polymerization Equilibrium. J. Chem. Soc. Chem. Commun., 1995, 711–712. [197] J.M. Kim, S. Jun, and R. Ryoo, Improvement of Hydrothermal Stability of mesoporous Silica using Salts: Reinvestigation for Time-dependent Effects. J. Phys. Chem. B, 1999, 103, 6200–6205.

596

Chemistry of Zeolites and Related Porous Materials

[198] Y.F. Shao, L.Z. Wang, J.L. Zhang, and M. Anpo, Novel Synthesis of High Hydrothermal Stability and Long-range Order MCM-48 with a Convenient Method. Microporous Mesoporous Mater., 2005, 86, 314–322. [199] D.T. On and S. Kaliaguine, Zeolite-coated Mesostructured Cellular Silica Foams. J. Am. Chem. Soc., 2003, 125, 618–619. [200] H.P. Lin, C.Y. Mou, and S.B. Liu, Formation of Mesoporous Silica Nanotubes, Adv. Mater., 2000, 12, 103–106. [201] A. Sayari, S. Hamoudi, and Y. Yang, Applications of Pore-expanded Mesoporous Silica. 1. Removal of Heavy Metal Cations and Organic Pollutants from Wastewater. Chem. Mater., 2005, 17, 212–216. [202] H.P. Lin and C.Y. Mou, Salt Effect in Post-synthesis Hydrothermal Treatment of MCH-41, Microporous Mesoporous Mater., 2002, 55, 69–80. [203] F.S. Xiao, Ordered Mesoporous Materials with Improved Stability and Catalytic Activity. Top. Catal., 2005, 35, 9–24. [204] F.S. Xiao, Y. Han, Y. Yu, X.J. Meng, M. Yang, and S. Wu, Hydrothermally Stable Ordered Mesoporous Titanosilicates with Highly Active Catalytic sites. J. Am. Chem. Soc., 2002, 124, 888–889. [205] Y. Liu, W.Z. Zhang, and T.J. Pinnavaia, Steam-stable MSU-S Aluminosilicate Mesostructures Assembled from Zeolite ZSM-5 and Zeolite Beta Seeds. Angew. Chem., Int. Ed., 2001, 40, 1255–1258. [206] Y. Liu and T.J. Pinnavaia, Assembly of Hydrothermally Stable Aluminosilicate Foams and Large-pore Hexagonal Mesostructures from Zeolite Seeds under Strongly Acidic Conditions. Chem. Mater., 2002, 14, 3–5. [207] Z.T. Zhang, Y. Han, F.S. Xiao, S.L. Qiu, L. Zhu, R.W. Wang, Y. Yu, Z. Zhang, B.S. Zou, Y.Q. Wang, H.P. Sun, D.Y. Zhao, and Y. Wei, Mesoporous Aluminosilicates with Ordered Hexagonal Structure, Strong Acidity, and Extraordinary Hydrothermal Stability at High Temperatures. J. Am. Chem. Soc., 2001, 123, 5014–5021. [208] H. Wang, Y. Liu, and T.J. Pinnavaia, Highly Acidic Mesostructured Aluminosilicates Assembled from Surfactant-mediated Zeolite Hydrolysis Products. J. Phys. Chem. B, 2006, 110, 4524–4526. [209] A. Sakthivel, S.J. Huang, W.H. Chen, Z.H. Lan, K.H. Chen, H.P. Lin, C.Y. Mou, and S.B. Liu, Direct Synthesis of Highly Stable Mesoporous Molecular Sieves containing Zeolite Building Units. Adv. Funct. Mater., 2005, 15, 253–258. [210] A.H. Lu and F. Schuth, Nanocasting Pathways to Create Ordered Mesoporous Solids, C. R. Acad. Sci., Ser. IIc: Chim., 2005, 8, 609–620. [211] M. Kang, S.H. Yi, H.I. Lee, J.E. Yie, and J.M. Kim, Reversible Replication between Ordered Mesoporous Silica and Mesoporous Carbon. Chem. Commun., 2002, 1944–1945. [212] B.Z. Tian, X.Y. Liu, H.F. Yang, S.H. Xie, C.Z. Yu, B. Tu, and D.Y. Zhao, General Synthesis of Ordered Crystallized Metal Oxide Nanoarrays Replicated by Microwave-digested Mesoporous Silica. Adv. Mater., 2003, 15, 1370–1374. [213] J. Roggenbuck and M. Tiemann, Ordered Mesoporous Magnesium Oxide with High Thermal Stability Synthesized by Exotemplating using CMK-3 Carbon. J. Am. Chem. Soc., 2005, 127, 1096–1097. [214] C.Z. Yu, B.Z. Tian, B. Fan, G.D. Stucky, and D.Y. Zhao, Salt Effect in the Synthesis of Mesoporous Silica Templated by Non-ionic Block Copolymers, Chem. Commun., 2001, 2726–2727. [215] B.Z. Tian, X.Y. Liu, B. Tu, C.Z. Yu, J. Fan, L.M. Wang, S.H. Xie, G.D. Stucky, and D.Y. Zhao, Self-adjusted Synthesis of Ordered Stable Mesoporous Minerals by Acid-Base Pairs, Nature Mater. (London), 2003, 2, 159–163.

Synthesis, Structure, and Characterization of Mesoporous Materials

597

[216] F.S. Xiao, Ordered Mesoporous Silica-based Materials Templated from FluorocarbonHydrocarbon Surfactant Mixtures and Semi-fluorinated Surfactants. Curr. Opin. Colloid Interface Sci., 2005, 10, 94–101. [217] L.F. Wang, K.F. Lin, Y. Di, D.L. Zhang, C.J. Li, Q. Yang, C.Y. Yin, Z.H. Sun, D.Z. Jiang, and F.S. Xiao, High-temperature Synthesis of Stable Ordered Mesoporous Silica Materials using Mesoporous Carbon as a Hard Template. Microporous Mesoporous Mater., 2005, 86, 81–88. [218] R.A. Pai, R. Humayun, M.T. Schulberg, A. Sengupta, J.N. Sun, and J.J. Watkins, Mesoporous Silicates prepared using Preorganized Templates in Supercritical Fluids. Science, 2004, 303, 507–510. [219] S. Zheng, L.A. Gao, Q.H. Zhang, and J.K. Guo, Synthesis, Characterization and Photocatalytic Properties of Titania-modified Mesoporous Silicate MCM-41, J. Mater. Chem., 2000, 10, 723–727. [220] A. Vinu, K.Z. Hossain, and K. Ariga, Recent Advances in Functionalization of Mesoporous Silica. J. Nanosci. Nanotechnol., 2005, 5, 347–371. [221] Y.H. Liu, H.P. Lin, and C.Y. Mou, Direct Method for Surface Silyl Functionalization of Mesoporous Silica, Langmuir, 2004, 20, 3231–3239. [222] F. de Juan and E. Ruiz-Hitzky, Selective Functionalization of Mesoporous Silica. Adv. Mater., 2000, 12, 430–432. [223] H.L. Castricum, M.C. Mittelmeijer-Hazeleger, A. Sah, and J. E. ten Elshof, Increasing the Hydrothermal Stability of Mesoporous SiO2 with Methylchlorosilanes – a ‘Structural’ Study, Microporous Mesoporous Mater., 2006, 88, 63–71. [224] A. Sayari and P. Liu, Non-silica Periodic Mesostructured Materials: Recent Progress, Microporous Mesoporous Mater., 1997, 12, 149–177. [225] D.L. Li, H.S. Zhou, and I. Honma, Design and Synthesis of Self-ordered Mesoporous Nanocomposite through Controlled in-situ Crystallization, Nature Mater. (London), 2004, 3, 65–72. [226] J.H. Zhu, W. Shen, H.L. Xu, Y.M. Zhou, C.Z. Yu, and D.Y. Zhao, Direct Synthesis of Titanium Substituted SBA-15 under Conventional Hydrothermal Conditions and their Catalysis Characterization. Acta. Chim. Sin., 2003, 61, 202–207. [227] F. Hoffmann, M. Cornelius, J. Morell, and M. Froba, Periodic Mesoporous Organosilicas (PMOs): Past, Present, and Future. J. Nanosci. Nanotechnol., 2006, 6, 265–288. [228] S. Inagaki, S. Guan, T. Ohsuna, and O. Terasaki, An Ordered Mesoporous Organosilica Hybrid Material with a Crystal-like Wall Structure. Nature (London), 2002, 416, 304–307. [229] S.A. Bagshaw and T.J. Pinnavaia, Mesoporous Alumina Molecular Sieves. Angew. Chem., Int. Ed. Engl., 1996, 35, 1102–1105. [230] F. Vaudry, S. Khodabandeh, and M.E. Davis, Synthesis of Pure Alumina Mesoporous Materials. Chem. Mater., 1996, 8, 1451–1464. [231] W. Deng, P. Bodart, M. Pruski, and B.H. Shanks, Characterization of Mesoporous Alumina Molecular Sieves Synthesized by Nonionic Templating. Microporous Mesoporous Mater., 2002, 52, 169–177. [232] H. Shibata, T. Ogura, T. Mukai, T. Ohkubo, H. Sakai, and M. Abe, Direct Synthesis of Mesoporous Titania Particles having a Crystalline Wall. J. Am. Chem. Soc., 2005, 127, 16396–16397. [233] Y.Q. Wang, X.H. Tang, L.X. Yin, W.P. Huang, Y.R. Hacohen, and A. Gedanken, Sonochemical Synthesis of Mesoporous Titanium Oxide with Wormhole-like Framework Structures. Adv. Mater., 2000, 12, 1183–1186. [234] J.C. Yu, L.Z. Zhang, and J.G. Yu, Direct Sonochemical Preparation and Characterization of Highly Active Mesoporous TiO2 with a Bicrystalline Framework. Chem. Mater., 2002, 14, 4647–4653.

598

Chemistry of Zeolites and Related Porous Materials

[235] Y.H. Yue, Z. Ma, W.M. Hua, and Z. Gao, Synthesis and Characterization of Mesoporous Titanium Dioxide, Acta. Chim. Sin., 2000, 58, 777–780. [236] B. Smarsly, D. Grosso, T. Brezesinski, N. Pinna, C. Boissiere, M. Antonietti, and C. Sanchez, Highly Crystalline Cubic Mesoporous TiO2 with 10-nm Pore Diameter made with a New Block Copolymer Template. Chem. Mater., 2004, 16, 2948–2952. [237] S.W. Boettcher, M.H. Bartl, J.G. Hu, and G.D. Stucky, Structural Analysis of Hybrid Titaniabased Mesostructured Composites, J. Am. Chem. Soc., 2005, 127, 9721–9730. [238] D.M. Antonelli and J.Y. Ying, Synthesis and Characterization of Hexagonally Packed Mesoporous Tantalum Oxide Molecular Sieves. Chem. Mater., 1996, 8, 874–881. [239] D.M. Antonelli and J.Y. Ying, Synthesis of a Stable Hexagonally Packed Meosporous Niobium Oxide Molecular Sieve through a Novel Ligand-asisted Templating Mechanism. Angew. Chem., Int. Ed. Engl., 1996, 35, 426–430. [240] Z.R. Tian, W. Tong, J.Y. Wang, N.G. Duan, V.V. Krishnan, and S.L. Suib, Manganese Oxide Mesoporous Structures: Mixed-valent Semiconducting Catalysts. Science, 1997, 276, 926– 930. [241] M. Tiemann and M. Froba, Mesostructured Aluminophosphates Synthesized with Supramolecular Structure Directors. Chem. Mater., 2001, 13, 3211–3217. [242] L.M. Wang, B.Z. Tian, H. Fan, X.Y. Liu, H.F. Yang, C.Z. Yu, B. Tu, and D.Y. Zhao, Block Copolymer Templating Synthese of Ordered Large-pore Stable Mesoporous Aluminophosphates and Fe-aluminophosphate Based on an ‘‘Acid-Base Pair’’ Route. Microporous Mesoporous Mater., 2004, 67, 123–133. [243] A. Bhaumik and S. Inagaki, Mesoporous Titanium Phosphate Molecular Sieves with Ionexchange Capacity. J. Am. Chem. Soc., 2001, 123, 691–696. [244] N. Mizuno, H. Hatayama, S. Uchida, and A. Taguchi, Tunable One-pot Syntheses of Hexagonal-, Cubic-, and Lamellar-Mesostructured Vanadium-Phosphorus Oxides. Chem. Mater., 2001, 13, 179–184. [245] S.D. Shen, B.Z. Tian, C.Z. Yu, S.H. Xie, Z.D. Zhang, B. Tu, and D.Y. Zhao, Synthesis of Highly Ordered Thermally Stable Cubic Mesostructured Zirconium Oxophosphate Templated by Tri-headgroup Quaternary Ammonium Surfactants, Chem. Mater., 2003, 15, 4046– 4051. [246] P.N. Trikalitis, K.K. Rangan, T. Bakas, and M.G. Kanatzidis, Single-crystal Mesostructured Semiconductors with Cubic Ia3d Symmetry and Ion-exchange Properties. J. Am. Chem. Soc., 2002, 124, 12255–12260. [247] F.O.M. Gaslain, J. Parmentier, V.P. Valtchev, and J. Patarin, First Zeolite Carbon Replica with a Well Resolved X-ray Diffraction Pattern. Chem. Commun., 2006, 991–993. [248] M. Kruk, M. Jaroniec, R. Ryoo, and S.H. Joo, Characterization of Ordered Mesoporous Carbons Synthesized using MCM-48 Silicas as Templates. J. Phys. Chem. B, 2000, 104, 7960–7968. [249] J. Lee, S. Yoon, T. Hyeon, S.M. Oh, and K.B. Kim, Synthesis of a New Mesoporous Carbon and its Application to Electrochemical Double-layer Capacitors. Chem. Commun., 1999, 2177–2178. [250] M. Kaneda, T. Tsubakiyama, A. Carlsson, Y. Sakamoto, T. Ohsuna, O. Terasaki, S.H. Joo, and R. Ryoo, Structural Study of Mesoporous MCM-48 and Carbon Networks Synthesized in the Spaces of MCM-48 by Electron Crystallography. J. Phys. Chem. B, 2002, 106, 1256–1266. [251] W.H. Zhang, C.H. Liang, H.J. Sun, Z.Q. Shen, Y.J. Guan, P.L. Ying, and C. Li, Synthesis of Ordered Mesoporous Carbons Composed of Nanotubes via Catalytic Chemical Vapor Deposition. Adv. Mater., 2002, 14, 1776–1778. [252] J.S. Lee, S.H. Joo, and R. Ryoo, Synthesis of Mesoporous Silicas of Controlled Pore Wall Thickness and their Replication to Ordered Nanoporous Carbons with Various Pore Diameters. J. Am. Chem. Soc., 2002, 124, 1156–1157.

Synthesis, Structure, and Characterization of Mesoporous Materials

599

[253] R. Ryoo, S.H. Joo, M. Kruk, and M. Jaroniec, Ordered Mesoporous Carbons. Adv. Mater., 2001, 13, 677–681. [254] J. Lee, S. Yoon, S.M. Oh, C.H. Shin, and T. Hyeon, Development of a New Mesoporous Carbon using an HMS Aluminosilicate Template. Adv. Mater., 2000, 12, 359–362. [255] S.S. Kim and T.J. Pinnavaia, A Low Cost Route to Hexagonal Mesostructured Carbon Molecular Sieves. Chem. Commun., 2001, 2418–2419. [256] J. Lee, K. Sohn, and T. Hyeon, Fabrication of Novel Mesocellular Carbon Foams with Uniform Ultralarge Mesopores. J. Am. Chem. Soc., 2001, 123, 5146–5147. [257] T.W. Kim, R. Ryoo, K.P. Gierszal, M. Jaroniec, L.A. Solovyov, Y. Sakamoto, and O. Terasaki, Characterization of Mesoporous Carbons Synthesized with SBA-16 Silica Template. J. Mater. Chem., 2005, 15, 1560–1571. [258] Y. Sakamoto, T.W. Kim, R. Ryoo, and O. Terasaki, Three-dimensional Structure of LargePore Mesoporous Cubic Ia3d Silica with Complementary Pores and its Carbon Replica by Electron Crystallography., Angew. Chem., Int. Ed., 2004, 43, 5231–5234. [259] F.Q. Zhang, Y. Meng, D. Gu, Y. Yan, C.Z. Yu, B. Tu, and D.Y. Zhao, A Facile Aqueous Route to Synthesize Highly Ordered Mesoporous Polymers and Carbon Frameworks with Ia3d Bicontinuous Cubic Structure. J. Am. Chem. Soc., 2005, 127, 13508–13509. [260] B.H. Han, W.Z. Zhou, and A. Sayari, Direct Preparation of Nanoporous Carbon by Nanocasting. J. Am. Chem. Soc., 2003, 125, 3444–3445. [261] Y. Meng, D. Gu, F.Q. Zhang, Y.F. Shi, H.F. Yang, Z. Li, C.Z. Yu, B. Tu, and D.Y. Zhao, Ordered Mesoporous Polymers and Homologous Carbon Frameworks: Amphiphilic Surfactant Templating and Direct Transformation. Angew. Chem., Int. Ed., 2005, 44, 7053–7059. [262] G.S. Attard, P.N. Bartlett, N.R.B. Coleman, J.M. Elliott, J.R. Owen, and J.H. Wang, Mesoporous Platinum Films from Lyotropic Liquid Crystalline Phases, Science, 1997, 278, 838–840. [263] D.H. Wang, W.L. Zhou, B.F. McCaughy, J.E. Hampsey, X.L. Ji, Y.B. Jiang, H.F. Xu, J.K. Tang, R.H. Schmehl, C. O’Connor, C.J. Brinker, and Y.F. Lu, Electrodeposition of Metallic Nanowire Thin Films using Mesoporous Silica Templates. Adv. Mater., 2003, 15, 130–133. [264] H.J. Shin, C.H. Ko, and R. Ryoo, Synthesis of Platinum Networks with Nanoscopic Periodicity using Mesoporous Silica as Template. J. Mater. Chem., 2001, 11, 260–261. [265] K. Lee, Y.H. Kim, S.B. Han, H.K. Kang, S. Park, W.S. Seo, J.T. Park, B. Kim, and S.B. Chang, Osmium Replica of Mesoporous Silicate MCM-48: Efficient and Reusable Catalyst for Oxidative Cleavage and Dihydroxylation Reactions. J. Am. Chem. Soc., 2003, 125, 6844– 6845. [266] Y.F. Shi, Y. Meng, D.H. Chen, S.J. Cheng, P. Chen, T.F. Yang, Y. Wan, and D.Y. Zhao, Highly Ordered Mesoporous Silicon Carbide Ceramics with Large Surface Areas and High Stability. Adv. Funct. Mater., 2006, 16, 561–567. [267] A. Vinu, K. Ariga, T. Mori, T. Nakanishi, S. Hishita, D. Golberg, and Y. Bando, Preparation and Characterization of Well-ordered Hexagonal Mesoporous Carbon Nitride. Adv. Mater., 2005, 17, 1648–1652. [268] J.M. Kim, S.K. Kim, and R. Ryoo, Synthesis of MCM-48 Single Crystals. Chem. Commun., 1998, 259–260. [269] D.Y. Zhao, J.Y. Sun, Q.Z. Li, and G.D. Stucky, Morphological Control of Highly Ordered Mesoporous Silica SBA-15. Chem. Mater., 2000, 12, 275–279. [270] O. Terasaki, T. Ohsuna, Z. Liu, Y. Sakamoto, and A.E. Garcia-Bennett, Mesoporous Cryst. Relat. Nano-Struct. Mater., 2004, 148, 261–288. [271] S.N. Che, S. Kamiya, O. Terasaki, and T. Tatsumi, The Formation of Cubic Pm3n Mesostructure by an Epitaxial Phase Transformation from Hexagonal p6mm Mesophase. J. Am. Chem. Soc., 2001, 123, 12089–12090.

600

Chemistry of Zeolites and Related Porous Materials

[272] Y.F. Lu, H.Y. Fan, A. Stump, T.L. Ward, T. Rieker, and C.J. Brinker, Aerosol-assisted Selfassembly of Mesostructured Spherical Nanoparicles. Nature (London), 1999, 398, 223–226. [273] Q. Cai, Z.S. Luo, W.Q. Pang, Y.W. Fan, X.H. Chen, and F.Z. Cui, Dilute Solution Routes to Various Controllable Morphologies of MCM-41 Silica with a Basic Medium. Chem. Mater., 2001, 13, 258–263. [274] H.P. Lin and C.P. Tsai, Synthesis of Mesoporous Silica Nanoparticles from a Lowconcentration C(n)TMAX-Sodium Silicate Components. Chem. Lett., 2003, 32, 1092–1093. [275] V.V. Guliants, M.A. Carreon, and Y.S. Lin, Ordered Mesoporous and Macroporous Inorganic Films and Membranes. J. Membr. Sci., 2004, 235, 53–72. [276] M. Ogawa, Formation of Novel Oriented Transparent Films of Layered Silica-Surfactant Nanocomposites. J. Am. Chem. Soc., 1994, 116, 7941–7942. [277] R.E. Williford, R.S. Addleman, X.S. Li, T.S. Zemanian, J.C. Birnbaum, and G.E. Fryxell, Pore Shape Evolution in Mesoporous Silica Thin Films: From Circular to Elliptical to Rectangular. J. Non-Cryst. Solids, 2005, 351, 2217–2223. [278] D.Y. Zhao, P.D. Yang, D.I. Margolese, B.F. Chmelka, and G.D. Stucky, Synthesis of Continuous Mesoporous Silica Thin Films with Three-dimensional Accessible Pore Structures. Chem. Commun., 1998, 2499–2500. [279] H. Miyata, T. Suzuki, A. Fukuoka, T. Sawada, M. Watanabe, T. Noma, K. Takada, T. Mukaide, and K. Kuroda, Silica Films with a Single-crystalline Mesoporous Structure. Nature Mater. (London), 2004, 3, 651–656. [280] P.C.A. Alberius, K.L. Frindell, R.C. Hayward, E.J. Kramer, G.D. Stucky, and B.F. Chmelka, General Predictive Syntheses of Cubic, Hexagonal, and Lamellar Silica and Titania Mesostructured Thin Films. Chem. Mater., 2002, 14, 3284–3294. [281] M. Ogawa, N. Shimura, and A. Ayral, Deposition of Thin Nanoporous Silica Layers on Solid Surfaces. Chem. Mater., 2006, 18, 1715–1718. [282] Q.S. Huo, J.L. Feng, F. Schuth, and G.D. Stucky, Preparation of Hard Mesoporous Silica Spheres. Chem. Mater., 1997, 9, 14–17. [283] P. Zhang, Y.L. Liu, N. Bai, Y.L. Fu, S.G. Li, X.J. Meng, H. Ding, and W.Q. Pang, Synthesis and Characterization of Mesoporous Co-SiO2 Spheres with High Surface Area. Chem. J. Chin. Univ.-Chin., 2002, 23, 1466–1469. [284] M. Grun, I. Lauer, and K.K. Unger, The Synthesis of Micrometer- and Submicrometer-size Spheres of Ordered Mesoporous Oxide MCM-41. Adv. Mater., 1997, 9, 254–257. [285] S.B. Yoon, K. Sohn, J.Y. Kim, C.H. Shin, J.S. Yu, and T. Hyeon, Fabrication of Carbon Capsules with Hollow Macroporous Core/Mesoporous Shell Structures, Adv. Mater., 2002, 14, 19–21. [286] Q.S. Huo, D.Y. Zhao, J.L. Feng, K. Weston, S.K. Buratto, G.D. Stucky, S. Schacht, and F. Schuth, Room Temperature Growth of Mesoporous Silica Fibers: A New High-surface-area Optical Waveguide. Adv. Mater., 1997, 9, 974–978. [287] F. Marlow, B. Spliethoff, B. Tesche, and D.Y. Zhao, The Internal Architecture of Mesoporous Silica Fibers. Adv. Mater., 2000, 12, 961–965. [288] F. Kleitz, F. Marlow, G.D. Stucky, and F. Schuth, Mesoporous Silica Fibers: Synthesis, Internal Structure, and Growth Kinetics. Chem. Mater., 2001, 13, 3587–3595. [289] P.D. Yang, D.Y. Zhao, B.F. Chmelka, and G.D. Stucky, Triblock-Copolymer-Directed Syntheses of Large-pore Mesoporous Silica Fibers, Chem. Mater., 1998, 10, 2033–2036. [290] J.F. Wang, C.-K. Tsung, W.B. Hong, Y.Y. Wu, J. Tang, and G.D. Stucky, Synthesis of Microporous Silica Nanofibers with Controlled Pore Architectures. Chem. Mater., 2004, 16, 5169–5181. [291] C.Z. Yu, B.Z. Tain, J. Fan, B. Tu, and D.Y. Zhao, Synthesis of Hierarchically Arrayed Silica Nanorods via Reverse Amphiphilic Block Copolymer Mesophases. Chem. J. Chin. Univ.Chin., 2003, 24, 5–8.

Synthesis, Structure, and Characterization of Mesoporous Materials

601

[292] H.P. Lin and C.Y. Mou, ‘‘Tubules-within-a-tubule’’ Hierarchical Order of Mesoporous Molecular Sieves in MCM-41. Science, 1996, 273, 765–768. [293] H.F. Yang, Q.H. Shi, B.Z. Tian, S.H. Xie, F.Q. Zhang, Y. Yan, B. Tu, and D.Y. Zhao, A Fast Way for Preparing Crack-free Mesostructured Silica Monoliths. Chem. Mater., 2003, 15, 536–541. [294] J.F. Wang, C.K. Tsung, R.C. Hayward, Y.Y. Wu, and G.D. Stucky, Single-crystal Mesoporous Silica Ribbons. Angew. Chem. Int. Ed., 2005, 44, 332–336. [295] Y.Y. Wu, G.S. Cheng, K. Katsov, S.W. Sides, J.F. Wang, J. Tang, G.H. Fredrickson, M. Moskovits, and G.D. Stucky, Composite Mesostructures by Nano-confinement. Nature Mater. (London), 2004, 3, 816–822. [296] A. Taguchi and F. Schuth, Ordered Mesoporous Materials in Catalysis. Microporous Mesoporous Mater., 2005, 77, 1–45. [297] S. Giri, B.G. Trewyn, M.P. Stellmaker, and V.S.Y. Lin, Stimuli-responsive Controlled-release Delivery System Based on Mesoporous Silica Nanorods Capped with Magnetic Nanoparticles. Angew. Chem., Int. Ed., 2005, 44, 5038–5044. [298] N.K. Mal, M. Fujiwara, and Y. Tanaka, Photocontrolled Reversible Release of Guest Molecules from Coumarin-modified Mesoporous Silica. Nature (London), 2003, 421, 350–353. [299] M. Hartmann, Ordered Mesoporous Materials for Bioadsorption and Biocatalysis. Chem. Mater., 2005, 17, 4577–4593. [300] J.W. Zhao, F. Gao, Y.L. Fu, W. Jin, P.Y. Yang, and D.Y. Zhao, Biomolecule Separation using Large Pore Mesoporous SBA-15 as a Substrate in High Performance Liquid Chromatography. Chem. Commun., 2002, 752–753. [301] R.S. Franchi, P.J.E. Harlick, and A. Sayari, Applications of Pore-expanded Mesoporous Silica. 2. Development of a High-capacity, Water-tolerant Adsorbent for CO2. Ind. Eng. Chem. Res., 2005, 44, 8007–8013. [302] L.X. Zhang, W.H. Zhang, J.L. Shi, Z. Hua, Y.S. Li, and J. Yan, A New Thioether Functionalized Organic-Inorganic Mesoporous Composite as a Highly Selective and Capacious Hg2þ Adsorbent. Chem. Commun., 2003, 210–211. [303] J.L. Shi, Z.L. Hua, and L.X. Zhang, Nanocomposites from Ordered Mesoporous Materials. J. Mater. Chem., 2004, 14, 795–806. [304] H.F. Yang, Q.Y. Lu, F. Gao, Q.H. Shi, Y. Yan, F.Q. Zhang, S.H. Xie, B. Tu, and D.Y. Zhao, One-step Synthesis of Highly Ordered Mesoporous Silica Monoliths with Metal Oxide Nanocrystals in their Channels. Adv. Funct. Mater., 2005, 15, 1377–1384. [305] H.F. Yang and D.Y. Zhao, Synthesis of Replica Mesostructures by the Nanocasting Strategy. J. Mater. Chem., 2005, 15, 1217–1231. [306] Y.Y. Wu, T. Livneh, Y.X. Zhang, G.S. Cheng, J.F. Wang, J. Tang, M. Moskovits, and G.D. Stucky, Templated Synthesis of Highly Ordered Mesostructured Nanowires and Nanowire Arrays. Nano Lett., 2004, 4, 2337–2342. [307] B.J. Scott, G. Wirnsberger, and G.D. Stucky, Mesoporous and Mesostructured Materials for Optical Applications. Chem. Mater., 2001, 13, 3140–3150. [308] W.P. Cai, M. Tan, G.Z. Wang, and L.D. Zhang, Reversible Transition between Transparency and Opacity for the Porous Silica Host Dispersed with Silver Nanometer Particles within its Pores. Appl. Phys. Lett., 1996, 69, 2980–2982. [309] S. Baskaran, J. Liu, K. Domansky, N. Kohler, X.H. Li, C. Coyle, G.E. Fryxell, S. Thevuthasan, and R.E. Williford, Low Dielectric Contant Mesoporous Silica Films through Molecularly Templated Synthesis. Adv. Mater., 2000, 12, 291–294. [310] F. Goettmann, A. Moores, C. Boissiere, P. Le Floch, and C. Sanchez, A Selective Chemical Sensor Based on the Plasmonic Response of Phosphinine-stabilized Gold Nanoparticles Hosted on Periodically Organized Mesoporous Silica Thin Layers. Small, 2005, 1, 636–639. [311] K.J. Edler, Current Understanding of Formation Mechanisms in Surfactant-templated Materials. Aust. J. Chem., 2005, 58, 627–643.

9 Porous Host–Guest Advanced Materials Zeolite molecular sieves have been used extensively. The conventional application fields of zeolites include ion-exchange, gas separation, adsorption, catalysis, and so on. However, in the past two decades, the use of zeolites as hosts to prepare host–guest assembly materials has attracted increasing interest. In particular, the mesoporous molecular sieves invented in the 1990s have laid new foundations for host–guest assembly. The compositions and structures of zeolitic materials are versatile, and, in principle, it is possible to prepare a particular guest species using a special zeolite host. In return, through variation of the guest species and the assembly approaches, various guest materials, which may exhibit versatile chemical-physical properties, can be prepared. Because the diameters of the channels and cages in zeolites fall in the nanometer scale, the guest particles confined in a zeolite should also be nanometer sized, and this lays foundation for the chemical preparation of guest species with quantum size effects.[1,2] According to the guest species used, host–guest composite materials based on molecular sieve hosts are classified into five major categories. The first category includes composites of zeolites with encapsulated metal clusters or metal ion clusters, and occasionally in the clusters there are nonmetal ligands such as carbonyl groups. The second category concerns host–guest composite materials of zeolites and dye molecules. The third category involves polymers or carbons such as fullerenes and carbon nanotubes in zeolitic materials. Whereas the fourth category consists of inorganic semiconductor particles in the channels and cages of zeolites, the fifth includes metal-complex-encapsulating zeolites. These host–guest materials not only exhibit interesting chemical-physical properties but also show great application potential in a variety of areas. In the past decade, attention has been attracted to microporous compounds formed by metal-organic coordination polymers. These compounds differ from the conventional

Chemistry of Zeolites and Related Porous Material – Synthesis and Structure Ruren Xu, Wenqin Pang, Jihong Yu, Qisheng Huo and Jiesheng Chen # 2007 John Wiley & Sons, (Asia) Pte Ltd

604

Chemistry of Zeolites and Related Porous Materials

zeolite molecular sieves in that their frameworks contain organic components. It is because of the existence of the organic components that these microporous compounds exhibit vast structural versatility. In principle, numerous frameworks can be designed through varying the organic ligands, and this opens new possibilities for the exploration of microporous materials with various pore architectures. Nevertheless, because of the organic components in the frameworks, the thermal stability of these materials is much inferior to that of the conventional zeolites. The as-synthesized metal-organic porous coordination polymers usually contain guest molecules in their crystal structures, and in fact these materials can also be regarded as host–guest materials. Therefore, in this chapter these materials will be described, together with other zeolite host–guest composite materials.

9.1 Metal Clusters in Zeolites 9.1.1

Definition of Metal Clusters

Generally, a metal cluster can be regarded as a tiny metal particle. However, close inspection of the internal structure and electron configuration of a metal cluster will reveal that metal clusters are distinct from metal particles. Correspondingly, their physicochemical properties are also very different from those of metal particles. A metal cluster may be naked, but, in most cases, they are enclosed by ligands, because otherwise they would not be stable. Because in a zeolite there are framework O and extraframework cations, and these O or extra-framework cations can protect the metal clusters in the channels of zeolites, the metal clusters may be stabilized.[3] The size of metal clusters and the number of atoms in the cluster can vary in a large range. The number of nuclei m can vary from 1 to thousands. In a zeolite, metal clusters can be located in different cages and defects. And the metal clusters in zeolites can be roughly classified as the following types (Figure 9.1): (1) Very small clusters (1 < m < 4) accommodated in the small cages of zeolites (such as the sodalite cage of faujasite) or in the straight channels of zeolites such as mordenite and zeolite L. (2) Those with lower nuclei number (1 < m < 10
40) and smaller sizes (< 1
1:3 nm). These clusters are located in the larger cages such as the supercage of faujasite, or in the cross-shaped intersection of two perpendicular channels. Their size is limited by the cages and the cross opening of channels.

Figure 9.1 The size and location of metal clusters in zeolites. Reproduced from [3]. Copyright (2002) Springer-Verlag

Porous Host–Guest Advanced Materials

605

(3) In some cases, the metal clusters are not confined by the cages of the zeolites, but are connected with one another to form cluster bunches. (4) Those with a size (usually reaching 2–3 nm) apparently exceeding the size of the largest zeolite cage but are still encapsulated by the bulk zeolite. (5) Those present in the defects of zeolite crystals. (6) Metal clusters formed on the external surface of zeolite crystals from metal atoms or small metal clusters escaping from the zeolite pores. These metal particles are close to the corresponding bulk metals in properties. 9.1.2

Preparation Approaches to Metal Clusters

Generally two approaches can be used to prepare metal clusters in zeolites. One involves direct evaporation and deposition of metals into the pores of zeolites, whereas the other is to load the metal-containing precursors into the zeolite pores followed by formation of metal clusters in the zeolite pores through decomposition or reduction.[3] According to the preparation techniques used, these two approaches can be further divided into the following sub-approaches: Metal Salt Impregnation and Reduction A solution of the metal salt is prepared and the solution is thoroughly mixed with the corresponding zeolite. In this way, the metal salt will penetrate into the zeolite pores. After impregnation, the solid zeolite is dried, calcined, and reduced in gaseous hydrogen to form metal clusters in zeolite pores. The metal salts used are usually chlorides, nitrates, or carboxylates. The impregnation method is suitable for the following cases: (a) the zeolite lacks ion-exchange capacity or the capacity is low; (b) the loaded metals (such as Mo) are present as anions that are not able to be ion-exchanged into the zeolite pores; (c) generation of protons needs to be avoided because H2-reduction after ionexchange usually produces protons in the zeolite channels. In some cases, the combination of the impregnation and the ion-exchange methods may reduce the reduction temperature, rendering the formation of metal clusters easier. Adsorption and Decomposition of 0-Valence Metal Compounds The oxidation state of the metals in organometallic and metal-carbonyl compounds can be regarded as 0. These compounds can be introduced into the zeolite pores through vapor deposition or through solution adsorption. After entering into the pores, these precursor compounds can be decomposed through thermal treatment or other physicochemical methods to form metal clusters in the zeolite pores. This technique requires that the precursor molecules be smaller than the pore diameter so that they are able to diffuse inside the pores. In addition, the vapor pressure of the precursors should not be too low because otherwise the evaporation cannot proceed smoothly. During decomposition, formation of intermediates which can diffuse outside the zeolite channels should be avoided, and the decomposition of organic ligands should not lead to easily depositing species to block the zeolite channels. Metal Vapor Preparation Method For metals with a relatively high vapor pressure, it is possible to load the metal clusters into the zeolite pores through direct evaporation. Rabo et al. first reported the preparation

606

Chemistry of Zeolites and Related Porous Materials

Figure 9.2 Apparatus for preparation of metal clusters in zeolites through vapor method. Reproduced from [4]. Copyright (2002) Springer-Verlag

of neutral or ionic sodium clusters through contact of carefully dehydrated zeolites with sodium vapor in vacuum. In fact, besides pure metal itself, some compounds such as NaN3 can also be used as the metal sources. Figure 9.2 shows the apparatus for the preparation of zeolite-encapsulated metal clusters through the vapor preparation method.[4] This apparatus consists of the vacuum pumping system, valves, joints, and tube reactor. For testing use, at the end of the reactor is connected an electron spin resonance (ESR) or a neutron diffraction sample tube which can be sealed conveniently after sample preparation for further characterization. Ion-exchange and Reduction This technique is commonly used for the preparation of metal clusters in zeolite pores. First, the metal ions are exchanged into the zeolite pores, and then these metal ions are reduced through various chemical-physical approaches to form metal clusters. Before ion-reduction, it is necessary to calcine the ion-exchanged zeolite in O2 atmosphere because calcination results in more fixed location of the cations, in favor of reduction and movement of the cations. The reduction of the ion-exchanged zeolites is usually realized through thermal treatment in H2 atmosphere, and, to avoid diffusion of the reduced metal atoms to condense outside the zeolite pores, the reduction temperature should be as low as possible depending on the reducibility of the metal ions. In some cases, in vacuum or in an inert atmosphere, thermal treatment will lead to spontaneous reduction of the exchanged metal ions to form metal atoms. The presence of extra-framework molecules or ions may facilitate this spontaneous reduction. For instance, in the presence of water molecules, at 773 K the Pd2þ ions in the channels of zeolite Y can be reduced to metal cluster particles through thermal treatment. The vacuum decomposition of [Pt(NH3)4]2þ complex in zeolite can also result in formation of metal platinum clusters. Metal clusters can also be prepared through reduction of ion-exchanged zeolites using strong reducing agents. Yoon and Kochi used NaX and NaY to react with n-butyllithium dissolved in n-hexane and they also obtained Na43þ metal clusters. Bimetallic Cluster Preparation Bimetallic clusters in zeolites can be obtained through the following routes: (1) impregnation of the zeolite with a mixture of two metal salts followed by reduction using hydrogen. The disadvantage of this method is that it is not easy to control the growth of the required metal clusters; (2) ion-exchange of two metal ions into a zeolite

Porous Host–Guest Advanced Materials

607

followed by H2 reduction. This approach is suitable for not only metal ions with similar reducibilities but also metal ions with different reducibilities, because the easily reduced metal ions would facilitate the dissociation of H2 so as to make the reduction of the lessreducible metal ions easier; (3) adsorption of the neutral complex of a second metal into the zeolite loaded with the first metal clusters, followed by decomposition of the metal complex. 9.1.3

Alkali Metal Clusters

Preparation of Alkali Metal Clusters It is well known that electrons can be dissolved in liquids. Similarly, zeolites can act as solid solvents to dissolve extra electrons. But in most cases, these extra electrons need to be attached to the metal clusters located in the zeolite channels or cages. Therefore, the corresponding metal clusters are also called ion clusters because of the presence of charges. Initially, the formation of ion clusters in zeolites was realized through physical approaches, for instance g-ray irradiation or high-frequency ultraviolet (UV) irradiation of cations in zeolites. In 1965, Kasai reported that after g-ray irradiation of dehydrated Na-Y zeolite in vacuum, the obtained pink sample showed ESR signals (Figure 9.3).[5] The signal splitting indicates the presence of Na43þ metal clusters in the zeolite. The metal cluster concentration obtained through physical approaches is usually low and the reaction involved is represented in Equation (9.1): 4Naþ þ e
! Na4 3þ

ð9:1Þ

Later on, Rabo and coworkers discovered that the exposure of Na-Y to sodium vapor at 580  C resulted in the same species, but in this case the sample was bright red in color, suggesting that the metal cluster concentration was relatively high. However, a sample obtained through similar treatment of Na-X appeared blue in color, which was confirmed

Figure 9.3 Na43þlocated in the sodalite cage of zeolite Y and its ESR spectrum at room temperature. Reproduced from [4]. Copyright (2002) Springer-Verlag

608

Chemistry of Zeolites and Related Porous Materials

to arise from formation of Na65þ ion clusters. UV-irradiation followed by adsorption of sodium vapor leads to formation of Na43þ ion clusters in sodalite structure. If a zeolite is mixed with fused-ring aromatic compounds followed by light excitation, the fused-ring aromatic would contribute its electrons to the alkali metal ions in the zeolite channels to form ion clusters. The preparation of ion clusters through this approach depends on the Si/Al ratio in zeolites and the nature of the cations and adsorbates. High-concentration ion clusters can form through direct evaporation of metals such as sodium into the pores of zeolites. However, this method requires high-temperature treatment (so that the alkali metal vapor can reach a required value), and the alkali metal to be strongly reducing. As a result, the use of this method would lead to destruction of the zeolite framework to various degrees. Furthermore, the distribution of ion clusters in zeolites obtained through this technique is not uniform. High-concentration metal ion clusters can also be prepared through chemical routes. For example, the interaction of reducing agents, which are formed through dissolution of metallic lithium in primary amine or dissolution of butyllithium into alkane solvents, with zeolites containing alkali metal cations results in formation of high-content metal ion clusters in the latter. Thorough adsorption of sodium azide dissolved in alcohol into zeolite pores followed by thermal decomposition can also produce high-concentration alkali metal ion clusters. In 1984, Edwards and coworkers reported the formation of Na43þ in zeolites Y and A and the formation of K43þ in Kþ-exchanged zeolite Y. Later on, they found that if the zeolite was Na-Y, the formed metal ion clusters are Na43þ no matter what (Na or K) the reaction vapor was; whereas if the zeolite used was K-Y, the obtained metal clusters will be K43þ. That is, the formed metal cluster species is not related with the vapor but simply depends on the type of cation in the zeolite used. In fact, through variation of the reaction condition, various Mþ ðn
1Þ ion clusters can be prepared. If M ¼ Na, n ¼ 2  6, whereas if M ¼ K, n ¼ 3,4. After the alkali metals enter the channels or cages of zeolites to form metal ion clusters, the electrons on the original metal atoms will be released to be shared by more than one metal atom. It has been confirmed that these free electrons actually occupy the holes formed by the metal atoms (ions). Therefore, these electrons are also called solid solvated electrons (in analogy with the solvated electrons formed by alkali metals in solvents such as liquid ammonia),[7] and the formed compounds are called solid electrides. Depending on zeolite structure and composition and preparation method or reaction conditions, the obtained ion clusters are distinct in size, composition, and distribution. In some cases, different ion clusters form in the same zeolite by using the same preparation approach. Correspondingly, the number of electrons on the ion clusters and the electron configurations are versatile. To characterize the electron configuration and the structures of ion clusters, various techniques have to be employed. Some ion clusters are colored. For example, the Na43þ appears to be red-brown, whereas the larger sodium ion clusters are blue-grey. Therefore, these ion clusters can be detected through UV-visible diffuse reflectance spectroscopy. ESR spectroscopy is usually used for detection of unpaired electrons, and it is suitable for systems with lower concentrations of unpaired electrons. The tetrahedral Na43þ ion cluster possesses one unpaired electron, and this unpaired electron would produce an ESR spectrum with 13 fine lines because the electron is affected by the sodium nucleus (nuclear spin quantum number I ¼ 3/2). However, if the system contains both Na54þ and Na65þ ion clusters, the ion clusters would lead to 16 and

Porous Host–Guest Advanced Materials

609

19 fine lines, respectively, in the ESR spectrum, making the experimentally observed ESR spectrum rather complicated. For the detection of metal ion clusters such as Na54þ, Cs42þ, etc. with no unpaired electrons, then ESR spectroscopy is invalid. For spin-paired systems, techniques such as 23Na magic angle spinning nuclear magnetic resonance (MAS NMR) and neutron diffraction can be employed to obtain relevant structural information. If the concentration of extra electrons in the system is high, spin-pairing to form ion clusters such as Na53þ seems to be unavoidable. Structures and Location of Alkali Metal Ion Clusters The exact location of ion clusters in zeolites is interesting. Generally it is possible to locate the ion clusters through X-ray or neutron diffraction techniques in combination with spectroscopic methods. Either in sodalite, zeolite A, or in zeolite Y, the formed Na43þ gives rise to identical ESR spectra, indicating that the Na43þ ion cluster is located in the same microenvironment, that is, in the sodalite cage. The sodalite cage of zeolites A and X can also accommodate K32þ ion clusters. Relevant studies indicate that other ion clusters can also form in the sodalite cage. However, alkali metals with a larger radius such as Rb and Cs would exist in different structure forms when they form ion clusters. For instance, Cs43þ clusters in the sodalite cages of zeolite A may be connected with one another to form a chain structure. It is also proposed that in the sodalite cages the clusters may not be isolated but interact with the ions or ion clusters in neighboring cages to form ion cluster groups. Figure 9.4 shows the interactions among the different metal ion clusters in zeolite cages. Theoretically, if the metal (ion) cluster concentration in a zeolite is high enough, the clusters would interact with one another, rendering the insulating zeolite a metallic conductor. When the content of Na metal in a zeolite is very high, the composite becomes black. Under there circumstances, the ESR fine structure disappears,

Figure 9.4 (a) The K4pþ (bottom) and the K3qþ (top) ion clusters in zeolite X; the clusters in neighboring sodalite cages may interact through the ion in the hexagonal prism in between to form a continuous cluster; or, the connecting atom (deep-colored) can be regarded as a K3rþ cluster; (b) an Rb8pþ cluster located in the center of the sodalite cage of zeolite; (c) a Cs4pþ cluster located in the sodalite cage of zeolite A; if the two terminal atoms or one of them is lost, a smaller cluster would form. Reproduced from [4]. Copyright (2002) SpringerVerlag

610

Chemistry of Zeolites and Related Porous Materials

Figure 9.5 Room-temperature ESR spectra of Na-Y (FAU) containing (a) 3, (b) 8, (c) 13, (d) 32 extra sodium atoms per unit cell. Reproduced from [4]. Copyright (2002) Springer-Verlag

and only one broad signal with g ¼ 2 is shown. This phenomenon arises from the electron exchange because the metal ion clusters in adjacent sodalite cages of the zeolite get very close to one another. Metal Ion Cluster Crystals and Cation Continuum For samples with a sodium concentration which is not too low, the ESR spectrum of Na/ Na-Y shows no fine structure of Na43þ but a single line without fingerprints (Figure 9.5). Initially, Edwards and coworkers thought that this single line arises from the metal nanoparticles in the zeolite.[6] However, subsequent study indicated that this single line might originate from the interaction between adjacent Na43þ ion clusters. Through ab initio molecular dynamic modeling calculations, Ursenbach et al. confirmed this postulation. The calculation suggested that extent of interactions between Na43þ clusters in adjacent sodalite cages of Na-Y is comparable to the reciprocal of observed 23Na ultrafine coupling constant of Na43þ. In this case, if each ion cluster has at least one adjacent ion cluster, the disappearance of the ultrafine structure of the ESR spectrum becomes obvious. On the basis of detailed investigation into the structures of metal-ioncontaining zeolites, Amstrong and coworkers pointed out that there exist Na43þ interaction networks in zeolites. The distance between metal ion clusters and the orientation of the clusters in zeolite cages depends on the zeolite hosts, and as a result the three-dimensional arrays can be described in different ways. They can be called cluster crystals or superlattices. The Na/ Na-Y cluster crystal formed by accommodation of one Na43þ cluster in each sodalite cage is written as in Equation (9.2): 56
8Na0 þ ðNaþ Þ56 ½Al56 Si136 O384 56
! ðNaþ Þ32 ðNa3þ 4 Þ8 ½Al56 Si136 O384 Y Y

ð9:2Þ

where [Al56Si136O384]Y stands for the zeolite Y framework. Because the Na43þ cluster can be present in zeolites with different structures,[5,6,8,9] it is possible to prepare cluster crystals with various geometric arrangements. In these cluster crystals, the centered cubic array formed by the encapsulation of sodalite framework is the simplest, 56


Porous Host–Guest Advanced Materials

611

Figure 9.6 Cluster crystals formed by Na43þ clusters located in the sodalite cages of zeolites with different structures. (a) Sodalite (SOD). (b) Zeolite Y (FAU). (c) Zeolite A (LTA). Although the array shown in (c) has not been prepared, the analogous potassium clusters shown in (d) is indeed available; in (d) the cluster crystal is actually composed of K43þ clusters in all the sodalite cages and the larger K124þ clusters in every other a-cages. Reproduced from [4]. Copyright (2002) Springer-Verlag

and it exists in the host–guest compound designated ‘black sodalite’ as shown in Equation (9.3):[10,11] 6
2Na0 þ ðNaþ Þ6 ½Al6 Si6 O24 sod 6
! ðNa3þ 4 Þ2 ½Al6 Si6 O24 sod

ð9:3Þ

Of course, more complex systems also exist (see Figure 9.6). Although there is no evidence for the existence of mixed metal clusters as yet, ESR signals show that there might be interactions between sodium and cesium clusters in zeolites. Properties and Applications of Alkali Metal Clusters The alkali metal clusters in zeolites attracted attention initially because of their intense color changes. It was proposed that these materials could find applications in cathode ray displays or in reading-and-writing device manufacture. Na43þ cluster can reversibly absorbs O2 and form O2
ion in Na-Y.[12] Other molecules such as SO2, N2, Ar, Kr, CO, CO2, NH3, benzene, n-hexane, and halogenoalkanes also interact with these metal clusters,[13,14] and, as a result, the metal clusters may catalyse the chemical reactions involving these molecules. In principle, the metal clusters in the zeolite channels and cages may act as basic catalysts. Martens et al. investigated the catalytic performance of sodium clusters in zeolites A, X, and Y for the isomerization reaction of (Z)-2-butene and for the hydrogenation reaction of (Z)-2-butene, acetylene, and benzene.[15] Later on, Hannus et al. also reported the catalytic performance of sodium clusters in a series of Y zeolites for 1-butene isomerization,[16] and they proposed that in this catalytic reaction process the butane isomerization intermediate is a carbanion because of the basicity of the metal clusters. Simon et al. reported that in Na-X the Na65þ clusters could facilitate the isomerization of cyclopropane to propene.[17] The metal ion clusters formed by pyrolysis of NaN3 in the cages of zeolite can catalyse branch-alkylation of toluene and the aldehyde condensation of ketones.

612

Chemistry of Zeolites and Related Porous Materials

Figure 9.7 (a) Quantum-wire arrays based on the one-dimensional channels of zeolite L; (b) randomly oriented K3pþ metal clusters which may induce charge transport along the K/K-L channels. Reproduced from [4]. Copyright (2002) Springer-Verlag

Because cation zeolites such as Na-X can adsorb a large amount of alkali metals to form metal clusters, a large number of extra electrons in unit volume can be produced in the compounds. When the concentration of these extra electrons reaches a certain value, it is possible for the metal-cluster-containing host–guest material to undergo transition from an insulator to a conductor. In particular, when there are crystallographically onedimensional channels, the formed one-dimensional metal clusters may become quantum wires (Figure 9.7).[18–21] In 1997, Anderson et al. reported that when more and more extra electrons are introduced into the zeolite hosts,[19] some zeolites indeed show an electrical conductivity increase. The loading of potassium into zeolite L increases the room-temperature conductivity of the latter by 10 000 times. Nevertheless, the measured temperature dependence of conductivity on temperature indicates that the conduction mechanism is characteristic of thermal activation, and therefore that the electrical conduction may involve the redox jump of the K32þ/K3þ process (see Figure 9.7). The K/K-A host–guest compound exhibits interesting ferromagnetic behavior,[22] and this magnetism may be related with the formation of a superlattice.[23] 9.1.4

Metal Clusters of Silver

In 1962, Ralek et al. reported that after dehydration–rehydration treatment, the silverexchanged zeolite A changed from white (hydrated) to yellow to orange and finally to brick-red in color.[24] In 1977, Kim and Seff observed a similar phenomenon for Ag-A single crystal when it was treated carefully at 400  C, and they revealed that after dehydration there exist Ag60 metal clusters in the Ag-A zeolite on the basis of XRD analysis (Figure 9.8).[25] Later, Jacobs et al. revealed that Ag32þ clusters may be present in the zeolite.[26]

Porous Host–Guest Advanced Materials

613

Figure 9.8 The Ag6 cluster in zeolite A, (a) located in the sodalite cage and it is surrounded by 8 Agþ ions; (b) located in the a-cage and surrounded by 14 Rbþ or Csþ ions. In (a) the positioning of the atoms may also be considered as linear Ag32þ clusters (black atoms). Reproduced with permission from [4]. Copyright (2002) Springer-Verlag

The dehydration–reduction of Agþ ions in the zeolite pores to form silver clusters may involve two steps. The first occurs at a temperature lower than 250  C, and, in this process, the silver ions oxidize the water molecules and are themselves reduced; the second step in the process is a nonaqueous reaction, involving the oxidation of framework oxygen ions, and the reaction Equation (9.4) is as follows: 2Agþ þ O2
½zeolite ! 2Ag0 þ 1=2O2 þ ½zeolite

ð9:4Þ

0

In the absence of coordination water molecules, isolated Ag atoms do not actually exist because isolated silver atoms are paramagnetic, and theoretically they may be detected through ESR spectroscopy. Nevertheless, the compounds show no ESR signals. Therefore, it is concluded that the Ag0 atoms formed through reduction interact with one another to form metal clusters.[27] In zeolites, silver clusters can also be obtained through reduction of the silver ions in the zeolites using alkali metals. Through H2 reduction, silver atom clusters or ion clusters with various sizes can be prepared using Ag-A or AgNa-A zeolites. In some cases, these clusters show ESR signals, indicating that there exist unpaired electrons. By means of physical methods such as g-ray irradiation, paramagnetic silver metal clusters may also be acquired at low temperatures.[28,29] Because in zeolites the silver clusters change their color when in contact with water, it has been suggested that these compounds may be used as water-vapor-sensing materials. Ozin et al. observed the vapor-pressure chromic, cathode-ray chromic, water chromic, photochromic, and thermochromic properties for silver-sodalite, and they proposed that these compounds hold promise for use as sensors.[30] In addition, the silver clusters in zeolites are also sensitive to other molecules, and therefore, they have great potential as chemical-sensing materials. 9.1.5

Noble Metal (Platinum, Palladium, Rhodium, Ruthenium, Iridium, Osmium) Clusters

Zeolite molecular sieves play an important role in catalysis reactions. The introduction of noble metals into zeolite pores to form bifunctional catalysts is of particular interest. To introduce noble metals into zeolites, first noble metal ions or cation complexes are

614

Chemistry of Zeolites and Related Porous Materials

exchanged into zeolites pores, and then the metal ions are reduced using reducing agents (mainly hydrogen) at a certain temperature to form noble metal clusters. To make sure that the metal clusters are homogeneously distributed in zeolite pores, in some cases the ion-exchanged zeolite must be thermally treated in O2 atmosphere prior to reduction. Differing from alkali metal ion clusters, the zero-valence noble metal clusters prepared through this technique vary over a considerable size range, and the number of atoms contained in the cluster varies as well. For example, small-angle X-ray scattering (SAXS) and transmission electron microscopic (TEM) analysis indicates that the Pt-Y zeolite prepared through hydrogen reduction at 300  C contains Pt metal clusters with a diameter of 0.6–1.3 nm.[31,32] Further TEM structure analysis reveals that almost all the Pt clusters are smaller than the size (1.2 nm) of the supercage of zeolite Y, and the particles are homogeneously distributed in the zeolite. It is reasonable to believe that these Pt clusters are located in the supercages of the Y zeolite.[33] The number of atoms in each cluster is probably 40, and the cluster is accommodated in the faujasite supercage in the form of a truncated tetrahedron. Of course, under different preparation conditions, other Pt clusters may also form in the sodalite cages of the Y zeolite or on the external surface of the crystal. Pd2þ ion and its reduced atom have a mobility in zeolite higher than the corresponding 2þ Pt and Pt. As a result, it is difficult for all the prepared Pd metal clusters to be located in the supercages of the host–guest Pd-Y zeolite. In general, smaller Pd metal clusters with a diameter of 0.6 nm can be formed in the supercages of zeolite Y under mild conditions. Other noble metal (Ru, Rh, Os, Ir) clusters in zeolite Y prepared through ion-exchange and hydrogen-reduction method are usually smaller than the Pt cluster under similar preparation conditions. Some noble metal (Ru, Ir) clusters can also be obtained through loading of metal carbonyl compounds in zeolite pores followed by pyrolysis. The carbonyl metal cluster anions of Ru such as [Ru6C(CO)16]2
and [H2Ru10(CO)25]2
can be loaded in mesoporous molecular sieves through direct absorption from organic solvents. For example, when the cluster ions in salt form are dissolved in a mixed solvent of diethyl ether and dichloromethane followed by mixing with dry detemplated MCM-41 in an appropriate ratio, the cluster metal anions can be sorbed into the mesoporous channels. Because the hydroxy groups in the mesopores are able to interact with the carbonyl group of the cluster anions through H-bonding, the two species are attracted to each other and the cluster anions are bound in the mesopores tightly. High-resolution electron microscopy clearly shows that the metal cluster anions are arrayed in the MCM-41 channels.[34] Under mild conditions, pyrolysis of the formed host–guest materials would lead to the decomposition of the cluster anions in the mesoporous channels to form highly dispersed metal clusters. The orderliness of the metal clusters is less than that of the cluster anions assembled into the channels, but they exhibit excellent catalytic performance for hydrogenation.[35] 9.1.6

Other Metal Clusters

The mercury-exchange zeolite X can adsorb a large quantity of mercury vapor, and this process can be described as in Equation (9.5) and (9.6): Hg0 þ Hg½zeolite2þ ! Hg2 ½zeolite2þ 0

xHg þ Hg2 ½zeolite

2þ

! Hgðxþ2Þ ½zeolite

ð9:5Þ 2þ

ð9:6Þ

Porous Host–Guest Advanced Materials

615

When Cd-A reacts with cadmium vapor, Cdn2þ (n ¼ 2
4) and Cd54þ clusters form in the zeolite, as revealed by structural and spectroscopic analysis.[36–38] Cadmium and zinc react with H-Y to form Cd22þ and Zn22þ cationic clusters.[39,40] It is of interest that when zinc vapor reacts with the small-pore H-SAPO-CHA (chabazite structure) molecular sieve, mononuclear univalent Znþ ions in the SAPO-CHA pores can be prepared.[41] Because there are unpaired electrons on the Znþ, the as-prepared compound shows ESR signals, and the magnetic moments of the unpaired electrons interact antiferromagnetically at low temperatures. The reason that no Znmþ n clusters are formed in this case may lie in that the generation of the large quantity of gaseous hydrogen prevents the cluster formation. Reports on other zeolite-metal cluster host–guest compounds are very rare. On the basis of single-crystal X-ray diffraction (XRD) analysis, Heo et al. pointed out that it is possible to grow In58þ and In32þ ion clusters in zeolite A.[42,43] After grinding of a mixture of hydrated cobalt nitrate [Co(NO3)26H2O] and Na-X, the cobalt nitrate can easily enter the pores of the X zeolite. Upon heating of the cobalt nitrate-loaded Na-X zeolite at 60  C, the cobalt nitrate decomposes to form Co3O4. If this host–guest material is further heated in H2 atmosphere at 550  C, the cobalt oxide located in the zeolite would be reduced to cobalt atoms, which undergo coarsening to form cobalt metal clusters.[44] However, X-ray diffraction and electron microscopy analysis indicate that the particle size of the cobalt clusters reaches 15–25 nm, much larger than the supercage size of the X zeolite. Further inspection reveals that the cobalt metal clusters are not dispersed on the external surface of the zeolite crystals but instead are homogeneously distributed in the bulk of the crystals. The only explanation of this phenomenon is that there exist a large number of defects in the zeolite crystals, and the cobalt metal clusters are accommodated in these defects.

9.1.7

Clusters of Metal Oxides or Oxyhydroxide

Upon heating, zeolites with di- or trivalent cations may generate ½Ml Om ðOHÞn pþ oxyhydroxide clusters through hydrolysis. Cations with higher oxidation states tend to form similar clusters more easily. Lanthanide-exchanged zeolite Y can be used as highperformance petroleum refinery catalysts after calcination. The thermal stability of these catalysts is very high, and this high thermal stability may originate from the charged Obridged lanthanide metal clusters formed in the sodalite cages of the zeolite.[45,46] On the basis of XRD analysis, Park and Seff pointed out that in the sodalite and supercages of completely dehydrated, partially dehydrated, and anhydrous La-X zeolites there exist various metal–oxygen cluster ions.[47] A typical example is the La4O44þ cluster located in the sodalite cage (Figure 9.9). Similar Pb4O4 cluster can also be formed in zeolite A. In fact, this Pb4O4 cluster can further interact with Pb2þ or Pb4þ ions to form larger clusters such as Pb8O4pþ. It has been evidenced that in sodalite there exists Pb2(OH)(H2O)33þ.[48] When zeolite Y with adsorbed Fe(CO)5 is thermally treated under O2 atmosphere, Fe6On clusters form, and, in these clusters, the Fe occupies the vertex of an octahedron.[49] The interaction of dehydrated Zn-A with Zn vapor results in body-centered cubic Zn9 On pþ clusters. In fact, in these clusters there may exist Zn–Zn bonds because of O deficiency.[50]

616

Chemistry of Zeolites and Related Porous Materials

Figure 9.9 Schematic representation of the structures of (a) La4O44þ, (b) Zn9 Opþ n and (c) MxX clusters in the SOD cage of zeolites X, A, and sodalite. In (b) the anionic positions (grey) are partially occupied. Reproduced from [4]. Copyright (2002) Springer-Verlag

9.2 Dyes in Zeolites Dye molecules tend to aggregate. In solution, dye molecules aggregate even at very low concentrations. After aggregation, the energy of excited dye molecules can be released easily through thermal relaxation, and therefore their optically active properties cannot be realized. If dye molecules are dispersed in the zeolite pores, the aggregation of molecules can be avoided, and as a result the dye molecules may exhibit excellent optical activities, such as lasing capacity etc.[51] There are four approaches to load dyes into porous molecular sieves: direct ionic exchange for cationic dyes,[52] vapor-phase deposition,[53] crystallization encapsulation,[54] and precursor in-situ synthetic method.[55] Hoppe et al. successfully loaded methylene blue into zeolite Na-Y through ionic exchange and crystallization encapsulation methods.[56] X-ray structure analysis indicates that in the host–guest material prepared through ionic exchange, the dye molecule is located at the center of the supercage of the zeolite, whereas in the compound prepared via the crystallization encapsulation technique the dye molecule is close to the opening window of the supercage. In addition, in the loading process, some of the methylene blue molecules undergo demethylation. Dye molecules loaded in Na-Y show intense fluorescence, but the luminescence efficiency differs for the dye molecules located in different positions. If AlPO4-5 molecular sieve with 1-D channels is directly mixed with solutions of dye molecules, the dye molecules will enter the channels of the molecular sieve with the solvent molecules. After separation and drying of the obtained solid material, the solvent molecules will vaporize and the dye molecules will be attached in the channels of the microporous crystals. Rurack et al. investigated the loading of 2,20 -bipyridine-3,30 - diol dye molecules into AlPO4-5 molecular sieve and the spectral properties of the formed host–guest composite.[57] It is found that the host–guest material exhibits intense fluorescence, and the crystal displays optical anisotropy, which indicates that the guest dye molecules loaded in the host channels are arranged linearly. Nevertheless, there is a tilting angle (about 22 ) between the axis of the guest dye molecule and the running direction of the crystal channel. Further spectroscopic analysis reveals that in the AlPO4-5 channels there exist a small number of water molecules and a tiny number of proton acid sites which affect the guest dye molecules in the channels to a certain degree

Porous Host–Guest Advanced Materials

617

but have no influence on the fluorescence properties of the guest molecules. Proflavin is a commonly used luminescent dye molecule. It is somewhat basic, and as a result this molecule is able to interact with proton to form a salt. Protonated proflavin can be exchanged into the pores of zeolites Y and L.[58] The fluorescence intensity of the proflavin dye molecules loaded in a zeolite increases with the loading amount, but decreases when the concentration reaches a certain value because when the dye molecules get too close to one another self-quenching occurs. In zeolite L the proflavin molecules are present in both monoprotonated and diprotonated forms, and the luminescence intensity is high, whereas in zeolite Y the proflavin molecules are mainly monoprotonated. Through adding a small amount of dye molecule 1-ethyl-4-{4-[p-(dimethylamino)phenyl]-1,3-butadienyl}-pyridinium perchlorate (hereinafter designated ‘pyridine 2’) into the synthetic system for AlPO4-5, Vietze et al. obtained AlPO4-5 crystals with encapsulated dye molecules.[59] The diameter of the dye molecule is about 0.6 nm, and this molecule can just be accommodated in the one-dimensional channel (diameter 0.74 nm) of the AlPO4-5 crystal. It is believed that the chain direction of the dye molecules is parallel to the axis of the zeolite molecular sieve (Figure 9.10); that is, the dye molecules are arranged along the microporous channels of the molecular sieve. When the crystallization time for the microporous crystals is short, the obtained crystals appear as hexagonal prisms, whereas prolongation of the reaction time leads to dumbbellshaped crystals. The inner part of the dumbbell crystals is still a hexagonal prism, but the two ends of the prism are attached with many thin prismatic needles. The prism of the inner part of the crystal is regular, with a diameter of about 8 mm; and because there are dye molecules inside them, the crystals may function as laser oscillating chambers. Optical microscopic observation reveals that the color of the dye is located in the prismatic part of the crystal, whereas the needles at the two ends of the crystal are colorless, indicating that the needles contain very small amounts of dye. For the whole synthesized crystal, the percentage of the dye molecules is between 0.1 and 0.01%. The zeolite–dye host–guest composite obtained through this method exhibits apparent optical anisotropy, and the crystal displays thermoelectrical properties, suggesting that the dye molecules in the AlPO4-5 channels are oriented not only along the channel direction but also in a head-to-tail manner. When pumped with an Nd:YAG laser (wavelength 532 nm), the dye-containing AlPO4-5 exhibits intense fluorescence. Spectral analysis indicates that the fluorescence emission peak is narrow, and in some cases several lines appear whereas

Image Not Available In The Electronic Edition

618

Chemistry of Zeolites and Related Porous Materials

some crystals give only one single line. The fluorescence varies with the pumping laser intensity and through observation of this variation it is known that the fluorescence is a laser process. Further analysis indicates that the fluorescent activity is shown only within a 34 mm region in the inner part of the dumbbell crystal, and this phenomenon is in agreement with the high content of dye molecules in this part. The lasing behavior of the dye/ AlPO4-5 composite suggests that this host–guest composite can be used as solid laser.[60] This approach for preparation of a microlaser through loading of dyes into regular channels is also valid for mesoporous materials. Yang, Wirnsberger, and colleagues used a similar method[61,62] and successfully loaded dye molecules into mesoporous silica to form microlaser materials. With block copolymer as a template, Vogel et al. successfully loaded rhodamine 6G dye molecules into mesoporous titania.[63] Owing to a dispersion effect of the block copolymer, the loaded 6G molecules in the mesoporous material overcome aggregation, and as a result they exhibit superior laser emission properties. These mesoporous host–guest materials can form films, and through lithography they can be patterned. Therefore, these materials may find applications in fabrication of

Figure 9.11 Framework structure of zeolite L. (a) Projection along the c axis; (b) projection perpendicular to the main channel direction, with A–E representing the positions of the cations; (c) 12-membered-ring main channel; (d) scanning electron microscopoe image of zeolite L crystals; (e) structures of pyronine (left) and oxonine (right). Reproduced with permisson from [66]. Copyright (2003) Wiley-VCH

Porous Host–Guest Advanced Materials

619

microlasers and other optically active devices. Besides dye molecules, other luminescent materials can also be loaded into zeolite channels to form laser materials. For instance, the loading of Nd3þ complexes in nano-sized zeolite cages leads to a host–guest composite material that emits intensely in the near-infrared (near-IR) region.[64] The loading of rhodamine dye molecules into microporous and mesoporous molecular sieves results in sensing materials. For example, after grafting onto the channel walls of the mesoporous MCM-41, rhodamine B sulfonate (rh B-sulfo) shows a fluorescence spectrum very sensitive to SO2 molecules. In the presence of SO2 the fluorescence is quenched, whereas when the SO2 is removed the fluorescence is recovered.[65] Not only can one type of dye molecule be loaded into the channels of molecular sieves as a guest, but also two or more different dye molecules can be loaded into one zeolite or relevant microporous and mesoporous material through special design. Calzaferri and coworkers assembled various dye molecules in zeolite L, taking advantage of the wide channels of this zeolite (Figure 9.11), and they found that these multi-component dyes exhibit efficient energy-transfer properties as guests in the channels. They also called these host–guest systems zeolite antennae.[66,67] As shown in Figure 9.12, the pyronine molecules are loaded in the central part of zeolite L microcrystal whereas the oxinine molecules are loaded at the ends of the zeolite channels.[68] In this case, because the absorption and emission wavelengths of pyronine are shorter than those of oxonine, after excitation of pyronine upon light absorption, the energy will be transferred among different pyronine molecules along the zeolite channel and finally the energy will be transferred to the oxinine molecules at the ends of the zeolite crystal where the energy will be released through spectral emission. Calzefferi et al. also suggested that if the double-dye-molecule-loaded zeolite crystal is attached onto the surface of a semiconductor, it is possible that the energy will be transferred from the

Figure 9.12 Schematic representation for loading of donor and acceptor dye molecules in zeolite L crystal structure. The dark rectangular squares at the two ends of the channel indicate the presence of acceptor dye, whereas the blank rectangular squares in the middle indicate the presence of donor dye. This diagram also shows the dye molecules and their electron transition moment orientation after the central part of the zeolite crystal is enlarged. Reproduced with permission from [68]. Copyright (2000) NOVA Science Publishers

620

Chemistry of Zeolites and Related Porous Materials

dye to the semiconductor, achieving the goal of exciting the semiconductor with shortwavelength light. In some cases, when the size of dye molecules is similar or larger than that of the pores of a microporous crystal, it is difficult to realize the host–guest assembly through direct loading. Nevertheless, an alternative approach is to load the dye molecules into a larger-pore zeolite, followed by alteration of the host zeolite structure through chemical-physical techniques, and in this way larger dye molecules can be loaded into smaller-pore zeolites. For instance, under mild conditions methyl blue and perylene can be loaded into VPI-5, which is converted into AlPO4-8 after thermal treatment at 120  C. This thermal treatment of the host–guest material will lead to an AlPO4-8 with the guest dye molecules tightly confined in the framework of the newly formed host.[69] Microcrystalline AlPO4-5 possesses one-dimensional circular channels that are suitable for loading with chain organic molecules. When optically active organic molecules are arranged in an orderly manner in the AlPO4-5 channels, they may exhibit special properties and functions. p-Nitroaniline (P-NA) has a relatively large dipole moment, and it shows nonlinear optical properties. However, the crystal formed by p-NA molecules possesses a symmetry center, and therefore the crystal exhibits no nonlinear optical property. To take advantage of the special feature of the p-NA molecule, it is necessary to array these molecules in three-dimensional space in such a way that no symmetry center arises. Zeolite Y has a symmetry center, and therefore loading of p-NA in zeolite Y cannot lead to nonlinear optical materials. However, AlPO4-5 crystal structure has no symmetry center, and if p-NA is loaded in AlPO4-5 it is possible to form a host–guest material that shows nonlinear optical properties; this has been confirmed. Through a vapor-transport deposition technique, p-NA can be loaded in the one-dimensional channels of AlPO4-5.[70,71] Raman spectroscopy indicates that the status of the p-NA molecule in the channel is affected by the features of the host AlPO4-5 crystal. The guest molecules can be arranged in the channels in two different ways. One is that the p-NA molecules are positioned head-to-tail to form a chain structure in the channel, whereas in the other way there are weak interactions among the p-NA molecules and the molecules are similar to those in a molten state of p-nitroaniline. However, in the second case, the p-NA molecules in the AlPO4-5 channels are still arrayed in a rather orderly manner. If the loading amount of p-NA in AlPO4-5 is small, there are no H-bonds between the p-NA molecules and the obtained host–guest composite shows no secondorder nonlinear optical properties (second-harmonic generation, abbreviated as SHG). When the loading amount reaches a certain value, the host–guest composite exhibits apparent second-harmonic generation, and IR spectroscopy indicates that in this case there are strong H-bonding interactions between the p-NA molecules in the channels. pNA/AlPO4-5 host–guest composite can exhibit not only nonlinear optical properties, but also thermoelectricity.[72] The loading of dye molecules in zeolites can also result in spectral hole-burning material. Hole-burning is defined as the phenomenon of absorption intensity decrease at the irradiation laser frequency when a material is irradiated with a laser light (Figure 9.13). Spectral hole-burning can be observed at low temperatures when thionine and methyl blue are loaded into zeolite X or Y through ion-exchange.[73] The holeburning wavenumber of the thionine/Y host–guest composite is 605:05 cm
1 , and the highest hole-burning temperature is 13 K. A higher hole-burning temperature (about

Porous Host–Guest Advanced Materials

621

Figure 9.13 Principle of spectral hole-burning. On the left side is the absorption spectrum before laser irradiation; whereas on the right side is the absorption spectrum after laser irradiation

80 K) has been observed for the composite formed by loading of phthalocyanine dye in zeolites.[74] It is worthwhile to point out that only in zeolites with water molecules can dye molecules exhibit apparent spectral hole-burning properties. This may be because the interactions between water molecules and the dye molecules in the pores of zeolites render the chromophoric groups of the dye molecules solvated, thus facilitating the electron–phonon interactions.

9.3 Polymers and Carbon Materials in Zeolites 9.3.1

Polymers in Zeolites

The organic monomers adsorbed into zeolite pores easily undergo polymerization to form polymers under proper conditions.[75,76] Attention has been attracted to polymeric materials with conducting properties in microporous and mesoporous molecular sieves. These polymers may exist as single chains due to confinement by the zeolite pores, and therefore they are very important for investigation into physico-chemical properties of polymers and for miniaturization of electronic devices. When acetylene molecules are adsorbed into zeolites, they may polymerize under appropriate conditions to form polymeric fragments with conjugated double bonds.[77] In zeolites, the polymerization of monomers sometimes requires the presence of oxidizing agents. For instance, Cu2þ and Fe3þ, etc., exchanged into the pores of zeolites Y and mordenite can act as an oxidant to facilitate the polymerization of successively adsorbed pyrrole or thiophene, forming polypyrrole or polythiophene.[78–80] Further oxidation of polypyrrole will render the polymer chains electrically conductive (Figure 9.14). Another common oxidant for polymerization is the water-soluble persulfat ion. Usually, oxidation polymerization occurs when a zeolite with adsorbed monomers is mixed with persulfate. Through this approach, polyaniline can be prepared in zeolites mordenite and Y.[81] The electrical conduction properties of polyaniline are closely related with the extent to which the molecule is oxidized and protonated, and, as a result, the structure and composition of the zeolite host may affect the properties considerably. Polyacrylonitrile is a very important polymer material. Using zeolites as a host template, acrylonitrile monomers can polymerize to form polyacrylonitrile in the zeolite

622

Chemistry of Zeolites and Related Porous Materials

Figure 9.14 Schematic representation of the formation of polymer from pyrrole in zeolite mordenite channels. Reproduced from [80]. Copyright (2001) Elsevier

channels.[82] First, the zeolite is dehydrated through evacuation, followed by contact of the dehydrated zeolite with the vapor of liquid acrylonitrile monomer, and then the monomer molecules will be adsorbed into the zeolite channels through diffusion. Polyacrylonitrile encapsulated in the zeolite framework can be obtained by appropriately heating the mixture of the acrylonitrile monomer-containing zeolite, persulfate, and sulfite solutions. After formation of the host–guest material, the zeolite framework can be removed through dissolution in HF solution, and the remaining polymer is characteristic of the normal polyacrylonitrile on the basis of analysis results (Figure 9.15). It has been found that the polymerization of acrylonitrile differs in zeolites with different structures. In zeolite Y, the relative molecular mass of the polymer can reach 19 000, whereas in the

Figure 9.15 Schematic representation of polymerization and pyrolysis of acrylonitrile in zeolite. 1. Monomer entering into zeolite channel; 2. monomer polymerization; 3. pyrolysis of polymer. Reproduced with permission from [82]. Copyright (1992) American Chemical Society

Porous Host–Guest Advanced Materials

623

zeolite mordenite the formed polymer has a relative molecular mass of only about 1000. Polymers cannot be formed in silicalite with an even smaller pore size because the pores of this zeolite is too small to allow for polymerization to occur. The polyacrylonitrile in zeolites can be carbonized through pyrolysis, and the carbonized matter after dissolution of the zeolite framework exhibits semiconducting properties with a conductivity of about 10
5 S=cm. Bein and coworkers investigated the formation of poly(methyl methacrylate) (PMMA) through polymerization of methyl methacrylate (MMA) in channels of microporous crystals including Na-Y, mordenite, zeolite b, and ZSM-5, and mesoporous molecular sieves such as MCM-41 and MCM-48.[83] MMA in zeolites can also polymerize as acrylonitrile does, and with an increase in the host channel size the polymerization degree is enhanced. Electron microscopic observation indicates that the polymerization reaction proceeds mainly inside the zeolite channels, because almost no polymers are observable on the external surface of the zeolite particles. This is further confirmed by the fact that the polymer/zeolite composite lacks the characteristic glass transition temperature for a bulk polymer. When the suspension of dehydrated H-Y zeolite in dichloromethane is mixed with divinyl ether, the latter will enter the channels of the zeolite rapidly and undergo polymerization to various degrees.[84] Protons are present in H-Y, and these protons may destroy the ether bonds of the divinyl ether, resulting in formation of polyethylene cations on the polymer chains. The presence of the polyethylene cations causes the host–guest composites to show various colors. The Na/H ratio in H-Y and the substituent groups on the divinyl ether influence the behavior of the polymerization reaction and the length of the polyethylene cations to a great extent. 9.3.2

Preparation of Porous Carbon using Zeolites

Porous carbon can be prepared through using zeolites as the guest materials. First, organic species such as acrylonitrile,[82] polyacrylonitrile,[85] poly(furfuryl alcohol),[86] or phenolic resin [87] are loaded into zeolite channels, and then the loaded organic species are carbonized through pyrolysis; the carbonized composites are mixed with an acid solution to dissolve the inorganic substrate, and carbon molecular sieves are formed consequently. The zeolites used include zeolites Y, mordenite, b, and L. According to the zeolites used, the organic species and the preparation conditions vary, and various porous carbon materials can be obtained. Except for a few examples, the pore structures of the porous carbons prepared through this approach are not uniform, and, therefore, these porous carbons cannot be used as molecular sieves as zeolites can. Through similar techniques, porous carbons can also be achieved using non-1D pore mesoporous molecular sieves, but in this case the obtained porous carbons have a pore size falling into the range 5  10 nm. Furthermore, unlike the porous carbons obtained from microporous zeolites, the porous carbon materials obtained from mesoporous molecular sieves possess uniform pore structures and sizes.[88–90] Actually, this type of porous carbon can be classified as molecular sieves because of its pore uniformity, and it may exhibit chemical-physical properties different from those of other porous materials. For instance, platinum nanoparticles can be homogeneously dispersed onto the internal surface of these uniform-pore carbon molecular sieves.[91]

624

Chemistry of Zeolites and Related Porous Materials

If one-dimensional channel mesoporous molecular sieves are used as the host for preparation of carbon materials, the obtained carbon is usually of linear structure because the carbonaceous material grows only in the one-dimensional channels, and interconnection to form a three-dimensional network is not possible. Wu and Bein reported evaporation of acrylonitrile monomers into mesoporous molecular sieves at room temperature followed by polymerization of the monomers in the presence of a catalyst.[92] The host–guest composite after polymerization underwent pyrolysis at various temperatures, forming one-dimensional graphitic material distributed in the mesoporous channels. These graphitic species can be regarded as one-dimensional conducting wires, and the experimental results indicate that the 1-D carbon/MCM-41 possesses electrical conductivity. Nevertheless, because of the insulating feature of the mesoporous SiO2, the electrical conductivity measured is far less than that of the product obtained through pyrolysis of bulk polyacrylonitrile at the same temperature. The polymer/zeolite host–guest precursor for preparation of porous carbon can also be obtained through direct contact of monomers in a carrier gas with zeolite molecular sieves followed by polymerization. For example, propylene can enter into zeolite Y under the carriage of N2 and polymerize to form polypropylene. After pyrolysis, the polypropylene undergoes carbonization, and the host zeolite framework of the carbonization product can be removed by dissolution in acids, leaving carbon material with characteristic pores.[93] However, the pore-size distribution of the porous carbons obtained through this approach is not uniform, and hence they are hardly used as molecular sieves for sieving small molecules. 9.3.3

Fullerenes Assembled in Zeolites

The van der Waals’ diameter of C60 is about 1 nm, larger than the pore opening of most microporous crystals. Therefore, it is rather difficult to load C60 into zeolite molecular sieves. However, VPI-5 is a 1-D channel microporous crystal with a relatively larger (about 1.25 nm) pore diameter. Its pore size is large enough to accommodate the C60 molecule. Hamilton et al. successfully loaded C60 molecules into VPI-5 channels through interactions of C60 benzene solution with VPI-5 molecular sieve under 50 atm at 50 C.[94] Because of the confinement of the molecular sieve framework, the C60 in VPI-5 exhibits properties very different from those of the parent C60. First of all, the C60 can prevent VPI-5 from conversion into AlPO4-8 under thermal treatment. More interestingly, upon laser irradiation, the C60 in VPI-5 emits relatively intense white light, whereas the parent C60 luminesces very weakly and the emission-wavelength range is rather narrow. C60 can also be loaded into the channels of mesoporous materials through solution impregnation. Drljaca et al. mixed a toluene solution of C60 with mesoporous SiO2 and obtained a host–guest composite containing solvated C60 molecules in the channels.[95] The color of this solid material is similar to that of the C60 toluene solution (pale purple). However, when the toluene solvent in the mesoporous channels is evaporated through thermal treatment, the sample becomes yellow. The appearance of this yellow color and the corresponding UV-visible diffuse reflectance spectrum indicate that the C60 molecules in the channels undergo aggregation. If the yellow-colored host–guest material is contacted with toluene again, the original pale purple color reappears, suggesting that the aggregation and dissolution of the C60

Porous Host–Guest Advanced Materials

625

molecules located in the mespoporous channels is a reversible process. If hydrophobic cyclodextrin molecules with strong encapsulating ability are added to the mesoporous SiO2 solid containing toluene/C60, these cyclodextrin molecules easily capture the C60 molecules in the mesoporous channels to form C60 aggregates in the space encapsulated by the cyclodetrin. In this case, the host–guest solid material becomes brown-yellow. C60 can also be directly loaded into the channels of mesoporous molecular sieves through thermal diffusion. Chen et al. mixed C60 and MCM-41 material with various features under vacuum and they observed the dehydroxylation behavior of the latter.[96] They discovered that the C60 apparently facilitates the dehydroxylation process. MCM-41 possesses a large number of hydroxy groups on the inner walls after detemplating, and these hydroxy groups exist as either H-bonded or isolated forms.[97] It has been revealed that when the loading amount of C60 is small, the isolated hydroxy groups are lost more readily than are the H-bonded ones, and the dehydroxylation behavior in the presence of C60 differs for MCM-41 materials with different hydroxy-group numbers or features. IR spectroscopic analysis indicates that, during dehydroxylation, the C60 captures the hydroxy groups of the MCM41 silanols to form C

H and C

OH bonds followed by condensation at higher temperatures to be removed as H2O molecules. 9.3.4

Carbon Nanotube Growth in Zeolites

Since Iijima reported carbon nanotube (Figure 9.16) preparation and its microstructure in 1991,[98] carbon nanotubes have attracted enormous attention. Because of their unique structures, these materials exhibit a variety of interesting physico-chemical properties, and they may play an important role in manufacture of nanodevices. Carbon nanotubes can be classified as single-wall and multi-wall forms. Single-wall carbon nanotubes can be regarded as being formed from curling of single-sheet graphite, whereas multi-wall carbon nanotubes may be considered as being formed from successive encapsulation of smaller single-wall nanotubes by larger ones. Regardless of whether it is single-wall or multi-wall, a complete carbon nanotube requires carbon cage fragments similar to fullerenes to close the ends. In some cases, the carbon nanotubes are not endclosed, and the valences of the carbon atoms at the ends may be saturated with other

Figure 9.16 High-resolution electron microscope images of carbon nanotubes first observed by Iijima (the fringes in the figure represent the walls of the carbon nanotubes). Reproduced with permission from [98]. Copyright (1991) Nature Publishing

626

Chemistry of Zeolites and Related Porous Materials

Figure 9.17 The structure of a single graphene sheet denoted with (n,m) integer pairs (the base vector of the unit cell of the graphene sheet is also indicated)

heteroatoms such as H, N, or O. When graphite single sheet is curled to form a single-wall carbon nanotube, various nanotubes with different diameters and symmetries may form depending on the curling manner. The structure of a particular carbon nanotube may be denoted using a pair of integers (n, m), and the vector C is defined as in Equation (9.7), C ¼ na1 þ ma2

ð9:7Þ

where a1 and a2 are the base vectors of the single graphene sheet, and n  m. Figure 9.17 illustrates a single graphene sheet with (n, m) integer pairs and the base vector. From the figure it is seen that when m ¼ 0, the formed carbon nanotube is a zigzag one, whereas when n ¼ m, the formed carbon nanotube is an armchair one. All other types of carbon nanotubes are chiral ones. Among the carbon nanotubes that can be capped by a hemisphere of C60,[99] those with a zigzag structure have an integer pair of (9,0) whereas those with an armchair structure have an integer pair of (5,5) (Figure 9.18). Because ja1 j ¼ ja2 j ¼ 0:246 nm, the magnitude of C can be represented as 0:246ðn2 þ nm þ m2 Þ1=2 , and the unit is nanometer. The diameter of a carbon nanotube represented by vector C is given in Equation (9.8). dt ¼ 0:246ðn2 þ nm þ m2 Þ1=2 =p

ð9:8Þ

Carbon nanotubes can be prepared through various techniques. The technique reported by Iijima is similar to that for the preparation of C60 and other fullerenes; that is, the graphite arc discharge method. The only difference lies in the fact that, for carbon

Figure 9.18 The structures of two carbon nanotubes which can be capped by half-C60 molecules. (a) Zigzag (9,0) nanotube; (b) armchair (5,5) nanotube. Reproduced with permission from [99]. Copyright (1999) Cambridge University Press

Porous Host–Guest Advanced Materials

627

nanotube preparation, there is a distance between the two electrodes, and the carbon is evaporated from the anode to be deposited on the cathode. This deposit contains carbon nanotubes. The carbon matter obtained through graphite vaporization and deposition on a quartz substrate using an electron beam in vacuum also contains carbon nanotubes. Of course, other techniques such as laser ablation and vacuum evaporation are also feasible for the preparation of carbon nanotubes through graphite vaporization. Pyrolysis of organic molecules such as benzene in the presence of H2 has also been used to prepare multi-wall carbon nanotubes. In some cases, addition of a certain amount of catalyst containing metals is conducive to the formation of carbon nanotubes. In addition, carbon nanotubes can also be obtained through electrochemical (electrolysis) methods. Using porous alumina as an external template, carbon nanotubes can be grown in the channels. However, the pore size of porous alumina is large, falling into the macropore (diameter > 50 nm) category, and therefore the diameter of the prepared carbon nanotubes is also large and they are usually multi-walled. Although there are many techniques for the preparation of carbon nanotubes, the preparation of single-wall carbon nanotube with uniform diameter is usually not easy. Carbon nanotubes with a diameter smaller than 1 nm are even more difficult to prepare. Because the properties of single-wall carbon nanotubes are different from those of multiwalled ones and these properties can be easily elucidated, it is more important to prepare single-wall carbon nanotubes. Theoretically, microporous crystals with one-dimensional channels are able to be used as external templates for the preparation of carbon nanotubes. Of course, the pore size of these microporous crystals should not be too small because otherwise even the thinnest carbon nanotube cannot be accommodated in the micropores. Tang et al. grew carbon nanotubes in AlPO4-5.[100,101] First, they prepared perfect AlPO4-5 single crystals hydrothermally using tripropylamine (TPA) as the template in the presence of F
ions. Secondly, they placed the AlPO4-5 single crystals in a vacuum system for pyrolysis at 500  800  C. The pyrolysed AlPO4-5 single crystals appear deep black, and they exhibit anisotropy for polarized light absorption. When the polarization direction of the polarized light coincides with the crystal channel running direction, the absorption reaches a maximum, whereas when the polarization direction is perpendicular to the channel axis the absorption is at its weakest.[102] Obviously, through pyrolysis, the tripropylamine molecules located in the AlPO4-5 micropore channels undergo decomposition and the product is a carbon-rich material. X-Ray photoelectron spectroscopy and elemental analysis indicate that this carbon-rich matter contains almost no N or H, and it can be regarded as being composed of carbon only. Although in principle the formation of carbon nanotubes in AlPO4-5 is possible, it is rather difficult to confirm the existence of carbon nanotubes in the pyrolysed product in the AlPO4-5 channels. The carbon material obtained through dissolution of the pyrolysed AlPO4-5 crystals in hydrochloric acid is very easily damaged to form graphite fragments under high-resolution electron microscopy. However, after close inspection, it is found that tiny single-wall carbon nanotubes are indeed present.[103] These observed carbon nanotubes may be more stable than others because they have fewer defects. The diameter of the carbon nanotube is about 0.42 nm, larger than the lattice spacing (0.34 nm) for the graphite fragments in the same TEM image. The evidence for the presence of carbon nanotubes in the pyrolysed AlPO4-5 crystal is from Raman spectroscopy. Figure 9.19 shows the Raman spectra of the AlPO4-5 crystals after treatment under various

628

Chemistry of Zeolites and Related Porous Materials

Figure 9.19 Raman spectra of AlPO4-5 after pyrolysis treatment at various temperatures. (a) as-Synthesized AlPO4-5 single crystals; (b) after pyrolysis at 400  C; (c) after pyrolysis at 500  C; (d) after pyrolysis at 550  C; (e) after calcination in O2 atmosphere at 600  C. Reproduced from [104]. Copyright (1999) Springer-Verlag

conditions. It is seen from the Figure that Raman signals corresponding to the template molecules appear in the spectrum of the as-synthesized TPA-containing AlPO4-5 crystals. When the sample is treated at 400  C, the Raman spectral peaks disappear completely, and in this case the template molecules decompose to form an amorphous material. If the crystals are calcined in air or in O2 atmosphere, the carbon material is removed completely, and in this case there are no Raman signals. Nevertheless, when the crystals are pyrolysed in vacuum at 500  550  C, the Raman spectrum shows a series of new peaks that are different from either TPA or graphite (diamond) signals. This phenomenon indicates that a new allotrope of carbon is formed. On the basis of the signal positions, it is inferred that this new carbon material possesses the features of a carbon nanotube.[104] Similar to the UV-visible absorption, the Raman spectrum of the carbon-nanotube-containing AlPO4-5 also exhibits anisotropy. The powder X-ray diffraction patterns for the as-synthesized AlPO4-5 crystals, the pyrolysed carbon-nanotube-containing AlPO4-5 crystals, and the AlPO4-5 crystals after complete template removal show no distinct change (Figure 9.20). However, the relative intensities of the diffraction peaks of the three materials differ, and with the removal of the template the diffraction intensities increase.[105] In addition, the diffraction positions are moved slightly toward higher angles. From the X-ray diffraction patterns it is concluded that template removal leads to slight lattice contraction of the AlPO4-5 crystals, but the formation of carbon nanotubes does not damage the basic structure of AlPO4-5. This indicates that the carbon nanotubes are strictly confined in the microporous channels of AlPO4-5. The carbon atoms of the carbon nanotubes formed after pyrolysis originate from the tripropylamine template molecules occluded in the AlPO4-5 crystal channels during the crystallization of AlPO4-5.

Porous Host–Guest Advanced Materials

629

Figure 9.20 Powder X-ray diffraction patterns of AlPO4-5 (Cu-Ka radiation, l ¼ 1.5418 A˚). (a) as-Synthesized template-containing sample; (b) pyrolysed sample containing carbon nanotubes; (c) completely detemplated sample. The inset shows the variation of the (001) reflection position.

The as-synthesized AlPO4-5 crystals are colorless, whereas, upon growth of carbon nanotubes through pyrolysis in vacuum, the crystals become dark brown in color. Furthermore, as the pyrolysis proceeds, this dark brown color is deepened. Through monitoring the degree of polarization (DOP) of the light absorption of the pyrolysed crystals, the number of carbon nanotubes can be determined. DOP is defined as in Equation (9.9): DOP ¼

jI?
I== j jI? þ I== j

ð9:9Þ

where I// (I?) represents intensity of transmitted light whose polarization direction is parallel with (perpendicular to) the c-axis of the AlPO4-5 crystal, and the DOP value varies within the range 01. If the DOP value is 0, the crystal is isotropical; and if the value is 1, the crystal is completely anisotropical. Figure 9.21 shows the plot of DOP value versus the AlPO4-5 pyrolysis temperature. From the Figure it is seen that pyrolysis of the AlPO4-5 crystals at a temperature below 700 K leads to no anisotropy of polarized light absorption. In fact, when the crystal is pyrolysed at a temperature slightly higher than 673 K for 4 h, the crystal color is dark, but in this case the crystal exhibits no lightpolarization property, suggesting that carbon nanotubes cannot form, although at this pyrolysis temperature the template molecules undergo decomposition. In the 703  753 K temperature range, the DOP value increases distinctly, indicating that a considerable number of carbon nanotubes are formed in this temperature range. Further increase of the pyrolysis temperature results in no obvious variation or even a slight decrease of the DOP value. It seems that the carbon nanotubes are stabilized in the AlPO4-5 crystal channels at elevated temperatures. The channel diameter of AlPO4-5 is 0.74 nm, and therefore the outer diameter of the formed carbon nanotubes must be smaller than this value. In fact, in consideration of the

630

Chemistry of Zeolites and Related Porous Materials

Figure 9.21 Plot of DOP versus pyrolysis temperature for AlPO4-5 crystals (left); polarized light images (right) of AlPO4-5 crystal: (a) as-synthesized, (b) pyrolysed at 673 K, (c) pyrolysed at 823 K (C stands for the direction of the crystal main channel axis).

carbon nanotube diameter, the van der Waals’ distance between the carbon nanotube and the crystal wall could be deducted. Figure 9.22 is a schematic of the structures of the three possible single-wall carbon nanotubes with a diameter of about 0.4 nm. They are designated (4,2), (3,3) and (5,0), respectively and the corresponding diameters are 0.419, 0.412, and 0.396 nm. If the van der Waals’ distances are added, the van der Waals’ diameters of the three carbon nanotubes are 0.759, 0.752, and 0.736 nm, respectively. It is clear that only the (5,0) carbon nanotube most fits the one-dimensional channel of the AlPO4-5 crystal. Owing to their structural features, carbon nanotubes exhibit unique electrical conduction properties. The carbon nanotubes prepared through common methods have a larger diameter and usually they show metallic or semiconducting properties. In AlPO4-5 crystals, the carbon nanotubes can grow along the microcrystal channels upon pyrolysis, and they are arrayed in an order manner. As a result, it is convenient to measure their conduction properties. The nanotube-containing AlPO4-5 crystals are vertically placed in

Figure 9.22 The structures of three possible single-wall carbon nanotubes with a diameter of about 0.4 nm, among which the diameter of the (4,2) tube is 0.419 nm, that of the (3,3) tube is 0.412 nm, whereas that of the (5,0) one is 0.396 nm

Porous Host–Guest Advanced Materials

631

the holes of a ceramic plate and fixed using epoxy resin. Both sides of the ceramic plate are polished so that the ends of the crystals are accessible and may undergo contact with the electrodes. In this way, the voltage–current curves and the plot of conductivity versus temperature are obtainable for the carbon nanotube/AlPO4-5 composite. The experimental results show that the template-containing AlPO4-5 and the completely detemplated AlPO4-5 crystals are typically insulators, whereas the carbon nanotube/AlPO4-5 crystal exhibits unique conduction properties. At room temperature, the measured conductivity is in 0:1 :cm
1 order of magnitude. The magnitude of this conductivity is smaller than that of metallic carbon nanotubes but is similar to that of semiconducting carbon nanotubes. As the temperature is decreased, the conductivity drops as well, indicating that the carbon nanotubes have semiconducting behavior. However, when the temperature is lowered to below 20 K, the carbon nanotube/AlPO4-5 crystals exhibit the Meissner effect; and this effect implies that the carbon nanotubes in the AlPO4-5 crystals are a one-dimensional superconductor at temperatures below 20 K.[106]

9.4 Semiconductor Nanoparticles in Zeolites Since late 1980s, increasing interest has been attracted to chemical fabrication of zerodimensional semiconductor clusters (or quantum dots) because these species exhibit unusual optoelectronic properties.[107] As the size of a semiconductor is decreased to the nanometer scale, the wavelength of electrons or holes in the crystal is comparable with the crystal size, and in this case the quantum-size effect occurs and the semiconductor bandgap varies with the crystal size. The smaller the crystal size, the wider the bandgap. This is the so-called blue-shift phenomenon. If the crystal size is small enough, strong exciton resonance appears. Because the pore diameter of a zeolite is usually smaller than 1.5 nm, the semiconductor particles grown in a zeolite microporous crystal host will exhibit an apparent quantum size effect. Metal-organic chemical-vapor deposition (MOCVD) technique can be used to prepare semiconductor nanoparticles in the pores of zeolites effectively. A typical example is the formation of compound semiconductor particles in zeolite Y through grafting of organometallic species followed by reaction with gases. This approach can be used to obtain II–VI-, IV–VI-, and III–V-type semiconductor compounds encapsulated in zeolite Y.[108–110] Because of the presence of charge-balancing protons, it is very easy for H-Y to react with the organometallic species in the channels as per Equation (9.10): ZO-H þ ðCH3 Þ4 Sn ! ZO-SnðCH3 Þ3 þ CH4

ð9:10Þ

where Z stands for zeolite framework. Quantitatively, the number of protons in zeolite channels corresponds to the number of organometallic molecules that can be grafted into the zeolite channels to form methane molecules. After formation of the grafting host– guest material, the grafted organometallic groups may react with the H2S introduced into the system to form hydrogen sulfide compounds; as per Equations (9.11) and (9.12): ZO-SnðCH3 Þn þ nH2 S ! ZO-SnðSHÞn þ nCH4

ð9:11Þ

ZO-Sn-ðSH=CH3 Þ þ H2 S ! ZOH þ HS-SnðSH=CH3 Þ

ð9:12Þ

632

Chemistry of Zeolites and Related Porous Materials

Afterward, the host–guest material is thermally treated at a certain temperature, and the hydrogen sulfide species in the microporous channels will condense to form sulfide clusters (Sn4S6). Of course, other types of metal sulfide nanoclusters such as Cd6Se4, Zn6S4, etc., can also be obtained through variation of the metal in the organometallic molecule or through reaction with hydrogen sulfide. In some cases, the newly formed protons can react with further introduced organometallic compounds, and, as a result, the amount of the loaded metal sulfide can be increased further. Through this approach, not only II–VI- but also III–V-type semiconductor, such as GaP, clusters can be prepared. These semiconductor clusters show distinct quantum-size effects, and an apparent blue shift is observed for the band edge of their electronic transition absorptions. In 1990, Canham reported the room-temperature luminescence of porous silicon.[111] This important discovery has attracted much attention towards an investigation in silicon with the quantum-size effect. Through chemical-vapor deposition, it is also possible to load silicon nanoparticles into zeolite channels and pores. First, disilane is treated with the protons in a zeolite and is grafted to the zeolite walls [Equation (9.13)]: ZO-H þ H3 SiSiH3 ! ZOSiH2 SiH3 þ H2

ð9:13Þ

Secondly, mild thermal treatment leads to decomposition of the grafted disilane molecules, and the decomposed product reacts with the grafted disilane to form clusters. Finally, clusters containing as many as 60 Si atoms are formed in the supercages of zeolite Y.[112,113] These clusters located in the zeolite cages emit orange-red light at room temperature. With a decrease of temperature, the emission intensity of the host–guest material is further enhanced (see Figure 9.23). Gao et al. used single-crystal silicon as the silicon source and prepared giant crystals of silicalite-I, and they deposited silicon nanoparticles in the zeolite channels through reaction of silane with the detemplated silicalite-I single crystals.[114] Further investigation indicates that the silicon nanoparticles located in the silicalite-I single crystals emit strong red light, and the luminescence spectral wavelength is 570 nm at room temperature, whereas it blue-shifts to 551 nm at 10 K. Besides silicon nanoclusters, it is also possible to prepare germanium and

Figure 9.23 Variation of the luminescence property for silicon nanoclusters loaded in zeolite Y. Reproduced with permission from [107]. Copyright (1996) Wiley-VCH

Porous Host–Guest Advanced Materials

633

silicon–germanium nanoclusters in zeolite Y through the vapor-deposition technique. Similar to compound semiconductor nanoclusters, these element semiconductor nanoclusters also exhibit apparent quantum-size effects, and their bandgap is distinctly blue-shifted in comparison with the corresponding bulk materials. Through chemical-vapor deposition, not only can semiconductor nanoclusters be prepared in microporous zeolites, but also larger semiconductor clusters in mesoporous molecular sieves can be obtained. Detemplated mesoporous silica possesses rich silanols on the channel walls, and these silanols easily react with organometallic molecules so as to graft the latter onto the mesoporous channel walls. It has been demonstrated that 200 wt% disilane can be loaded in detemplated MCM-41.[112] After thermal treatment, these grafted disilane species decompose to form silicon nanoclusters. Because the silicon content is high, the silicon nanoclusters in the mesoporous channels are actually able to connect with one another to form nanowires. As is the case for nanosemiconductors in microporous crystals, the nanosemiconductor clusters located in mesoporous molecular sieves also exhibit quantum-size effects. Their bandgaps and emission energies are correlated with the loading amount of semiconductor and the particle size. Through chemical-vapor deposition, Ge nanowires can also be formed in mesoporous channels.[115] It has been shown that the reaction of Cd2þ-ion-exchanged zeolites with H2S gas results in formation of cadmium sulfide nanoparticles. In the sodalite cage of faujasite, nanoclusters such as Cd4S4 can be formed, as confirmed by extended X-ray fine-structure analysis and X-ray powder diffraction structural analysis.[116] Owing to the quantum-size effects of nanoparticles, the UV-visible diffuse reflectance spectrum of the cadmium sulfide nanoparticles loaded in zeolite microporous channels is distinctly different from that of bulk CdS. The bandgap of bulk CdS semiconductor corresponds to the absorption edge of electronic transition. However, in microporous zeolite crystals, if the concentration of the Cd4S4 nanoparticles formed is small, the related electronic absorption spectrum blue-shifts markedly, indicating that as the nanoparticles are formed the semiconductor bandgap is widened. When the number of Cd2þ ions in the system is increased, the concentration of the formed Cd4S4 nanoparticles is increased as well. In this case, because of the interactions between semiconductor nanoparticles located in adjacent sodalite cages, the absorption edge shifts gradually toward the longer wavelength region. Nevertheless, for molecular sieves with different structures, the degree to which the nanoparticles interact with one another varies. The interactions in zeolite A are weaker than in zeolite Y. Detailed structural analysis indicates that the cadmium sulfide clusters are formed only in the sodalite cages (Figure 9.24) and not in the supercages of faujasite. This is probably because the coordination of the framework oxygen atoms of the sodalite cage stabilize the formed nanoclusters. It is too ideal for one Cd4S4 nanocluster to occupy one sodalite cage. In the treatment of zeolite with H2S many mesoporous defects will be produced, and these mesoporous defects may accommodate larger cadmium sulfide clusters. Therefore, in the microporous crystal with a particular structure, there may exist many different cadmium sulfide particles with various sizes. However, the content of these cadmium sulfide clusters located defects is usually limited, and, as a result, they have little effect on the electronic spectral properties of the composite. Detailed composition and structural analysis indicates that the nanoclusters contain not only Cd and S, but also O. In fact, these clusters can be written as Cd4(S,O)4.

634

Chemistry of Zeolites and Related Porous Materials

Figure 9.24 Structure of Cd4S4 cluster (a) and its location in sodalite cage (b). Reproduced with permission from [116]. Copyright (1989) American Chemical Society

Besides cadmium sulfide, other semiconductor particles can also be loaded in zeolite microporous crystals. Moller et al. prepared cadmium sulfide nanoparticles in zeolite Y through a similar approach.[117] Nevertheless, structural analysis indicates that the formed cluster particles are actually rather complex, and apart from cadmium sulfide clusters, there exist other nanoclusters such as Cd4O4 or Cd2O2Se in the channels of zeolite Y. These nanoclusters are not isolated, and they strongly interact with the framework oxygen of the zeolite. The low-temperature phase of bulk silver sulfide is a semiconductor with a monoclinic structure, and its room-temperature bandgap is 1 eV. Because of its unique luminescence property, silver sulfide nanoparticles have attracted much interest recently. The technique to grow silver sulfide nanoparticles usually involves formation of Agþ-loaded zeolite through ion-exchange of Agþ ions into a particular zeolite (such as zeolite A). Silver sulfide nanoparticles can be obtained through reaction of the loaded zeolite with H2S gas. The silver sulfide particle size in zeolite A may be controlled through varying the amount of Agþ ions initially exchanged into the zeolite channels. Nevertheless, if the the silver sulfide nanoclusters are located only in the a-cages of zeolite A, their size should not exceed the diameter of the cage (about 1.5 nm). At higher concentrations, the silver nanoclusters in adjacent cages may interact with one another, leading to variation of the UV-visible absorption and emission spectra.[118,119] PbI2 exhibits semiconducting properties. This compound may be loaded into zeolite channels through vapor transportation. When dehydrated Na-A zeolite is sealed with PbI2 in an evacuated quartz tube followed by thermal treatment at 420  C for 24 h, the PbI2 will enter into Na-A through a vapor-deposition process.[120] The loading amount of the guest in the host can be varied through control of the ratio of zeolite over PbI2 in the quartz tube. UV-visible diffuse reflectance spectroscopy reveals that as the loading amount of the guest is increased, PbI2 forms (PbI2)4 clusters initially in the zeolite supercages followed by growth to afford (PbI2)5 clusters. HgI2 is also a typical semiconductor material, and it is also easy to vaporize after thermal treatment. Therefore, HgI2 may be loaded into various zeolites through a vaportransportation approach. Because of the confinement of the zeolite framework, HgI2 in zeolites may exhibit distinct quantum-size effects. The electronic transition absorption spectrum of HgI2 loaded in AlPO5-5 single crystals shows an apparent blue-shift, and in

Porous Host–Guest Advanced Materials

635

the meantime the composite exhibits anisotropic properties.[121] This anisotropy suggests that the HgI2 in AlPO4-5 1-D channels is probably present as chains. Element semiconductors include Se, Te, Ge, and Si, etc. Selenium has a low melting point (230  C), and at lower temperatures it gives high vapor pressures. Therefore, it is possible to load selenium in a zeolite through vapor deposition at a temperature as low as 150  C when selenium is sealed in a vacuum system with the dehydrated zeolite. Parise et al. investigated the incorporation of Se into zeolites A, X, Y, AlPO4-5, and mordenite.[122] Element selenium exists as different forms in different channel structures. In linear-channel molecular sieves (AlPO4-5 and mordenite), it is present as chains running along the channel direction,[123–125] whereas in the smaller-cage zeolite A, it is located in the cages as Se8 ring molecules. However, in larger-cage zeolites such as X and Y, there exist both selenium ring molecules and spiral chains. The distance between the adjacent selenium atoms in each selenium chain located in a zeolite is shorter than that of the Se

Se bond length of bulk selenium crystal, because the interactions between selenium chains are weakened when they are accommodated in zeolite channels. Through a polarized Raman spectroscopic study on Se-incorporated AlPO4-5 single crystals,[126] Poborchii et al. discovered that in the main channels there are a small number of Se chains besides the spiral Se single chains and Se8 ring molecules. In mordenite, the state of selenium in the channels varies to a certain extent depending on the types of cations in the zeolite and the preparation method of the host–guest material.[127] When the Na is replaced by K in mordenite, the ratio of chained Se over Se6 ring in the loaded guest species is decreased; and in the meantime, if the guest is loaded with the vapor-deposition method, the obtained guest Se is more ordered than the guest molecules obtained through liquid Se injection. The case for Te is similar to that for Se; that is, the guest Te may exist in the host as either chains or as ring forms depending on the host’s channel structure. Polarized Raman spectroscopy and electronic transition spectroscopy indicate that the spiral S and Te chains are able to be loaded in mordenite through the vapor-deposition technique,[128] and, in the meantime, S6, S8, and Te6 elemental ring molecules may also exist. III–V-Type compound semiconductors have been gaining more attention recently. It has been reported that III–V-type semiconductor nanoparticles or nanowires can be grown in mesoporous channels through chemical vapor deposition. The preparation of III–V semiconductor nanoclusters follows the same principle as for silicon and germanium clusters. First, the organometallic compound of Al, Ga, or In (such as trimethylindium) is grafted onto the mesoporous channel walls through vapor-deposition reaction, and secondly phosphine is introduced to react with the alkyl metal to form III–V semiconductor compound particles. Because the reaction of phosphine proceeds at a rather high temperature (about 300  C), there is no need to pyrolyse the reaction product.[129] Spectroscopic and high-resolution electron microscopic analysis indicate that the III–V semiconductor clusters prepared through this technique are distributed not only in the mesoporous channels but are also partly deposited on the external surface of the mesoporous molecular sieve. Through reaction of adsorbed (CH3)3Ga with PH3, it is possible to grow GaP nanoclusters in zeolite pores and channels. Srdanov et al. investigated the assembly of GaAs in mesoporous MCM-41 and the optical properties of the assembled host–guest composite material.[130] They used tert-butylarsine and trimethylgallium as the arsenic and gallium sources, respectively, and deposited gallium

636

Chemistry of Zeolites and Related Porous Materials

arsenide directly in the MCM-41 channels at 700  C through organometallic chemical vapor deposition. The electronic transition absorption spectrum of the host–guest composite compound formed through deposition shows an apparent blue shift, indicative of a quantum-size effect. The luminescence property of the composite depends on pore size of the host MCM-41 material. Further analysis indicates that the particle size distribution of the deposited GaAs nanoparticles is wide, and the particles are present not only in the MCM-41 channels but also on the external surface of the mesoporous molecular sieve. The GaAs particles on the external surface of mesoporous material are larger.

9.5 Metal Complexes in Molecular Sieves Because of the confinement and separation of zeolite frameworks, metal complexes in molecular sieve channels exhibit physical-chemical properties different from those in solution and in the solid state.[131] The molecular sieve frameworks protect the metal complexes and the latter show enhanced thermal stability and antioxidation properties. The methods of preparation of metal complexes in zeolites are versatile, but they may be classified as the following: (1) direct synthesis of porous framework compounds encapsulating metal complexes using the metal complex as a template; (2) the so-called ship-in-bottle approach, where small ligands complex with the metal ion to form a larger metal complex in the zeolite channels; (3) loading of volatile metal complexes into zeolite channels through vapor transportation; (4) grafting of complexes onto the walls of molecular sieves; this approach is suitable for preparation of complex/mesoporous molecular sieve host–guest composites. In fact, as the study of this topic is deepened, the assembly of host–guest systems involving molecular sieves will be extended to a great extent, and the functionality of the assembly-formed composite materials will gain more attention. Some composites that are potentially applicable in areas such as catalysis, luminescence, and biomimetics will be developed continuously. In the meantime, the structures, properties, and the mechanisms for their optochemical and optophysical interactions will be revealed effectively. 9.5.1

Incorporation of Metal–Pyridine Ligand Complexes

Pyridine Complexes In N-containing heterocyclic aromatic hydrocarbons the extended p-bonding renders the electrons delocalized. These compounds possess not only basicity but also strong coordination ability toward metals. As ligands, they complex metal ions, especially transition metal ions, to form versatile compounds with rich special properties. As early as the 1950s and 1960s, copper–pyridine complex was used as a homogeneous catalyst for oxidation-coupling reactions, and the reactants included acetylene,[132] phenols,[133] and so on. Copper(II) tetrapyridine complex may catalyse the oxidation of 2,6-dimethylphenol to oxides of phenols in the presence of O2. If the complex is loaded or dispersed on a solid support or into a porous crystal, it may be possible to obtain composite heterogeneous catalysts with superior catalytic performance, and, in the meantime, the shape selectivity of porous crystals may be taken advantage of. Ukisu et al.[134] successfully incorporated the copper complex of a substituted pyridine into the pores

Porous Host–Guest Advanced Materials

637

of zeolite Y, and they investigated the state and catalytic performance of the loaded complex. They first dealuminated zeolite Y to form an Si-OH-enriched microporous crystal, and then they used 2-[2-(trichlorosilanyl)ethyl]pyridine to react with the Si-OH groups. The former was thus grafted onto the walls of the Y zeolite, forming Py-NaY composite. When this Py-NaY composite was contacted with Cu2þ-containing solution, Cu-Py-NaY composite was formed. X-Ray absorption fine structure analysis indicated that the oxidation state of the oxidized copper was 2þ, whereas that of the reduced copper was þ1. Dai and Lunsford reported that the Cu2þ of [Cu(Py)4]2þ located in the cages of zeolite Y was reduced to Cuþ after evacuation to allow dehydration, but the Cuþ was able to be oxidized to Cu2þ again in the presence of O2.[135] Bo¨hlmann et al. loaded copper pyridine complex into mesoporous MCM-41 and investigated the electron paramagnetic properties of the composite.[136] The loading method they used was ion-exchange; that is, the cation complex of Cu2þ and pyridine was first prepared, followed by ion-exchange of the complex into Al-containing MCM-41 channels. Po¨ppl and Kevan incorporated copper–pyridine complex into Al-free MCM-41 channels, and they found that in this case the interactions between the complex and the MCM-41 walls were weak.[137] Yamada investigated the substitution reaction of copper(II) bis(acetylacetonate) loaded into zeolite NaX by pyridine.[138] ESR signal variation was clearly observed after the substitution reaction, and this substitution reaction could not proceed in solution or on silica gel. The static electric field generated by the zeolite cations tends to reduce the stability of the Cu(acac)2þ complex. Bipyridine Complexes 2,20 -Bipyridine has two aromatic rings and two coordinating N atoms, and therefore it is very easy for this ligand to chelate a transition metal ion. Furthermore, the two aromatic rings of this molecule make the electrons more delocalized. As a result, bipyridine complexes of transition metals usually possess properties more interesting than those of pyridine complexes. Similar to pyridine complexes, bipyridine complexes can be incorporated into porous crystal hosts to form host–guest composites through various techniques. The most common method is the so-called ship-in-bottle approach, that is, transition metal ions are introduced into the channels or pores of porous crystals through ion-exchange, followed by coordination of the metals by successively incorporated bypyridine ligands. The formed complexes are so bulky that they cannot escape from the porous crystal windows as is the case for a large ship in a bottle which cannot escape from the bottleneck. The complexes of pyridine and ruthenium possess unique luminescent and catalytic properties, and therefore their assemblage as guest molecules has been extensively investigated. Dewilde et al. first reported the incorporation of tris(2,20 -bipyridine)ruthenium(II) complex into zeolite Y through the ship-in-bottle approach.[139] They first introduced [RuIII(NH3)6]3þ into zeolite Y through ion-exchange to form [RuIII(NH3)6]-Y, and then heated the mixture of bipyridine and [RuIII(NH3)6]-Y. In this case, the RuIII was reduced to RuII and [RuII(bpy)3]-Y, which has spectral properties completely different from that in solution, was formed. After incorporation, the complex exhibits versatile optophysical behaviors which depend on the hydration degree, loading amount, and so on. Later on, Quayle et al. reported the oxidation of the [RuII(bpy)3]-Y to [RuIII(bpy)3]-Y using Cl2.[140] The RuIII complexes loaded in zeolites are able to decompose water into O2.

638

Chemistry of Zeolites and Related Porous Materials

Calzaferri and coworkers described the assemblage of [Ru(bpy)3]2þ into the supercages of zeolite Y in detail.[141] They found that generally the first ship-in-bottle reaction may introduce [Ru(bpy)3]2þ into 50% of the supercages of zeolite Y, whereas multiple loadings may incorporate the complex into 65% of the supercages. Attempts to load [Ru(bpy)3]2þ into more than 65% of the supercages would lead to the formation of other complexes such as [Ru(bpy)n(NH3)6
2n)2þ etc. Ledrey and Dutta investigated, in detail, the catalytic performance of [RuIII(bpy)3]3þ loaded in zeolite for the oxidation of water,[142] and it was found that when the pH of the system was lower than 4, no O2 was formed. They believed that the RuIII complex first binds H2O to form OH radicals which evolve to H2O2, and the latter then interacts with the unchanged RuIII to form O2. The same research team[143] prepared [Ru(bpy)2(H2O)2]-Y and oxidized it to 2þ
[(bpy)2(Oz)RuIV


O] in air (where Oz stands for zeolite-framework O). They also 2þ prepared Ru(bpy)3] -zeolite composite material with different loading amounts, and investigated the interactions between the guest species located in different supercages.[144] In the absence of identical guest complexes in cages surrounding a cage containing a particular complex, the fluorescence lifetime of the host–guest composite is similar to that of the complex in solution, whereas when the loading amount of the guest reaches a certain value, the luminescence intensity of the host–guest composite decreases, and, in the meantime, the fluorescence decay rate is increased and the lifetime is shortened. Excited–ground state or excited–excited state energy (or electron) transfers may occur for guest complexes. Maruszewki characterized the multi-pyridine complex of ruthenium loaded in zeolite Y cages using absorption and resonance Raman spectroscopy.[145] The complexes involved included [Ru(bpy)2]2þ, [Ru(bpy)(bpz)]2þ, [Ru(byp)2(dmb)]2þ, and so on. Photoelectrontransfer reaction may occur between [Ru(bpy)3]2þ loaded in zeolite cages and methyl violet.[146] After excitation by light illumination, the excited state *[Ru(II)(bpy)3]2þ generated through charge-transfer transition transfers its electron to the methyl violet in an adjacent cage to form MVþ  ion radical, whereas the complex itself is converted into [RuIII(bpy)3]3þ ion [Figure 9.25(a)]. To maintain the charge balance with the zeolite

Figure 9.25 Electron-transfer reaction-chain based on zeolite channel. (a) Photoinduced electron transfer between [Ru(bpy)3]2þ and methyl violet; (b) redox chain of [Ru(bpy)2(bpz)]nþ/DQ55nþ/PVS. Reproduced with permission from [80]. Copyright (2001) Elsevier

Porous Host–Guest Advanced Materials

639

framework, for every electron transferred, an Naþ ion must be moved simultaneously. Therefore, the mobility of the Naþ ion affects the electron-transfer rate and the recovery of the electron. The lifetime of the MVþ  cationic radical is several hours. Raman spectroscopy indicates that the radical has strong interactions with the zeolite framework. Kincald incorporated the complex of ruthenium with bipyridine and 2,20 -bipiperazine (abbreviated as bpz), Ru(bpy)2bpz, ruthenium 5-methyl-2,20 -bipyridine (abbreviated as mmb) complex, and N,N0 -trimethyl-2,20 -bipyridinediium (abbreviated as DQ552þ) into the adjacent supercages of zeolite Y to form a photochemical system in which the [RuII(bpy)2(bpz)]2þ acts as a photosensor.[147] Upon illumination with light, an electron is promoted from the ground state to the excited state to form *[Ru(bpy)2(bpz)]2þ, and then this excited state loses an electron which is transferred to the DQ552þ, the latter being reduced to DQ55þ  , whereas the complex is oxidized to [Ru(bpy)2(bpz)]3þ. The oxidized complex may obtain another electron from the [Ru(mmb)3]2þ in an adjacent cage to become its reduced state. In this way, the [Ru(mmb)3]2þ does not interact with DQ552þ directly to undergo redox reaction, and the electron-transfer process is accomplished through the photosensor-separated electron donor–acceptor system [Figure 9.25(b)]. After reduction, the electron acceptor DQ552þ interacts with the propylviologensulfonate, PVS, in the solution and reduces the latter. The PVS-reduction efficiency of this double-complex system is higher than that of the monocomplex-zeolite Y system formed from [Ru(mmb)3]2þ or [Ru(bpy)2(bpz)]2þ alone. Kozuka and coworkers incorporated [Ru(bpy)3]2þ into a silica gel membrane prepared through the sol–gel technique, and investigated the luminescence properties of the composite.[148] They hydrolysed tetraethyl orthosilicate mixed with [Ru(bpy)3]Cl2 6H2O O and directly obtained the silica film encapsulating the complex molecules. It was found that the luminescence arising from recovery of ground state from excited state which corresponds to metal-ligand charge transfer transition in the complex on the sol-gel film is red-shifted, whereas after thermal treatment the luminescence is blue-shifted. However, when the heating temperature is over 200  C, the luminescence is red-shifted again. With an increase in drying temperature, the fluorescence lifetime is increased, whereas the luminescence efficiency is decreased. They attributed the luminescence efficiency decrease to the quenching effect of O2 in air toward the complex molecules. Besides forming unique complexes with ruthenium, bipyridine can also form functional coordination compounds with other metals. Pinnavaia and colleagues investigated the incorporation of a bipyridine complex of manganese(II) into mesoporous MCM-41 and they found that the composite is an effective catalyst for the oxidation of styrene.[149] Through electron-spin-resonance spectroscopy (ESR) in combination with UV-visible and IR spectroscopic analysis, Kevan and coworkers revealed the state and related properties of the manganese-bipyridine complex loaded in MCM-41 channels.[150] They discovered that when the loading amount is low, the guest complex exists as monomolecules in the channels, whereas if the loading concentration exceeds a particular value, the guest complex molecules undergo aggregation, and the ESR signal fine-structure disappears. The manganese ion in the guest complex molecule may undergo reversible redox reaction, and the oxidation state may vary between þ2 and þ4. This lays the foundation for the host–guest composite material to be used as a redox catalyst. Knops-Gerrits et al. investigated the luminescence properties of the zeolite X, Y, and EMT-encapsulated complex formed from MnII and bipyridine or o-phenanthroline (abbreviated as phen).[151]

640

Chemistry of Zeolites and Related Porous Materials

The structure of zeolite X is identical with that of zeolite Y but the numbers of negative charges contained in the framework and the cations (Naþ) are different for these two zeolites; the topology of EMT is different from that of zeolite X or Y and the number of its framework negative charges and cations is also less, but its pore opening size is comparable with that of zeolite X or Y. [Mn(bpy)3]2þ in all zeolites shows charge-transfer transition absorption (495 nm), whereas the charge-transfer transition absorption for [Mn(phen)3]2þ is broadened because of steric hindrance which causes distortion. The luminescence efficiency of [Mn(bpy)3]2þ incorporated in zeolite X is enhanced, whereas the luminescence wavelength is red-shifted; when this molecule is loaded in the more spacious EMT, its luminescence wavelength is red-shifted and the corresponding luminescence efficiency is reduced. The luminescence of [Mn(phen)3]2þ in zeolites is weak and the signal is widened. Quayle et al. exchanged FeII into NaY under the protection of N2, and then mixed the exchanged product with bipyridine and heated the mixture to form [FeII(bpy)3]-Y composite compound.[152] When the content of FeII is smaller than one FeII per supercage, the coordination is the most effective. Whereas if the FeII content is too high, the coordination reaction will drive the excess of FeII to be located in between two complex molecules and to bind the ligands through extra p-bondings. When the FeII loading amount is low, chlorine gas may oxidize [Fe(bpy)3]2þ to [Fe(bpy)3]3þ, and the oxidation conversion is 90%. At high loading, this oxidation is still very much incomplete, probably because the chlorine gas cannot enter into the zeolite pores easily or because there is not enough room for the formation of chloride ion. Umemura et al. used various techniques to investigate the assembly and states of [FeL3]2þ in zeolite Y [L ¼ ethylenediamine, 2-(aminoethyl)pyridine, 2,20 -bipyridine, 1,10-o-phenanthroline, 4,40 -dimethyl-2,20 -bipyridine, and 5,6-dimethyl-1,10-o-phananthroline].[153] It was found that the former four ligands can form guest complexes with the iron ions in the zeolite channels because their volumes are small, whereas the latter two are not able to form coordination compounds with the FeII ion in the zeolite Y cages because of their larger size. The structure of the guest complexes formed by ethylenediamine and 2(aminoethyl)pyridine undergo no distortion, whereas in the guest complexes formed by 2,20 -bipyridine and 1,10-o-phenanthroline the FeII is in its low-spin state, suggesting that the squeezing of the zeolite framework enhances the coordination ability of the ligand toward the FeII center. 9.5.2

Incorporation of Metal–Schiff Base Complexes


Imines are C


NH group-containing compounds formed through reaction of ammonia with an aldehyde or a ketone. However, this type of imine is very unstable, and it is easy for these compounds to hydrolyse to compounds containing a carbonyl group. However, the substituted imines synthesized from condensation of a primary amine and an aldehyde or ketone are much more stable [Equation (9.14)]: 0
RCHO þ R0 NH2 ! RCH


NR þ H2 O

ð9:14Þ

These substituted imines are the so-called Schiff bases. Although substituted imines are more stable than imines, they still undergo reversible hydrolysis or polymerization. Nevertheless, if the substituting alkyl groups on the C and N atoms are replaced by

Porous Host–Guest Advanced Materials

641

aromatic groups, the imines thus formed are further stabilized. Because of the difference in substituted groups, the types of Schiff bases are versatile. N,N0 -Di(salicylal)ethylenediimine is a typical Schiff base formed through condensation of salicylaldehyde and ethylenediamine, and this compound is abbreviated as SALEN. SALEN is a widely used chelating ligand; not only are its two N-atoms capable of coordinating but also its two OH groups are able to participate in coordination. Balkus Jr. et al. successfully prepared Rh(SALEN) complex in zeolites X and Y. They incorporated RhIII into the zeolite through ion-exchange and heated the mixture of the zeolite and SALEN at 140  C for 13 h.[154] The reaction product was cooled, and washed copiously with chloroform, and brown-yellow Rh(SALEN)-Y or Rh(SALEN)-X was obtained. In 1991, Bedioui et al. reported incorporation of CoIII-SALEN complex into zeolite Y and investigated the electrochemical properties of the composite. Co(SALEN)-Y has two electrochemical signals, corresponding to CoIII/CoII and CoII/CoI redox pairs, respectively.[155] These potential values are the same as those of the signals of the monomer Co(SALEN)3þ in solution. In addition, there is another pair of redox signal for Co(SALEN)3þ between the typical signals for CoIII/CoII and CoII/CoI. Bedioui et al. attributed this pair of signals to the CoIII/CoII redox reaction of Co(SALEN)3þ, which has strong interactions with the zeolite framework. In other words, Co(SALEN)3þ exists in zeolite Y in two different coordination states, but the detailed coordination modes are not clear. It was revealed that the assembly product Co(SALEN)-Y may be used as a good oxidation–reduction catalyst. Bedioui et al. also prepared [MnIIISALEN]þ and [FeIIISALEN]þ complexes in zeolite Y.[156] Electrochemical analysis results indicated that, depending on preparation methods, two types of complexes with different coordination modes are present in the zeolite channels. In addition, the incorporated complex is accessible by small molecules, especially molecular oxygen, in solution, and the latter may become activated. This suggests that the zeolite-loaded [FeIIISALEN]þ and [MnIIISALEN]þ are able to act as biomimetic molecules to participate in oxygen transportation or activation. Their advantage lies in the fact that they do not form dimers as in solution. Bessel and Rolison reported the electrochemical behavior of [Co(SALEN)]2þ and Fe(bpy)3]2þ in zeolite Y.[157] They prepared an electrode using the complex–zeolite composite and carbon powder, and tested the electrochemical behaviors of the electrode and the composite dispersed in a solution. It was found that the electrochemical behaviors of these two materials differ to a great extent. After several cycles, the former loses all the electrochemical signals, whereas the latter continuously shows the signals. They believed that the electrochemical signals arise from the complex attached onto the zeolite external surface (defects or external supercages), whereas the complex inside the zeolite channel does not participate in electron transfer of the electrochemical process. In fact, there has been dispute on whether the electrochemical signals arise from electron transfer in zeolite channels or from those complexes on the zeolite external surface. Both views can find experimental support.[158,159] Pd(SALEN) exhibits good catalytic activity when used as homogeneous and heterogeneous hydrogenation catalysts, but the selectivity is not satisfactory. To solve this problem, Kowalak et al. incorporated Pd(SALEN) into zeolites X and Y and used the composite for a selective hydrogenation catalyst.[160] They discovered that the composite shows high selectivity for the hydrogenation of alkenes; the main product components obtained through hydrogenation of a 1:1 (v/v) mixture of hexene and cyclohexene are

642

Chemistry of Zeolites and Related Porous Materials

hexane and E-2-hexene, whereas cyclohexane is not observable in the product. This suggests that channels of zeolite X or Y limit the formation of larger molecules so that the reaction selectivity is enhanced. 9.5.3

Incorporation of Porphyrin and Phthalocyanine Complexes

Porphin (Porphyrin) is a heterocyclic compound containing 20 carbon atoms and can be regarded as being constructed from 4 pyrrole rings. This compound is formed through condensation of form aldehyde and pyrrole. Another common macrocyclic N-containing compound is phthalocyanine, which can be formed through polymerization of odicyanobenzene under acidic conditions. The structures of these two heterocyclic compounds are shown in Figure 9.26. Porphins with substituent groups on the macrocycle are named as derivatives of porphyrin. Because the properties of the substituents differ, the corresponding porphyrin molecules have different properties. Although, in most cases, the coordination mode of a porphyrin is the formation of a coordination plane by the 4 N atoms with the metal center, a porphyrin may also chelate a metal ion in a bidentate, tridentate, or a nonplanar tetradentate mode. In these coordination modes, the metal ion is located outside the plane. The macrocycle phthalocyanine contains 8 N atoms, but usually only the four N-atoms on the inner side of the cycle are able to coordinate. In fact, in most cases the synthesis of phthalocyanine is realized in the presence of a metal ion as the template. It is also possible to attach various substituents on the phthalocyanine macrocycle. As for porphyrin, when coordinating to a metal ion, the H-atoms of the two NH groups on the inner side of the phthalocyanine cycle are replaced. The incorporation of metal porphyrin and phthalocyanine complexes into porous crystals has been gaining increasing interest. The properties of the complexes located in zeolite channels or cages are usually different from those of the compounds in solution, and they may find applications in areas such as catalysis, photochemistry, electrochemistry, and biomimetics. Nakamura et al. successfully synthesized tetramethylporphyrin (TMP) complexes of iron and manganese in NaY zeolite.[161] They investigated the catalytic properties of the composite for oxidation of cyclohexane in the presence of H2O2. The results indicate that the catalytic activity of [Fe(TMP)]-Y and [Mn(TMP)]-Y is enhanced in comparison with the corresponding FeII and MnII exchanged Y zeolite, and the catalytic product consists

N N NH

NH HN

N

N

N

N N

HN N

Figure 9.26 Structures of porphin (Porphyrin) and phthalocyanine. Both possess cyclic delocadized p bonds, and therefore both are aromatic to a certain degree

Porous Host–Guest Advanced Materials

643

mainly of cyclohexanol and cyclohexanone. Liu et al. silylated the mesoporous MCM-41 channel walls using 3-aminopropyl(triethoxy)silane and then introduced the ruthenium complex of a porphyrin carbonyl.[162] The porphyrin they used was tetrakis(4-chlorophenyl)porphyrin. The 4 N-atoms of the porphyrin and one carbonyl occupy 5 coordination sites of the ruthenium and the remaining coordination site was occupied by a solvent molecule, and it is very easy for the solvent molecule to be replaced by the NH2 group of the silylation agent so that the complex is tightly attached onto the MCM41 channel walls. The assembly composite compound possesses good catalytic properties for oxidation reactions. The research results also indicate that the catalytic performance is optimal if the Ru weight content is in the range 0.1  0.8%. Unlike in solution, the incorporated complex does not form dimers because it is attached to the channel wall. This is one of the reasons that the catalytic activity of the composite material is high. Wang et al. synthesized tetrachlorotetramethylporphyrin (TCTMP) and tetrabromotetramethylporphyrin (TBTMP), the size of which is comparable with the inner diameter of NaX zeolite.[163] They encapsulated the porphyrins into Co2þion-exchanged NaY supercages through a solid–liquid-phase step-by-step synthetic technique, and found that the decomposition temperature of the encapsulated porphyrin molecules is increased by about 70  C. For the oxidation reaction of styrene by H2O2,[164] the porphyrin-loaded zeolite exhibits a catalytic conversion 12-times that of the pristine metal-porphyrin. Through step-by-step sealed synthesis, tetraphenyltetrabenzoporphine may also be incorporated into NaY supercages. Wang et al. investigated the catalytic activities of tetraphenyltetrabenzoporphyrinzinc(II) [TPTBP-Zn(II)], NaY and the porphyrin-loaded NaY (TPTBP-Zn(II)-NaY) for the oxidation reaction of styrene by H2O2,[165] and they found that tetraphenyltetrabenzoporphyrinzinc(II) has no catalytic activity, whereas the free zeolite shows some activity but the composite of zeolite and the porphyrin exhibits much enhanced activity. The fact that the catalytic activity of the metal-porphyrin loaded in a zeolite is enhanced may be because the reactants in the zeolite cages are electrostatically interacted by both the metal-porphyrin and the zeolite so that the degree of styrene activation is increased. It was also found experimentally that, when acting as a catalyst, the number of porphyrin molecules should not be high because there should be enough free space for the reactants to come into contact with the catalytic sites to lower the activation energy of the reaction, and to increase the conversion. The size of phthalocyanine is larger than that of porphine, and there is still dispute on whether phthalocyanine is able to enter the pores of microporous crystals. Nevertheless, it is possible for faujasite, which has an inner cage diameter of 1.3 nm, to accommodate phthalocyanine molecules, and there have been occasional reports in the literature on the incorporation of phthalocyanine–metal complexes into faujasite cages.[166] Paez-Mozo et al. investigated the incorporation of cobalt–phthalocyanine complex into zeolite Y and the physicochemical properties of the composite in detail.[167] However, they found that besides the cobalt–phthalocyanine complex, unremovable impurity compounds, which are tightly bound to the acidic sites of the zeolite, are also present in the zeolite cages. Balkus, Jr. et al. also prepared and characterized the complexes of CoII and CuII with hexadecafluorophthalocyanine (F16Pc) incorporated in zeolites.[168] They used two incorporation methods. One is the commonly used ion-exchange-reaction approach whereas the other is the direct synthesis of zeolites in the presence of the complexes. Middle IR and UV-visible spectroscopy and powder X-ray diffraction in combination

644

Chemistry of Zeolites and Related Porous Materials

with elemental analysis indicate that the MF16Pc complex is located inside the zeolite cages. Electrochemical analysis also shows CoII/CoI and CuII/CuI redox signals, which are not observable in solution. Metal–phthalocyanine compounds are large in size so that the incorporation of these molecules into microporous molecular sieves is limited to a certain degree. The appearance of mesoporous molecular sieves laid the foundation for incorporation of metal–phthalocyanine complexes into molecular sieves. The incorporation of metal– phthalocyanine compounds into mesoporous molecular sieves may be realized through two pathways. One is the addition of the metal–phthalocyanine complex into the synthetic system of the mesoporous material, the formation of which directly occludes the metal–phthalocyanine with the surfactant template in the mesoporous channels; another route is to introduce the metal–phthalocyanine into the mesoporous channels through impregnation of the metal–phthalocyanine solution with detemplated mesoporous material. Depending on the incorporation technique and conditions, the existing state and dispersion degree of metal–phthalocyanine in mesoporous channels vary to a certain extent.[169–171] Zinc–phthalocyanine (ZnPc) is dispersed in the template medium of mesoporous channels as single molecules after incorporation into MCM-41 through the direct synthesis approach, because its absorption spectrum is very similar to that of the complex dissolved in dimethylformamide (DMF) solution. Through impregnation, the incorporation of ZnPc into mesoporous channels exists not only as single molecules but also as molecular dimers. 9.5.4

Incorporation of Other Metal Complexes

Through solution- and solid-state ion-exchange, it is possible to exchange Cu2þ into mesoporous molecular sieve MCM-41 to form Cu-MCM-41 composite.[172] Po¨ppl et al. used ESR and electron spin-echo modulation techniques to investigate the coordination state of Cu2þ in Cu-MCM-41 and its interactions with adsorbates D2O and NH3. The research results indicate that locations of the copper ions in samples obtained through solution exchange differ to a great extent from those in samples obtained through solid exchange. Ethylenediamine is a common aliphatic chain chelate, and it forms very stable chelating compounds with a variety of transition metals. Howe and Lunsford incorporated CoII-ethylenediamine complex into zeolites X and Y to form composite compounds that can adsorb oxygen.[173] In both zeolite cages the oxygen adduct [CoII])(en)2O2]2þ may be formed, and this complex adduct is stable up to 70  C in the presence of oxygen. The ESR parameters of the adduct are similar to those for the adduct in solution. Nonaromatic N-containing heterocyclic compounds have attracted much interest recently as ligands. The aromaticity of these macrocyclic compounds is weaker than aromatic heterocyclic compounds so that their configurations can be varied greatly upon coordination, and their coordination modes are also versatile. 1,4,7-Triazacyclononane is a typical nonaromatic N-containing heterocyclic compound. Its three N-atoms may participate in coordination to metal ions. Bein and coworkers incorporated [Mn(tmtacn)]2þ, the MnII complex of 1,4,7-trimethyl-1,4,7-triazacyclononane (tmtacn) into zeolite Y supercages,[174] and characterized the existing state of the complex by ESR spectroscopy. They found that the incorporated complex is suitable for use as a catalyst for epoxidation reaction with H2O2 as the oxidant.

Porous Host–Guest Advanced Materials

645

Dioxotetramine macrocyclic compounds are another type of extensively investigated N-containing nonaromatic macrocyclic ligands. The complexes formed by these ligands with metal ions exhibit various unique properties. Taking advantage of the large channel size of MCM-41, the CuII complexes (140Cu and 14T2Cu) formed by the dioxotetramine macrocyclic ligand 1,4,8,11-tetraazacyclotetradecane-5,7-dione (abbreviated 140) and the substituted dioxotetramine macrocyclic ligand 1,11-bis(2-thienylmethyl)-1,4,8,11tetraazacyclotetradecane-5,7-dione (abbreviated 14T2) may be loaded into the channels of the silica form of MCM-41.[175] Diffuse reflectance UV-visible and ESR spectroscopy indicate that, after loading, the absorption peak of 140Cu remains unchanged whereas that of 14T2Cu blue-shifts by 19 nm, suggesting that the interaction of 14T2Cu with MCM-41 is stronger than that of 140Cu. After incorporation, the ESR spectra for both 140Cu and 14T2Cu show anisotropy. Cobaltocene cation (Cp2Coþ) is rather rigid, and it is stable even under hydrothermal conditions. Using cobaltocene cation as a template, the zeolites nonasil and ZSM-51 (NON) can be synthesized. The size of cobaltocene matches that of the cage of NON’s structure,[176,177] therefore this template is tightly encapsulated by the zeolite framework of NON. Cobaltocene can also act as a template for the synthesis of AlPO4-16 and AlPO4-5.[178] Through using methylated cobaltocene cation (Cp*2Coþ) as a template, UTD-1, a totally new structure zeolite, has been synthesized. This is also the first highsilica zeolite with 14-membered rings.[179] The methylcobaltocene cations in UTD-1 can be removed through washing. Honma et al. reported the synthesis of mesoporous M41S silica using a ferrocene–quaternary ammonium derivative [ferrocenyl-(CH2)11Nþ(CH3)3] as template.[180] Enzymes in biological system is constructed from protein, and many enzymes contain transition metals. These polypeptide-chain-bound or coordinated metal ions play unique roles in catalysis. Therefore, there have been continual attempts to synthesize metal– amino acid complexes to mimic natural metal-containing enzymes. Weckhuysen et al. incorporated copper(II) histidine complex into zeolite Y and they found that the incorporated complex exhibited excellent catalytic performance for oxidation.[181] Differing from the commonly used ion-exchange followed by coordination method, the technique they adopted was to synthesize the Cu(His)22þ complex first and then directly ionexchange it into NaY zeolite. Through ESR analysis, they found that the amino N, carboxylate O, and imidazole ring N of one histidine participate in coordination to the CuII ion whereas only the amino N and the carboxylate O of another histidine coordinate to the metal. The sixth coordination site of the CuII ion may accept an extra ligand. In catalytic oxidation reactions, this sixth coordination site may activate the oxidant. It has been revealed that using tert-butyl peroxide as the oxidant, the conversions of 1-pentanol, benzyl alcohol, and cyclohexene are 12, 56, and 28%, respectively, in the presence of the composite as a catalyst. The main products are pentanoic acid, phenylacetylaldehyde, and 1,2-cyclohexanediol, respectively, and the selectivity is rather high. Therefore, the [Cu(His)2]2þ-Y composite can be regarded as an effective enzyme-mimicking compound. The structures and compositions of microporous crystals are becoming more versatile, and this has laid foundations for incorporation of complexes with various structures and functionalities in microporous crystals. The previously reported complex incorporation was mainly limited to the use of zeolites as the hosts. The aluminophosphate microporous

646

Chemistry of Zeolites and Related Porous Materials

crystal family discovered in the 1980s possesses versatile structures as well, but the incorporation of complexes in aluminophosphate hosts has been rarely reported. It is not surprising that complex incorporation in microporous aluminophosphates will extend the porous host–complex guest assembly chemistry. Mesoporous molecular sieves M41S (including MCM-41 and MCM-48) have the advantages of large pore size (>1.5 nm) and of being capable of accommodating large molecules. Complex molecules with a large size may enter or be loaded into the channels or cages of M41S mesoporous molecular sieves to form composite materials with special functionalities [150] such as high catalytic performance. The channel size of mesoporous molecular sieves is large, and after introduction of complex molecules there is still enough room for guest molecules to pass through, and, as a result, diffusion in these materials may not be affected when they are used as catalysts. Therefore, it is envisioned that mesoporous molecular sieves can be widely used to accommodate complex molecules to form high-performance catalysts. Evans et al. reported the grafting of aminosilane onto the walls of mesoporous silica,[182] and it was found that, after grafting, the amino groups of the aminosilane exhibit strong coordination ability, and they may coordinate to many metal ions such as Mn2þ, Cu2þ, Co2þ, and Zn2þ to form complexes. Evans et al. investigated the physical-chemical properties of the complex/mesoporous silica host–guest composite materials and their use as catalysts for the oxidation of aromatic amines. It was discovered that the manganesecontaining host–guest material showed the highest catalytic activity, and the activity of the copper-containing compound is the second highest, followed by those of the cobaltand zinc-containing compounds. There is an apparent induction period for the reaction involving the latter two compounds as catalysts. It is also of significance to incorporate complex molecules into microporous crystals to form photochemically or photophysically active centers. Because of the separation by the host framework, the complexes located in the channels or cages of microporous crystals are isolated. If the isolated centers with oxidation or reduction features are loaded in the connected and adjacent cages of a microporous crystal, redox pairs may be formed. Electron transfer may occur on these redox pairs under the excitation of light, and therefore photochemical reactions may proceed effectively. This is important for the utilization of solar energy. In addition, this type of assembly system may also be used to simulate the electron transfer process of oxidation– reduction in biological systems. Apart from forming catalysts for photochemical reactions, some rare-earth ion complexes may also form efficient luminescent materials after incorporation into microporous crystals. Alvaro et al. loaded a europium complex into zeolite Y,[183] mordenite, and ZSM-5. Because of their confinement in the zeolite framework, the chance for the luminescent centers to decay nonradiationally is reduced, and as a result the lifetime is increased in comparison with that in solution. In the meantime, upon formation of the complex, the luminescence intensity of Eu3þ ion is distinctly increased. Therefore, it is possible to prepare valuable composite luminescent materials using microporous crystals as hosts and complexes as guests. The frameworks of inorganic porous materials are usually rigid, and the framework atoms do not react with guest molecules under normal conditions. Therefore, the channels and cages of porous materials may be ideal templates for the formation of complexes that have shapes and sizes matching the channels and cages. Some complex

Porous Host–Guest Advanced Materials

647

molecules that are not obtainable from solution or are difficult to purify may be synthesized using porous substrates as templates. The products may be isolated through removal of the porous substrate walls.

9.6 Metal–Organic Porous Coordination Polymers Metal–organic framework compounds are a class of new porous materials which have attracted academic interest during the past decade. These materials are composed of metal ions and organic ligands bridging the metal ions, and they are also called metal– organic framework (MOF) coordination polymers.[184,185] Here we should distinguish between metal–organic compounds from metallorganic (or organometallic) compounds. The former contain no M

C bonds usually, whereas the latter require the presence of M

C bonds. Framework metal–organic coordination polymers usually contain various channels, and these channels differ from those of zeolites in shape, size, and adsorption properties. It is worthwhile to point out that through design of organic ligands with functional groups it is possible to obtain MOF molecular sieves with functionalized channel walls. Many as-synthesized metal–organic molecular sieves accommodate guest molecules in their channels or pores, and these guest molecules may be the solvent of the synthetic system or templates used as for the synthesis of zeolites. These guest molecules may be removed through heating or evacuation, but in some cases the removal of the guest species may lead to collapse of the framework. The thermal stability of metal– organic molecular sieves is lower than that of inorganic framework porous materials, and therefore their application in conventional high-temperature catalysis is limited. However, in nonconventional fields the applications of these compounds have been showing great potential. 9.6.1

Transition Metal–Multicarboxylate Coordination Polymers

Multidentate ligands formed through attachment of two or more carboxylate groups onto a benzene or a larger aromatic ring may easily connect metal ions into three-dimensional coordination polymers. In a mixed-solvent system of water and ethanol, Chui et al. synthesized a 3-D framework porous compound [Cu3(TMA)2 (H2O)3]n from a semiaqueous solvothermal system using benzene-1,3,5-tricarboxylic acid (TMA, sometimes also abbreviated as BTC) and copper ions as reactants.[186] This compound contains [Cu2(O2CR)4] (R stands for aromatic ring) structural units which are interconnected to form a 3-D channel system with channel diameter of about 1 nm (Figure 9.27). The assynthesized compound contains guest water molecules in the channels, but these molecules may be removed through thermal treatment, or may be replaced by other guest molecules such as pyridine. This porous coordination polymer is thermally stable up to about 240  C. TMA can coordinate to not only Cu2þ to form the typical 3-D framework polymer but also to Co2þ to form a porous host–guest compound Co(TMA)(py)n with a different structure. The guest pyridine molecules in this compound may be removed from the channels without the collapse of the framework, and the pyridine-removed porous material adsorbs other guest molecules.[187] Polymerization of TMA with a Ni2þ macrocyclic complex leads to a porous coordination polymer with a unique structure. In the

648

Chemistry of Zeolites and Related Porous Materials

Image Not Available In The Electronic Edition

structure of this compound the macrocyclic molecule remains, and this macromolecule also participates in coordination polymerization.[188] The guest molecules located in the channels of the porous coordination polymer are water and/or pyridine. 1,3,5,7-Adamantanetetracarboxylic acid (denoted ATC) has four carboxylate groups and it is also an ideal multidentate ligand. The reaction of this ligand with Cu2þ under hydrothermal conditions (190  C) results in a coordination polymer Cu2(ATC) 6H2O. This compound possesses large channels, and the guest molecules may be driven out without apparent change of the host framework. The compound exhibits excellent microporous adsorption properties.[189] The reaction of 4,40 ,400 -benzene-1,3,5-triyltribenzoic acid (denoted BTB) and copper(II) nitrate in a mixed solvent of ethanol, DMF, and water at 65  C for 1 d gives rise to an interwoven coordination polymer Cu3(BTB)2(H2O)3  (DMF)9(H2O)2.[190] This compound possesses channels with a diameter of 1.6 nm (Figure 9.28), and the channels contain guest DMF and water molecules. Upon removal of the guest molecules, the compound exhibits excellent adsorption properties. 9.6.2

Coordination Polymers with N-containing Multidentate Aromatic Ligands

The other common donor coordination atom besides O atom is N. Many aromatic rings may contain one or more N-heteroatoms. These aromatic rings may interconnect with one another to form larger ligands, and more than one N-atom in the compound may participate in coordination to metal ions. 4,40 -Bipyridine (bpy) is an N-containing bidentate ligand widely used to form coordination polymers. Noro et al. used bpy as a

Porous Host–Guest Advanced Materials

649

Image Not Available In The Electronic Edition

ligand and synthesized a series of framework compounds containing guest anions and water molecules in the reaction system of copper ions and AF6-type anions (A ¼ Si, Ge, and P). The framework structures of these compounds may be controlled through varying the guest anion.[191] The framework geometry of porous 3-D framework compounds formed by some Ncontaining ligands and metals is affected by the guest species in the channels to a great degree. Biradha and Fujita used 2,4,6-tris(4-pyridyl)triazine (denoted TPT) as the ligand, and synthesized the host–guest compounds [(ZnI2)3(TPT)2  5.5C6H5NO2] and [(ZnI2)3(TPT)2 5.5C6H5CN] from a solution of ZnI2 and nitrobenzene or cyanobenzene.[192] They discovered that in the presence of the guest nitrobenzene or cyanobenzene molecules, the whole framework of the host is swollen, whereas when the guest molecules are removed, the host framework is apparently shrunken. The guest species in this compound may also be replaced by other molecules. In some cases, the magnetic properties of transition metal porous coordination polymers depend on the guest species occluded in the framework channels of the compounds. The framework of the porous coordination polymer Fe2(azpy)4(NCS)4 (guest) formed from E4,40 -azobipyridine interconnection with Fe2þ ions is rather flexible. With variation of the guest species, the geometry of this framework is altered greatly so that high spin–low spin flip occurs for the Fe2þ metal centers, and consequently the macromagnetic properties of the compound also undergo flipping.[193] Lin and coworkers used an axially chiral bridging ligand (R)-6,60 -dichloro-2,20 -dihydroxy-1,10 -binaphthyl-4,40 -bipyridine, L, to construct homochiral porous MOFs.[194] They obtained crystals of [Cd3Cl6L3]4DMF6MeOH3H2O by slow diffusion of diethyl ether into a mixture of the ligand and CdCl2 in MeOH/DMF. The structure of this compound is a noninterpenetrating 3-D network with very large chiral channels of 1.6  1.8 nm crosssection. It was revealed that the compound contains 54.4% void space that is accessible to guest molecules, and X-ray powder diffraction demonstrated that the framework structure was maintained upon removal of all the solvent molecules. Interestingly, Ti-(OPr)i4 can react with the chiral dihydroxy groups in the channels of the compound to afford a Lewis acidic material that is catalytically active for the addition of ZnEt2 to aromatic aldehydes to form chiral secondary alcohols.

650

9.6.3

Chemistry of Zeolites and Related Porous Materials

Coordination Polymers with N- and O-containing Multidentate Ligands

Some aromatic cyclic molecules not only contain an N-heteroatom on the aromatic ring but also have carboxylate group(s) on one or more of the carbon atoms. Therefore, these ligands can form coordination polymers through coordination of not only their N but also their carboxylate groups. Zhao et al. used pyridine-2,6-dicarboxylic acid (H2dipc) as ligand and synthesized a coordination polymer with an empirical formula [Ln(dipc)3Mn1.5(H2O)nH2O] from Mn2þ and rare earth metal ion Ln3þ (Ln ¼ Pr, Gd, Er) system under hydrothermal conditions.[195] This polymeric compound possesses one-dimensional channels with a channel diameter of 0.6 nm, and the guest water molecules are distributed in the channels. Structural analysis indicates that these guest water molecules may be removed from the channels without collapse of the framework, and therefore this polymer may be used as a microporous crystal molecular sieve. Because there are unpaired electrons on the framework metal ions, the compound also exhibits magnetic properties. It is common for open-framework coordination polymers formed from transition metal ions and ligands to exhibit magnetic properties. The magnetic moments on adjacent manganese ions of the 3-D framework porous compound Mn(PDB)H2O (PDB ¼ pyrpyridine-3,4-dicarboxylic acid) formed by coordination of Mn2þ ions with PDB show antiferromagnetic interactions at low temperatures.[196] The framework channels of this compound are occupied by water molecules that also coordinate to the Mn2þ ions. Su et al. used a tripodal ligand [N0 -(carboxymethyl)benzimidazol-2-ylmethyl]bis(benzimidazol-2-ylmethyl)amine (HAcntb) and Cu2þ and synthesized [Cu6(Acntb)6](ClO4)6 nH2O porous coordination polymer from an ethanolic solution.[197] This compound possesses channels with a diameter of about 1.1 nm, and in the channels there are occluded guest water molecules and perchlorate ions. Depending on synthetic conditions, the guest water molecules vary in number. Chiral porous coordination polymers have great application potential because they may be used as chiral catalysts, and there have been continual attempts to introduce chirality into coordination polymers with various approaches. A simple strategy is to use a chiral ligand, and after the formation of coordination polymers the chirality is automatically introduced into the polymer with the ligand. Quitenine [(8S,9R)-9-hydroxy-60 -Methoxy cinchonan-3-carboxylic acid (HQA)] is a chiral ligand molecule. If quitenine is treated with Cd(OH)2 in a hydrothermal system containing racemic 2-butanol, a host–guest coordination polymer Cd(QA)2 with chiral channels crystallizes.[198] In the chiral channels of this compound there exist chiral butanol guest molecules, suggesting that the chiral channels are able to separate the guest enantiomers. 9,9-Diethyl-2,7-bis(4-pyridylethynyl)fluorene, and chiral 9,9-bis[(S)-2-methylbutyl]-2,7bis(4-pyridylethynyl)fluorene were also used to react with copper nitrate to prepare coordination polymers with grid channels. If the ligands are chiral, the obtained corresponding coordination polymers are also chiral.[199] Another method to prepare chiral coordination polymers is to use special guest molecules as templates that may also induce the formation of chiral coordination polymers. Kepert et al. used ethylene glycol and propylene glycol as templates to induce the formation of metal (Co, Ni) benzenetricarboxylate porous coordination polymers, the framework of which exhibits chirality.[200] Through mixing of an N-containing ligand and an O-containing one in the same reaction system, porous coordination polymers with two or more ligands may be

Porous Host–Guest Advanced Materials

651

obtained. Under there circumstances, each ligand contains either N or O, but the whole coordination polymer contains both N and O. Through solvothermal reaction (150  C) of a 1-D chain coordination polymer [Co(bpdc)(H2O)2] H2O (bpdc ¼ 2,20 -bipyridyl-3,30 dicarboxylic acid) in DMF solution, the chain structure is converted into a porous 3-D framework compound [Co3(bpdc)3bpy]  4DMF H2O.[201] The solvent water and DMF molecules are accommodated in the channels as guests. It is interesting that the conversion from 1-D polymer to 3-D framework is reversible. The channels of 3-D framework compounds are shape selective, and they may act as ideal sites for shapeselective catalysis. After reactions, the products may be isolated through reversible conversion of the host framework into the 1-D framework. Kitaura et al. obtained a new lamellar compound [Cu(dhbc)2(4,40 -bpy)]H2O from a diethyl ether solution of copper nitrate, 4,40 -bipyridine, and 2,5-dihydroxybenzoic acid (dhbc).[202] There are strong p–p interactions between the layers of the compound, and because of these interactions the compound behaves as a 3-D framework structure with channels containing guest water molecules that are removable reversibly. After dehydration, the host material exhibits adsorption properties, but its adsorption capacity for nitrogen is affected by the gas pressure. 9.6.4 2þ

Zinc-containing Porous Coordination Polymers

Zn ion has strong affinity toward O- and N-containing ligands. Therefore, it is very easy for Zn2þ to form coordination polymers with O- and N-containing multidentate ligands. Many of the previously reported porous framework coordination polymers contain Zn2þ as the central atom. Yaghi and coworkers synthesized a series of porous framework coordination polymers MOF-n (MOF ¼ metal–organic framework) from solvothermal or semi-aqueous solvothermal systems using multicarboxylate compounds as ligands and Zn2þ as the central metal ion. The desorption and adsorption properties of the assynthesized compounds have also been elucidated.[203,204] These coordination polymers formed from multicarboxylates and zinc are thermally stable up to about 300  C, and consequently they are potentially applicable in many areas. In some cases, Zn2þ may be bridged by O to form ZnxOy clusters that are interconnected by organic ligands to form coordination polymers. Through using Zn4O as a structural unit, Yaghi and coworkers prepared a series of coordination polymers with various channel sizes. In the Zn4O unit, the Zn is located at the corner of the vertex of a tetrahedron, whereas the center of the tetrahedron is occupied by the O atom. After formation of coordination polymers, the Zn on the vertex of the tetrahedron is connected with adjacent Zn4O clusters through ligands to form 3-D interwoven framework structures. The size of the channels or cages of the framework may be controlled through choosing various lengths of ligands (Figure 9.29). Further investigation reveals that these compounds exhibit H2-storage capacities at low (78 K) or room temperature and at a certain pressure (20 bar).[205] Some polymer compounds may also show methane-storage capacity.[206] Through coordination of O-bridged trinuclear Zn2þ clusters with chiral ligands, chiral porous coordination polymers may be obtained.[207] Because the two enantiomers of the ligand may be isolated separately, the obtained porous coordination polymers may also be present as the separate enantiomers. The guest water molecules in the polymer channels may be reversibly adsorbed and desorbed, and this lays the foundation for these

652

Chemistry of Zeolites and Related Porous Materials

Image Not Available In The Electronic Edition

compounds to be used as molecular sieves. Further investigation indicates that these enantiomerically pure chiral porous metal–organic molecular sieves are able to separate enantiomers and are catalytically active. Zinc ion and its O-bridged clusters may also be coordinated by two or more mixed ligands to form various coordination polymers. Chen et al. synthesized a new porous framework complex Zn3(BDC)(BTC)2 2NH(CH3)2 2NH2(CH3)2 from a mixed ligand system containing 1,4-benzenedicarboxylic acid (BDC) and 1,3,5-benzenetricarboxylic acid (BTC) through a semi-aqueous solvothermal reaction approach.[208] The compound contains dimethylamine guest molecules. Sun et al. reported the liquid–liquid diffusion and solvothermal synthesis of two coordination polymers Zn(ISN)2 2H2O and InH(BDC)2 (ISN ¼ isonicotinic acid, BDC ¼ benzenedicarboxylic acid), the topological structure of which is analogous to that of quartz.[209] In these two compounds there exist spiral chiral channels. 9.6.5

Adsorption Properties and H2 Storage of MOFs

The as-synthesized MOF compounds usually contain guest species such as water and other solvent molecules in their pores and/or channels. Upon thermal treatment, these guest molecules may be driven out off the framework structures of the MOFs, and in some cases the framework structures of the thermally treated MOFs are maintained. The stable MOFs after removal of guest species are porous and may adsorb a variety of other molecules that are smaller than the pore size of the MOFs. H2 is a clean energy source, and it is very promising for H2 to be used in fuel cell operations. However, the storage of H2 in a high capacity (both in weight and in volume) is still rather challenging. Recently, the use of MOFs as H2-storage media has been extensively investigated, and it has been found that some MOFs exhibit significant H2-storage capacities. So far as is known, most of the reported MOFs with H2-storage capacities are composed of ZnII or CuII and organic linkers.[210] After the successful synthesis, Yaghi and coworkers investigated, in detail, the adsorption properties of their MOF compounds composed of ZnII and multicarboxylate

Porous Host–Guest Advanced Materials

653

Table 9.1 Sorption capacity for MOF-n (n ¼ 2
5) [203]

MOF-n MOF-2

MOF-3

MOF-4 MOF-5

Guest N2 CO2 CH2Cl2 CHCl3 C6H6 C6H12 N2 Ar CO2 CH2Cl2 CHCl3 CCl4 C6H6 C6H12 C2H5OH N2 Ar CH2Cl2 CHCl3 CCl4 C6H6 C6H12

Guest/formula unit 0.63 0.70 0.82 0.64 0.56 0.47 0.74 0.71 0.95 0.69 0.56 0.45 0.50 0.41 4.60 22.87 28.75 11.00 8.86 7.38 7.88 6.38

Guest/unit cell 2.52 2.80 3.28 2.56 2.24 1.88 0.74 0.71 0.95 0.69 0.56 0.45 0.50 0.41 18 183 230 88 71 59 63 51

Pore cross-section (A˚)

Apparent surface area (m2/g)

Pore volume (cm3/g)

7

270

8

140

14 12

2900

0.094 0.086 0.229 0.222 0.220 0.221 0.038 0.036 0.038 0.065 0.065 0.065 0.064 0.065 0.612 1.04 1.03 0.93 0.94 0.94 0.94 0.92

linkers.[203] Table 9.1 lists the adsorption capacities measured by Yaghi and colleagues for MOF-n (n ¼ 2–5). Type-I isotherms typical for zeolites and related microporous materials have been observed for MOF-2, -3, and -5. After evacuation, MOF-5 possesses a free pore volume of 55–60% of its crystal. Metal coordinative unsaturation is present in the channels of MOF-4, and, as a result, a more complex sorption processes may occur as observed for ethanol uptake in MOF-4. Yaghi and coworkers are among the research groups who first described H2 storage by MOFs, and because of H2-uptake properties, the research in MOFs has been further intensified recently. In their Science paper, Yaghi and coworkers reported their discovery that metal–organic framework-5 (MOF-5) adsorbed hydrogen up to 4.5 wt% at 78 K and 1.0 wt% at room temperature and a pressure of 20 bar.[205] Inelastic neutron scattering spectroscopy indicates the presence of two binding sites associated with hydrogen binding to zinc and the BDC linker, respectively. It is also found that the topologically similar isoreticular IRMOF-6 and -8 with cyclobutylbenzene and naphthalene linkers, respectively, exhibit approximately double and quadruple the adsorption capacity for MOF-5 at room temperature and 10 bar. Following this first report in H2 adsorption by MOFs, Yaghi and coworkers have also tested the H2-uptake properties of other MOF compounds, among which is MOF-505 with an as-synthesized composition of [Cu2(bptc)(H2O)2(dmf)3(H2O)] where bptc stands for 3,30 ,5,50 -biphenyltetracarboxylate. Upon removal of the guest molecules, this compound takes up 2.47 wt% H2 at 77 K and 760 Torr.

654

Chemistry of Zeolites and Related Porous Materials

A chiral zinc–organic framework compound [Zn2(L)]  4H2O has been synthesized through hydrothermal reaction of ZnCl2 and 4,40 -bipyridine-2,20 ,6,60 -tetracarboxylic acid (H4L).[211] Structural analysis indicates that the framework of the compound is a 5connected network with a 4466 topology comprising ZnII bound to L4
anion. The chirality is generated by the helical chains of hydrogen-bonded guest water molecules in the channels of the compound, and removal of these guest water molecules from the crystal leads to a porous compound Zn2L that is thermally stable and chemically inert. The BET surface area of the guest-removed Zn2L is 312.7 m2/g and the compound also exhibits significant gas-storage capacities for H2 (1.08 wt % at 4 bar and 77 K) and for methane (3.14 wt % at 9 bar and 298 K). The adsorption behavior of Zn2L toward other organic solvent vapors such as benzene, chloroform, toluene, etc. has also been investigated, and it is revealed that the adsorption is dominated by absorbate–absorbate or absorbate–absorbent interactions. Yaghi and coworkers investigated the H2 saturation uptake of a series MOF compounds including HKUST-1 composed of CuII and benzene-1,3,5-tricarboxylate, the isoreticular (IRMOF) IRMOF-1, -6, -11, and -20, MOF-177, and -74, all of which are constructed from ZnII and a organic linker.[212] It is found that, at low pressures, the H2 uptake is not proportional to the surface area of the MOF compounds, whereas at high pressures the saturation H2-adsorption capacity is linearly correlated with the surface area of the MOFs. For MOF-74, the uptake at saturation (26 bar) is 2.3 wt% and 3.5 wt% for IRMOF-11 (34 bar), whereas for MOF-177 and IRMOF-20, saturation is reached between approximately 70 and 80 bar, and the corresponding H2 uptakes are 7.5 and 6.7 wt%, respectively. Results are displayed in Figure 9.30. Garberoglio et al. have modeled adsorption of light gases including H2 on a number of MOF materials using molecular simulations.[213] Good agreement between simulations and experiments is observed for some cases but very poor agreement also exists in other cases. Their calculations indicate that at room temperature none of the tested materials is

Figure 9.30 Plot of saturation H2 uptake versus Langmuir surface area. Reproduced with permission from [212]. Copyright (2006) American Chemical Society

Porous Host–Guest Advanced Materials

655

able to store significant amounts of hydrogen for use in fuel cell vehicles. Nevertheless, IRMOF-14 exhibits very high H2 uptake at 77 K, and the total uptake of this material may reach 15 wt%. Yang and Zhong also performed Monte Carlo simulation and density functional theory calculation on adsorption of hydrogen in MOF-505 to provide insight into molecular-level details of the underlying mechanisms.[214] Their calculation results show that metal oxygen clusters are preferential adsorption sites for hydrogen, and the strongest adsorption of hydrogen is found in the directions of coordinatively unsaturated open metal sites. The H2-storage capacity of MOF-505 at room temperature and moderate pressures is predicted to be low. Zhou and coworkers reported an interesting interweaving MOF,[215] Cu3(TATB)2(H2O)3 (TATB stands for 4,40 ,400 -s-triazine-2,4,6-triyltribenzoate). Despite the interweaving feature of its framework, the MOF compound exhibits N2- and H2-adsorption capacities after removal of the guest water molecules in the channels. Although the surface area (3800 m2/g) of this MOF material is lower than those of IRMOF-20 and MOF-177 (4346 and 4526 m2/g, respectively); the compound obtained by Zhou’s group has greater hydrogen-adsorption capacity (1.9 wt% vs 1.356 and 1.25 wt%, respectively for the latter two at 77 K and 760 Torr). Hydrogen uptake has also been compared with another two interpenetrating MOFs, IRMOF-9, and IRMOF-13.6. Cu3(TATB)2 possesses a higher hydrogen uptake than either (1.17 and 1.73 wt%, respectively). The authors attribute this to the presence of accessible unsaturated metal centers and the existence of pores and channels in a size range well suited to the dihydrogen molecule.

References [1] G.D. Stucky and J.E. MacDougall, Quantum Confinement and Host/Guest Chemistry: Probing a New Dimension. Science, 1990, 247, 669–678. [2] G.A. Ozin, Nanochemistry - Synthesis in Diminishing Dimensions. Adv. Mater., 1992, 4, 612–649. [3] P. Gallezot, Preparation of Metal Clusters in Zeolites. In: Molecular Sieves, Vol. 3., SpringerVerlag, Berlin, Heidelberg, 2002, 257–305. [4] P.P. Anderson, Ionic Clusters in Zeolites. In: Molecular Sieves. Vol. 3. Springer-Verlag, Berlin, Heidelberg, 2002, 307–338. [5] P.J. Kasai, Electron Spin Resonance Studies of g- and X-ray-irradiated Zeolites. J. Chem. Phys., 1965, 43, 3322–3327. [6] M.R. Harrison, P.P. Edwards, J. Klinowski, J.M. Thomas, D.C. Johnson, and C.J. Page, Ionic and Metallic Clusters of the Alkali Metals in Zeolite Y. J. Solid State Chem., 1984, 54, 330–341. [7] P.P. Edwards, P.A. Anderson, and J.M. Thomas, Dissolved Alkali Metals in Zeolites. Acc. Chem. Res., 1996, 29, 23–29. [8] D.C. Johnson, J. Klinowski, C.J. Page, M.R. Harrison, P.P. Edwards, M.J. Sienko, and J.M. Thomas, Ionic and Metallic Clusters in Zeolites. J. Chem. Soc., Chem. Commun., 1984, 982–984. [9] W.G. Hodgson, J.S. Brinen, and E.F. Williams, Electron Spin Resonance Investigation of Photochromic Sodalites. J. Chem. Phys., 1967, 47, 3719–3723. [10] R.M. Barrer and J.M. Cole, Interaction of Sodium Vapour with Synthetic Sodalite: Sorption and Formation of Colour Centers. J.Phys. Chem. Solids, 1968, 29, 1755–1758.

656

Chemistry of Zeolites and Related Porous Materials

[11] N.P. Blake V.I. Srdanov, G.D. Stucky, and H. Metiu., An investigation of the Electronic and Optical Properties of Dehydrated Sodalite fully doped with Na. J. Chem. Phys., 1996, 104, 8721–8729. [12] J.A. Rabo, C.L. Angell, P.H. Kasai, and V. Schomaker, Studies of Cations in Zeolites: Adsorption of Carbon Monoxide; Formation of Ni Ions and Na43þ Centers. Discuss. Faraday Soc., 1966, 41, 328–349. [13] T.Y. Ben, C. Naccache, M. Che, and A.J. Tench, ESR Studies of 17O
2 and SO
2 Generated from Na3þ4 Centres in NaY Zeolites. Chem. Phys. Lett., 1974, 24, 41–44. [14] U. Westphal and G. Geismar, Alkalimetall-cluster im Zeolith Y Darstellung, Eigenschaften, Reaktionen. Z. Anorg. Allg. Chem., 1984, 508, 165–175. [15] L.R.M. Martens, P.J. Grobert, and P.A. Jacobs, Preparation and Catalytic Properties of Ionic Sodium Clusters in Zeolites. Nature (London), 1985, 315, 568–570. [16] I. Hannus, I. Kirisci, A. Be´res, J.B. Nagy, and H. Forster, Characterization of Alkali Metal Cluster Containing Faujasites by Thermal, IR, ESR, Multi-NMR and Test Reaction Studies. Stud. Surf. Sci. Catal., 1995, 98, 81–88. [17] M.W. Simon, J.C. Edwards, and S.L. Suib, Characterization and Catalytic Studies on Defect Sites Formed upon the Thermal Decomposition of Sodium Ionic Clusters in NaX Zeolite. J. Phys. Chem., 1995, 99, 4698–4709. [18] P.A. Anderson, R.G. Bell, C.R.A. Catlow, F.L. Chang, A.J. Dent, P.P. Edwards, I. Gameson, I. Hussain, A. Porch, and J.M. Thomas, Matrix-bound Nanochemical Possibilities. Chem. Mater., 1996, 8, 2114–2120. [19] P.A. Anderson, A.R. Armstrong, A. Porch, P.P. Edwards, and L.J. Woodall, Structure and Electronic Properties of Potassium-loaded Zeolite L. J.Phys. Chem. B, 1997, 101, 9892– 9900. [20] M.J. Kelly, A Model Electronic Structure for Metal-intercalated Zeolites. J. Phys.: Condens. Matter., 1995, 7, 5507–5519. [21] P.A. Anderson, A.R. Armstrong, and P.P. Edwards, Ionization and Delocalization in Potassium Zeolite L: a Combined Neutron Diffraction and Electron Spin Resonance Study. Angew. Chem., Int. Ed. Engl., 1994, 33, 641–643. [22] Y. Ikemoto, T. Nakano, Y. Nozue, O. Terasaki, and S. Qiu, Optical Studies of Nanoscale Materials Incorporated in the Space of Zeolite Crystals. Mater. Sci. Eng., B, 1997, 48, 116–121. [23] Y. Maniwa, H. Kira, F. Shimizu, and Y. Murakami, Development of Superstructure in Potassium-loaded Zeolite LTA, Kx-K12Si12Al12O48. J. Phys. Soc. Jpn., 1999, 68, 2902–2905. [24] M. Ralek, P. Jiru, O. Grubner, and H. Beyer, Molekularsiebe mit Farbiger Indizierung des Wassergehaltes. Collect. Czech Chem. Commun., 1962, 27, 142–143. [25] Y. Kim and K. Seff, Structure of a Very Small Piece of Silver Metal. The Octahedral Silver (Ag6) Molecule. Two Crystal Structures of Partially Decomposed Vacuum-dehydrated Fully Silver(1þ) Ion-exchanged Zeolite A. J. Am. Chem. Soc., 1977, 99, 7055–7057. [26] P.A. Jacobs, J.B. Uytterhoeven, and H.K. Beyer, Some Unusual Properties of Activated and Reduced AgNaA Zeolites. J. Chem. Soc., Faraday Trans. 1, 1979, 75, 56-64. [27] T. Sun and K. Seff, Silver Clusters and Chemistry in Zeolites. Chem. Rev., 1994, 94, 857–870. [28] B. Xu and L. Kevan, Formation of Silver Ionic Clusters and Silver Metal Particles in Zeolite Rho Studied by Electron Spin Resonance and Far-infrared Spectroscopies. J. Phys. Chem., 1991, 95, 1147–1151. [29] N. Narayana and, L. Kevan, Characterization of Silver Species: Locations and Environments in A, X, and Y Zeolites by Electron Spin Echo Spectroscopy. J. Chem. Phys., 1982, 76, 3999–4005.

Porous Host–Guest Advanced Materials

657

[30] G.A. Ozin, J. Godber, and A. Stein, Very High Density Optical Storage of Data in Silver Sodalites. US Patent 1990 4, 942, 119. [31] J. de Graaf, A.J. van Dillen, K.P. de Jong, and D.C. Koningsberger, Preparation of Highly Dispersed Pt Particles in Zeolite Y with a Narrow Particle Size Distribution: Characterization by Hydrogen Chemisorption, TEM, EXAFS Spectroscopy, and Particle Modeling. J. Catal., 2001, 203, 307–321. [32] P. Gallezot, A. Alarcon-Diaz, J.A. Dalmon, A.J. Renouprez, and B. Imelik, Location and Dispersion of Platinum in PtY Zeolites. J. Catal., 1975, 39, 334–349. [33] I.Y. Chan, R. Csencsits, M.A. O’Keefe, and R. Gronsky, Computer-Simulated Images of Platinum Clusters in the Channels of Y Zeolites: Zone-Axis Results. J. Catal., 1987, 103, 466–473. [34] W.Z. Zhou, J.M. Thomas, D.S. Shephard, B.F.G. Johnson, D. Ozkaya, T. Maschmeyer, R.G. Bell, and Q.F. Ge, Ordering of Ruthenium Cluster Carbonyls in Mesoporous Silica. Science, 1998, 280, 705–708. [35] D.S. Shephard, T. Maschmeyer, B.F.G. Johnson, S.J.M. Thomas, G. Sankar, D. Ozkaya, W. Zhou, R.D. Oldroyd, and R.G. Bell, Bimetallic Nanoparticle Catalysts Anchored inside Mesoporous Silica. Angew. Chem., Int. Ed. Engl., 1997, 36, 2242–2245. [36] L.B. McCusker and K. Seff, Cadmium(I) and dicadmium(I). Crystal Structures of Cadmium(II)-exchanged Zeolite A Evacuated at 500 degree C and of its Cadmium Sorption Complex. J. Am. Chem. Soc., 1979, 101, 5235–5239. [37] S.B. Jang, U.S. Kim, Y. Kim, and K. Seff, Crystal Structures of Fully Dehydrated Cd(II)Exchanged Zeolite A and of its Cadmium Sorption Complex Containing Cd2þ, Cdþ, Cd22þ, and Cd20. J. Phys. Chem., 1994, 98, 3796–3800. [38] A. Goldbach, P.D. Barker, P.A. Anderson, and P.P. Edwards, The Clusters Cd22þ and Cd42þ in Zeolite A. Chem. Phys. Lett., 1998, 292, 137–142. [39] T. Sprang, A. Seidel, M. Wark, F. Rittnera, and B. Boddenbe, Cadmium Ion Exchange in Zeolite Y by Chemical Vapour Deposition and Reaction. J. Mater. Chem., 1997, 7, 1429– 1432. [40] A. Seidel, F. Rittner, B. Boddenberg, Chemical Vapor Deposition of Zinc in Zeolite HY. J. Phys. Chem. 1998, 102, 7176–7812. [41] Y. Tian, G.D. Li, and J.S. Chen, Chemical Formation of Mononuclear Univalent Zinc in a Microporous Crystalline Silicoaluminophosphate. J. Am. Chem. Soc., 2003, 125, 6622–6623. [42] N.H. Heo, H.C. Choi, S.W. Jung, M. Park, and K. Seff, Complete Redox Exchange of Indium for Tlþ in Zeolite A. Crystal Structures of Anhydrous Tl12-A and In10-A  In. Indium Appears as In2þ, Inþ, and In0. The Clusters (In5)8þ and (In3)2þ are Proposed. J. Phys. Chem. B, 1997, 101, 5531–5539. [43] N.H. Heo, S.H. Kim, H.C. Choi, S.W. Jung, and K. Seff, Crystal Structure of Indium-exchanged Zeolite A Containing Sorbed Disulfur. J. Phys. Chem. B, 1998, 102, 17–23. [44] I. Hussain, I. Gameson, P.A. Anderson, M. Slaski, P.P. Edwards, and A. Dyer, A Route to the Dispersion of Ultrafine Cobalt Particles on Zeolite Na-X through Salt Occlusion and Reduction. J. Chem. Soc., Dalton Trans., 1996, 775–781. [45] J.A. Rabo and P.H. Kasai, Caging and Electrolytic Phenomena in Zeolites. Prog. Solid State Chem., 1975, 9, 1–19. [46] J.V. Smith, J.M. Bennett, and E.M. Flanigen, Dehydrated Lanthanum-exchanged Type Y Zeolite. Nature (London), 1967, 215, 241–244. [47] H.S. Park and K. Seff, Crystal Structures of Fully La3þ-exchanged Zeolite X: an Intrazeolitic La2O3 Continuum, Hexagonal Planar and Trigonally Monocapped Trigonal Prismatic Coordination. J. Phys. Chem. B, 2000, 104, 2224–2236.

658

Chemistry of Zeolites and Related Porous Materials

[48] K. Sahl, Refinement of the Crystal Structure of Bicchulite, Ca2(Al2SiO6)(OH)2. Z. Kristallogr., 1980, 152, 13–21. [49] A. Carlsson, T. Oku, J.O. Bovin, G. Karlsson, Y. Okamoto, N. Ohnishi, and O. Terasaki, The Structure of Iron Oxide Implanted Zeolite Y, Determined by High-resolution Electron Microscopy and Refined with Selected Area Electron Diffraction Amplitudes. Chem. Eur. J., 1999, 5, 244–249. [50] J.E. Readman, I. Gameson, J.A. Hriljac, P.P. Edwards, and P.A. Anderson, Synthesis and Structure of Zinc Oxide Clusters Encapsulated in Zeolite LTA. Chem. Commun., 2000, 595–596. [51] G. Schulz-Ekloff, D. Wohrle, B. van Duffel, and R.A. Schoonheydt, Chromophores in Porous Silicas and Minerals: Preparation and Optical Properties. Microporous Mesoporous Mater., 2002, 51, 91–138. [52] G. Calzaferri and N. Gfeller, Thionine in the Cage of Zeolite L. J. Phys. Chem., 1992, 96, 3428–3435. [53] S.D. Cox, T.E. Gier, G.D. Stucky, and J.D. Bierlein, Inclusion Tuning of Non-linear Optical Materials: Switching the SHG of p-Nitroaniline and 2-Methyl-p-nitroaniline with Molecular Sieve Hosts. J. Am. Chem. Soc., 1988, 110, 2986–2987. [54] S. Wohlrab, R. Hoppe, G. Schulz-Ekloff, and D. Wo¨hrle, Encapsulation of Methylene Blue into Aluminophosphate Family Molecular Sieves. Zeolites, 1992, 12, 862– 865. [55] N. Herron, G.D. Stucky, and C.A. Tolman, Shape Selectivity in Hydrocarbon Oxidations using Zeolite Encapsulated Iron Phthalocyanine Catalysts. J. Chem. Soc., Chem. Commun., 1986, 1521–1522. [56] R. Hoppe, G. Schulz-Ekloff, D. Wohrle, C. Kirschhock, H. Fuess, L. Uytterhoeven, and R. Schoonheydt, Incorporation of Methylene Blue in NaY Zeolite at Crystallographically Defined Positions. Adv. Mater., 1995, 7, 61–64. [57] K. Rurack, K. Hoffmann, W. Al-Soufi, and U. Resch-Genger, 2,20 -Bipyridyl-3,30 -diol Incorporated into AlPO4-5 Crystals and its Spectroscopic Properties as Related to Aqueous Liquid Media. J. Phys. Chem. B, 2002, 106, 9744–9752. [58] V. Ganesan and R. Ramaraj, Spectral Properties of Proflavin in Zeolite-L and Zeolite-Y. J. Lumin., 2001, 92, 167–173. [59] U. Vietze, O. Kraub, F. Laeri, G. Ihlein, F. Schu¨th., B. Limburg, and M. Abraham, Zeolitedye Microlasers. Phys. Rev. Lett., 1998, 81, 4628–4631. [60] G. Ihlein, F. Schuth, O. Krauß, U. Vietze, and F. Laeri , Alignment of a Laser Dye in the Channels of the AlPO4-5 Molecular Sieve. Adv. Mater., 1998, 10, 1117–1119. [61] P. Yang, G. Wirnsberger, H.C. Huang, S.R. Cordero, M.D. McGehee, B. Scott, T. Deng, G.M. Whitesides, B.F. Chmelka, S.K. Buratto, and G.D. Stucky, Mirrorless Lasing from Mesostructured Waveguides Patterned by Soft Lithography. Science, 2000, 287, 465–467. [62] G. Wirnsberger and G.D. Stucky, Microring Lasing from Dye-doped Silica/Block Copolymer Nanocomposites. Chem. Mater., 2000, 12, 2525–2527. [63] R. Vogel, P. Meredith, I. Kartini, M. Harvey, J.D. Riches, A. Bishop, N. Heckenberg, M. Trau, and H. Rubinsztein-Dunlop, Mesostructured Dye-doped Titanium Dioxide for Microoptoelectronic Applications. Chem Phys Chem, 2003, 4, 595–603. [64] Y. Wada, T. Okubo, M. Ryo, T. Nakazawa, Y. Hasegawa, and S. Yanagida, High Efficiency Near-IR Emission of Nd(III) Based on Low-vibrational Environment in Cages of Nanosized Zeolites. J. Am. Chem. Soc., 2000, 122, 8583–8584. [65] M. Ganschow, M. Wark, D. Wohrle, and G. Schulz-Ekloff, Anchoring of Functional Dye Molecules in MCM-41 by Microwave-assisted Hydrothermal Cocondensation. Angew. Chem., Int. Ed., 2000, 39, 160–163.

Porous Host–Guest Advanced Materials

659

[66] M.M. Yatskou, M. Meyer, S. Huber, M. Pfenniger, and G. Calzaferri, Electronic Excitation Energy Migration in a Photonic Dye-zeolite Antenna. Chem Phys Chem., 2003, 4, 567–587. [67] M. Pauchard, S. Huber, R. Meallet-Renault, H. Maas, R. Pansu, and G. Calzaferri, Time- and Space-resolved Luminescence of a Photonic Dye-zeolite Antenna. Angew. Chem., Int. Ed., 2001, 40, 2839–2842. [68] G. Calzaferri, Dye Molecules in Zeolite L Nano Crystals for Efficient Light Harvesting. In: Photofunctional Zeolites, et. M. Anpo, NOVA Science Publishers, New York, 2000, 205–218. [69] Y.M. Jin and H. Chon, A Novel Method for Encapsulation of Dyes into AlPO4-8 Molecular Sieve. Chem. Commun., 1996, 135–136. [70] S.D. Cox, T.E. Gier, and G.D. Stucky, Second Harmonic Generation by the Self-aggregation of Organic Guests in Molecular Sieve Hosts. Chem. Mater., 1990, 2, 609–619. [71] W. Hill, F. Marlow, and J. Kornatowski, Raman Spectroscopic Investigation of the Physical State of para-Nitroaniline in Channels of the AlPO4-5 Molecular Sieve. Appl. Spectrosc., 1994, 48, 224–227. [72] G.J. Klap, S.M. van Klooster, M. Wubbenhorst, J.C. Jansen, H. van Bekkum, and J. van Turnhout., Polarization Reversal in AlPO4-5 Crystals Containing Polar or Nonpolar Organic Molecules: a Scanning Pyroelectric Microscopy Study. J. Phys. Chem. B, 1998, 102, 9518–9524. [73] M. Ehrl, H.W. Kindervater, F.W. Deeg, C. Brluchle, and R. Hoppe, Optical Spectroscopy of Thiazine and Oxazine Dyes in the Cages of Hydrated and Dehydrated Faujasite-type Zeolites: Molecular Dynamics in a Nanostructured Environment. J. Phys. Chem., 1994, 98, 11756–11763. [74] M. Ehrle, F.W. Deeg, C. Brauchle, O. Franke, A. Sobbi, G. Schulz-Ekloff, and D. Wo¨hrle High-temperature Non-photochemical Hole-burning of Phthalocyanine-zinc Derivates Embedded in a Hydrated AlPO4-5 Molecular Sieve. J. Phys. Chem., 1994, 98, 47–52. [75] C. Pereira, G.T. Kokotailo, and R.J. Gorte. Acetylene Polymerization in a H-ZSM-5 Zeolite. J. Phys. Chem., 1991, 95, 705–709. [76] S.D. Cox and G.D. Stucky, Polymerization of Methylacetylene in Hydrogen Zeolites. J. Phys. Chem., 1991, 95, 710–720. [77] P. Tsai, R.P. Cooney, J. Heaviside, and P. Hendra, Resonance Raman Spectra of Polyacetylene on Zeolite and Alumina Surfaces. Chem. Phys. Lett., 1978, 59, 510–513. [78] T. Bein and P. Enzel, Polypyrrole Chains in Zeolite Channels. Angew. Chem., 1989, 101, 1737–1738. [79] P. Enzel and T. Bein, Intrazeolite Synthesis of Polythiophene Chains. J. Chem. Soc., Chem. Commun., 1989, 1326–1327. [80] D.E. De Vos and P.A. Jacobs, Zeolite-based Supramolecular Assemblies. In: Introduction to Zeolite Science and Practice, ed. H. Van Bekkum, E.M. Flanigen, P.A. Jacobs, and J.C. Jansen, Elsevier, Amsterdam, 2001, 957–985. [81] C. Wu and T. Bein, Conducting Polyaniline Filaments in a Mesoporous Channel Host. Science, 1994, 264, 1757–1759. [82] P. Enzel and T. Bein, Poly(acrylonitrile) Chains in Zeolite Channels: Polymerization and Pyrolysis. Chem. Mater., 1992, 4, 819–824. [83] K. Moller, T. Bein, and R.X. Fischer, Entrapment of PMMA Polymers Strands in Micro- and Mesoporous Materials. Chem. Mater., 1998, 10, 1841–1852. [84] A. Graeser and S. Spange, Novel Polyvinyl Ether-HY Zeolite Hybrid Materials: General Features. Chem. Mater., 1998, 10, 1814–1819. [85] T. Kyotani, T. Nagai, S. Inoue, and A. Tomita, Formation of New Type of Porous Carbon by Carbonization in Zeolite Nanochannels. Chem. Mater., 1997, 9, 609–615.

660

Chemistry of Zeolites and Related Porous Materials

[86] Z. Ma, T. Kyotani, Z. Liu, O. Terasaki, and A. Tomita, Very High Surface Area Microporous Carbon with a Three-dimensional Nano-array Structure: Synthesis and its Molecular Structure. Chem. Mater., 2001, 13, 4413–4415. [87] S.A. Johnson, E.S. Brigham, P.J. Olliver, and T.E. Mallouk, Effect of Micropore Topology on the Structure and Properties of Zeolite Polymer Replicas. Chem. Mater., 1997, 9, 2448–2458. [88] R. Ryoo, S. Joo, and J. Jun, Synthesis of Highly Ordered Carbon Molecular Sieves via Template-mediated Structural Transformation. J. Phys. Chem. B, 1999, 103, 7743–7746. [89] J.W. Lee, S.H. Yoon, T.H. Hyeon, S.M. Oh, and K.B. Kim, Synthesis of a New Mesoporous Carbon and its Application to Electrochemical Double-layer Capacitors. Chem. Commun., 1999, 2177–2178. [90] S.B. Yoon, J.Y. Kim, and J.S. Yu, Synthesis of Highly Ordered Nanoporous Carbon Molecular Sieves from Silylated MCM-48 using Divinylbenzene as Precursor. Chem. Commun., 2001, 559–560. [91] S.H. Joo, S.J. Choi, I. Oh, J. Kwak, Z. Liu, O. Terasaki, and R. Ryoo, Ordered Nanoporous Arrays of Carbon Supporting High Dispersions of Platinum Nanoparticles. Nature (London), 2001, 412, 169–172. [92] C.G. Wu and T. Bein, Conducting Carbon Wires in Ordered, Nanometer-Sized Channels. Science, 1994, 266, 1013–1015. [93] J. Rodriguez-Mirasol, T. Cordero, L.R. Radovic, and J.J. Rodriguez, Structural and Textural Properties of Pyrolytic Carbon Formed within a Microporous Zeolite Template. Chem. Mater., 1998, 10, 550–558. [94] B. Hamilton, J.S. Rimmer, M. Anderson, and D. Leigh, Broad-band Visible Light Emission from C60 Molecules Confined in Molecular Cage Materials. Adv. Mater., 1993, 5, 583–585. [95] A. Drljaca, C. Kepert, L. Spiccia, C.L. Raston, C.A. Sandoval, and T.D. Smith, Qualitative Test for Supramolecular Complexation of C60 Using a Mesoporous Silica. Chem. Commun., 1997, 195–196. [96] J.S. Chen, Q. Li, H. Ding, W.Q. Pang, and R.R. Xu, Infrared Study on the Dehydroxylation of C60-loaded MCM-41. Langmuir, 1997, 13, 2050–2054. [97] J.S. Chen, Q.H. Li, R.R. Xu, and F.S. Xiao, Distinguishing the Silanol Groups in the Mesoporous Molecular Sieve MCM-41. Angew. Chem., Int. Ed. Enlg., 1995, 34, 2694– 2696. [98] S. Iijima, Helical Microtubules of Graphitic Carbon. Nature (London), 1991, 354, 56–58. [99] P.J.F. Harris, Carbon Nanotubes and Related Structures. Cambridge University Press, 1999, 4-15. [100] Z.K. Tang, H.D. Sun, J. Wang, J.S. Chen, and G.D. Li, Mono-sized Single-wall Carbon Nanotubes Formed in Channels of AlPO4-5 Single Crystal. Appl. Phys. Lett., 1998, 73, 2287–2289. [101] G.D. Li, Z.K. Tang, N. Wang, and J.S. Chen, Structural Study of the 0.4-nm Single-walled Carbon Nanotubes Aligned in Channels of AlPO4-5 Crystal. Carbon, 2002, 40, 917–921. [102] Z.M. Li, Z.K. Tang, H.J. Liu, N. Wang, C.T. Chan, R. Saito, S. Okada, G.D. Li, J.S. Chen, and ˚ Carbon Nanotubes Aligned N. Nagasawa, Polarized Absorption Spectra of Single-walled 4 A in Channels of an AlPO4-5 Single Crystal. Phys. Rev. Lett., 2001, 87, 127401-1–4. ˚ Carbon Nanotube Arrays. [103] N. Wang, Z.K. Tang, G.D. Li, and J.S. Chen, Single-walled 4 A Nature (London), 2000, 408, 50–51. [104] H.D. Sun, Z.K. Tang, J. Chen, and G. Li, Synthesis and Raman Characterization of Monosized Single-wall Carbon Nanotubes in One-dimensional Channels of AlPO4-5 Crystals. Appl. Phys. A, 1999, 69, 381–384. [105] G.D. Li, Z.K. Tang, N. Wang, K.H. Wong, and J.S. Chen, Generating the Narrowest Singlewalled Carbon Nanotubes in the Channels of AlPO4-5 Single Crystals. Stud. Surf. Sci. Catal., 2001, 135, 269–303.

Porous Host–Guest Advanced Materials

661

[106] Z.K. Tang, L. Zhang, N. Wang, X.X. Zhang, G.H. Wen, G.D. Li, J.N. Wang, C.T. Chan, and P. Sheng, Superconductivity in 4 Angstrom Single-walled Carbon Nanotubes. Science, 2001, 292, 2462–2465. [107] C.L. Bowes, A. Malek, and G.A. Ozin, Chemical Vapor Deposition Topotaxy in Porous Hosts. Chem. Vap. Deposition, 1996, 2, 97–103. [108] M.R. Steele, P.M. Macdonald, and G.A. Ozin, Topotactic Metal Organic Chemical Vapor Deposition in Zeolite Y: Structure and Properties of CH3MY from MOCVD Reactions of (CH3)2MHY, where M ¼ Zn, Cd. J. Am. Chem. Soc., 1993, 115, 7285–7292. [109] G.A. Ozin, M.R. Steele, and A.J. Holmes, Intrazeolite Topotactic MOCVD. 3-Dimensional Structure-controlled Synthesis of II-VI Semiconductor Nanoclusters. Chem. Mater., 1994, 6, 999–1010. [110] J.E. Mac Dougall, H. Eckert, G.D. Stucky, N. Herron, Y. Wang, K. Moller, T. Bein, and D. Cox, Synthesis and Characterization of Group III-V Semiconductor Clusters: Gallium Phosphide GaP in Zeolite Y. J. Am. Chem. Soc., 1989, 111, 8006–8007. [111] L.T. Canham, Silicon Quantum Wire Array Fabrication by Electrochemical and Chemical Dissolution of Waters. Appl. Phys. Lett. , 1990, 57, 1046–1050. [112] O. Dag, A. Kuperman, and G.A. Ozin., Nanostructures: New Forms of Luminescent Silicon. Adv. Mater., 1995, 7, 72–78. ¨ . Dag, A. Kuperman, and G.A. Ozin, Germanium Nanoclusters: Chemical Vapor Deposition [113] O of Digermane in Zeolite Y and Mordenite. Adv. Mater., 1994, 6, 147–150. [114] F.Gao, G. Zhu, X. Li, B. Li, S. Qiu, and O. Terasaki, Synthesis of a High-quality Host Material: Zeolite MFI Giant Single Crystal from Monocrystalline Silicon Slice. J. Phys. Chem. B., 2001, 105, 12704–12708. [115] R. Leon, D. Margolese, P.M. Petroff, and G. Stucky, Nanocrystalline Ge Filaments Alex in the Pores of a Mesosilicate. Phys. Rev. B: condens. Matter., 1995, 52, R2285–R2288. [116] N. Herron, Y. Wang, M.M. Eddy, G.D. Stucky, D. E. Cox, K. Moller, and T. Bein, Structure and Optical Properties of CdS Superclusters in Zeolite Hosts. J. Am. Chem. Soc., 1989, 111, 530–540. [117] K. Moller, M.M. Eddy, G.D. Stucky, N. Herron, and T. Bein, Stabilization of Cadmium Selenide Molecular Clusters in Zeolite Y: EXAFS and X-ray Diffraction Studies. J. Am. Chem. Soc., 1989, 111, 2564–2571. [118] R. Seifert, A. Kunzmann, and G. Calzaferri, The Yellow Colour of Silver-containing Zeolite A. Angew. Chem., Int. Ed., 1998, 37, 1521–1524. [119] D. Bruhwiler, R. Seifert, and G. Calzaferri, Quantum-sized Silver Sulfide Clusters in Zeolite A. J. Phys. Chem. B., 1999, 103, 6397–6399. [120] Z.K. Tang, Y. Nozue, O. Terasaki, and T. Goto, Frenkel Excitons in Ordered PbI2 Clusters Incorporated into Zeolite. Mol. Cryst. Liq. Cryst., 1992, 218, 61–66. [121] Z.K. Tang, M.T. Loy, J.S. Chen, and R.R. Xu, Absorption Spectra of Se and HgI2 Chains in Channels of AlPO4-5 Single Crystal. Appl. Phys. Lett., 1997, 70, 34–36. [122] J.B. Parise, J.E. MacDougall, N. Herron, R. Farlee, A.W. Sleight, Y. Wang, T. Bein, K. Moller, and L.M. Moroney, Characterization of Se-loaded Molecular Sieves A, X, Y, AlPO-5, and Mordenite.Inorg. Chem., 1988, 27, 221–228. [123] K. Tamura, S. Hosokawa, H. Endo, S. Yamasaki, and H. Oyanagi, The Isolated Se Chains in the Channels of Mordenite Crystal. J. Phys. Soc. Jpn., 1986, 55, 528–533. [124] Z.K. Tang, M.M. Loy, T. Goto, J.S. Chen, and R. Xu., Polarized Raman Spectra of Se Chains Isolated in Channels of AlPO4-5 Single Crystal. Solid State Commun., 1997, 101, 333–336. [125] I.L. Li, J.P. Zhai, P. Launois, S.C. Ruan, and Z.K. Tang, Geometry, Phase Stability, and Electronic Properties of Isolated Selenium Chains Incorporated in a Nanoporous Matrix. J. Am. Chem. Soc., 2005, 127, 16111–16119.

662

Chemistry of Zeolites and Related Porous Materials

[126] V.V. Poborchii, A.V. Kolobov, J. Caro, V.V. Zhuravlev, and K. Tanaka., Polarized Raman Spectra of Selenium Species Confined in Nanochannels of AlPO4-5 Single Crystals. Chem. Phys. Lett., 1997, 280, 17–23. [127] V.V. Poborchii, A.V. Kolobov, H. Oyanagi, G.S. Romanov, and K. Tanaka, Structure of Selenium Incorporated into Nanochannels of Mordenite: Dependence on Ion Exchange and Method of Incorporation. Chem. Phys. Lett., 1997, 280, 10–16. [128] V.V. Poborchii, Polarized Raman and Optical Absorption Spectra of the Mordenite Single Crystals Containing Sulfur, Selenium, or Tellurium in the One-dimensional Nanochannels. Chem. Phys. Lett., 1996, 251, 230–234. [129] J.R. Agger, M.W. Anderson, M.E. Pemble, O. Terasaki, and Y. Nozue, Growth of Quantum-confined Indium Phosphide inside MCM-41. J. Phys. Chem. B, 1998, 102, 3345–3353. [130] V.I. Srdanov, I. Alxneit, G.D. Stucky, C.M. Reaves, and S.P. DenBaarsl, Optical Properties of GaAs Confined in the Pores of MCM-41. J. Phys. Chem. B, 1998, 102, 3341– 3344. [131] X.Z. You, Q.J. Meng, and W.S. Han, Progress in Coordination Chemistry, Higher Education Press, Beijing, 2000, 222. [132] A.S. Hay, Oxidative Coupling of Acetylenes. J. Org. Chem., 1960, 25, 1275–1276. [133] A.S. Hay, H.S. Blanchand, G.F. Endres, and J.W. Eustance, Polymerization by Oxidative Coupling. J. Am. Chem. Soc., 1959, 81, 6335–6336. [134] Y. Ukisu, A. Kazusaka, and M. Nomura, Preparation and Catalysis of a Copper–Pyridine Complex Encapsulated into a Zeolite Supercage. J.Mol. Catal., 1991, 70, 165–174. [135] P.S.E. Dai and J.H. Lunsford, Spectroscopic Characterization of Tetrakis(pyridine)copper(II) Complexes in Zeolite Y. Inorg. Chem., 1980, 19, 262–264. [136] W. Bo¨hlmann, K. Schandert, A. Po¨ppl, and H.-C. Semmelhack, Synthesis and Electron Spin Resonance Studies of MCM-41 Doped with Copper Pyridine Complexes. Zeolites, 1997, 19, 297–304. [137] A. Po¨ppl and L. Kevan, Study of Adsorbate Interactions of Cu(II) Ion-exchanged into Siliceous MCM-41 Silica Tube Material by Electron Spin Resonance and Electron Spin Echo Modulation Spectroscopy. Langmuir, 1995, 11, 4486–4490. [138] Y. Yamada, Electron Spin Resonance Studies of Bis(acetylacetonato)copper(II) Adsorbed on X-type Zeolite. II. The Substitution Reaction of Ligands with Pyridine. Bull. Chem. Soc. Jpn., 1972, 45, 64–68. [139] W. Dewilde, G. Peeters, and J.H. Lunsford, Synthesis and Spectroscopic Properties of Tris(2,2’-bipyridine)ruthenium(II) in Zeolite Y. J. Phys. Chem., 1980, 84, 2306–2310. [140] W.H. Quayle and J.H. Lunsford, Tris(2,2’-bipyridine)ruthenium(III) in Zeolite Y: Characterization and Reduction on Exposure to Water. Inorg. Chem., 1982, 21, 97–103. [141] P. Laine´, M. Lans, G. Calzaferri, Limits of the in situ Synthesis of Tris(2,2’-bipyridine)ruthenium(II) in the Supercages of Zeolite Y. Inorg. Chem., 1996, 35, 3514–3518. [142] J.M. Ledney and P.K. Dutta, Oxidation of Water to Dioxygen by Intrazeolitic Ru(bpy)33þ. J. Am. Chem. Soc., 1995, 117, 7687–7695. [143] P.K. Dutta and S.K. Das, Oxidizing Properties of Zeolite-encapsulated Oxobis(2,2’-bipyridine)ruthenium(IV) Complexes Formed by Air Oxidation of Bis(2,2’-bipyridine)aquaruthenium(II). J. Am. Chem. Soc., 1997, 119, 4311–4312. [144] M. Sykora, J.R. Kincail, P.K. Dutta and N.B. Castagnola, On the Nature and Extent of Intermolecular Interactions between Entrapped Complexes of Ru(bpy)32þ in Zeolite Y. J. Phys. Chem. B, 1999, 103, 309–320. [145] K. Maruszewki, D.P. Strommen, K. Handrich, and J.R. Kincaid, Synthesis and Spectroscopic Properties of Zeolite-entrapped Bis-heteroleptic Ruthenium(II) Polypyridine Complexes. Inorg. Chem., 1991, 30, 4579–4582.

Porous Host–Guest Advanced Materials

663

[146] P.K. Dutta and J.A. Incavo, Photoelectron Transfer from Tris(2,2’-bipyridine)ruthenium (II) to Methylviologen in Zeolite Cages: A Resonance Raman Spectroscopic Study. J. Phys. Chem., 1987, 91, 4443–4446. [147] M. Sykora and J.R. Kincald, Photochemical Energy Storage in a Spatially Organized Zeolitebased Photoredox System. Nature (London), 1997, 387, 162–164. [148] P. Innocenzi, H. Kozuka, and T. Yoko, Fluorescence Properties of the Ru(bpy)32þ Complex Incorporated in Sol-gel-Derived Silica Coating Films. J. Phys. Chem. B, 1997, 101, 2285–2291. [149] S.S. Kim, W. Zhang, and T.J. Pinnavaia, Catalytic Oxidation of Styrene by Manganese(II) Bipyridine Complex Cations Immobilized in Mesoporous Al-MCM-41. Catal. Lett., 1997, 43, 149–154. [150] Z.H. Luan, J. Xu, and L. Kevan, Manganese-bipyridine Complex Incorporated into Mesoporous MCM-41 Molecular Sieves. Chem. Mater., 1998, 10, 3699–3706. [151] P.P.H.J.M. Knops-Gerrits, F.C. De Schryver, M. van de Auweraer, H. van Mingroot, X.Y. Li, and P.A. Jacobs, Emission from Zeolite-occluded Manganese-diimine Complexes. Chem. Eur. J., 1996, 2, 592–597. [152] W.H. Quayle, J.H. Lunsford, G. Peeters, G.L. De Roy, and E.F. Vansant, Synthesis and Spectroscopic Properties of Divalent and Trivalent Tris(2,2’-dipyridine)iron Complexes in Zeolite Y. Inorg. Chem., 1982, 21, 2226–2231. [153] Y. Umemura, Y. Minai, and T. Tominaga, Structural Distortion of 6-Coordinated Fe(II) Complexes in Zeolite Y. J. Phys. Chem. B, 1999, 103, 647–652. [154] K.J. Balkus, Jr, A.A. Welch, and B.E. Gnade, The Preparation and Characterization of Rh(III) SALEN Complexes Encapsulated in Zeolites X and Y. Zeolites, 1990, 10, 722–729. [155] F. Bedioui, E. De Boysson, J. Devynck, and K.J. Balkus, Jr, Electrochemistry of Zeolite Encapsulated Cobalt SALEN Complexes in Acetonitrile and DMSO Solutions. J. Chem. Soc., Faraday Trans., 1991, 87, 3831–3834. [156] Z. Gaillon, N. Sajot, F. Bedioui, J. Devynck, and K.J. Balkus, Jr, Electrochemistry of Zeolite Encapsulated Complexes. Part 3. Characterization of Iron and Manganese SALEN Entrapped in Y Faujasite Type Zeolite. J. Electroanal. Chem., 1993, 345, 157–167. [157] C.A. Bessel and D.R. Rolison, Topological Redox Isomers: Surface Chemistry of Zeoliteencapsulated Co(salen) and [Fe(bpy)3]2þ Complexes. J. Phys. Chem. B, 1997, 101, 1148–1157. [158] D.R. Rolison, C.A. Bessel, M.D. Baker, C. Seneratne, and J. Zhang, Reply to the Comment on ‘Zeolite-modified Electrodes: Intra- versus Extrazeolite Electron Transfer’. J. Phys. Chem., 1996, 100, 8610–8611. [159] F. Bedioui, J. Devynck, and K.J. Balkus, Jr, Comment on ‘Zeolite-modified Electrodes: Intraversus Extrazeolite Electron Transfer’. J. Phys. Chem., 1996, 100, 8607–8609. [160] S. Kowalak, R.C. Weiss, and K.J. Balkus, Jr, Zeolite Encapsulated Pd(salen), a Selective Hydrogenation Catalyst. J. Chem. Soc., Chem. Commun., 1991, 57–58. [161] M. Nakamura, T. Tatsumi, and H.-O. Tominaga, Synthesis and Catalytic Capability of Zeolite-encapsulated Iron and Manganese Tetramethylporphine Complexes. Bull. Chem. Soc. Jpn., 1990, 63, 3334–3336. [162] C.J. Liu, S.G. Li, W.Q. Pang, and C.M. Che, Ruthenium Porphyrin Encapsulated in Modified Mesoporous Molecular Sieve MCM-41 for Alkene. Chem. Commun., 1997, 65–66. [163] X.Q. Wang, Y.X. Liang, Y.W. Liu, and L.X. Yu, Synthesis and Characterization of Cobalt(II) Porphyrin in the Supercages of Faujasite, Chem. J. Chin. Univ., 1993, 14, 19–20. [164] X.Q. Wang, S. Gao, Y.W. Liu, L.X. Yu, Y. Li, F.H. Li, and X.Z. Cao, Preparation, Characterization and Catalytic Performance of Zeolite Containing Cobalt Porphyrin, Chem. J. Chin. Univ., 1994, 15, 789–793.

664

Chemistry of Zeolites and Related Porous Materials

[165] X.Q. Wang, S. Gao, C.S. Cao, L.X. Yu, and X.Z. Cao, Catalytic Property of Metal-porphyorin Encapsulated in the Supercage of Zeolite, Chin. J. Catal., 1996, 17, 343–345. [166] F. Bedioui, Zeolite-encapsulated and Clay-intercalated Metal Porphyrin, Phthalocyanine and Schiff-base Complexes as Models for Biomimetic Oxidation Catalysts: an Overview. Coord. Chem. Rev., 1995, 144, 39–68. [167] E. Paez-Mozo, N. Gabriunas, F. Lucaccioni, D.D. Acosta, P. Patrono, A. La Ginestra, P. Ruiz, and B. Delmon, Cobalt Phthalocyanine Encapsulated in Y-Zeolite – A Physicochemical Study. J. Phys. Chem., 1993, 97, 12819–12827. [168] K.J. Balkus, Jr, A.G. Gabrielov, S.L. Bell, F. Bedioui, L. Roue, and J. Devynck, Zeolite Encapsulated Co(II) and Cu(II) Perfluorophthalocyanines: Synthesis and Characterization. Inorg. Chem., 1994, 33, 67–72. [169] M. Ganschow, D. Wohrle, and G. Schulz-Ekloff, Incorporation of Differently Substituted Phthalocyanines in the Mesoporous Molecular Sieve Si-MCM-41. J. Porphyrins Phthalocyanines, 1999, 3, 299–309. [170] M. Wark, A. Ortlam, M. Ganschow, G. Schulz-Ekloff, and D. Wo¨hrle, Monomeric Encapsulation of Phthalocyanine-dye Molecules in the Pores of Si-MCM-41 and Ti-MCM-41. Ber. Bunsen-Ges Phys. Chem., 1998, 102, 1548–1553. [171] R. Hoppe, A. Ortlam, J. Rathousky, G. Schulz-Eklof, and A. Zukal, Synthesis of Titaniumcontaining MCM-41 Mesoporous Molecular Sieves in the Presence of Zinc Phthalocyanine and Rhodamine B. Microporous Mater., 1997, 8, 267–273. [172] A. Po¨ppl, M. Newhouse, and L.E. Kevan, Electron Spin Resonance and Electron Spin Echo Modulation Studies of Cupric Ion Ion-exchanged into Siliceous MCM-41. J. Phys. Chem., 1995, 99, 10019–10023. [173] R.F. Howe and J.H. Lunsford, Oxygen Adducts of Cobalt(II)-ethylenediamine Complexes in X- and Y-type Zeolites. J. Phys. Chem., 1975, 79, 1836–1842. [174] D.E. De Vos, J.L. Meinershagen, and T. Bein, Highly Selective Epoxidation Catalysts Derived from Intrazeolite Trimethyltriazacyclononane-manganese Complexes. Angew. Chem., Int. Ed. Engl., 1996, 35, 2211–2213. [175] X.C. Cao, G.D. Li, J.S. Chen, S.J. Ma, and R.R. Xu, Preparation and Characterization of 140Cu/MCM-41 14T2Cu/MCM-41, Chem. J. Chin. Univ., 1999, 20, 25–27. [176] G. van de Goor, B. Lindlar, P. Behrens, and J. Felsche, Solvent-free Synthesis of Clathrasils using Metal-organic Complexes as Structure-directing Agents. J. Chem. Soc., Chem. Commun., 1995, 2559–2561. [177] P. Behrens, G. van de Goor, and C.C. Freyhardt, Structure-determining C-H...O-Si Hydrogen Bonds in Cobalticinium Fluoride Nonasil. Angew. Chem., Int. Ed. Engl., 1995, 34, 2680–2682. [178] K.J. Balkus, Jr, A.G. Gabrielov, and S. Shepelev, Synthesis of AlPO4-5 and AlPO4-16 using Cobalticinium Hydroxide. Microporous Mater., 1995, 3, 489–495. [179] C.C. Freyhardt, M. Tsapatsis, R.F. Lobo, K.J. Balkus, Jr, and M.E. Davis, A Highsilica Zeolite with a 14-Tetrahedral-atom Pore Opening. Nature (London), 1996, 381, 295–298. [180] I. Honma and H.S. Zhou, Self-assembling Functional Molecules in Mesoporous Silicate Materials: Optical Properties and Mesophase of Dye-doped M41S. Adv. Mater., 1998, 10, 1532–1536. [181] B.M. Weckhuysen, A.A. Verberckmoes, I.P. Vannijvel, J.A. Pelgrims, P.L. Buskens, P.A. Jacobs, and R.A. Schoonheydt, Zeolite Encaged Cu (Histidine) Complexes as a Mimic of Natural Cu-enzymes. Angew. Chem., Int. Ed. Engl., 1995, 34, 2652–2654. [182] J. Evans, A.B. Zaki, M. El-Sheikh, and S.A. El-Safty, Incorporation of Transition-metal Complexes in Functionalized Mesoporous Silica and their Activity toward the Oxidation of Aromatic Amines. J. Phys. Chem., B, 2000, 104, 10271–10281.

Porous Host–Guest Advanced Materials

665

[183] M. Alvaro, V. Forues, S. Garcia, H. Garcia, and J.C. Scaiano, Intrazeolite Photochemistry. 20. Characterization of Highly Luminescent Europium Complexes Inside Zeolites. J. Phys. Chem. B, 1998, 102, 8744–8750. [184] A. Stein, Advances in Microporous and Mesoporous solids – Hightlights of Recent Progress. Adv. Mater., 2003, 15, 763–775. [185] S.L. James, Metal-organic Frameworks. Chem. Soc. Rev., 2003, 32, 276–288. [186] S.S.Y. Chui, S.M. F. Lo, J.P.H. Charmant, A.G. Orpen, and I.D. Williams, A Chemically Functionalizable Nanoporous Material [Cu3(TMA)2(H2O)3]n. Science, 1999, 283, 1148–1150. [187] O.M. Yaghi, G. Li, and H. Li, Selective Binding and Removal of Guests in a Microporous Metal-organic Framework. Nature (London), 1995, 378, 703–706. [188] H.J. Choi and M.P. Suh, Self-assembly of Molecular Brick wall and Molecular Honeycomb from Nickel(II) Macrocycle and 1,3,5-Benzenetricarboxylate: Guest-dependent Host Structures. J. Am. Chem. Soc., 1998, 120, 10622–10628. [189] B.L. Chen, M. Eddaoudi, T.M. Reineke, J.W. Kampf, M. O’Keeffe, and O.M. Yaghi, Cu2(ATC)  6H2O: Design of Open Metal Sites in Porous Metal-organic Crystals (ATC: 1,3,5,7-Adamantane tetracarboxylate). J. Am. Chem. Soc., 2000, 122, 11559–11560. [190] B.L. Chen, M. Eddaoudi, S.T. Hyde, M. O’Keeffe, and O.M. Yaghi, Interwoven Metalorganic Framework on a Periodic Minimal Surface with Extra-large Pores. Science, 2001, 291, 1021–1023. [191] S. Noro, R. Kitaura, M. Kondo, S. Kitagawa, T. Ishii, H. Matsuzaka, and M. Yamashita, Framework Engineering by Anions and Porous Functionalities of Cu(II)/4,40 -bpy Coordination Polymers. J. Am. Chem. Soc., 2002, 124, 2568–2583. [192] K. Biradha and M. Fujita, A Springlike 3D-Coordination Network that Shrinks or Swells in a Crystal-to-crystal Manner upon Guest Removal or Readsorption. Angew. Chem., Int. Ed., 2002, 41, 3392–3395. [193] G.J. Halder, C.J. Kepert, B. Moubaraki, K.S. Murray, and J.D. Cashion, Guest-dependent Spin Crossover in a Nanoporous Molecular Framework Material. Science, 2002, 298, 1762–1765. [194] C.D. Wu, A. Hu, L. Zhang, and W. Lin, A Homochiral Porous Metal-organic Framework for Highly Enantioselective Heterogeneous Asymmetric Catalysis, J. Am. Chem. Soc., 2005, 127, 8940–8941. [195] B. Zhao, P. Cheng, Y. Dai, C. Cheng, D.Z. Liao, S.P. Yan, Z.H. Jiang, and G.L. Wang, A Nanotubular 3D Coordination Polymer Based on a 3d-4f Heterometallic Assembly. Angew. Chem., Int. Ed., 2003, 42, 934–936. [196] W. Chen, Q. Yue, C. Chen, H.M. Yuan, W. Xu, J.S. Chen, and S.N. Wang, Assembly of a Manganese(II) Pyridine-3,4-dicarboxylate Polymeric Network Based on Infinite Mn–O–C Chains. Dalton Trans., 2003, 28–30. [197] C.Y. Su, X.P. Yang, B.S. Kang, and T.C. W. Mak, Th-Symmetric Nanoporous Network Built of Hexameric Metallamacrocycles with Disparate Cavities for Guest Inclusion. Angew. Chem., Int. Ed., 2001, 40, 1725–1728. [198] R.G. Xiong, X.Z. You, B.F. Abrahams, Z.L. Xue, and C.M. Che, Enantioseparation of Racemic Organic Molecules by a Zeolite Analogue. Angew. Chem., Int. Ed., 2001, 40, 4422–4425. [199] N.G. Pschirer, D.M. Ciurtin, M.D. Smith, U.H.F. Bunz, and H.C. zur Loye, Noninterpene˚ 2 Prepared by using trating Square-grid Coordination Polymers with Dimensions of 25  25 A 0 N,N -type Ligands: the First Chiral Square-grid Coordination Polymer. Angew. Chem., Int. Ed., 2002, 41, 583–585. [200] C.J. Kepert, T.J. Prior, and M.J. Rosseinsky, A Versatile Family of Interconvertible Microporous Chiral Molecular Frameworks: the First Example of Ligand Control of Network Chirality. J. Am. Chem., Soc., 2000, 122, 5158–5168.

666

Chemistry of Zeolites and Related Porous Materials

[201] L. Pan, H. Liu, X. Lei, X. Huang, D.H. Olson, N.J. Turro, and J. Li., IRPM-1: A Recyclable Nanoporous Material Suitable for Ship-in-bottle Synthesis and Large Hydrocarbon Sorption. Angew. Chem., Int. Ed., 2003, 42, 542–546. [202] R. Kitaura, K. Seki, G. Akiyama, and S. Kitagawa, Porous Coordination-polymer Crystals with Gated Channels Specific for Supercritical Gases. Angew. Chem., Int. Ed., 2003, 42, 428–431. [203] M. Eddaoudi, H. Li, and O.M. Yaghi, Highly Porous and Stable Metal-Organic Frameworks: Structure Design and Sorption Properties. J. Am. Chem. Soc., 2000, 122, 1391–1397. [204] L.N. Rosi, M. Eddaoudi, J. Kim, M. O’Keeffe, and O.M. Yaghi, Infinite Secondary Building Units and Forbidden Catenation in Metal-organic Frameworks. Angew. Chem., Int. Ed., 2002, 41, 284–287. [205] N.L. Rosi, J. Eckert, M. Eddaoudi, D.T. Vodak, J. Kim, M. O’Keeffe, and O.M. Yaghi, Hydrogen Storage in Microporous Metal-organic Frameworks. Science, 2003, 300, 1127–1129. [206] M. Eddaoudi, J. Kim, N. Rosi, D. Vodak, J. Wachter, M. O’Keefe, and O.M. Yaghi, Systematic Design of Pore Size and Functionality in Isoreticular MOFs and their Application in Methane Storage. Science, 2002, 295, 469–472. [207] J.S. Seo, D. Whang, H. Lee, S.I. Jun, J. Oh, Y.J. Jeon, and K. Kim, A Homochiral Metalorganic Porous Material for Enantioselective Separation and Catalysis. Nature (London), 2000, 404, 982–986. [208] W. Chen, J.Y. Wang, C. Chen, Q. Yue, H.M. Yuan, J.S. Chen, and S.N. Wang, Photoluminescent Metal-organic Polymer Constructed from Trimetallic Clusters and Mixed Carboxylates. Inorg. Chem., 2003, 42, 944–946. [209] J. Sun, L. Weng, Y. Zhou, J. Chen, Z. Chen, Z. Liu, and D. Zhao, QMOF-1 and QMOF-2: Three-dimensional Metal-organic Open Frameworks with a Quartzlike Topology. Angew. Chem., Int. Ed., 2002, 41, 4471–4473. [210] B. Panella, M. Hirscher, H. Pu¨tter, and U. Muller, Hydrogen Adsorption in Metal–organic Frameworks: Cu-MOFs and Zn-MOFs Compared. Adv. Funct. Mater., 2006, 16, 520–524. [211] X. Lin, A.J. Blake, C. Wilson, X.Z. Sun, N.R. Champness, M.W. George, P. Hubberstey, R. Mokaya, and M. Schroder, A Porous Framework Polymer Based on a zinc(II) 4,40 Bipyridine-2,20 ,6,60 -tetracarboxylate: Synthesis, Structure, and ‘Zeolite-Like’ Behaviors. J. Am. Chem. Soc., 2006, 128, 10745–10753. [212] A.G. Wong-Foy, A.J. Matzger, and O.M. Yaghi, Exceptional H2 Saturation Uptake in Microporous Metal-organic Frameworks. J. Am. Chem. Soc., 2006, 128, 3494–3495. [213] G. Garberoglio, A.I. Skoulidas, and J.K. Johnson, Adsorption of Gases in Metal Organic Materials: Comparison of Simulations and Experiments. J. Phys. Chem. B, 2005, 109, 13094–13103. [214] Q. Yang and C. Zhong, Understanding Hydrogen Adsorption in Metal-organic Frameworks with Open Metal Sites: a Computational Study. J. Phys. Chem. B, 2006, 110, 655–658. [215] D.F. Sun, S.Q. Ma, Y.X. Ke, D.J. Collins, and H.C. Zhou, An Interweaving MOF with High Hydrogen Uptake. J. Am. Chem. Soc., 2006, 128, 3896–3897.

Further Reading Important monographs, proceedings, and journals on molecular sieves and porous materials

Important monographs [1] D.W. Breck, Zeolite Molecular Sieves, Structure, Chemistry and Use. John Wiley & Sons, New York, London, Sydney, Toronto, 1974 [2] Dalian Institute of Chemical Physics, Chinese Academy of Sciences. Zeolite Molecular Sieves. Science Press, Beijing, 1978 (Chinese) [3] A. Dyer, An Introduction to Zeolite Molecular Sieves. John Wiley & Sons Ltd, Chichester, 1988 [4] R. Szostak, Handbook of Molecular Sieves. Van Nostrand Reinhold, New York, 1992 [5] R. Szostak, Molecular Sieves, Principles of Synthesis and Identification. Blackie Academic & Professional, London, 1998 [6] R. Szostak, Handbook of Molecular Sieves – Structures. Springer, New York, 2006 [7] F. Schu¨th, S.W.S. Kenneth, and J. Weitkamp, Handbook of Porous Solids. John Wiley & Sons, New York, 2002 [8] H.G. Karge and J. Weitkamp, Molecular Sieves Vol. 1, Synthesis. Springer, Berlin, Heidelberg, New York, Tokyo, 1998 [9] H.G. Karge and J. Weitkarp, Molecular Sieves Vol. 2, Structures and Structure Determination. Springer, Berlin, Heidelberg, 1999 [10] H.G. Karge and J. Weitkarp, Molecular Sieves Vol. 3, Post-Synthesis Modification I. Springer, 2002 [11] H.G. Karge and J. Weitkarp, Molecular Sieves Vol. 4, Characterization I. Springer, 2004 [12] H.G. Karge and J. Weitkarp, Molecular Sieves Vol. 5, Characterization II. Springer, 2006 [13] R.M. Barrer, Hydrothermal Chemistry of Zeolites. Academic Press, London, New York, 1982 [14] P.A. Jacobs and J.A. Martens, Synthesis of High-silicon Aluminosilicate Zeolites. Stud. Surf. Sci. Catal., 33, 1987

Chemistry of Zeolites and Related Porous Material – Synthesis and Structure Ruren Xu, Wenqin Pang, Jihong Yu, Qisheng Huo and Jiesheng Chen # 2007 John Wiley & Sons, (Asia) Pte Ltd

668

Chemistry of Zeolites and Related Porous Materials

[15] H.K. Beyer, H.G. Karge, I. Kirics, and J.B. Nagy, (Eds). Catalysis by Microporous Materials. Stud. Surf. Sci. Catal. 94, 1994 [16] R.R. Xu, Z. Gao, and Y. Xu, Progress in Zeolite Science – A China Perspective. World Scientific Press, Singapore, New Jersey, London, Hong Kong, 1995 [17] H. Chon, S.I. Woo, and S.E. Park, Recent Advance and New Horizons in Zeolite Science and Technology. Elsevier, Amsterdam, 1996 [18] J. Weitkamp and L. Puppe, Catalysis and Zeolites, Fundamentals and Applications. Springer, Berlin, 1999 [19] H. Van Bekkum, P.A. Jacobs, E.M. Flanigen, and J.C. Jansen, Introduction to Zeolite Science and Techology, Elsevier, Amsterdam, 137. 2001 [20] H. Ghobarkar, O. Scha¨f, Y. Massiani, and P. Knauth, The Reconstruction of Natural Zeolites, A New Approach to Announce Old Materials by Their Synthesis. Springer, New York, 2004 [21] C.R.A. Catlow, R.A. van Santen, and B. Smit, Computer Modelling of Microporous Materials, Elsevier, 2004

Manuals and Atlases [1] Verified Synthesis of Zeolitic Materials. Synthesis Commission of the International Zeolite Association, 2nd Edn, ed. H. Robson, Elsevier, 2001 [2] Atlas of Zeolite Framework Types. Structure Commission of the International Zeolite Association, 5th Edn, eds Ch. Baerlocher, W.M. Meier, and D.H. Olson, Elsevier, 2001 [3] Collection of Simulated XRD Powder Patterns for Zeolites. Structure Commission of the International Zeolite Association, 4th Edn, ed. M.M.J. Treacy and J.B. Higgins, Elsevier, 2001

Proceedings of International Zeolite Conferences (IZC) [1] Molecular Sieves. Proceedings of the lst IZC, London, U.K., 1967, Society of Chemical Industry London, 1968 [2] Molecular Sieves I and II. Proceedings of the 2nd IZC, Worcester, U.S.A., 1970, Adv. Chem. Ser., 101 and 102, 1971 [3] Molecular Sieves. Proceedings of the 3rd IZC, Zu¨rich, Switzerland, 1973, ed. W.M. Meier and J.B. Uytterhoeven, Adv. Chem. Ser., 121, 1973 [4] Molecular Sieves - II. Proceedings of the 4th IZC, Chicago, U.S.A., 1977, ed. J.R. Katzer, ACS Symp. Ser., 40, 1977 [5] Proceedings of the 5th Intenational Conference on Zeolites. Proceedings of the 5th IZC, Naples, ltaly, 1980, ed. L.V.C. Rees, Heyden, London, Philadelphia, Rheine, 1980 [6] Proceedings of the 6th International Conference on Zeolites. Proceedings of the 6th IZC, Reno, U.S.A., 1983, ed. D. Olson and A. Bisio, Butterworths, Guildford, 1984 [7] New Developments in Zeolites Science and Technology. Proceedings of the 7th IZC, Tokyo, Japan, 1986, ed. Y. Murakami, A. Iijima, and J.W. Ward, Stud. Surf. Sci. Catal., 28, 1986 [8] Zeolites: Facts, Figures, Future. Proceedings of the 8th IZC, Amsterdam, Netherlands, 1989, ed. P.A. Jacobs and R.A. Van Santen, Stud. Surf. Sci. Catal., 49, 1989 [9] Proceedings from the 9th IZC, Montreal, Canada, 1992, ed. R. von Ballmoos, J.B. Higgins, and M.M.J. Treacy, Butterworth-Heinemann, Boston, London, 1992 [10] Zeolites and Related Microporous Materials. State of the Art 1994, Proceedings of the 10th IZC, Garmisch-Partenkirchen, Germany, 1994, ed. J. Weitkamp, H.G. Karge, H. Pfeifer, and W. Holderich, Stud. Surf. Sci. Catal., 84. 1994

Further Reading

669

[11] Progress in Zeolite and Microporous Materials. Proceedings of the 11th IZC, Seoul, Korea, 1996, ed. Chon. Hakze, Ihm Son-ki and Uh Young Sun, Stud. Surf. Sci. Catal., 105, 1996 [12] Proceedings of the 12th IZC, ed. M.M.J. Treacy, B.K. Marcus, M.E. Bisher, and J.B. Higgins, MRS, Baltimore, U.S.A., 1998 [13] Zeolite and Mesoporous Materials of the Dawn of the 21st Century. Proceedings of the 13th IZC, Montpellier, France, 2001, ed. A. Galarneau, F. Di Renzo, F. Fajula, and J. Verdrine, Stud. Surf. Sci. Catal., 135, 2001 [14] Recent Advances in the Science and Technology of Zeolites and Related Materials. Proceedings of the 14th IZC, Cape Town, South Africa, 2004, ed. E. Sreen, L.H. Callanan, and M. Claeys, Stud. Surf. Sci. Catal., 154, 2004

Proceedings of important international symposiums on different topics (1) Synthesis [1] [2] [3] [4]

Zeolites, Synthesis, Structure, Technology and Application. Stud. Surf. Sci. Catal., 24B, 1985 Synthesis of High-silicon Aluminosilicate Zeolites. Stud. Surf. Sci. Catal., 33, 1986 Innovation in Zeolite Materials Science. Stud. Surf. Sci. Catal., 37, 1988 Zeolite Synthesis. ACS Symp. Ser., 398, 1989

(2) Characterization [1] Characterization of Porous Solids. Proceedings of the IUPAC Symposium (COPS I), 1987. Stud. Surf. Sci. Catal. 39, 1987 [2] Characterization of Porous Solids II. Proceedings of the IUPAC Symposium (COPS II), 1990. Stud. Surf. Sci. Catal., 62, 1990 [3] Characterization of Porous Solids III. Proceedings of the IUPAC Symposium (COPS III), 1993. Stud. Surf. Sci. Catal., 87, 1993 [4] Characterisation of Porous Solids V. Proceedings of the 5th International Symposium on the Characterisation of Porous Solids (COPS-V), Heidelberg, Germany, 1999. Stud. Surf. Sci. Catal. 128, 1999 [5] Characterization of Porous Solids VI. Proceedings of the 6th International Symposium on the Characterization of Porous Solids (COPS-VI), Allicante, Spain, 2002. Stud. Surf. Sci. Catal., 144, 2002 [6] Characterization of Porous Solids VII. Proceedings of the 7th International Symposium on the Characterization of Porous Solids (COPS-VII), Aix-en-Provence, France, 2005. Stud. Surf. Sci. Catal., 160, 2005

(3) Structure [1] Structure and Reactivity of Modified Zeolites. Proceedings of an International Conference, Prague, 1984. Stud. Surf. Sci. Catal., 18, 1984

(4) Catalysis and adsorption [1] Catalysis by Zeolites. Proceedings of an International Symposium, 1980. Stud. Surf. Sci. Catal., 5, 1980 [2] Zeolites as Catalysts, Sorbents and Detergent Builders. Applications and Innovations. Proceedings of an International Symposium, Wu¨zburg, 1988. Stud. Surf. Sci. Catal., 46, 1988

670

Chemistry of Zeolites and Related Porous Materials

[3] Catalysis and Adsorption by Zeolites. Proceedings of ZEOCAT 90, Leipzig, 1990. Stud. Surf. Sci. Catal., 65, 1990 [4] Zeolite Chemistry and Catalysis. Proceedings of an International Symposium, Czechoslovakia, 1991. Stud. Surf. Sci. Catal., 69, 1991 [5] Catalysis by Microporous Materials. Proceedings of ZEOCAT’95, Hungary, 1995. Stud. Surf. Sci. Catal., 94, 1995 [6] Zeolites: A Refined Tool for Designing Catalytic Sites. Proceedings of the International Symposium, Canada, 1995. Stud. Surf. Sci. Catal., 97, 1995

(5) Mesoporous materials [1] Mesoporous Molecular Sieves 1998. Proceedings of the 1st International Symposium, USA, 1998. Stud. Surf. Sci. Catal., 117, 1998 [2] Nanoporous Materials II. Proceedings of the 2nd Conference on Access in Nanoporous Materials, Canada, 2000. Stud. Surf. Sci. Catal., 129, 2000 [3] Proceedings of the 2nd International Symposium on Mesoporous Molecular Sieves (ISMMS), Microporous and Mesoporous Materials, 44–45, ed. L. Bonneviot, S. Giasson, S. Kaliaguine, and M. Sto¨cker, Elsevier, Amsterdam, 2001. [4] Nanoporous Materials III. Proceedings of the 3rd International Symposium on Nanoporous Materials, Canada, 2002. Stud. Surf. Sci. Catal., 141, 2002 [5] Nanotechnology in Mesostructured Materials. Proceedings of the 3rd International Mesostructured Materials Symposium, Korea, 2002. Stud. Surf. Sci. Catal., 146, 2000 [6] Mesoporous Crystals and Related Nano-Structured Materials. Proceedings of the Meeting on Mesoporous Crystals and Related Nano-Structured Materials, Stockholm, Sweden, 2004. Stud. Surf. Sci. Catal., 148, 2000 [7] Nanoporous Materials IV. Proceedings of the 4th International Symposium on Nanoporous Materials, Niagara Falls, Ontario, Canada, 2005, Stud. Surf. Sci. Catal., 156, 2005

(6) Advanced materials [1] Innovation in Zeolite Materials Science. Proceedings of an International Symposium, Nieuwpoort, 1987. Stud. Surf. Sci. Catal., 37, 1987 [2] Advanced Zeolite Science and Applications. Stud. Surf. Sci. Catal., 85, 1993 [3] Porous Materials in Environmentally Friendly Processes. Proceedings of the 1st International FEZA Conference, Hungary, 1999. Stud. Surf. Sci. Catal., 125, 1999

(7) ZMPC (Japan) [1] Chemistry of Microporous Crystals. Proceedings of the International Symposium on Chemistry of Microporous Crystals, Tokyo, 1990. Stud. Surf. Sci. Catal., 60, 1990 [2] Zeolites and Microporous Crystals. Proceedings of the International Symposium on Zeolites and Microporous Crystals, Japan, 1993. Stud, Surf. Sci. Catal., 83, 1993 [3] Zeolite and Microporous Crystals. Proceedings of the International Symposium on ZMPC, 1997. Microporous Mesoporous Mater., 21/4–6, 1998 [4] Zeolite and Microporous Crystals. Proceedings of the International Symposium on ZMPC, 2001. Microporous Mesoporous Mater., 48, 2001 [5] Zeolite and Microporous Crystals. Proceedings of the International Symposium on ZMPC, 2006, Microporous Mesoporous Mater., in press, 2007

Further Reading

671

(8) Proceedings of FEZA (Federation of the European Zeolite Associations) [1] Porous Materials in Environmentally Friendly Processes. Proceedings of the 1st International FEZA Conference, Eger, Hungary, 1999. Stud. Surf. Sci. Catal., 125, 1999 [2] Zeolites and Ordered Mesoporous Materials: Progress and Prospects. The 1st FEZA School on Zeolites, Prague, Czech Republic, 2005. Stud. Surf. Sci. Catal., 157, 2005 [3] Impact of Zeolites and Other Porous Materials on the New Technologies at the Beginning of the New Millennium. Proceedings of the 2nd International FEZA Conference, Taormina, Italy, 2002. Stud. Surf. Sci. Catal., 142, 2002 [4] Proceedings of the 3rd International FEZA Conference, Prague, Czech Republic, 2005. Stud. Surf. Sci. Catal., 158, 2005

Important international journals [1] Zeolites, ed. L.V.C. Rees and R. von Ballmoos, Butterworth, [2] Heinemann, Stoneham, MA, USA, 1981–1993 [3] Microporous Materials, ed. J. Weitkamp, Elsevier, Amsterdam, London, New York, Tokyo, 1993–1997 [4] Microporous and Mesoporous Materials, Ed-in-chief M. Sto¨cker, Founding ed. J. Weitkamp; Regional ed. S.L. Suib, R.W. Thompson, and K. Kuroda, Elsevier, Amsterdam, London, New York, Tokyo, as from 1998

The above three are the official publications of the International Zeolite Association. Besides, lots of papers on molecular sieves and porous materials have been frequently published in some journals on inorganic chemistry (Inorg. Chem., J. Chem. Soc., Dalton Trans., etc.), physical chemistry (J. Phys. Chem. and Langmuir, etc.), material chemistry (Chem. Mater. and J. Mater. Chem. etc.), solid-state chemistry (J. Solid State Chem. and Solid, State, Sci., etc.), catalytic chemistry (J. Catal. Appl. Catal., A, Curr. Opin. Colloid. Interface Sci., etc.), and some famous communication journals, such as Chem. Commun. and Angew. Chem., etc. Some important creative letters and reviews have also been published in Nature, Science, Chem. Rev., Chem. Soc. Rev., and Acc. Chem. Res., etc.

Index 2-D hexagonal mesophase 498–505 3-D hexagonal mesophase 482, 491, 577 AASBUs 406–14 Adsorption isotherms 354, 355 Adsorption of hydrogen 655 Adsorption properties 352, 381 Aging 130–5, 296–300 Aging temperature 136–7 AlPO4-11 (AEL) synthesis 179–80 AlPO4-5 (AFI) rational synthesis 432–3 structure 66–7 synthesis 178–9 with encapsulated dye molecules 617 AlPO-CJ11 structure 78, 79 AlPO-CJ19 structure 80 AlPO-CJ4 structure 74–5 AlPO-CJB1 structure 78–80 AlPO-CSC structure 87–8, 455–7 AlPO-DETA structure 75–6 AlPO-ESC structure 87–8

AlPO-HDA structure 76, 77 synthesis 441–3 AlPO-PDA synthesis 442 Aluminoarsenates synthesis 193 Aluminoborates anionic framework 198 positive framework 198–9 Aluminophosphates anionic framework 72–88 composition 33 structure design 412–4, 426–8 structural construction regularity 153–7 Aluminosilicates gel primary gel 131 secondary gel 131–5 structure and aging 296–300 Aluminosilicates composition 33 Aluminium source 125 Ammonolysis 347 Amorphous SiO2 preparation 283 structure 279–80 AMS-n 508 Anionic surfactants 537

Chemistry of Zeolites and Related Porous Material – Synthesis and Structure Ruren Xu, Wenqin Pang, Jihong Yu, Qisheng Huo and Jiesheng Chen # 2007 John Wiley & Sons, (Asia) Pte Ltd

674

Index

Application of zeolite coatings corrosion-resistant coatings 253 for pervaporation 255–7 hydrophilic and antimicrobial coatings 254 Assembly of 2-D nets allowed operators 404 enumeration of structures 404–6 operators 402–3 sheet conformation 403–4 ASU-16 structure 100–1 synthesis 210 ASU-31 structure 101 ASU-32 structure 101–2 BEC structure 56 synthesis 458 Beta (BEA) structure 54–6 synthesis 177–8, 213–4 Block copolymer surfactants 477, 490, 538 Building units cage 25, 27–9 chain 29–30 layer 30–2, 401–2 periodic building units 32–3 primary building units 23–4 secondary building unites (SBUs) 24–6 Building-block built-up approach AlPO-CSC 455–7 C60 624 Caged mesostructures 508–20 Cancrinite (CAN) structure 51–2 Carbon nanotube growth in zeolites 625–31 Cationic surfactants 535–7 Cations distribution and position in framework 34 templating effect in crystallization 139–44 Chabazite (CHA) structure 52 Channel and surface modification cation exchange method 380–1 channel modification method 381–3 external surface modification 383–91 internal surface modification 383

Channel dimension 43–6 Charge density matching 151–3 Charge-balancing effect 318 Chelating dealumination 365, 366 Chemical dealumination 364, 371 Chemical modification of mesoporous silica 558–61 Chemical-liquid deposition (CLD) 387–91 Chemical-vapor deposition (CVD) 633–8, 241, 384–5, 387 Chiral building blocks 225 Chiral catalytic centers 218–219 Chiral catalytic materials 221–6 assembly of chiral catalytic centers 218–9 coordination and condensation of chiral building blocks 225–6 germanates 219–21 phosphates 222–3 silicates 218–26 uranyl molybdates 221–2 Chiral mesoporous silica 581–2 Chiral metal complexes 443–54 Chiral open frameworks 92, 93, 97 Chiral porous coordination polymer 651–2, 654 Chirality transfer 443–54 CIT-1(CON) structure 60, 62 synthesis 458 CIT-5 (CFI) synthesis 203 structure 56–7 CLD-modified HZSM-5 387, 389, 390 Cloverite (-CLO) synthesis 204 structure 69–71 Cluster crystal 610–1 CMK-n 568–71 Co-condensation 560–1 Combinatorial synthesis 168–72, 454–5 Condensation reaction of silicate and aluminate ions 294–6 Coordination polymer 647–55 Coordination sequences (CSQ) 42 Critical micelle concentration (CMC) 480 Crystal transition 165–6 Crystallization field 125–8 Crystallization kinetics 326–37 Crystallization process 285–7 Crystallization temperature 137–8

Index Cubic channel mesostructures 505–8 Cubic-hexagonal intergrowth 508–511 CVD-modified SiHM catalyst 386 CZP structure 71 Data mining 430–3 De novo molecular design 434–5 Dealumination liquid phase method 364 hydrothermal and chemical methods 371–3 high-temperature 362–3 vapor phase method 370–1 Decision tree 431 Deformed mesophases 520–5 Degree of polarization (DOP) 629 Detemplating 345, 347, 348 extraction 348–50 Dry gel conversion 166 Dual-phase transition mechanism 305–6 Dyes in zeolites 616–20 ECR-34(ETR) structure 62, 63 EMT structure 50–1 Energy minimization 437 Evaporation-induced self-assembly (EISA) 534–5 Extra-large pore materials structure 88, 92–3, 95, 97, 100 synthesis 201–11 Faujasite (FAU) structure 50 synthesis of Linde-Y type 173–5 synthesis of Low-silica type X (LSX) FDU-1 synthesis 511–2 FDU-12 synthesis 518–20 FDU-4 structure 100 synthesis 209 FDU-5 synthesis 508 FJ-1 synthesis 209 Fluoride ion 161–4 Forbidden zone 415–26

174

675

Formation mechanism mesoporous silica 478–489 charge density matching 485–7 cooperative formation mechanism 483–5 folding sheets mechanism 486–7 generalized liquid crystal templating mechanism 487 original liquid crystal templating (LCT) mechanism 482–3 true liquid crystal templating mechanism (LCT) 487–8 microporous zeolites 285–7 Framework density 47 FSM-16 synthesis 486–7 Fullerenes assembled in zeolites 624 Gallium-containing zeolites 374 Galloarsenates synthesis 193 Gallophosphates structure 88–92 synthesis 192 Gas separation membranes 242 LTA 242 MFI 242 Gelation of silica sol 280–2 Gemini surfactants 535–6 Genetic algorithm 428 Germanates 100–1 Grafting 559–60 Heteroatoms coating zeolites 373, 378 Hexagonal mesoporous silica materials 498–505 Hierarchical porous silica materials 531–3 High-resolution electron microscopy 627 High-temperature calcination 345, 346 High-temperature rapid crystallization 167 High-temperature vapor-phase treatment 378 Host-guest interaction energy 435–54, 620 Hydro(solvo)thermal synthesis 122–4 Hydrogen uptake 653–4 Hydrogen-bonding interaction 435–54 Hydrogen-storage 652–5 Hypothetical zeolite database 429–30 IM-12(UTL) structure 65–6

676

Index

IM-12(UTL) synthesis 459 Indium phosphates structure 92–3 synthesis 194 Indium sulfides structure 101–2 Inorganic-organic hybrid materials 563–4 Interaction between organic and inorganic 475–478 Intersecting channel zeolites synthesis 212–5 Ion clusters location 609–10 Ion-exchange modification 351 Ion-exchange of zeolites under microwave irradiation 160–1 Ionic liquids 167 Iron phosphates structure 95–6 Isomorphous substitution 33, 368, 373–8 demetallation 378–9 gas-solid 377–8 liquid-solid 374–7 ITQ-15 (UTL) synthesis 459 ITQ-17 structure 56 ITQ-21 synthesis 458 ITQ-22 (IWW) synthesis 213, 458–9 structure 60, 61 ITQ-24(IWR) structure 65 ITQ-29 (LTA) synthesis 453–4 JDF-20 structure synthesis JLU-10 structure JLU-7 structure JLU-8 structure JLU-9 structure

76–8 205 450–2 444–7 447–8 448–50

KIT-6 506 KSW-2 522, 524–5 Loop configuration 41–2 Lowenstein’s Rule 33, 412 LTA structure 49–50 synthesis 172–3 LTL structure 51 synthesis 178 M41S materials 469–71 Macroporous material templating synthesis 529–31 MCM-41 498–500, 619 MCM-48 505–7 MCM-50 470, 479, 485 Mesopore size control 526–7 Mesoporous carbon 568 Mesoporous carbon as template 540 Mesoporous fiber 579–80 Mesoporous material synthesis through acid-base pair 555–6 Mesoporous metal oxides 565–7 Mesoporous metals 571–2 Mesoporous nanoparticles 575 Mesoporous phosphates 567 Mesoporous spheres and balls 577–9 Mesoporous thin film 576–7 Metal cluster in pore preparation approaches 605–7 alkali metal cluster 607 bimetallic cluster 606 cadmium 615 noble metal cluster 613–4 mercury 614 Metal cluster ion in pore alkali metal ion cluster 609–12 bifunctional catalyst 613 electrical conductivity 612 ESR spectra 609–13 metal-containing mesoporous silica-based materials 562–3 Na43þ 607–8 Metal complexes in pore catalytic performance of the loaded complex 636–7 epoxidatin reaction 644–6

Index nonaromatic macrocyclic ligand 645 redox pair 646 Metal-organic framework (MOF) 8–9, 651–5 Coordination polymer 647–9 Metal-Schiff base hydrogenation of alkenes 641 phthalocyanine complex 642–3 salen 641 selective hydrogenation catalyst 641 silylation agent 643 Methylene blue 616 Microporous chlorides 200 Microporous nitrides 201 Microporous sulfides 200 Microwave radiation 346 Microwave synthesis AlPO4-5 159–60 NaA 158–9 MIL-31 structure 89 Mixed ligand system 652 Mixed surfactants 538–9 Molecular engineering 13–14 Molecular simulation 324–5, 654–5 Mordenite (MOR) structure 52–3 synthesis 175–6 Morphologies of mesoporous silicas 573 MSU-n 525, 542, 547 Multicarboxylate linker 652–3 Nanocrystals and ultrafine zeolite particles 235–41 controlled crystallization condition 238–9 controlled crystallization in microreactor 239–40 controlled crystallization of sol 236–7 nanozeolite catalytic materials 240 ND-1 structure 93 synthesis 207 Nickel phosphates structure 96–7 Non equilibrium thermodynamics 118 Non-ionic (neutral) surfactants 537 Non-silica mesoporous materials 561–2 NTHU-1 structure 90 synthesis 207 Nucleation 300–5

677

Si-ZSM-48 302–304 Si-ZSM-5 302–304 One-dimensional superconductor 631 Ordered mesoporous materials 468–71 Organic chelate of silicon 283 Outer space synthesis 167 Oxide-modified HZSM-5 zeolite 382 Oxidative detemplating 347 Periodic mesoporous organosilicas (PMOs) 564–5 Phase transformation and control in mesoporous materials 525–6 Physical chemistry of mesostructure assembly 491–4 Pillared layered microporous materials 215–7 Polymer in zeolites 621–23 Polymeric surfactant 538 Polymerization state 269–75, 284–5 aluminate 284–5 polysilicate ions 269–77 in potassium silicate solution 270–1 in sodium salt solution 269–70, 271 in tetrabutylammonium silicate (TBAS) aqueous solution 275–7 in tetraethylammonium silicate (TEAS) aqueous solution 274–5 in tetramethylammonium silicate (TMAS) aqueous solution 272–3 Preparation of porous carbon 623–4 Pseudo-boehmite 284 Quantum wire

612

Raman spectroscopy 627 Rational synthesis 430–59 Removal of surfactant from mesoporous silica 539–40 Ring number 43 SAPO-31 synthesis 180–1 SAPO-34(CHA) synthesis 181 composition 33–4 SBA-1 synthesis 512–3 SBA-15 synthesis 500–3

678

Index

SBA-16 synthesis 517–8 SBA-2 synthesis 509–511 SBA-3 synthesis 503–5 SBA-6 synthesis 514–7 SBA-8 synthesis 520–3 SDA cleavage 168 Secondary synthesis 164–5, 350, 377 Semiconductor clusters in zeolites Cd4S4 633–4 HgI2 634 III-V semiconductor nanocluster 635 PbI2 634 Se, Te, Ge and Si 635 silver sulfide 634 Semiconductor nanoparticles blue-shift phenomenon 631 luminescence 632 metal-organic chemical-vapor deposition (MOCVD) 631 optoelectronic property 631 zero-dimensional semiconductor clusters 631 Sensor chemical-sensing material 613 water-vapor-sensing material 613 Shape-selective adsorption 384, 385 Shape-selective catalysis 651 Sharpless catalysts 219 Ship-in-bottle strategy 11 Si-addition 368, 370 Silica gel preparation 282–3 structure 279–280 Silica mesophases 541–2 Silica sol gelation 280 preparation 279 structure 277–80 Siliceous mesostructured cellular foams (MCFs) 531 Silicon enrichment 364 Silicon-addition 367 Silicon-enrich zeolites 366 Simulated annealing 399–401, 406–12

Single crystal mesoporous material bulk material dissolution 234–5 clear homogeneous system 234 F
ions systems 230–3 influence of nucleation suppressors 227 solvothermal conditions 228–30 two silica sources 233 Single-walled carbon nanotubes 627–31 Sodalite (SOD) structure 48–9 Sodalite cage 609–10 Solid hydrogel transformation 287–9 Solid laser 618 Solution-mediated transport mechanism 294–305 Solvent polarity (ETN) 121 Solvent-extraction method 348 Space-filling effect 39, 317–8 Special aggregation morphology AlPO4-5 fibers 250 silicate-1 microspheres 250 cellular structures 249–50 SSZ-23(STT) structure 57–9 SSZ-53(SFH) synthesis 452–3 structure 62, 63, 64–5 SSZ-59(SFN) synthesis 452–3 structure 63, 64–5 Structure-directing effect (SDE) 39, 307–26 Structure-type code 20–3 Supercage of Y zeolites 614 Surfactant effective packing parameter: g 489–491 Surfactant micelle and lyotropic liquid crystal 479–481 Synthesis of mesoporous materials at high temperatures 556–7 parameters 550–555 Systematic enumeration 401–6 Template for zeolites anions 322–4 cations 307–8 F
ion 320–1 metal complex 321–2 organic compounds 308–20 salts 322–4 water 322–4

Index Titanium-containing zeolites 181–2, 377 True templating effect 40–1, 311–3 Two-step calcination 346 Type material 20 UCSB-6(SBS) structure 71–2 ULM-16 synthesis 206 ULM-5 structure 89 synthesis 206 Ultra-stabilization 361, 362, 363 USY (ultra-stable Y zeolite) 363–73 UTD-1 (DON) synthesis 201, 203 structure 59–60 Vanadium phosphates 97–9 Vertex symbol 42 VPI-5 (VFI) synthesis 203 structure 67–8, 69 VSB-1 synthesis 208 VSB-5 structure 96–7 synthesis 208 Window size for caged mesostructures

527–9

679

Xe-adsorption-dynamic curves 382 Zeolite extra-large micropore 5–6 high-silica 4 low-silica 3–4 macropore 7–8 mesopore 6–7 natural 2–3 Zeolite films on stable supports 241–8 layer-by-layer (LBL) 243 a-axis oriented MFI zeolite films 245 b-axis oriented MFI zeolite films 244–5 patterned zeolite films 247–8 spin-on zeolite films 245–7 self-supporting zeolitic crystalline 241 low dielectric constant films 251 Zinc phosphates dimension build-up mechanism 197 structure 93–5 structure characters 195 synthetic approach 196 ZnHPO-CJ1 synthesis 211 ZSM-11 (MEL) structure 54 ZSM-5 (MFI) structure 53 synthesis 176–7