Modern Cyclophane Chemistry

Rolf Gleiter, Henning Hopf (Eds.) Modern Cyclophane Chemistry

Modern Cyclophane Chemistry. Edited by Rolf Gleiter, Henning Hopf Copyright © 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 3-527-30713-3

Further Titles of Interest: from WILEY-VCH N. Krause, A. S. K. Hashmi (Eds.)

Modern Allene Chemistry Two Volumes

2004, ISBN 3-527-30671-4 R. Mahrwald (Ed.)

Modern Aldol Reactions Two Volumes

2004, ISBN 3-527-30714-1 T. Takeda (Ed.)

Modern Carbonyl Olefination Methods and Applications

2004, ISBN 3-527-30634-X D. Astruc (Ed.)

Modern Arene Chemistry Concepts, Synthesis, and Applications

2002, ISBN 3-527-30489-4

Rolf Gleiter, Henning Hopf (Eds.)

Modern Cyclophane Chemistry

Professor Dr. Rolf Gleiter Institute of Organic Chemistry University of Heidelberg Im Neuenheimer Feld 270 69120 Heidelberg Germany Professor Dr. Henning Hopf Institute of Organic Chemistry Technical University of Braunschweig Hagenring 30 38106 Braunschweig Germany

n This book was carefully produced. Nevertheless, authors, editors and publisher do not warrant the information contained therein to be free of errors. Readers are advised to keep in mind that statements, data, illustrations, procedural details or other items may inadvertently be inaccurate.

Library of Congress Card No.: applied for A catalogue record for this book is available from the British Library. Bibliographic information published by Die Deutsche Bibliothek Die Deutsche Bibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data is available in the Internet at . © 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim All rights reserved (including those of translation in other languages). No part of this book may be reproduced in any form – by photoprinting, microfilm, or any other means – nor transmitted or translated into machine language without written permission from the publishers. Registered names, trademarks, etc. used in this book, even when not specifically marked as such, are not to be considered unprotected by law. Printed in the Federal Republic of Germany Printed on acid-free paper Cover design: The drawings for the cover were provided by Björn Hellbach Typesetting K+V Fotosatz GmbH, Beerfelden Printing betz-druck gmbh, Darmstadt Bookbinding Litges & Dopf Buchbinderei GmbH, Heppenheim ISBN

3-527-30713-3

V

Contents Preface

XV

List of Contributors

XVII

1

Cyclophynes

1.1 1.2 1.2.1 1.2.2 1.3 1.4 1.5 1.6 1.7

Yoshito Tobe and Motohiro Sonoda Introduction 1 Orthocyclophynes and Related Systems 2 Planar Dehydrobenzoannulenes 3 Nonplanar Orthocyclophynes 13 Metacyclophynes and Related Systems 17 Paracyclophynes and 1,3,5-Bridged Cyclophynes Concluding Remarks 35 Acknowledgement 36 References 36

2

Hetera (Cyclo-)phanes

2.1 2.1.1 2.1.2 2.1.3 2.2 2.2.1 2.2.1.1 2.2.1.2 2.2.1.3 2.2.1.4 2.2.1.5 2.2.1.6 2.2.1.7 2.2.2 2.2.2.1

Fritz Vögtle, Gregor Pawlitzki, and Uwe Hahn Introduction 41 Selection of Topics Presented 41 Definitions 42 Why Conduct a Survey on Heteraphanes? 49 Heteraphanes 51 Planar Chiral and Helical Chiral Phanes 51 Design of Phanes with Planar and Helical Chirality 51 Hetera [2.2]Metacyclophanes 53 Planar Chiral [2.2]Metacyclophanes 55 Planar Chiral Hetera [n]Para- and Hetera [n]Metacyclophanes 56 Dioxa [2.2]Phanes and Oxaza [2.2]Phanes 57 Enantiomer Separations 57 Strongly Helical Heteraphanes 58 Catenanes, Rotaxanes, and Knotanes of the Heteraphane Type 59 Template Synthesis of Rotaxanes Using Cyclophane Wheels 59

1

28

41

Modern Cyclophane Chemistry. Edited by Rolf Gleiter, Henning Hopf Copyright © 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 3-527-30713-3

VI

Contents

2.2.3 2.3 2.4 2.5

Higher Order [n]Rotaxanes via Non-ionic Template Effect 60 Combination of Anionic and Non-ionic Template 61 Molecular Knots and Similar Macrocycles of the Heteraphane Type 61 Further Heteraphanes, Metallaphanes and Supramolecular Phanes Conclusions 62 Acknowledgement 74 References 74

3

Highly Strained Cyclophanes

2.2.2.2 2.2.2.3 2.2.2.4

81

3.6 3.6.1 3.6.2 3.6.3 3.7

Takashi Tsuji Introduction 81 [n]Metacyclophanes 81 Synthesis 81 Structures and Physical Properties 83 Reactions of Strained [n]Metacyclophanes 84 Thermal and Photochemical Reactions 84 Addition Reactions 85 Reactions with Electrophiles 86 Reactions with Nucleophiles 87 [n]Paracyclophanes 89 Synthesis 90 Structures and Physical Properties 91 Reactions of Strained [n]Paracyclophanes 92 Thermal and Photochemical Reactions 92 Reactions with Electrophiles 94 Diels-Alder and Other Reactions 94 Kinetic Stabilization of [4]Paracyclophane Systems 95 Aromaticity of Bent Benzene Rings 96 Cyclophanes containing Polycyclic Aromatic Rings: (2,7)Pyrenophanes 96 [1.1]Paracyclophanes 98 Synthesis 99 Kinetic Stabilization of [1.1]Paracyclophane Systems 100 Structures and Physical Properties 100 References and Notes 102

4

Superphanes 105

4.1 4.2 4.2.1 4.3 4.4 4.4.1

Rolf Gleiter and Rolf Roers Introduction 105 [n2]Cyclopropenonophanes 106 Synthesis 106 Superbridged Cyclopropenyliophanes 109 C4-Superphanes 109 Properties of Cyclobutadieno Superphanes 115

3.1 3.2 3.2.1 3.2.2 3.2.3 3.2.3.1 3.2.3.2 3.2.3.3 3.2.3.4 3.3 3.3.1 3.3.2 3.3.3 3.3.3.1 3.3.3.2 3.3.3.3 3.3.3.4 3.4 3.5

62

Contents

4.4.2 4.5 4.6 4.7 4.8 4.9

Oxidative Demetallations 116 C5-Superphanes 117 Superbridged Benzene Rings 120 Concluding Remarks 126 Acknowledgement 127 References 127

5

Carbon-Bridged Ferrocenophanes

131

5.4 5.4.1 5.4.2 5.4.3 5.5 5.6

Joon-Seo Park and T. Randall Lee Introduction 131 Nomenclature 132 Mononuclear Carbon-Bridged Ferrocenophanes 133 [1]Ferrocenophanes 133 [2]Ferrocenophanes 134 [3]Ferrocenophanes 136 [4]Ferrocenophanes 140 [5]Ferrocenophanes 144 [m]Ferrocenophanes (m>5) 146 Multiply-Bridged Mononuclear Ferrocenophanes ([m]nFerrocenophanes) 147 Multinuclear Ferrocenophanes 150 [0.0]Ferrocenophanes 150 [1n]Ferrocenophanes 151 mn]Ferrocenophanes 153 Summary 154 References 154

6

Endohedral Metal Complexes of Cyclophanes

6.1 6.2 6.2.1 6.2.2 6.2.3 6.3 6.3.1 6.3.2 6.3.3 6.4 6.4.1 6.4.2 6.5 6.5.1 6.5.2 6.5.3

Rolf Gleiter, Bernhard J. Rausch, and Rolf J. Schaller Introduction 159 Singly-Bridged Group IVB Metallocenes 160 Singly-Carbon-Bridged Group IVB Metallocenes 160 Singly-Silicon-Bridged Group IVB Metallocenes 161 Singly-Boron- and Phosphorous-Bridged Group IVB Metallocenes 162 Doubly-Bridged Group IVB Metallocenes 163 Doubly-Carbon-Bridged Group IVB Metallocenes 163 Doubly-Silicon-Bridged Group IVB Metallocenes 165 Structural Features of Doubly-Bridged Group IVB Metallocenes 166 Endohedral Group VIB and VIIIB Metal p-Complexes 168 Bridged Bis[benzene]chromium Complexes 168 Bridged Metallocenophanes of Group VIIIB Metals 170 Cavities as Hosts for Cations of Group IB, IIIA and IVA Metals 176 Endohedral Silver Complexes with Cyclophanes 176 Silver Complexes with p-Prismands 180 Group IIIA and IVA Complexes of Cyclophanes 181

5.1 5.2 5.3 5.3.1 5.3.2 5.3.3 5.3.4 5.3.5 5.3.6 5.3.7

159

VII

VIII

Contents

6.5.4 6.6 6.7 6.8

Group IIIA Complexes with p-Prismands 182 Concluding Remarks 184 Acknowledgments 184 References 184

7

Intramolecular Reactions in Cyclophanes

189

7.5 7.5

Henning Hopf Introduction 189 Reactions between the Benzene Rings of Cyclophanes The Pseudo-gem Effect 198 Intramolecular Reaction between Functional Groups in Cyclophanes 201 Conclusions 207 References 208

8

Reactive Intermediates from Cyclophanes

8.1 8.2 8.3 8.4 8.5 8.6

Wolfram Sander Thermolysis of [2.2]Paracyclophanes 211 Photolytical Cleavage of [2.2]Paracyclophanes 217 Cleavage of [2.2]Paracyclophanes via Electron Transfer 220 Cleavage of Cyclophanes with Unsaturated Bridges 221 Cleavage of Cyclophanes with Carbonyl Groups in the Bridge 223 References 226

7.1 7.2 7.3 7.4

9

9.1 9.2 9.2.1 9.2.2 9.2.2.1 9.2.2.2 9.3 9.3.1 9.3.1.1 9.3.1.2 9.3.1.3 9.3.2 9.3.3 9.4 9.5 9.6

190

211

X-ray Crystal Structures of Porphyrinophanes as Model Compounds for Photoinduced Electron Transfer 229

Hermann Irngartinger and Thomas Oeser Introduction 229 Porphyrin–Quinone Cyclophanes 230 Double-Bridged Porphyrin–Quinone Cyclophanes 230 Single-Bridged Porphyrin–Quinone Cyclophanes 232 Phenyl-Spacered Porphyrin–Quinone Cyclophanes 232 Naphthalene-Spacered Porphyrin–Quinone Cyclophanes 235 Porphyrin-Aromatic-Ring Cyclophanes 238 Single Bridge from Opposite meso-Positions 238 Substituted Phenyl Rings as Cyclophane Components 238 Polycyclic Aromatic Ring-Systems as Cyclophane Components 243 Aromatic Heterocycles as Cyclophane Components 245 Single Bridge from Opposite non-meso-Positions 248 Capped Porphyrins 249 Porphyrinophanes with Fullerene Hosts 253 Concluding Remarks 254 References 255

Contents

10

Ultraviolet Photoelectron Spectra of Cyclophanes

10.1 10.2 10.3 10.4 10.5 10.6 10.7 10.7.1 10.7.2 10.7.3 10.8

Heidi M. Muchall 259 Introduction 259 [2.2]Paracyclophane 263 Modified Bridges in [2.2]Paracyclophane 265 Conjugating Substituents in [2.2]Paracyclophane Donor–Acceptor Cyclophanes 268 Heterocyclophanes 270 Miscellaneous Compounds 270 [6]Phanes with Higher Aromatic Systems 270 Cyclopropenophanes 271 Metacyclophanediynes 272 References 273

11

11.1 11.2 11.3 11.4 11.4.1 11.4.2 11.4.3 11.5 11.6 11.6.1 11.6.2 11.7 11.8 11.9 11.10 11.10.1 11.10.2 11.10.3 11.10.4 11.10.5 11.11

UV/Vis Spectra of Cyclophanes

266

275

Paul Rademacher Introduction 275 Characteristic Properties of Cyclophanes with Implications for their Electronic Spectra 275 [n]Cyclophanes 277 [m.n]Paracyclophanes 279 [1.1]Paracyclophane 279 [2.2]Paracyclophane 280 [m.n]Paracyclophanes 284 Multibridged [2n]Cyclophanes and Related Compounds 285 [m.n]Arenophanes 288 [m.n]Naphthalenophanes 288 [n.n]Pyrenophanes 289 Fluorinated Cyclophanes 290 Heterocyclophanes 290 Multi-layered Cyclophanes 297 Donor–Acceptor Cyclophanes 298 Donor–Acceptor Substituted [n.n]Paracyclophanes 298 [n.n]Paracyclophane Quinhydrones 300 [n.n]Metacyclophane Quinhydrones 303 Oligooxa[m.n]paracyclophane Quinhydrones and Related Compounds 304 Donor–Acceptor Substituted [n.n]Cyclophanes with Extensive p-Electron Systems 305 References 307

12

Electronic Circular Dichroism of Cyclophanes

12.1 12.2

Stefan Grimme and Arnold Bahlmann Introduction 311 Theoretical Methods 312

311

IX

X

Contents

12.3 12.3.1 12.3.2 12.4 12.4.1 12.4.1.1 12.4.1.2 12.4.2 12.4.2.1 12.4.2.2 12.4.2.3 12.4.3 12.4.3.1 12.4.3.2 12.4.4 12.4.4.1 12.5 12.6 12.7

Excited States of Model Compounds 314 Boat-type Deformation in Benzene 314 Interacting Benzene Fragments 319 Theoretical and Experimental CD Spectra of Cyclophanes [n]Cyclophanes 323 9,12-Dimethyl-4-oxa[7]paracyclophane 323 [6]Paracyclophane-8-carboxylic Acid 324 [2.2]Paracyclophanes 325 [2.2]Paracyclophane 325 4-Fluoro-[2.2]paracyclophane 327 4-Methyl[2.2]paracyclophane 329 [2.2]Metacyclophanes 331 1-Thia[2.2]metacyclophane 331 1-Thia-10-aza[2.2]metacyclophane 332 Cyclophanes with Two Different Aromatic Rings 333 14,17-Dimethyl[2](1,3)azuleno[2]paracyclophane 333 Conclusions 334 Acknowledgement 335 References 335

13

Fully Conjugated Beltenes (Belt-Like and Tubular Aromatics)

13.1 13.2 13.3 13.4 13.5 13.6 13.7

Rainer Herges Introduction 337 Approaches Towards Fully-Conjugated Beltenes 341 Belt-like Benzoannulenes (8) 342 [0n]Paracyclophanes 7 344 Möbius Belts 354 Conjugated Belts from Fullerenes 355 References 356

14

14.1 14.1.1 14.1.2 14.1.3 14.2 14.3 14.3.1 14.3.2 14.4 14.4.1 14.4.2

Molecular Electrochemistry of Cyclophanes

323

337

359

Bernd Speiser Introduction 359 Electron Transfer and Molecular Electrochemistry 359 Molecular Electrochemistry of Cyclophanes 359 Methods of Molecular Electrochemistry 360 Molecular Electrochemistry of Hydrocarbon Cyclophanes 361 Molecular Electrochemistry of Functionalized Cyclophanes 363 Substituted Cyclophanes 363 Cyclophanes with Non-metallic Heteroatoms as Ring Members 364 Molecular Electrochemistry of Organometallic Cyclophane Derivatives 367 A Possible Classification of Organometallic Cyclophane Derivatives 367 Metallocenophanes and Metallametallocenophanes 367

Contents

14.4.3 14.4.4 14.5 14.6

Cyclophanes Cyclophanes Conclusions References

as p-Complex Ligands 370 as Supramolecular Components 374 376 376

15

NMR Spectra of Cyclophanes

15.1 15.2 15.2.1 15.2.2 15.2.3 15.3 15.4 15.5 15.6 15.7 15.8 15.9

Ludger Ernst and Kerstin Ibrom Introduction and Scope 381 [n]Phanes 382 [n]Metacyclophanes 382 [n]Paracyclophanes 385 Other [n]Phanes 387 [2.2]Phanes 389 [3.3]Phanes 396 [m.n]Phanes (m > 2, n ³ 2) 400 [mn]Phanes 406 Other Phanes 409 Conclusion 412 References 412

381

16

Strained Heteroatom-Bridged Metallocenophanes

16.1 16.2 16.2.1 16.2.1.1 16.2.1.2 16.2.1.3 16.2.1.4 16.2.2 16.2.3 16.2.4 16.3 16.3.1 16.3.2 16.3.2.1 16.3.2.2 16.3.2.3 16.3.2.4 16.4 16.4.1 16.4.2 16.5

Ian Manners and Ulf Vogel Introduction 415 Synthesis and Structure 416 [1]Ferrocenophanes 416 Group 13 Bridged [1]Ferrocenophanes 417 Group 14 Bridged [1]Ferrocenophanes 417 Group 15 Bridged [1]Ferrocenophanes 419 Group 16 Bridged [1]Ferrocenophanes 419 Other [1]Metallocenophanes 419 [2]Ferrocenophanes 420 Other [2]Metallocenophanes 422 Ring-Opening Polymerization of Strained Ferrocenophanes 422 Stoichiometric Insertion and Ring-Opening Reactions 422 Ring-Opening Polymerizations (ROP) 423 Thermal ROP 423 Living Anionic ROP of Silicon-Bridged [1]Ferrocenophanes 424 Transition Metal-Catalyzed ROP of [1]Ferrocenophanes 425 Other Initiation Methods for ROP 426 Properties and Applications of Ring-Opened Polyferrocenes 426 Polyferrocenylsilanes 426 Other Polymetallocenes via ROP 430 References 431

415

XI

XII

Contents

17

17.1 17.2 17.3 17.4 17.5 17.5.1 17.5.2 17.5.3 17.6 17.7 17.8 17.9 18

18.1 18.2 18.3 18.4 18.4.1 18.4.2 18.4.3 18.4.4 18.5 18.6 19

19.1 19.1.1 19.2 19.3 19.3.1 19.4 19.4.1 19.5

Cyclophanes as Templates in Stereoselective Synthesis

435

Valeria Rozenberg, Elena Sergeeva, and Henning Hopf Introduction 435 Chiral [2.2]Paracyclophanes: Nomenclature and Stereochemical Assignment 435 The Resolution of Representative Mono- and Disubstituted [2.2]Paracyclophanes 437 Monosubstituted [2.2]Paracyclophanes as Chiral Inductors 437 Disubstituted [2.2]Paracyclophanes as Chiral Inductors 440 Ortho- and syn-latero-Substituted Derivatives 440 Pseudo-ortho-Derivatives 449 Pseudo-gem-Derivatives 452 Chiral Templates from Substituted [2.2]Paracyclophanes as Building Blocks 454 Stereoselective Reactions in the Side Chain of the Paracyclophanyl Moiety 457 Concluding Remarks 458 References 459 Vapor-Based Polymerization of Functionalized [2.2]Paracyclophanes: A Unique Approach towards Surface-Engineered Microenvironments

Doris Klee, Norbert Weiss, and Jörg Lahann Introduction 463 Synthesis of Functionalized [2.2]Paracyclophanes 464 CVD Polymerization of Functionalized [2.2]Paracyclophanes Immobilization of Bioactive Substances via Functionalized Poly(p-xylylenes) 472 Introduction 472 Surfaces for Blood Contact 475 Surfaces for Tissue Contact 476 Surface Engineering of Microfluidic Devices 478 Conclusions 481 References 482 From Cyclophanes to Molecular Machines

463

466

485

Amar H. Flood, Yi Liu, and J. Fraser Stoddart Introduction 485 Control over the Location and Motion of Moving Parts in Molecular Machines 485 The Creation of the Tetracationic Cyclophane 486 Host–Guest Chemistry with the Tetracationic Cyclophane 488 Location Control: Host–Guest Chemistry 488 Catenane Chemistry with the Tetracationic Cyclophane 492 Going for Gold – The Story of Olympiadane 495 Rotaxane Chemistry with the Tetracationic Cyclophane 497

Contents

19.5.1 19.6 19.6.1 19.6.2 19.6.3 19.6.4 19.6.5 19.7 19.7.1 19.7.2 19.7.3 19.7.4 19.8 19.9 19.10

Location Control – Catenanes and Rotaxanes 499 Switchable Rotaxanes, Catenanes and Pseudorotaxanes 499 Controllable Molecular Shuttles 500 Switchable Catenanes 502 Switchable Rotaxanes 503 Photochemically Switchable Pseudorotaxanes 503 Control of Motion 504 Electronic Devices Containing Molecular Switches 505 A [2]Catenane-Based Electronic Device 506 Bistable [2]Rotaxane Electronic Devices 508 Memory Devices 510 Miniaturization of the Crossbar 510 Mechanical Devices with Molecular Machines 513 Conclusions 515 References 516

20

Molecular Recognition Studies with Cyclophane Receptors in Aqueous Solutions 519

20.1 20.2 20.2.1 20.2.2 20.2.3 20.2.4

20.3 20.3.1 20.3.2 20.3.3 20.4 20.5 20.6 20.7 20.8

François Diederich Introduction 519 Complexation of Aromatic Solutes 520 Polar Effects in Cyclophane Complexation 520 Complexation of Polycyclic Aromatic Hydrocarbons (PAHs) in Water 521 The Combination of Apolar Binding and Ion Pairing Leads to Very High Substrate Selectivity 523 Solvent Dependency of Cyclophane-Arene Complexation and the Nonclassical Hydrophobic Effect: Enthalpic Driving Forces in Aqueous Solution 525 Steroid Recognition by Cyclophane Receptors 527 The Search for Cyclophanes with Larger Preorganized Cavity-Binding Sites 527 Steroid Complexation by Cyclophanes 9 and 10 529 Double-decker Cyclophanes for Efficient Steroid Complexation: Dissolution of Cholesterol in Water 533 Catalytic Cyclophanes 536 Dendritically Encapsulated Cyclophanes (Dendrophanes) 541 Conclusions 543 Acknowledgments 543 References 544

Subject Index

547

XIII

XV

Preface [2.2]Paracyclophane, first prepared in 1949 by Brown and Farthing and then systematically investigated by Cram and his co-workers from 1951 onward, is now a molecule with a history spanning half a century. A history, however, as confirmed inter alia by this monograph, that is not only active and living, but also increasingly extending into novel fields and applications. This is rare for any organic compound and certainly rare for hydrocarbons (excluding such household chemicals such as natural gas and gasoline). So, what are the reasons for this unabated interest? The first reason was already obvious to the pioneer workers in this area: The interaction between layered p-systems combined with an unusual three-dimensional structure. Indeed, most of the early studies in cyclophane chemistry were connected to the problem of the “bent and battered benzene rings” (so the title of a classical review article by Cram) and the consequence this structural feature had on the chemical behavior of the cyclophanes. Obviously, the distance between the aromatic subunits (the “decks”) could be altered by changing the nature of the molecular bridges, and one can say that this classical period of cyclophane chemistry culminated in the preparation of superphane by Boekelheide and certainly with the preparation of [1.1]paracyclophane by Tsuji. What – secondly – had become more and more obvious during these studies was the realization of what may be called the “phane concept”. Just as benzene is the epitome of a planar (flat) aromatic molecule, [2.2]paracyclophane is the example par excellence of a three-dimensional aromatic molecule. Not only did leaving the p-plane lead to the realization that a new, widely variable class of chiral molecules had become available – with all the effects on, for example, stereoselective synthesis – but the bridged aromatic concept could be connected effortlessly to the simultaneously evolving areas of crown ether, cryptand, cavitand, and supramolecular chemistry. To put it another way: numerous supramolecular structures are cyclophanes (and it may be added in passing that – viewed from this angle – there is also a huge group of naturally occurring cyclophanes). Thirdly, the zwitter character of the cyclophanes became evident during this period. Whereas, for example, adamantane or cubane are only aliphatic molecules, and benzene or the condensed aromatic systems or even the fullerenes are only (highly) unsaturated compounds, the cyclophanes are both: aromatics and aliphatModern Cyclophane Chemistry. Edited by Rolf Gleiter, Henning Hopf Copyright © 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 3-527-30713-3

XVI

Preface

ics (as already indicated by the euphonious name), making the use of both types of chemistries applicable. As far as the aromatic part of the cyclophanes is concerned, this not only meant the application of the typical reactions of classical aromatic chemistry to these compounds, but also the preparation of reactive intermediates derived from the aromatic rings: radical cations, radical anions, etc. Furthermore there was also the unique possibility of combining cyclophane chemistry with the archetypical layered inorganic chemistry, as exemplified by the classics ferrocene and dibenzene chromium. The cyclophanes, however, offer the additional topologically interesting and important possibility of bonding the metal to the interior or the exterior of the organic ligand. Incidentally, the chemical behavior of the molecular bridges in cyclophanes has not been studied to the same extent as the reactivity of the aromatic subunits. Finally, the crucial test for the significance of a molecular assembly is its practical application. After all, if the properties of a chemical compound are really “interesting”, there must also be practical, industrial applications. With the increased use of cyclophanes as monomers for the preparation of new polymers by CVD techniques with unusual and characteristic properties, polymers not just of importance as chemically very stable surfaces, e.g. for printed circuits, but also as new materials for medical application and drug delivery, cyclophanes are finding more and more applications in materials science. Together with their use in all kinds of optical devices, and as ligands in stereoselective synthesis, we are observing a very rich field developing here. It is our aim with this volume to show these different developments in cyclophane chemistry. That we succeeded was, of course, only possible because we found a group of excellent contributors and cyclophane specialists willing to give us their time and share their knowledge. As far as the practical side is concerned, we are very grateful to Drs. Gudrun Walter, Bettina Bems from Wiley/VCH, Mary Korndorffer for copy editing and Petra Krämer and C. Mlynek for drawing the numerous and often very complex structures. We dedicate this volume to the memory of Donald Cram and Virgil Boekelheide.

Heidelberg/Braunschweig April 2004

Rolf Gleiter Henning Hopf

XVII

List of Contributors Arnold Bahlmann Universität Münster Organisch-Chemisches Institut Corrensstraße 40 48149 Münster Germany

Rolf Gleiter Universität Heidelberg Organisch-Chemisches Institut Im Neuenheimer Feld 270 69120 Heidelberg Germany

François Diederich ETH Zürich Department of Chemistry and Applied Biosciences Wolfgang-Pauli-Straße 10 8093 Zürich Switzerland

Stefan Grimme Universität Münster Organisch-Chemisches Institut Corrensstraße 40 48149 Münster Germany

Ludger Ernst Technische Universität Braunschweig NMR-Labor der Chemischen Institute Hagenring 30 38106 Braunschweig Germany Amar H. Flood University of California, Los Angeles Department of Chemistry and Biochemistry 405 Milgard Avenue CA 90095-1569 Los Angeles USA

Uwe Hahn Rheinische Friedrich-WilhelmsUniversität Bonn Kekulé-Institut für Organische Chemie und Biochemie Gerhard-Domagk-Str. 1 53121 Bonn Germany Rainer Herges Universität Kiel Institut für Organische Chemie Otto-Hahn-Platz 4 24098 Kiel Germany

Modern Cyclophane Chemistry. Edited by Rolf Gleiter, Henning Hopf Copyright © 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 3-527-30713-3

XVIII

List of Contributors

Henning Hopf Technische Universität Braunschweig Institut für Organische Chemie Hagenring 30 38106 Braunschweig Germany Kerstin Ibrom Technische Universität Braunschweig Institut für Organische Chemie Hagenring 30 38106 Braunschweig Germany Hermann Irngartinger University of Heidelberg Institute of Organic Chemistry Im Neuenheimer Feld 270 69120 Heidelberg Germany Doris Klee RWTH Aachen Department of Textile and Macromolecular Chemistry 52074 Aachen Germany Joerg Lahann University of Michigan Departments of Chemical Engineering, Material Science and Engineering, and Macromolecular Science and Engineering 3414 G.G. Brown 2300 Hayward Ann Arbor MI – 48109-2139 USA T. Randall Lee University of Houston Department of Chemistry 4800 Calhoun Road Houston, TX 77204-5003 USA

Yi Liu University of California, Los Angeles Department of Chemistry and Biochemistry 405 Milgard Avenue CA 90095-1569 Los Angeles USA Ian Manners University of Toronto Department of Chemistry 80 St. George Street Toronto, Ontario M5S 3H6 Canada Heidi M. Muchall Concordia University Department of Chemistry and Biochemistry and Centre for Research in Molecular Modeling Montreal, Quebec H4B 1R6 Canada Thomas Oeser University of Heidelberg Institute of Organic Chemistry Im Neuenheimer Feld 270 69120 Heidelberg Germany Joon-Seo Park University of Houston Department of Chemistry 4800 Calhoun Road Houston, TX 77204-5003 USA Gregor Pawlitzki Rheinische Friedrich-WilhelmsUniversität Bonn Kekulé-Institut für Organische Chemie und Biochemie Gerhard-Domagk-Str. 1 53121 Bonn Germany

List of Contributors

Paul Rademacher Universität Duisburg-Essen Institut für Organische Chemie Universitätsstr. 5–7 45117 Essen Germany Bernhard J. Rausch Clariant BL Pharma Germany Molecules Synthesis Centre Industriepark Höchst, D 569 65926 Frankfurt/Main Germany Rolf Roers Bayer MaterialScience AG Building B 103 51368 Leverkusen Germany Valeria Rozenberg Russian Academy of Science A. N. Nesmeyanov Institute of Organoelement Compounds Vavilova str., 28 119991 Moscow Russia Wolfram Sander Ruhr University Bochum Department of Organic Chemistry II 44780 Bochum Germany Rolf J. Schaller EMS-Dottikon AG Hembrunnstr. 17 5605 Dottikon Switzerland Elena Sergeeva Russian Academy of Science A. N. Nesmeyanov Institute of Organoelement Compounds Vavilova str., 28 119991 Moscow Russia

Motohiro Sonoda Osaka University Graduate School of Engineering Science, Division of Frontier Materials Science 1-3 Machikaneyama Toyonaka 560-8531 Japan Bernd Speiser Universität Tübingen Institut für Organische Chemie Auf der Morgenstelle 18 72076 Tübingen Germany J. Fraser Stoddart University of California, Los Angeles Department of Chemistry and Biochemistry 405 Milgard Avenue CA 90095-1569 Los Angeles USA Takashi Tsuji Hokkaido University Division of Chemistry, Graduate School of Science Sapporo 060-0810 Japan Yoshito Tobe Osaka University Graduate School of Engineering Science, Division of Frontier Materials Science 1-3 Machikaneyama Toyonaka 560-8531 Japan Ulf Vogel University of Toronto Department of Chemistry 80 St. George Street Toronto, Ontario M5S 3H6 Canada

XIX

XX

List of Contributors

Fritz Vögtle Rheinische Friedrich-WilhelmsUniversität Bonn Kekulé-Institut für Organische Chemie und Biochemie Gerhard-Domagk-Str. 1 53121 Bonn Germany

Norbert Weiss RWTH Aachen Department of Textile and Macromolecular Chemistry 52074 Aachen Germany

1

1

Cyclophynes Yoshito Tobe and Motohiro Sonoda

1.1

Introduction

The chemistry of cyclophynes having carbon–carbon triple bond bridges has been one of the most actively investigated fields in modern cyclophane chemistry, particularly in connection with the evolving fields of carbon-rich materials and shapepersistent macrocyclic compounds. In accordance with the remarkable progress in this field over the last decade, a number of reviews has already been published based on the authors’ own perspectives [1–9]. Since, in this review, the authors wish to provide a comprehensive survey of cyclophyne chemistry, some overlaps are inevitable. The properties of cyclophynes are characterized by the geometric and electronic properties of triple bonds and the substitution pattern of the aromatic rings. With regard to the geometrical properties, the macrocyclic frameworks of cyclophynes can be expanded by incorporation of triple bonds because of their linearity. The substitution pattern of the aromatic rings, on the other hand, fixes the direction of the bridging triple bonds, defining the whole molecular shape. As a result, a variety of two- and three-dimensional architectures can be built by connecting aromatic rings with triple bond linkages. The ortho substitution pattern would lead to conjugated cyclophynes, also regarded as dehydrobenzoannulenes (DBAs), with either planar or nonplanar shape depending principally on their ring size. Although planar DBAs had been studied extensively during the late 1950s through mid-1970s in connection with the aromaticity/antiaromaticity of annulenes, these are currently being investigated with renewed interest regarding their potential as opto-electronic materials. On the other hand, nonplanar macrocycles of this type have been studied with regard to their conformation, chirality, and their potential application to liquid crystalline and sensing materials. Some highly unsaturated members of this type of compound have been shown to serve as precursors of ordered carbon materials. Since the meta substitution pattern would make the size of the cyclophynes larger than the ortho analogs, keeping the planarity of the macrocyclic rings, a number of shape-persistent metacyclophynes have been synthesized in order to investigate their supramolecular properties such as guest-binding and self-association in solution, solid state, and at solid–liqModern Cyclophane Chemistry. Edited by Rolf Gleiter, Henning Hopf Copyright © 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 3-527-30713-3

2

1 Cyclophynes

uid and air–liquid interfaces. To form cyclic structures of para substitution pattern, macrocycles adopt a cylindrical shape, and with 1,3,5-trisubstitution, a cagelike structure can be constructed. These three-dimensional molecules have been studied in view of the host–guest chemistry of curved p-systems, strain of triple bonds, and their transformation into all-carbon molecules. Concerning the electronic properties of C–C triple bonds, acetylene linkages are capable of transmitting electronic perturbation efficiently from one end of conjugated p-systems to the other because of the isotropic distribution of the p electrons along the C–C axis. This characteristic is responsible for the unique optoelectronic properties predicted for planar cyclic compounds like DBAs. Because C–C triple bonds have a high energy content, it is well known that structurally unhindered alkynes react violently, frequently with explosion. Similarly, diynes are known to undergo topochemical polymerization to form poly(butadiyne)s, potential materials with nonlinear optical properties. The high potential of the triple bonds toward spontaneous chemical reactions has been exploited to provide the transformation of some cyclophynes into all-carbon and carbon-rich materials. In addition, because triple bonds serve as good ligands for transition metals, a number of metal complexes of cyclophynes with unique structures has been readily prepared. Besides the interest from materials science, recent development of new synthetic methods, particularly those based on transition metal-catalyzed bond formation between sp–sp2 and sp–sp carbons, have been playing a crucial role in the rapid growth of this field. This chapter covers the recent development of the twoand three-dimensionally shaped cyclophynes from the above points of view. For convenience, the molecules are classified into three types: (1) orthocyclophynes with either planar or nonplanar shape; (2) metacyclophynes; and (3) paracyclophynes and 1,3,5-bridged cyclophynes, on the basis of the substitution pattern. Those compounds with mixed substitution pattern such as ortho, para and meta, para are included in one of the above three types depending on their most characteristic structural and/or functional properties.

1.2

Orthocyclophynes and Related Systems

Fully conjugated planar orthocyclophynes are regarded as dehydrobenzoannulenes (DBAs) [2, 3, 4, 6]. 12-, 14- and 18-membered macrocycles of this type are investigated intensively in connection with their opto-electronic properties, theoretical and experimental inspection of aromaticity, and as substructures of hitherto unknown two-dimensional carbon networks, graphyne and graphdiyne [9]. On the other hand, since the macrocyclic frameworks of larger orthocyclophynes are nonplanar, these molecules are investigated in view of their structure and the chirality of twisted p-systems [7]. In addition, highly unsaturated compounds of this class are studied as possible precursors to nanocarbon materials such as carbon onions and tubes.

1.2 Orthocyclophynes and Related Systems

Fig. 1.1

1.2.1

Planar Dehydrobenzoannulenes

It was predicted that the two-dimensional carbon network, graphyne, would exhibit semiconductive and nonlinear optical properties [10]. Recent calculations from first principles confirmed that graphyne should be a semiconductor with a moderate band gap and should be quite stable once synthesized [11]. It was predicted that potassium-intercalated graphyne would become metallic with a remarkably short distance (4.30 Å when each 12-membered ring holds a potassium atom) between the intercalant layers [12]. In connection with the nonlinear optical properties of graphyne, hyperpolarizabilities of ladder oligomers of dehydrobenzo[12]annulene ([12]DBA) have been estimated, based on semiempirical calculations, indicating that these can be promising third-order nonlinear optical materials [13]. Force field calculations for networks related to graphdiyne revealed that they would exhibit negative Poisson’s ratios, an unusual mechanical property caused by the rotation of triangles [14]. Most cyclophynes were synthesized by using the Stephens-Castro and Sonogashira reactions for the C(sp2)–C(sp) bond formation or copper(II)-mediated oxidative coupling for C(sp)–C(sp) bond formation. Remarkable improvements in the synthesis of cyclophynes have been achieved as exemplified by the synthesis of the known [12]DBA 1 a. These include (1) the Stephens-Castro reaction of the preformed copper(I) acetylide [15]; (2) one-step synthesis by the Sonogashira coupling using acetylene [16]; (3) selective construction of the 12-membered ring based on a masking/protection protocol for both aryl iodide using dialkyltriazene group and alkyne terminal using trialkylsilyl groups [17]; and (4) the alkyne metathesis method [18], which has been developed recently (Scheme 1.1). The first two methods were also utilized for the synthesis of thiophene analogs 2 a [16] and 3 [19]. By stepwise construction of the requisite precursor followed by intramolecular Sonogashira coupling, Haley synthesized bisDBAs 4 and 5 a [17]. An another derivative of the latter compound 5 b was also prepared by an in situ deprotection–cyclization method to prove the theoretical prediction for hyperpolarizability of this sys-

3

4

1 Cyclophynes

Fig. 1.2

Scheme 1.1

1.2 Orthocyclophynes and Related Systems

Scheme 1.2

tem (Scheme 1.2) [20]. Not only the parent hydrocarbon 1 a but also a number of its derivatives 1 b–d have been prepared by the metathesis method, except for those with substituents adjacent to the alkyne bridges. Similarly, the metathesis method was applied to the synthesis of bisDBA 4 and orthometacyclophyne 6 [18]. Since the late 1980s, Youngs and Tessier have been developing new frontiers in the chemistry of the [12]DBA system [6]. A number of transition metal complexes of 1 a were prepared such as Co(0), Cu(I), Ag(I), and Ni(0) complexes [21]. Significantly among these metal complexes, the Ni(0) complex becomes semiconductive [22]. For example, the Ni(0) complex doped with potassium [2.2.2]cryptate (1 : 2 ratio) exhibits conductivity of a moderate semiconductor regime (2 ´ 10–3 S cm–1). The hexamethoxy derivative 1 g, hexa-alkoxy derivatives 1 e–f having long alkyl chains, and their Ni(0) complexes were also prepared, as well as the quinone 7 and thiophene derivatives 2 b–c [23]. In spite of the expectation based on their structural similarities to hexasubstituted triphenylene derivatives, DBAs 1 e–f and 2 b–c with long alkyl chains did not exhibit liquid crystalline properties. A dihydro derivative of 1 a, cyclophanediyne 8, was prepared by Iyoda et al. using modified McMurry coupling conditions, followed by dehydroxylation [24]. Although the X-ray structure analysis of 8 revealed its short transannular distance (the shortest C(sp)–C(sp) distance: 2.77 Å), it was inert to photoirradiation. On the other hand, heating 8 at 250 8C in the presence of cyclohexa-1,4-diene resulted in the formation of dibenzonaphthocyclooctatetraene, a product of the Bergman cyclization. The corresponding transannular distances of the oxygen- and sulfurbridged diynes 9 a and 9 b, respectively, are longer than that of 8 (3.44 and 3.50 Å, respectively), and it is reported that these compounds are inert to thermal and photochemical cyclization [25]. Hexa-ethynyl derivatives 1 i–j were prepared as the key intermediates for the synthesis of antikekulene 12, an antiaromatic counterpart of kekulene [26]. Thus

5

6

1 Cyclophynes

the first intramolecular CpCo-catalyzed cyclotrimerization of 1 i and 1 j yielded the triphenylene-based cyclophynes 10 a and 10 b, respectively, whose second cycloisomerization furnished the corresponding pentaphenylene-based cyclophynes 11 a and 11 b (Scheme 1.3). The final cyclization to the target 12 is yet to be realized. Besides this spectacular synthetic endeavor, the parent hydrocarbon 1 h was found to form crystals with highly organized channels including tetrahydrofuran induced by multiple alkyne C–H. . .O hydrogen bonds [27]. Perethynyl-substituted dibenzodehydro[12]annulene 13 a was prepared by Rubin et al. by oxidative coupling of a differentially substituted hexa-ethynylbenzene derivative [28]. Hydrocarbon 13 a can be regarded as a constituent unit of a two-dimensional network with different topology from that of graphdiyne and as a monomer of ladder polymers constructed from the dibenzodehydro[12]annulene ring, called acynes. In this connection, Anthony prepared tetra-ethynyl-substituted [14]DBA 14 having acenaphthene units as a possible monomer unit of the related ladder polymers [29]. In connection with the topochemical polymerization of butadiynes, Komatsu et al. prepared octafluoro[12]DBA 13 c, whose reduction potential was more negative than that of the parent hydrocarbon 13 b [30]. In the two crystal forms of 13 c, the molecules are stacked in a slanted manner, while in the cocrystal of 13 b and 13 c (1 : 2), the molecules are stacked in a columnar arrangement forming a sandwich-like complex. In spite of the favorable packing geometries for topochemical polymerization, none of the crystals underwent photochemical polymerization, probably because of the high potential energy of the polymerization product. As a quantitative measure of aromaticity of DBAs, the nuclear independent chemical shift (NICS) values at the center of the macrocyclic ring of 1 a, as well as

Scheme 1.3

1.2 Orthocyclophynes and Related Systems

Fig. 1.3

those for [14]annulene 15, [16]annulene 16, and [18]annulene 17 a, were estimated by the DFT method, indicating that 1 a and 16 are moderately antiaromatic while 15 and 17 a are weakly aromatic [31]. Similarly we can predict that highly strained [10]annulene 18 is fully aromatic [31 c]. In order to investigate the effect of strain on the tropicity of DBA rings, tetrakisdehydro[12]annulene 19 was generated by photochemical [2+2] fragmentation of the propellane precursor 20 in an argon matrix at low temperature and was characterized by comparison of UV and IR spectra with those calculated by the DFT method (Scheme 1.4) [32]. Unfortunately, since the [12]annulene system 19 was too reactive to observe its tropicity in solution, the next higher homolog 21 was produced by the same method from the homologous precursor 22. The calculated 1H NMR chemical shift by the DFT method and experimental observations revealed that strained pentakisdehydro[14]annulene 21 was as diatropic as the corresponding tetrakisdehydro derivative 22. Interestingly, the 14-membered DBA ([14]DBA) 15 underwent topochemical polymerization induced by photoirradiation or by applying pressure (20 000 psi) [33]. From the packing diagram of the crystal, trans–trans geometry of the polymer chain was suggested. This represents the first example of the topochemical polymerization of a conjugated cyclic diyne derivative, although the reactions of acyclic butadiyne derivatives have been well known to form poly(butadiyne)s, which are known to exhibit interesting optical properties. In connection with the theoretical prediction regarding the tropicity of dehydrobenzo[14]annulenes, Haley prepared a number of monobenzo, dibenzo, and monothiopheno derivatives such as 23–27 and studied their diatropicities based on the 1H NMR chemical shifts [34]. As a result, fusion of a benzene ring to the dehydro[14]annulene ring induces stepwise loss of aromaticity with increasing number of the fused rings. The relative aromaticity (RA) was defined by the chemical shifts relative to those of a nonconjugated reference compound. The experimental RAexpt values are in accord with the theoretical values (RANICS) derived from the calculated NICS values. To further confirm the aromaticity of the [14]annulene system, hybrid systems 28 and 29 formed by fusion of the [14]DBA ring to dimethyldihydropyrene (DHP) unit were prepared [35]. The chemical shift of the probe methyl group in the DHP system of mono-DBA-fused compound 28 was compared with appropriate references. However, the effect of benzoannelation on aromaticity of the [14]DBA ring is not conclusive probably because of the small chemical shift change. BisDBA fused system 29 cannot be used for the examination of aromaticity because

7

8

1 Cyclophynes

Fig. 1.4

the effect of one DBA unit on the DHP probe should counteract that of the other DBA. Since Mitchell et al. developed reversible photochemical isomerization of fused DHPs with the corresponding [2.2]metacyclophane systems, the isomerization of 29 was examined, because its absorption band extended to 600–800 nm region, which may be used as “read out” without scrambling of information. However, photochemical isomerization was not successful because the cyclophane form reverts very easily to the starting material 29. A hybrid system 30 between [14]DBA 15 and [2.2]paracyclophane was synthesized in order to investigate the global transannular delocalization through the overall p system [36]. By comparison of the UV absorption with those of appropriate acyclic reference compounds, it was concluded that 30 sustained p-electron delocalization through the circuit composed of the DBA and [2.2]paracyclophane systems.

1.2 Orthocyclophynes and Related Systems

Scheme 1.4

Haley developed an in situ deprotection–Sonogashira coupling method using o(ethynylphenyl)butadiyne derivatives, which are otherwise too reactive to handle, as a butadiyne building block of DBAs, utilizing the reactivity difference between TMS and TIPS groups towards deprotection [37]. With this method, synthesis of a variety of dehydrobenzo[18]annulene ([18]DBA) derivatives was achieved, including the parent hydrocarbon 17 a (its first characterization) and derivatives such as 17 b and 17 c with electron-donating and accepting substituents at specific positions [38]. The first hyperpolarizabilities (b) of several donor–acceptor substituted derivatives of 17 a were determined in order to examine their second-order nonlinear optical properties [39]. The hyperpolarizability of some derivatives was comparable to that of the known reference compound, 4-dimethylamino-4’-nitrostilbene, while that of others was superior. This method was also applied to the synthesis of [16]DBA 16 [40], thiopheno[18]annulenes 31–33 [41], and the [18]DBAcrown ether hybrid 34 [42]. The latter compound was prepared with the intention of organizing the DBA units by formation of a supramolecular poly(rotaxane) structure by complexation with a polyammonium ion at the crown ether unit. Moreover, larger DBAs 35 and 36 with hexatriyne linkage(s) were prepared by the same protocol using an o-(ethynylphenyl)hexatriyne derivative [40]. The Haley method was also successfully adopted as the crucial step for the synthesis of a number of graphdiyne structures such as “bow tie” 37, “boomerang” 38, “diamond” 39, “half wheel” 40, and “trefoil” 41 [37, 43]. All the partial graphdiynes are stable up to ca. > 200 8C where they decomposed within a narrow range of temperatures as observed by DSC, forming black, insoluble materials which were difficult to characterize. The ultimate goal, a “perfect wheel” 42, is yet to be synthesized. Organometallic DBA derivatives have been studied extensively by Bunz et al. [3]. Ferrocene derivatives of [18]DBAs 43 a–b and [24]DBAs 44 have been synthesized by oxidative coupling of the corresponding diethynylferrocene [44]. Since 43 a and 43 b exhibited similar oxidation potentials, it was deduced that the through bond/ space interactions between the ferrocene units were small. Similarly, (CpCo)cyclobutadienyl derivatives 45 a–b and 46 a were obtained from a diethynylcyclobutadiene complex. Selective deprotection of the TMS/TIPS-protected tetraethynylcyclobutadiene complex followed by oxidative coupling furnished perethynylated [18]- and [24]annulene derivatives 45 c and 46 b, respectively, which would serve as building blocks for the organometallic graphdiynes.

9

10

1 Cyclophynes

Fig. 1.5

Ferroceno[14]annulenes 47 and 48 and (CpCo)cyclobutadieno[14]annulenes 49 and 50 were prepared to study the aromaticity/antiaromaticity of the ferrocene and (CpCo)cyclobutadiene rings, respectively [45]. Since the p electrons in the fused [14]annulene rings in 47 and 48 are more localized than the corresponding benzoannulenes, it is deduced that ferrocene is more aromatic than benzene. Moreover, according to the same ring current criterion, the (CpCo)cyclobutadiene ring is also more aromatic than benzene, even slightly so than ferrocene. The larger homologs, ferroceno[18]annulene 51 a–c and (CpCo)cyclobutadieno[18] annulene 52 a–d were prepared to explore the possible transformation of organometallic DBAs into onion- or tube-like carbon structures by their explosive decomposition [46], the observations first being reported by Vollhardt (Section 1.2.2). Among several organometallic [18]DBAs, only 51 b, 52 b, and 52 c underwent explosive decomposition. While the materials formed from the former two compounds were amorphous, that derived from 51 b exhibited distinct nanostructure as revealed by the TEM observations. In connection with the ultimate two-dimensional network, the synthesis of partial structures, cyclobutadieno “bow tie” 53, cyclobutadieno and ferroceno “half square” 54 and 55, and cyclobutadieno “three-quarters square” 56, has been accomplished [47]. The target, “perfect square” 57, has not so far been prepared.

1.2 Orthocyclophynes and Related Systems

Fig. 1.6

11

12

1 Cyclophynes

Fig. 1.7

Fig. 1.8

1.2 Orthocyclophynes and Related Systems

Fig. 1.9

1.2.2

Nonplanar Orthocyclophynes

Orthocyclophynes constructed from more than four benzene rings adopt nonplanar conformations and have been investigated in view of their structure and the chirality of twisted p-systems and of their unique properties caused by the twisted structures. Indeed, the X-ray crystal structure of [24]orthocyclophanetetrayne 58 revealed that it possessed twisted, saddle-like conformation of the macrocyclic ring [15]. Moreover, the crystal of the cyclic hexamer 59 adopts a helical conformation with the principal axis extending through the two peripheral rings and between the two eclipsed tolan moieties [15]. The solid state conformation of the even larger decamer 60 is irregular and no longer helical [48]. Butadiyne-bridged macrocycle 61 was also shown to possess a twisted macrocyclic ring [49]. The mixed ethynylene–butadiynylene macrocycles 62 a–b, 63, and 64, the latter two being prepared using Haley’s protocol, were also shown to adopt nonplanar saddle-like conformations [37, 50]. The barrier for flipping of 62 a was estimated by a variable-temperature NMR study to be 9.3 kcal mol–1. Perhaps the most exciting finding, reported by Vollhardt et al., for these highly unsaturated macrocycles is that 62 b exploded violently with a flash of orange light to yield ordered carbon materials which were shown by TEM to possess tube- and onion-like structures [50]. Moreover, not only the corresponding cobalt complex 65 but also simple acyclic complexes such as 66

13

14

1 Cyclophynes

were shown to form carbon onions with diameters of 100–800 Å and multiwalled nanotubes with diameters of 60–400 Å upon annealing at 800 8C [51]. A number of twisted macrocycles were synthesized by Fallis et al. by dimerization of terminal alkynes using copper(I)-mediated oxidative coupling. In certain cases, however, intramolecular cyclization competed with the dimerization. Thus in the case of diethynylterphenyl 67, both intramolecular cyclization and dimerization took place, yielding monomers 68 and dimers 69, respectively, whose distribution depended on substituents and reaction conditions (Scheme 1.5) [52]. The carboxy derivative of monomer 68 was shown by X-ray structural analysis to possess highly strained triple bonds with distortion angles of 16.3–16.78. In contrast, similar oxidative coupling of monomer 70 a–c, whose backbone was extended compared to that of terphenyl 67, resulted in the selective formation of cyclic dimers 71 a–c. The X-ray crystal structure of 71 a revealed that the central benzene rings

Fig. 1.10

1.2 Orthocyclophynes and Related Systems

were aligned to a slightly offset but overlapping geometry with inter-ring distance of 3.57 Å. It is noteworthy that the dialkoxy derivative 71 c exhibits liquid crystalline behavior, suggesting the occurrence of a unique liquid crystalline phase based on twisted p frameworks. Similarly, the isomeric cyclophynes 72 and 73 with meta-bridged central benzene rings were prepared by oxidative dimerization [53]. These isomers did not interconvert because of steric constraints. Large octatetrayne-bridged cyclophynes 74 [54] and 75 [55] were prepared by the in situ deprotection–oxidative coupling method without isolating unstable phenylbutadiyne intermediates. The twisted thiophenophynes 76 a–c were prepared by Marsella by intermolecular Sonogashira coupling of the terminal alkyne and iodothiophene derivatives [56]. Since the efficiency of the cyclization depends on the substituents of the central benzene ring, the mixed cyclophyne 76 b being formed most efficiently, the presence of phenyl/perfluorophenyl interactions was invoked. Baxter prepared the pyridine analog 77 of twisted cyclophyne 71 a and investigated its properties as a potential fluorescence sensor for heavy metals such as Pd(II) and Hg(I) [57]. He also prepared the macrocycle 78 and its isomer 79 having central bipyridine rings [58]. While molecular mechanics calculations for 78 suggest that it possesses a twisted conformation similar to that of 71 a, molecular modeling of 79 by semiempirical molecular orbital calculations suggests the presence of a few minimum energy conformers. The former 78 binds Co(II), Ni(II), Zn(II), and Cu(II) ions as detected by UV absorption and fluorescence, while its isomer 79 binds only Cu(II) and Ag(I). Phenanthroline-based

Fig. 1.11

15

16

1 Cyclophynes

Scheme 1.5

macrocycle 80 was prepared using Cu(II)-templated oxidative coupling of the monomer unit [59]. The initial product was a copper(II) complex from which free macrocycle 80 was liberated by treatment with KCN. The barrier for twisting of the free macrocycle 80 and its Cu(II) complex was determined by variable temperature NMR spectra to be 9.4 and 13.6 kcal mol–1, respectively. The [6]helicene-based double-helical cyclophyne 81 was prepared by Fox in enantiomerically pure form and its UV and CD spectra have been investigated [60]. Since the two helicene units are orthogonal with respect to each other, electronic delocalization through the entire p system seems to be inhibited. Double-helical binaphthyl-based cyclophynes 82 [61] (enantiomerically pure) and bithiophenebased scaffold for double-helical ladder polymers 83 [62] (racemic) were prepared by the Sonogashira reaction. Diederich prepared chiral binaphthol-based cyclophynes such as 84 and 85 by oxidative coupling of the monomer units and studied their binding properties toward pyranoside derivatives [63]. The trimeric macrocycle 84 forms 1 : 1 complexes with glucopyranoside derivatives in CDCl3 with moderate discrimination of not only the enantiomers but also the anomers. On the other hand, the tetrameric phosphate receptor 85 was capable of binding a pyranose derivative even in a protic solvent containing up to 20% v/v CD3OD in CD3CN.

1.3 Metacyclophynes and Related Systems

Fig. 1.12

1.3

Metacyclophynes and Related Systems

The general structures for phenylacetylene macrocycles 86 and diethynylbenzene macrocycles 87 may represent prototypical metacyclophynes. In general, when the number (n) of the constituent units is small (n = 3), the macrocycles must be unstable because of bond angle distortion. When n is large (n ³ 7), the macrocycles should adopt nonplanar conformations. On the other hand, medium-sized macrocycles (n = 4–6) should be stable and adopt approximate planar conformations. Metacyclophynes have been extensively studied during the last decade, because they have a persistent flat shape and can functionalize at both the interior and/or exterior of the macrocyclic framework. Aspects of their supramolecular chemistry that have been studied include guest-binding ability utilizing internal binding sites, construction of supramolecular structures using peripheral functionalities, self-as-

17

18

1 Cyclophynes

Fig. 1.13

Fig. 1.14

sociation in solution, liquid crystalline phase, and solid state, and organization at various interfaces [1, 5, 7]. Most of the syntheses of this class of compounds have been achieved by the Sonogashira type C(sp2)–C(sp) bond formation and by the Hay or Eglinton methods

1.3 Metacyclophynes and Related Systems

for C(sp)–C(sp) bond formation. By using a combination of selective protection/ masking of a terminal acetylene moiety by TMS group and aryl iodide by triazene group, respectively, Moore succeeded in the size-selective synthesis of a variety of nanoscale macrocycles such as 88 [64]. An attempt at the solid phase synthesis of a derivative of the phenylacetylene macrocycle 86 (n = 6) was not successful presumably because of catenation of the macrocycle to the polymer support [65]. One-step synthesis based on alkyne metathesis was investigated by Bunz to furnish derivatives of 86 (n = 6), albeit in low yields [66]. A comprehensive survey regarding the synthetic strategy for construction of unsaturated macrocycles was reported recently by Schlüter [7]. Noncovalent template-directed, size-selective synthesis of porphyrin-containing metacyclophynes such as 89 and 90 has been developed by Saunders using coordination of pyridine derivatives of appropriate size and geometry to metalloporphyrin unit of the substrates [67]. This method was employed to prepare even larger porphyrin macrocycles such as the 108-membered 91 [68] and the 144-membered 92 [69], the largest shape-persistent macrocycle so far known. On the other hand, covalent template-directed method was employed by Höger et al. for the synthesis of ethynylene-butadiynylene macrocycles such as 108 a described later, giving the desired macrocycle more selectively than the macrocyclization without using a template [70]. A new method for oxidative coupling of a terminal acetylene was developed by Bäuerle based on efficient formation of a self-assembled platinum intermediate 93, which then furnished butadiyne-bridged terthiophenophyne 94 by oxidative elimination of the metal (Scheme 1.6) [71]. Highly strained small metacyclophynes are not accessible by these coupling reactions. Thus phenylacetylene trimer 95 and tetramer 96 a–b were prepared by bromination–dehydrobromination sequences of the corresponding cyclophenes, which were obtained by using the McMurry reaction [72]. This method was applied to the synthesis of twisted biphenyl-based cyclophanediynes 97 a–b [73]. Xray structural analysis of 95, 96 a and 97 b revealed that the triple bond of the former was deformed by 21.48, while the largest distortion of the latter two was only 12.38 and 9.58, respectively. Other methods for the synthesis of metacyclophynes include the McMurry coupling for dienetetrayne 98 [74] and the Williamson ether synthesis for oxygen- or sulfur-bridged compounds 99 a and 99 b [75]. Moore and coworkers synthesized acetylene-bridged [26]metacyclophynes 86 having substituents at the periphery of the desired benzene ring and developed pioneering works on the properties derived from their association by weak noncovalent interactions [1]. Namely, self-association behavior of 100 a–f was investigated in both apolar (chloroform) and polar solvents such as acetone, acetonitrile, and THF by 1H NMR spectroscopy and vapor pressure osmomeric (VPO) analyses [76]. In chloroform, compounds 100 a–d self-associate to form dimeric aggregate caused by a p–p stacking interaction, while in polar solvents 100 d forms higher aggregates owing to solvophobic interactions. The peripheral substituents play crucial role in the association behavior in both apolar and polar solvents. In chloroform, the ester-substituted compound 100 a exhibits the stronger tendency to self-associate than the mixed ester–alkoxy hybrid derivatives 100 b–c, presum-

19

20

1 Cyclophynes

Fig. 1.15

1.3 Metacyclophynes and Related Systems

Scheme 1.6

ably because of favorable polar–p interactions in the former which bears only electron-withdrawing groups. Similarly, in polar solvents, ester 100 d aggregates much more strongly than the ethers 100 e–f; the reason for the substituent effect on the solvophobically driven aggregation is uncertain. The imine derivatives 101 also stacked in polar solvents to form aggregates, indicating that this group does not disturb solvophobically driven p–p stacking [77]. In acetone, 101 tends to form a dimer more strongly than higher aggregates, presumably because of electrostatic interactions between the imine groups. Yamaguchi et al. synthesized chiral ana-

Fig. 1.16

21

22

1 Cyclophynes

logs 102 incorporating enantiomerically pure chiral benzo[c]phenanthrene units and investigated the stereoselectivity in their self-association in solution [78]. Enantiomerically pure macrocycles aggregate more strongly than the corresponding racemic macrocycle and the M,M,M isomers aggregate more strongly than M,P,M isomer, indicating that the macrocycles differentiate between enantiomers and diastereomers. [26]Metacyclophynes 100 g–i bearing alkoxy substituents, however, form discoid nematic liquid crystalline phases in a relatively wide range, at temperatures higher than 120 8C [79]. In solid state, phenol derivative 100 j forms channels with diameter of about 9 Å, which are filled with solvent molecules (12 ethanol, 6 methanol, and 3 water), because of the two-dimensional hydrogen bonds [80]. Each layer of 100 j stacks in ABCABC sequence. On the basis of the surface pressure-area isotherm, it was deduced that amphiphilic macrocycles such as 100 k and 100 l with segregated hydrophilic and hydrophobic groups formed Langmuir–Blodgett

Fig. 1.17

1.3 Metacyclophynes and Related Systems

Fig. 1.18

(LB) layers adopting the edge-on orientation [81]. The layer formed from 100 k was transferred onto fused silica and Si(100) surfaces, forming a well-organized two-dimensional array with slightly tilted edge-on configuration. [26]Metacyclo100 m possessing interior methoxy groups was synthesized by Kawase and Oda and was shown to bind ammonium ions selectively, in spite of the large cavity relative to the ion [82]. X-ray structural analysis of 100 m indicated that the distance between the diagonal oxygen atoms was 8.1 Å. In the crystal, it forms a displaced sheet which stacks to form open channels with internal pores of 4.5 Å. By contrast, [24]metacyclophyne 96 b with a smaller binding site binds a variety of alkali metal cations with considerably large association constants [72 c]. Association behavior of hexagonal butadiyne-bridged cyclophynes 104 a–b, “big brothers” of 100 a and 100 d, and its square analogs 103 a–b have been investigated in detail by Tobe et al. using 1H NMR and VPO in both apolar and polar solvents [83]. In polar solvents, these aggregate stronger than in chloroform due to the solvophobic interactions, with association constants for the formation of higher aggregates several times larger than those for dimerization, suggesting a nucleation mechanism. They also self-associate more strongly in aromatic solvents such as toluene and o-xylene than in chloroform. Moreover, 104 a–b tend to aggregate more readily than the corresponding acetylene-bridged cyclophynes 100 a and 100 d, owing to the withdrawal of the electron density from the aromatic rings by the butadiyne moieties. Cyano derivative 104 c and the pyridine analogs 105 and 106 were also prepared [84]. Interestingly, these form heteroaggregates with the

23

24

1 Cyclophynes

corresponding cyclophynes 103 a and 104 a due to dipole–dipole interactions, with distinct recognition of the ring size of their partners. The macrocycle 104 c, 105, and 106 bearing an interior binding group bind organic cations such as tropylium and guadinium ions to form both 1 : 1 and 2 : 1 (host : guest) complexes. The hybrids 107 a–b between 100 a and 104 a, prepared by Tour, aggregate in chloroform, albeit only weakly [85]. Since 107 b does not possess electron-withdrawing groups, the presence of hydrogen bonds between the hydroxy groups was invoked. Much larger macrocycles 108 a–f, which possess substituents not only at the corner phenyl group but also on the rotating phenyl groups placed between the corners, have been synthesized by Höger et al. [5, 86]. The characteristic feature of this system is that the amphiphilic substituents on the rotating rings can modify the properties of macrocycles in response to the external environment. Thus in the crystal of 108 b obtained from pyridine, two alkoxy groups are pointing outward and the cavity is self-occupied by the other two alkoxy groups as well as the solvent (6 pyridine molecules), two of which are hydrogen-bonded to the internal phenolic groups [87]. On the other hand, in the crystal obtained from THF, all alkoxy groups point outwards, forming distinct channels which are filled with the solvent (12 THF molecules), four of which are hydrogen bonded to the hydroxy groups of 108 b. The internal hydroxy groups of 108 b also act as the binding functionality in complexation with a tetraamine guest, which fits the cavity of 108 b to form maximum hydrogen bonds [88]. Interestingly, macrocycle 108 c bearing octadecyl groups on the rotating phenyl rings formed a nematic mesophase between 185–207 8C [89]. Since the related macrocycle 108 d having octadecyl groups on the corner did not form liquid crystals and, in the X-ray structure of 108 c, all alkyl chains self-occupied the cavity, it was deduced that such a structure, with flexible side groups pointing inwards, must be responsible for the formation of this novel thermotropic mesophase. Although most of the macrocycles of this type did not associate in solution, the heavily alkylated derivative 108 e formed solvophobically driven aggregates in dichloromethane/hexane mixture [90]. More strikingly, cyclophyne 108 f with two poly(styrene) chains of narrow dispersity, stacked with each other in cyclohexane to form long tubular aggregates up to 60 nm length [91]. With regard to the organization on surfaces, compound 108 e was reported to adsorb on the surface of highly oriented pyrolytic graphite (HOPG) in highly ordered face-on configuration, which was observed by STM [90]. Macrocycles 109 a–c having bipyridine units were prepared by Schlüter et al. [92]. X-ray structural analysis revealed that these form layered structures and channels filled mainly with solvent molecules and partly with the side chains. The stacking sequences (109 a AB, 109 b ABC, and 109 c ABCD pattern) differed and depend on the direction of the flexible side chain in the crystals. Ruthenium complexes of 109 a and 109 c were prepared, in which both metal atoms were located in the exocyclic position. Moreover, a number of terpyridine-based large macrocycles such as 110 a–c and 111 a–c were synthesized [93]. In analogy with 109 a–c, crystals of 110 a–c form layered structures and channels filled with solvent molecules and flexible side chains. Ordered monolayers of 110 c adsorbed on HOPG

1.3 Metacyclophynes and Related Systems

Fig. 1.19

were observed by STM although the resolution of the images was not high presumably because of low attraction of the substrate on the surface. With regard to the host–guest chemistry of metacyclophynes, Yoshida et al. prepared [24]metacyclophanetetrayne 112 containing two pyridine rings facing each other [94]. By complexation with SbCl5, the fluorescence of 112 was enhanced, while it did not bind SbCl3 and zinc halides. Oxygen-bridged macrocycle 113 formed a complex with C60 fullerene with a moderate binding constant [95]. In contrast, the exocyclic pyridine-containing macrocycles 114 a and 114 b were pre-

25

26

1 Cyclophynes

Fig. 1.20

pared with the aim of constructing supramolecular structures by chelating them to transition metals [96]. The related pyridinophanes 115 a–c formed complexes with ruthenium porphyrin [97]. The same macrocycle 115 c self-assembled with platinum(II) to form a double-decker complex, which in the crystal formed channels filled with two solvent molecules (ClCH2CH2Cl) [98]. From the same point of view, a phenanthrolin-based macrocycle 116 was prepared by Schmittel [99]. By self-assembly of two molecules of 116 with four copper(I) ions and two molecules of linear rod-like units, which also possessed two phenanthrolin units, a large structure called a nanobox with > 5000 Å3 void, containing about 240 molecules of acetone, was prepared. Redox active macrocycles 117 a–b were prepared and their oxidation potentials were determined [100]. Since both compounds exhibited three reversible oxidation

1.3 Metacyclophynes and Related Systems

Fig. 1.21

potentials, one of which was a 2-electron and one a 3-electron oxidations, respectively, these molecules can accommodate many holes with independent potentials. Potentially redox active cyclophynes having carbazole rings 118 [101] and oligothiophene moieties 119 a–b [102] were also prepared. Godt utilized the rigid shape of the meta phenylene-ethynylene-substituted benzene unit to prepare extremely large macrocycles such as 120 and catenane 121, which would not otherwise be easily accessible [103]. Some of the related large macrocycles exhibited thermotropic nematic and smectic mesophases.

27

28

1 Cyclophynes

Fig. 1.22

1.4

Paracyclophynes and 1,3,5-Bridged Cyclophynes

Since paracyclophynes adopt a cylindrical shape with curved p-conjugated systems, they have been studied in view of the noncovalent interactions of curved psystems and strain of the triple bonds, particularly in the small p-systems. In contrast to ortho- and metacyclophynes, since it is difficult to synthesize cyclophynes of this type by transition metal-mediated coupling reactions, a number of methodologies have been developed to construct the macrocyclic framework efficiently. [2n]Paracyclophynes (n = 6, 7, 8, 9) 122 a–d were prepared by Kawase and Oda by bromination/dehydrobromination of the corresponding olefinic paracyclophanes, which were in turn obtained by using the McMurry reaction [104]. As the ring size becomes smaller, notable downfield shifts of the sp carbon signals were observed in the 13C NMR spectra due to increasing strain. Since semiempirical mo-

1.4 Paracyclophynes and 1,3,5-Bridged Cyclophynes

Fig. 1.23

lecular orbital calculations predict that 122 a and 122 c should possess nanosize cavities with diameters of 13.1 and 17.4 Å, respectively, the formation of novel inclusion complexes with relatively large nonpolar guests was expected. Indeed, 122 a formed a 1 : 1 inclusion complex with hexamethylbenzene in which the guest molecule was situated at the center of the cavity tilted 98 from the horizontal plane of the macrocyclic ring, as revealed by X-ray analysis [105]. On the other hand, 122 c formed crystals including four molecules of toluene in the cavity which was more elliptical than that of 122 a. A pair of toluene molecules is located between two molecules of 122 c with short interatomic distances, indicating that they act as an adhesive by CH–p nteractions. Since the cavity of 122 a may fit C60 fullerene, 1 : 1 complex formation of 122 a with C60 was investigated [106]. Indeed, 122 a binds C60 strongly even in benzene with a binding constant of 1.6 ´ 104 L mol–1. X-ray crystal structural analysis of the inclusion complex of 122 a

29

30

1 Cyclophynes

with a C60 derivative revealed that the C60 cage was not deeply embedded in the cavity but floating on the bowl-shaped conformation of 122 a, because the cavity is a little too small for C60 to be fully embedded. Oxygen-bridged paracyclophyne 123 also binds C60 in toluene but with less strength [95]. Butadiyne-bridged [46]cyclophyne 124 was synthesized by the group of Tsuji using photochemical valence isomerization of the corresponding Dewar benzene valence isomer 125 [107]. Semiempirical molecular orbital calculations predict that it has a cylindrical cavity with a diameter of about 15 Å. Although the individual triple bonds were not much deformed (bending angle: about 98), compound 124 was air-sensitive and decomposed gradually within several days. Related metaparacyclophyne 126 and orthoparacyclophyne 127 were also successfully prepared by photochemical valence isomerization of the corresponding Dewar benzene isomers 128 and 129 [108]. While both diastereomers of 128 underwent smooth isomerization to 126, only the dl-isomer of 129 was converted into 127 and meso-129 decomposed on photolysis. In view of the lability of 124, hexatriyne-bridged [63]paracyclophyne 130 would not be stable enough for isolation. Accordingly, its formation was detected by la-

Fig. 1.24

1.4 Paracyclophynes and 1,3,5-Bridged Cyclophynes

ser-desorption time-of-flight mass spectroscopy (LDTOF MS) of its precursor 131 having propellane units by [2+2] cycloreversion of the cyclobutene rings [109]. Conversion of the ions of 130 into the carbon cluster C36 has been investigated in connection with the related study for paracyclophanepolyynes 136 a–b described below. Cyclophynes of [n.n]paracyclophane type would become more reactive as the number of triple bonds in the bridge increases, because of the enhanced reactivity of conjugated polyynes and of the bond angle strain. [6.6]Paracyclophanetetrayne 132, prepared by novel dimerization of a quinodimethane type intermediate, possesses relatively strained triple bonds with bending angles of 8–118 but it is stable enough for full characterization [110]. Dithia[7.7]cyclophanetetrayne 133 is much less strained [75 b]. On the other hand, fully unsaturated [8.8]paracyclopha-

Fig. 1.25

31

32

1 Cyclophynes

neoctayne 134 eluded isolation in oxidative decomplexation of its precursor, the cobalt complex 135 [111]. The formation of the higher analog, [12.12]paracyclophanedodecayne 136 a, was only indicated in the negative mode LDTOF MS of its precursor 137 a, from which stepwise loss of indan fragments took place by [2+2] cycloreversion [112]. Interestingly, the corresponding chloro derivative 136 b, generated similarly from 137 b, lost chlorine atoms to form ultimately the C36 anion, a carbon cluster attracting particular interest. These observations suggest possible use of reactive polyynes for size-selective, and hopefully geometry-selective, formation of carbon clusters in connection with the formation of C60 from highly unsaturated 1,3,5-bridged cyclophynes described below, even though the structure of the above C36 anion has not yet been confirmed. To construct the three-dimensional structure of 1,3,5-bridged cyclophynes, the length of the bridge and proper choice of precursor turned out to be critical owing to competing intra- versus intermolecular reactions. For example, while trithia[7.7.7](1,3,5)cyclophanehexayne 138 was obtained in a good yield by intramolecular cyclization of dimeric precursor 139, attempted coupling from monomeric units 140 and 141 failed (Scheme 1.7) [113]. Similarly, Cu(II)-mediated oxidative coupling of tris(ethynylphenyl) compounds 142 and 144 afforded highly strained metacyclophynes 143 [114] and 145 [115], respectively, by intramolecular coupling, rather than giving the respective cage-shaped dimers (Scheme 1.8). On the other hand, large three-dimensional macrocycles were readily obtained by oxidative dimerization of the corresponding monomer units. The first member of this class, [16.16.16](1,3,5)cyclophanehexa-eneoctadecayne 146 was synthesized

Fig. 1.26

1.4 Paracyclophynes and 1,3,5-Bridged Cyclophynes

Scheme 1.7

Scheme 1.8

by Rubin et al. [116]. Interestingly, in the ion cyclotron resonance mass spectrum of 146, dehydrogenation down to C60H–14 formed by the loss of four hydrogen atoms was observed. Moreover, Rubin and Tobe prepared independently the precursors 147 [117] and 148 a–b [118] to generate C60H6 (149 a) or C60Cl6 (149 b), which would form C60 fullerene efficiently by cyclization of reactive polyyne chain with loss of six hydrogen or chlorine atoms. Indeed, in the LDTOF MS of 147 and 148 a–b, not only the formation of 149 a–b by extrusion of carbon monoxide and indan, respectively, but also their further transformation into C60 ion were observed, indicating a possible method for size- and shape-selective synthesis of carbon clusters of the fullerene family from highly unsaturated cage-like cyclophynes (Scheme 1.9). This method was also applied to the generation of diazafullerene C58N2, albeit in low efficiency of dehydrogenation from the precursor 150 [119]

33

34

1 Cyclophynes

Fig. 1.27

Scheme 1.9

1.5 Concluding Remarks

Fig. 1.28

and the formation of C78H18 (151), a possible precursor to higher fullerene like C78 [120]. Extremely large, three-dimensional, phenylacetylene macrocycles 152 and 153 were successfully prepared by Moore et al. [121]. The final step of the synthesis proceeded in high yields were intramolecular Sonogashira coupling of bis-seco intermediates, which were manipulated using the same masking/protention protocol employed for the synthesis of planar macrocycles such as 88.

1.5

Concluding Remarks

In connection with the concurrent growth of chemistry of carbon-rich materials with opto-electronic functions, triggered by the discovery of fullerenes and nanotubes, and the emerging supramolecular chemistry of nano-size materials, enormous progress, which is much more than that envisaged as a revival of annulene chemistry, has been achieved during the last decade. The unique structural and electronic properties of cyclophynes, and their facility for manipulation, ensure that their potential to provide new materials will continue.

35

36

1 Cyclophynes

1.6

Acknowledgement

This work was supported by CREST, Japan Science and Technology Agency (JST).

1.7

References 1

2

3 4 5 6 7 8 9

10 11

12 13 14 15

16

(a) J. S. Moore, Acc. Chem. Res. 1997, 30, 402–413; (b) D. Zhao, J. S. Moore, Chem. Commun. 2003, 807–818. M. M. Haley, Synlett 1998, 557–565; (b) M. M. Haley, J. J. Pak, S. C. Brand, Top. Curr. Chem. 1999, 201, 81–130; (c) J. A. Marsden, G. J. Palmer, M. M. Haley, Eur. J. Org. Chem. 2003, 2355–2369. U. H. F. Bunz, Top. Curr. Chem. 1999, 201, 131–161. U. H. F. Bunz, Y. Rubin, Y. Tobe, Chem. Soc. Rev. 1999, 28, 107–119. S. Höger, J. Polym. Sci. Part A: Polym. Chem. 1999, 37, 2685–2698. W. J. Youngs, C. A. Tessier, J. D. Bradshaw, Chem. Rev. 1999, 99, 3153–3180. C. Grave, A. D. Schlüter, Eur. J. Org. Chem. 2002, 3075–3098. S. Toyota, M. Iyoda, F. Toda, Annu. Rep. Prog. Chem., Sect. B 2002, 98, 359–407. G. J. Bodwell, T. Satou, Angew. Chem. 2002, 114, 4175–4178; Angew. Chem. Int. Ed. 2002, 41, 4003–4006. R. H. Baughman, H. Eckhardt, M. Kertesz, J. Chem. Phys. 1987, 87, 6687–6699. (a) N. Narita, S. Nagai, S. Suzuki, K. Nakano, Phys. Rev. B 1998, 58, 11009; (b) N. Narita, S. Nagai, S. Suzuki, K. Nakano, Phys. Rev. B 2000, 62, 11146. N. Narita, S. Nagai, S. Suzuki, Phys. Rev. B 2001, 64, 245408. Y. Zhou, S. Feng, Solid State Commun. 2002, 122, 307–310. J. N. Grima, K. E. Evans, Chem. Commun. 2000, 1531–1532. D. Solooki, J. D. Bradshaw, C. A. Tessier, W. J. Youngs, R. F. See, M. Churchill, J. D. Ferrara, J. Organomet. Chem. 1994, 470, 231–236. M. Iyoda, A. Vorasingha, Y. Kuwatani, M. Yoshida, Tetrahedron Lett. 1998, 39, 4701–4704.

17

18 19

20 21

22

23

24

25

J. M. Kehoe, J. H. Kiley, J. J. English, C. A. Johnson, R. C. Petersen, M. M. Haley, Org. Lett. 2000, 2, 969–972. O. S. Miljanic, K. P. C. Vollhardt, G. D. Whitener, Synlett 2003, 29–34. D. Solooki, J. D. Bradshaw, C. A. Tessier, W. J. Youngs, Organometallics 1994, 13, 451–455. M. Sonoda, Y. Sakai, T. Yoshimura, K. Hirose, Y. Tobe, unpublished results. (a) J. D. Ferrara, C. Tessier-Youngs, W. J. Youngs, J. Am. Chem. Soc. 1985, 107, 6719–6721; (b) J. D. Ferrara, C. Tessier-Youngs, W. J. Youngs, Organometallics 1987, 6, 676–678; (c) A. Djebli, J. D. Ferrara, C. Tessier-Youngs, W. J. Youngs, J. Chem. Soc., Chem. Commun. 1988, 548–549; (d) J. D. Ferrara, A. Djebli, C. Tessier-Youngs, W. J. Youngs, J. Am. Chem. Soc. 1988, 110, 647–649; (e) J. D. Ferrara, C. Tessier-Youngs, W. J. Youngs, Inorg. Chem. 1988, 27, 2201– 2202. (a) J. D. Ferrara, C. Tessier-Youngs, W. J. Youngs, J. Am. Chem. Soc. 1988, 110, 3326–3327; (b) J. D. Ferrara, A. A. Tanaka, C. Fierro, C. A. TessierYoungs, W. J. Youngs, Organometallics 1989, 8, 2089–2098. (a) J. D. Kinder, C. A. Tessier, W. J. Youngs, Synlett 1993, 149–150; (b) W. J. Youngs, J. D. Kinder, J. D. Bradshaw C. A. Tessier, Organometallics 1993, 12, 2406–2407; (c) D. Zhang, C. A. Tessier, W. J. Youngs, Chem. Mater. 1999, 11, 3050–3057. M. Iyoda, K. Fuchigami, A. Kusaka, T. Yoshida, M. Yoshida, H. Matsuyama, Y. Kuwatani, Chem. Lett. 2000, 860–861. H. Wandel, O. Wiest, J. Org. Chem. 2002, 67, 388–393.

1.7 References 26

27 28

29

30

31

32

33

34

35

36

37

C. Eickmeier, H. Junga, A. J. Matzger, F. Scherhag, M. Shim, K. P. C. Vollhard, Angew. Chem. 1997, 109, 2194– 2199; Angew. Chem. Int. Ed. Engl. 1997, 36, 2103–2108. A. J. Matzger, M. Shim, K. P. C. Vollhardt, Chem. Commun. 1999, 1871–1872. J. D. Tovar, N. Jux, T. Jarrosson, S. I. Khan, Y. Rubin, J. Org. Chem. 1997, 62, 3432–3433. G. J. Palmer, S. R. Prakin, J. E. Anthony, Angew. Chem. 2001, 113, 2577– 2580; Angew. Chem. Int. Ed. 2001, 40, 2509–2512. T. Nishinaga, N. Nodera, Y. Miyata, K. Komatsu, J. Org. Chem. 2002, 67, 6091– 6096. (a) J. Jusélius, D. Sundholm, Phys. Chem. Chem. Phys. 2001, 3, 2433–2437; (b) I. Alkorta, I. Rozas, J. Elguero, Tetrahedron 2001, 57, 6043–6049; (c) A. J. Matzger, K. P. C. Vollhardt, Tetrahedron Lett. 1998, 39, 6791–6794. (a) Y. Tobe, I. Ohki, M. Sonoda, H. Niino, T. Sato, T. Wakabayashi, J. Am. Chem. Soc. 2003, 125, 5614–5615; (b) I. Hisaki, T. Eda, M. Sonoda, Y. Tobe, Chem. Lett. 2004, 33, 620–621. K. P. Baldwin, A. J. Matzger, D. A. Scheiman, C. A. Tessier, K. P. C. Vollhardt, W. J. Youngs, Synlett 1995, 1215– 1218. (a) H. S. Blanchette, S. C. Brand, H. Naruse, T. J. R. Weakley, M. M. Haley, Tetrahedron 2000, 56, 9581–9588; (b) A. J. Boydston, M. M. Haley, Org. Lett. 2001, 3, 3599–3601; (c) A. J. Boydston, M. M. Haley, R. V. Williams, J. R. Armantrout, J. Org. Chem. 2002, 67, 8812–8819. (a) D. B. Kimball, M. M. Haley, R. H. Mitchell, T. R. Ward, Org. Lett. 2001, 3, 1709–1711; (b) D. B. Kimball, R. H. Mitchell, T. R. Ward, S. Bandyopadhyay, R. V. Williams, J. R. Armantrout, M. M. Haley, J. Org. Chem. 2002, 67, 8798–8811. A. J. Boydston, L. Bondarenko, I. Dix, T. J. R. Weakley, H. Hopf, M. M. Haley, Angew. Chem. 2001, 113, 3074–3077; Angew. Chem. Int. Ed. 2001, 40, 2986–2989. (a) M. M. Haley, S. C. Brand, J. J. Pak, Angew. Chem. 1997, 109, 863–866; Angew. Chem. Int. Ed. Engl. 1997, 36, 836–838;

38 39

40

41 42

43 44

45

46

47

48

49

(b) W. B. Wan, S. C. Brand, J. J. Pak, M. M. Haley, Chem. Eur. J. 2000, 6, 2044–2052. J. J. Pak, T. J. R. Weakley, M. M. Haley, J. Am. Chem. Soc. 1999, 121, 8182–8192. A. Sarkar, J. J. Pak, G. W. Rayfield, M. M. Haley, J. Mater. Chem. 2001, 11, 2943–2945. (a) M. M. Haley, M. L. Bell, J. J. English, C. A. Johnson, T. J. R. Weakley, J. Am. Chem. Soc. 1997, 119, 2956–2957; (b) M. L. Bell, R. C. Chiechi, C. A. Johnson, D. B. Kimball, A. J. Matzger, W. B. Wan, T. J. R. Weakley, M. M. Haley, Tetrahedron 2001, 57, 3507–3520. A. Sarkar, M. M. Haley, Chem. Commun. 2000, 1733–1734. J. J. Pak, T. J. R. Weakley, M. M. Haley, D. Y. K. Lau, J. F. Stoddart, Synthesis 2002, 1256–1260. W. B. Wan, M. M. Haley, J. Org. Chem. 2001, 66, 3893–3901. (a) M. Altmann, J. Friedrich, F. Beer, R. Reuter, V. Enkelmann, U. H. F. Bunz, J. Am. Chem. Soc. 1997, 119, 1472–1473; (b) U. H. F. Bunz, G. Roidl, M. Altmann, V. Enkelmann, K. D. Shimizu, J. Am. Chem. Soc. 1999, 121, 10719–10726. (a) M. Laskoski, W. Steffen, M. D. Smith, U. H. F. Bunz, Chem. Commun. 2001, 691–692; (b) M. Laskoski, M. D. Smith, J. G. M. Morton, U. H. F. Bunz, J. Org. Chem. 2001, 66, 5174–5181. M. Laskoski, W. Steffen, J. G. M. Morton, M. D. Smith, U. H. F. Bunz, J. Am. Chem. Soc. 2002, 124, 13814–13818. (a) M. Laskoski, G. Roidl, M. D. Smith, U. H. F. Bunz, Angew. Chem. 2001, 113, 1508–1511; Angew. Chem. Int. Ed. 2001, 40, 1460–1463; (b) M. Laskoski, W. Steffen, J. G. M. Morton, M. D. Smith, U. H. F. Bunz, Angew. Chem. 2002, 114, 2484–2488; Angew. Chem. Int. Ed. 2002, 41, 2378–2382. K. P. Baldwin, R. S. Simons, J. Rose, P. Zimmermann, D. M. Hercules, C. A. Tessier, W. J. Youngs, J. Chem. Soc., Chem. Commun. 1994, 1257–1258. L. Guo, J. D. Bradshaw, C. A. Tessier, W. J. Youngs, J. Chem. Soc., Chem. Commun. 1994, 243–244.

37

38

1 Cyclophynes 50

51

52

53 54

55

56 57 58

59

60 61

62 63

64

65 66

P. I. Dosa, C. Erben, V. S. Iyer, K. P. C. Vollhardt, I. M. Wasser, J. Am. Chem. Soc. 1999, 121, 10430–10431. R. Boese, A. J. Matzger, K. P. C. Vollhardt, J. Am. Chem. Soc. 1997, 119, 2052–2053. (a) S. K. Collins, G. P. A. Yap, A. G. Fallis, Angew. Chem. 2000, 112, 393–396; Angew. Chem. Int. Ed. 2000, 39, 385–388; (b) S. K. Collins, G. P. A. Yap, A. G. Fallis, Org. Lett. 2000, 2, 3189–3192. M. A. Heuft, S. K. Collins, A. G. Fallis, Org. Lett. 2003, 5, 1911–1914. M. M. Haley, M. L. Bell, S. C. Brand, D. B. Kimball, J. J. Pak, W. B. Wan, Tetrahedron Lett. 1997, 38, 7483–7486. M. A. Heuft, S. K. Collins, G. P. A. Yap, A. G. Fallis, Org. Lett. 2001, 3, 2883– 2886. M. J. Marsella, Z.-Q. Wang, R. J. Reid, K. Yoon, Org. Lett. 2001, 3, 885–887. P. N. W. Baxter, Chem. Eur. J. 2003, 9, 2531–2541. (a) P. N. W. Baxter, J. Org. Chem. 2001, 66, 4170–4179; (b) P. N. W. Baxter, Chem. Eur. J. 2002, 8, 5250–5264. M. A. Heuft, A. G. Fallis, Angew. Chem. 2002, 114, 4702–4705; Angew. Chem. Int. Ed. 2002, 41, 4520–4522. J. M. Fox, D. Lin, Y. Itagaki, T. Fujita, J. Org. Chem. 1998, 63, 2031–2038. D. L. An, T. Nakano, A. Orita, J. Otera, Angew. Chem. 2002, 114, 179–181; Angew. Chem. Int. Ed. 2002, 41, 171–173. M. J. Marsella, I. T. Kim, F. Tham, J. Am. Chem. Soc. 2000, 122, 974–975. S. Anderson, U. Neidlein, V. Gramlich, F. Diederich, Angew. Chem. 1995, 107, 1722–1726; Angew. Chem. Int. Ed. Engl. 1995, 34, 1596–1599. (a) J. S. Moore, J. Zhang, Angew. Chem. 1992, 104, 873–874; Angew. Chem. Int. Ed. Engl. 1992, 31, 922–924; (b) J. Zhang, D. J. Pesak, J. L. Ludwick, J. S. Moore, J. Am. Chem. Soc. 1994, 116, 4227–4239. D. B. Shortell, L. C. Palmer, J. M. Tour, Tetrahedron 2001, 57, 9055–9065. P.-H. Ge, W. Fu, W. A. Herrmann, E. Herdweck, C. Campana, R. D. Adams, U. H. F. Bunz, Angew. Chem. 2000, 112, 3753–3756; Angew. Chem. Int. Ed. 2000, 39, 3607–3610.

67

68

69

70

71

72

73 74

75

76

(a) A. Vidal-Ferran, C. M. Müller, J. K. M. Sanders, J. Chem. Soc., Chem. Commun. 1994, 2657–2658; (b) S. Anderson, H. L. Anderson, J. K. M. Sanders, J. Chem. Soc. Perkin Trans. 1 1995, 18, 2247–2254; (c) S. Anderson, H. L. Anderson, J. K. M. Sanders, J. Chem. Soc. Perkin Trans. 1 1995, 18, 2255–2267; (d) S. Anderson, H. L. Anderson, J. K. M. Sanders, Acc. Chem. Res. 1993, 26, 469– 475. J. Li, A. Ambroise, S. I. Yang, J. R. Diers, J. Seth, C. R. Wack, D. F. Bocian, D. Holten, J. S. Lindsey, J. Am. Chem. Soc. 1999, 121, 8927–8940. S. Rucareanu, O. Mongin, A. Schuwey, N. Hoyler, A. Gossauer, W. Amrein, H.-U. Hediger, J. Org. Chem. 2001, 66, 4973–4988. (a) S. Höger, A.-D. Meckenstock, H. Pellen, J. Org. Chem. 1997, 62, 4556– 4557; (b) S. Höger, A.-D. Meckenstock, Chem. Eur. J. 1999, 5, 1686–1691. G. Fuhrmann, T. Debaerdemaeker, P. Bäuerle, Chem. Commun. 2003, 948– 949. (a) T. Kawase, N. Ueda, H. R. Darabi, M. Oda, Angew. Chem. 1996, 108, 1658– 1660; Angew. Chem. Int. Ed. Engl. 1996, 35, 1556–1558; (b) T. Kawase, N. Ueda, M. Oda, Tetrahedron Lett. 1997, 38, 6681– 6684; (c) T. Kawase, Y. Hosokawa, H. Kurata, M. Oda, Chem. Lett. 1999, 745– 746. K. Utsumi, T. Kawase, M. Oda, Chem. Lett. 2003, 32, 412–413. M. Srinivasan, S. Sankararaman, I. Dix, P. G. Jones, Org. Lett. 2000, 2, 3849–3851. (a) Y. Yamaguchi, S. Kobayashi, T. Wakamiya, Y. Matsubara, Z. Yoshida, J. Am. Chem. Soc. 2000, 122, 7404–7405; (b) S. Kobayashi, S. Wakamoto, Y. Yamaguchi, T. Wakamiya, K. Sugimoto, Y. Matsubara, Z. Yoshida, Tetrahedron Lett. 2003, 44, 1807–1810; (c) G. J. Bodwell, T. J. Houghton, D. Miller, Tetrahedron Lett. 1998, 39, 2231–2234. (a) J. Zhang, J. S. Moore, J. Am. Chem. Soc. 1992, 114, 9701–9702; (b) A. S. Shetty, J. Zhang, J. S. Moore, J. Am. Chem. Soc. 1996, 118, 1019–1027; (c) S.

1.7 References

77 78 79

80 81

82 83

84

85 86

87

88 89

Lahiri, J. L. Thompson, J. S. Moore, J. Am. Chem. Soc. 2000, 122, 11315–11319. D. Zhao, J. S. Moore, J. Org. Chem. 2002, 67, 3548–3554. K. Nakamura, H. Okubo, M. Yamaguchi, Org. Lett. 2001, 3, 1097–1099. (a) J. Zhang, J. S. Moore, J. Am. Chem. Soc. 1994, 116, 2655–2656; (b) O. Y. Mindyuk, M. R. Stetzer, P. A. Heiney, J. C. Nelson, J. S. Moore, Adv. Mater. 1998, 10, 1363–1366. D. Venkataraman, S. Lee, J. Zhang, J. S. Moore, Nature 1994, 371, 591–593. (a) A. S. Shetty, P. R. Fischer, K. F. Stork, P. W. Bohn, J. S. Moore, J. Am. Chem. Soc. 1996, 118, 9409–9414; (b) O. Y. Mindyuk, M. R. Stetzer, D. Gidalevitz, P. A. Heiney, J. C. Nelson, J. S. Moore, Langmuir 1999, 15, 6897–6900. Y. Hosokawa, T. Kawase, M. Oda, Chem. Commun. 2001, 1948–1949. (a) Y. Tobe, N. Utsumi, K. Kawabata, K. Naemura, Tetrahedron Lett. 1996, 37, 9325–9328; (b) Y. Tobe, N. Utsumi, A. Nagano, K. Naemura, Angew. Chem. 1998, 110, 1347–1349; Angew. Chem. Int. Ed. Engl. 1998, 37, 1285–1287; (c) Y. Tobe, N. Utsumi, A. Nagano, M. Sonoda, K. Naemura, Tetrahedron 2001, 57, 8075– 8083; (d) Y. Tobe, N. Utsumi, K. Kawabata, A. Nagano, K. Adachi, S. Araki, M. Sonoda, K. Hirose, K. Naemura, J. Am. Chem. Soc. 2002, 124, 5350–5364. Y. Tobe, A. Nagano, K. Kawabata, M. Sonoda, K. Naemura, Org. Lett. 2000, 2, 3265–3268. C.-H. Lin, J. Tour, J. Org. Chem. 2002, 67, 7761–7768. (a) S. Höger, V. Enkelmann, Angew. Chem. 1995, 107, 2917–2919; Angew. Chem. Int. Ed. Engl. 1995, 34, 2713–2716; (b) S. Höger, A.-D. Meckenstock, S. Müller, Chem. Eur. J. 1998, 4, 2423– 2434; (c) S. Höger, K. Bonrad, L. Karcher, A.-D. Meckenstock, J. Org. Chem. 2000, 65, 1588–1589. S. Höger, D. L. Morrison, V. Enkelmann, J. Am. Chem. Soc. 2002, 124, 6734–6736. D. L. Morrison, S. Höger, Chem. Commun. 1996, 2313–2314. S. Höger, V. Enkelmann, K. Bonrad, C. Tschierske, Angew. Chem. 2000, 112,

90

91

92

93

94

95

96 97

98

99

100 101 102

2356–2358; Angew. Chem. Int. Ed. 2000, 39, 2268–2270. S. Höger, K. Bonrad, A. Mourran, U. Beginn, M. Möller, J. Am. Chem. Soc. 2001, 123, 5651–5659. S. Rosselli, A.-D. Ramminger, T. Wagner, B. Silier, S. Wiegand, W. Häußler, G. Lieser, V. Scheumann, S. Höger, Angew. Chem. 2001, 113, 3234–3237; Angew. Chem. Int. Ed. 2001, 40, 3137– 3141. (a) O. Henze, D. Lentz, A. D. Schlüter, Chem. Eur. J. 2000, 6, 2362–2367; (b) O. Henze, D. Lentz, A. Schäfer, P. Franke, A. D. Schlüter, Chem. Eur. J. 2002, 8, 357–365. C. Grave, D. Lentz, A. Schäfer, P. Samori, J. P. Rabe, P. Franke, A. D. Schlüter, J. Am. Chem. Soc. 2003, 125, 6907–6918. S. Kobayashi, Y. Yamaguchi, T. Wakamiya, Y. Matsubara, K. Sugimoto, Z. Yoshida, Tetrahedron Lett. 2003, 44, 1469–1472. Y. Yamaguchi, S. Kobayashi, N. Amita, T. Wakamiya, Y. Matsubara, K. Sugimoto, Z. Yoshida, Tetrahedron Lett. 2002, 43, 3277–3280. S.-S. Sun, A. J. Lees, Organometallics 2001, 20, 2353–2358. (a) K. Campbell, R. McDonald, N. R. Branda, R. R. Tykwinski, Org. Lett. 2001, 3, 1045–1048; (b) K. Campbell, R. McDonald, R. R. Tykwinski, J. Org. Chem. 2002, 67, 1133–1140. K. Campbell, C. J. Kuehl, M. J. Ferguson, P. J. Stang, R. R. Tykwinski, J. Am. Chem. Soc. 2002, 124, 7266–7267. (a) M. Schmittel, H. Ammon, Synlett 1999, 750–752; (b) M. Schmittel, H. Ammon, V. Kalsani, A. Wiegrefe, C. Michel, Chem. Commun. 2002, 2566– 2567. M. Mayor, J.-M. Lehn, J. Am. Chem. Soc. 1999, 121, 11231–11232. S. Maruyama, H. Hokari, T. Wada, H. Sasabe, Synthesis 2001, 1794–1799. J. Krömer, I. Rios-Carreras, G. Fuhrmann, C. Musch, M. Wunderlin, T. Debaerdemaeker, E. Mena-Osteritz, P. Bäuerle, Angew. Chem. 2000, 112, 3623– 3628; Angew. Chem. Int. Ed. 2000, 39, 3481–3486.

39

40

1 Cyclophynes 103 (a) Ö. Ünsal, A. Godt, Chem. Eur. J.

104

105

106

107 108

109 110

111

1999, 5, 1728–1733; (b) A. Godt, S. Duda, Ö. Ünsal, J. Thiel, A. Härter, M. Roos, C. Tschierske, S. Diele, Chem. Eur. J. 2002, 8, 5094–5106; (c) M. R. Shah, S. Duda, B. Müller, A. Godt, A. Malik, J. Am. Chem. Soc. 2003, 125, 5408–5414. (a) T. Kawase, H. R. Darabi, M. Oda, Angew. Chem. 1996, 108, 2803–2805; Angew. Chem. Int. Ed. Engl. 1996, 35, 2664–2666; (b) T. Kawase, N. Ueda, K. Tanaka, Y. Seirai, M. Oda, Tetrahedron Lett. 2001, 42, 5509–5511. T. Kawase, Y. Seirai, H. R. Darabi, M. Oda, Y. Sarakai, K. Tashiro, Angew. Chem. 2003, 115, 1659–1662; Angew. Chem. Int. Ed. 2003, 42, 1621–1624. T. Kawase, K. Tanaka, N. Fujiwara, H. R. Darabi, M. Oda, Angew. Chem. 2003, 115, 1662–1666; Angew. Chem. Int. Ed. 2003, 42, 1624–1628. M. Ohkita, K. Ando, T. Suzuki, T. Tsuji, Chem. Commun. 2001, 2570–2571. (a) M. Ohkita, K. Ando, K. Yamamoto, T. Suzuki, T. Tsuji, Chem. Commun. 2000, 83–84; (b) M. Ohkita, K. Ando, T. Suzuki, T. Tsuji, J. Org. Chem. 2000, 65, 4385–4390. Y. Tobe, T. Morinaka, M. Sonoda, unpublished results. H. Hopf, P. G. Jones, P. Bubenitschek, C. Werner, Angew. Chem. 1995, 107, 2592–2594; Angew. Chem. Int. Ed. Engl. 1995, 34, 2367–2368. M. M. Haley, B. L. Langsdorf, Chem. Commun. 1997, 1121–1122.

112 Y. Tobe, R. Furukawa, M. Sonoda, T.

113 114 115 116

117

118

119

120 121

Wakabayashi, Angew. Chem. 2001, 113, 4196–4198; Angew. Chem. Int. Ed. 2001, 40, 4072–4074. H. E. Ensley, S. Mahadevan, J. Mague, Tetrahedron Lett. 1996, 37, 6255–6258. S. K. Collins, G. P. A. Yap, A. G. Fallis, Org. Lett. 2002, 4, 11–14. Y. Tobe, J. Kishi, I. Ohki, M. Sonoda, J. Org. Chem. 2003, 68, 3330–3332. (a) Y. Rubin, T. C. Parker, S. I. Kahn, C. L. Holliman, S. W. McElvany, J. Am. Chem. Soc. 1996, 118, 5308–5309; (b) Y. Rubin, Chem. Eur. J. 1997, 3, 1009–1016. Y. Rubin, T. C. Parker, S. Pastor, S. Jalistagi, C. Boulle, C. L. Wilkins, Angew. Chem. 1998, 110, 1353–1356; Angew. Chem. Int. Ed. Engl. 1998, 37, 1226–1229. (a) Y. Tobe, N. Nakagawa, K. Naemura, T. Wakabayashi, T. Shida, Y. Achiba, J. Am. Chem. Soc. 1998, 120, 4544–4545; (b) Y. Tobe, N. Nakagawa, J. Kishi, M. Sonoda, K. Naemura, T. Wakabayashi, T. Shida, Y. Achiba, Tetrahedron 2001, 57, 3629–3636. Y. Tobe, H. Nakanishi, M. Sonoda, T. Wakabayashi, Y. Achiba, Chem. Commun. 1999, 1625–1626. Y. Tobe, R. Umeda, M. Sonoda, unpublished results. (a) Z. Wu, S. Lee, J. S. Moore, J. Am. Chem. Soc. 1992, 114, 8730–8732; (b) Z. Wu, J. S. Moore, Angew. Chem. 1996, 108, 320–322; Angew. Chem. Int. Ed. Engl. 1996, 35, 297–299.

41

2

Hetera (Cyclo-)phanes Fritz Vögtle, Gregor Pawlitzki, and Uwe Hahn

2.1

Introduction 2.1.1

Selection of Topics Presented

In this survey we first sum up achievements made in the author’s group during the past few years in the field of rigid cyclophane propellers. These comprise chiral [2.2]metacyclophanes, the chirality of which is mostly due to a heteroatom in the bridge part of the molecule. “Tailoring” and “tuning” of the chirality of cyclophanes was successfully approached in this way. [n]Para- and [n]metacyclophanes (n = bridge length) and phanes containing characteristic aliphatic parts, such as adamantane (“aliphanes”, see below), are also described. Macrobi- and tri-cyclic phanes as well as larger ring phanes up to gigantocyclic ones are likewise compared. Catenanes and rotaxanes of various types including the amide-based cyclophane family have been made accessible in preparative yields by new template syntheses. New phane-type molecular knots with amide bridging units were developed exhibiting topological chirality. No doubt the emergence and rational design and synthesis of new (cyclo-)phane scaffolds – and the easier and quicker availability of their building blocks – during the past few years has accelerated and facilitated the development of new supramolecular architectures. It seems difficult in the framework of such a volume to supplement the author’s own research results with all the other international work on hetera(cyclo-)phanes, with the same rigour. However, as we wanted to set our own results in the context of other’s work, we decided to present other contributions in the form of tables of the various families of characteristic compound structures with corresponding references indicated there. In such a way, the reader gains a fascinating graphical survey of what has been achieved during the past years. As (cyclo-)phanes “live” with their formulae which immediately reveal ring size, as well as their hetero-, hetera-, carbo-, carba-character, the reader can access the literature of compound families he is interested in.

Modern Cyclophane Chemistry. Edited by Rolf Gleiter, Henning Hopf Copyright © 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 3-527-30713-3

42

2 Hetera (Cyclo-)phanes

Much progress has been made recently particularly in the development of metalla and supramolecular phanes, where some formulae given in the last part of this contribution also disclose the principles and the possibilities for future. 2.1.2

Definitions

Heteraphanes are (cyclic) molecules that are composed of one (or more) arene unit(s) (“aromatic” ring like benzene, pyridine, thiophene) that is bridged by an aliphatic unit (of bridge length 0–? atoms) in which at least one atom is a heteroatom [1–3]. Fig. 2.1 shows examples of a heteraphane, a heterophane and a mixed hetero/heteraphane in order to demonstrate the difference. “Phane” is used as the stem name for such ring compounds. In Fig. 2.2 we use the prefix “cyclo” (or “benzeno”) if a benzene ring is the arene core. “Carbaphane” is used for those (carbo- or hetero-)phanes with bridges lacking heteroatoms (Fig. 2.3). Some other definitions have been formed and accepted during the recent years: in “aliphanes” [1], the arene part of the (areno-)phanes is replaced by a characteristic aliphatic ring. Fig. 2.4 shows characteristic examples. In the following, we will exclude the more trivially ortho-bridged benzenes, even if they can be viewed as “orthocyclophanes” [1]. Whereas overviews on cyclophanes and heterophanes have already been published, heteraphanes to our knowledge seem never to have been collected and compared [4].

a Thiophenophane a Oxaphane a [n+3]Phane

Concrete phane name: 2-Thia[n+2](1,3)benzenophane

Fig. 2.1

Some “heteraphanes” and related phane-species

2.1 Introduction

Fig. 2.2

Some heterophanes

43

44

2 Hetera (Cyclo-)phanes

Fig. 2.2

(cont.)

2.1 Introduction

Some carbaphanes (benzenophanes, cyclophanes) bearing no heteroatom in the bridges. At the same time these compounds are carbophanes (no heteroatom in the arene core)

Fig. 2.3

45

46

2 Hetera (Cyclo-)phanes

Fig. 2.3

(cont.)

2.1 Introduction

Fig. 2.3

(cont.)

Fig. 2.4

Some “aliphanes”

47

48

2 Hetera (Cyclo-)phanes

Fig. 2.4

(cont.)

2.1 Introduction

2.1.3

Why Conduct a Survey on Heteraphanes?

1) It is difficult to find these scattered compounds in the literature. Many are not named as phanes and even less characterized as heteraphanes. 2) There is a large variety of compounds described in different contexts. Unless they are collected in one article, they cannot be compared properly. 3) The synthetic methods leading to heteraphanes are usually C–X-bond formations (X = heteroatom), which means that there is a rich knowledge available in carrying through and optimizing macrocyclization reactions of this type [5]. 4) Many crown ethers, cryptands and macrocyclic ligand systems related to natural products [1] belong to this family of phane compounds (Fig. 2.5). 5) The chemistry of “metallaphanes” as well as of “supramolecular phanes” has recently developed impressively quickly. In “metallaphanes” (list see below, Fig. 2.18), a metal center forms one or several of the bridge’s hetero atoms. Also by means of hydrogen bond formation, macrocyclization can be done through supramolecular forces, which leads to “supramolecular phanes” (list see below, Fig. 2.19). 6) Wheels for rotaxanes and macrocyclic rings for catenanes and knots mainly consist of heteraphanes (and heterophanes), as their template syntheses rely on heteroatoms in the bridges (or/and in the core) [5–7]. 7) Heteroatoms in the bridges often lead to lower symmetry of the molecules and that is why they can be chiral provided that conformational rigidity/mobility of the molecule is properly designed. 8) Solubility can be influenced to some extent by implementation of heteroatoms. 9) ”Nanocycles” of nm-size, “gigantocycles” (³ 50) and “ultracycles” (> 100 ring members) [8] can often be made easier than defined macrocyclic hydrocarbons of same ring member number (Fig. 2.6). 10) It is impressive how manifold the research areas are, in the context of which heteraphanes have been synthesized. To sum up, even if heteraphanes may be less eye-catching than the corresponding hydrocarbons (like e.g. “Kekuléne” [1]) or heterophanes (e.g. “sexipyridine” [1]) they offer advantages that justify putting them together here for the first time as a family of compounds. In such a way, they can help to design new phane compounds and help to easier find new synthetic strategies for such macrocycles. Of course, in this rather short article not every synthesized heteraphane can be put forward; this contribution therefore contains a selection of newer and more recent work around the years 2000–2002 and in such a way should help to find literature that is otherwise difficult to scan.

49

50

2 Hetera (Cyclo-)phanes

MeO

Fig. 2.5

A selection of heteraphanes related to natural products [1].

2.2 Heteraphanes

Fig. 2.5

(cont.)

Fig. 2.6

A representative of giganto-/ultracyclic heteraphane [8].

2.2

Heteraphanes 2.2.1

Planar Chiral and Helical Chiral Phanes 2.2.1.1 Design of Phanes with Planar and Helical Chirality

Helical molecules are well known in nucleic acid, peptide, and sugar chemistry. The different interaction of left- and right-handed screw-shaped molecules with left- and right-circularly polarized light, respectively, is a fundamental process that can be studied by electronic circular dichroism (CD) measurements. An advantage of small helical molecules compared with biological ones is that their usually rigid structure (cf. hexahelicene) permits more precise calculations and conclusions compared to flexible aliphatic molecules containing stereogenic centers [66, 67].

51

52

2 Hetera (Cyclo-)phanes

The design of new planar chiral or helical chiral molecules with property differences that although small can be tailored, is a permanent challenge in organic chemistry. The spectroscopic and chiroptic examination of a series of related molecules should lead to a deeper knowledge about structure and strain and the CD and electronic properties. Structure–chiroptics relationships are important for the qualitative and quantitative understanding of chirality and the interaction of circularly polarized light with chiral molecules. The examination of model substances may in future allow the prediction of (chir-)optical properties and the determination of the absolute configuration. As far as synthesis is concerned, such knowledge allows the design of novel helical molecules with tailored chiral properties. The formulae 62–66 (Fig. 2.7) show our concept to reduce the size and complexity of screw molecules while preserving the propeller sense. By replacing benzene rings through aliphatic bridges one finally ends up with the dinucleic planar chiral [2.2]metacyclophane 65 skeleton and with the mononucleic [n]meta- or [n]paracyclophanes 66. We therefore used the rigid [2.2]metacyclophane skeleton (65, Fig. 2.7), which leads to small planar chiral compounds by suitable substitution and to small helical chiral molecules by insertion of heteroatoms into the cyclophane bridges (see 67 b, 68–71, Fig. 2.8), in such a way ending up with hetera [2.2]phanes. An advantage of this choice is the limited number of atoms, that made a reliable calculation of the CD spectra possible, and the good solubility and high crystallization tendency of [2.2]metacyclophanes allowing X-ray crystallographic examination of torsion effects and absolute configurations.

Evolution of more and more simple helical and planar chiral molecules starting from the quaterphenylophane 62 by continuous formal distraction of benzene rings; these molecules are chiral without having stereogenic centers (if n in 66 is small) [67]

Fig. 2.7

2.2 Heteraphanes

2.2.1.2 Hetera [2.2]Metacyclophanes

Introduction of heteroatoms into the [2.2]metacyclophane’s bridges (Fig. 2.8), thus leading to symmetry breaking and to heteraphanes, allows a controlled increase of the twisting of the connected benzene rings against each other. In addition the internal deformation of the two-bladed propellers should be the more pronounced the closer the two benzene rings are pressed together by short cyclophane bridges -CH2-X- and -CH2-Y- in 67 (Fig. 2.8). Therefore we synthesized a number of monohetera- and dihetera [2.2]metacyclophanes [68] in which the carbon–heteroatom distance determines the molecular strain. Fig. 2.8 shows [2.2]metacyclophanes 68–70 in the order of increasing strain. The C-X bond length decreases from X=S to X=O, so that the 1-oxa [2.2]metacyclophane 70 [69] is the sterically most strained monohetera [2.2]metacyclophane with the highest boat-shaped deformation in the benzene rings. The thiazacyclophanes of type 71 are comparatively easy to synthesize in one (cyclization) step [68], so that we were able to prepare a range of differently substituted representatives that allowed us to study the influence of intra-annular substituents R on the CD and other properties [70].

Fig. 2.8

Two-wing propeller molecules

53

54

2 Hetera (Cyclo-)phanes

The chirality-inducing step is the formal substitution of one or more carbon atoms in the bridge of the [2.2]metacyclophane hydrocarbon 67 b to yield the heterocycles 67 a (X =/ Y, Fig. 2.8). The length of the bridges and consequently helicity and strain can be influenced within wide limits. The first member of the [2.2]metacyclophane family to be synthesized was the achiral hydrocarbon 67 b itself [71]; the first molecule with heteroatoms in the bridge was 1,10-dithia [2.2]metacyclophane 67 c [72]. 1-Thia- 68 and 1-oxa [2.2]metacyclophane 70 were prepared many years later [69] by intramolecular ring closure methods. Considering that these cyclophanes are ring-strained, the syntheses of 67 c and 71 and most cyclophanes mentioned in the following were carried out under high-dilution conditions [73], by the use of selected bases and by taking advantage of the cesium effect [74]. All chiral representatives 68–71 could be separated into their enantiomers using chiral column materials. Comparisons of experimental and theoretical CD spectra allowed the assignment of the absolute configuration of 68 and 71 [75]. A slight shortening of the cyclophane bridges was also achieved by the synthesis of the N-tosylazacyclophane 69 b [76]. In this case the tosyl group affected the chiroptical studies adversely, caused by the additional (tosyl) chromophore. Subsequent removal of the tosyl group in order to yield the free amine 69 a failed because of special reactivity due to the clamped molecule’s geometry and strain. This problem was circumvented by using a different protection-group strategy. Introduction of the N-activating trifluoroacetyl group allowed cyclization and subsequent cleavage of the activating group and yielded the free amine 69 a [77].

The N-activating effect of the trifluoroacetyl group also allowed for the first time two-component cyclizations of suitable building blocks yielding 1,10-diazacyclophane 72 a [78] and 1-thia-10-aza [2.2]metacyclophanes 71 a, b. The 1,10-diazacyclophane 72 b is, according to calculations and NMR studies, one of the most strained compounds in the [2.2]metacyclophane series. The close proximity of the arene rings is indicated by a strongly highfield-shifted inner proton signal [d(Hi) = 3.81] compared to the [2.2]metacyclophane hydrocarbon 67 b [d(Hi) = 4.25]. The even higher strained (because the C-O is shorter than the C-N bond length) 1-oxa-10-aza [2.2]metacyclophane 73 a was detected by mass spectrometry after comparable synthetic procedures, but isolation has failed so far. We tried to improve the synthetic processes for such highly strained molecules (repulsive steric

2.2 Heteraphanes

Hetera[2.2]metacyclophanes: representatives 73 a–d are helical chiral. Dioxa[2.2]metacyclophane does not yet exist

Fig. 2.9

Fig. 2.10 Planar chiral carba[2.2]metacyclophanes

interactions between the intra-annular positions 8 and 16, see Fig. 2.10) combined with new N-activating group strategies. Nevertheless, all efforts to synthesize the highest strained 1,10-dioxa [2.2]metacyclophane 74 in preparative amounts failed.

2.2.1.3 Planar Chiral [2.2]Metacyclophanes

[2.2]Metacyclophanes 65, substituted in one or more of the positions 4, 14 or 6, 12 (cf. Fig. 2.10) are planar chiral and were used by Lehner, Schlögl, Ugi, Ruch, and Derflinger for studies towards quantitative correlations with regard to chirality functions [79]. Thus, 75 b should exhibit a double Cotton effect intensity compared to 65 b, provided that R is the same. Spectroscopic and theoretical results were published in detail [80]. Attempts to introduce substituents into the carbon skeleton of diaza [2.2]metacyclophanes 72 to render it chiral were unsuccessful. Because of the higher steric strain the synthetic yields are lower than for 71 a [78].

55

56

2 Hetera (Cyclo-)phanes

Fig. 2.11 The achiral 1,10-dihetera[2.2]metacyclophanes 67 and 72

2.2.1.4 Planar Chiral Hetera [n]Para- and Hetera [n]Metacyclophanes

We synthesized a number of dithiabridged paracyclophanes [81] with aliphatic and partly unsaturated bridges. These were obtained under dilution conditions by the use of suitably functionalized cyclization substrates (dibromides [82], dithiols [83]). The introduction of unsaturated elements (76 a, b) enabled the examination of the influence of etheno and ethyno bridges towards the deformation of the aromatic parts and chiroptical properties. NMR- and X-ray crystallographic studies revealed C2-symmetry of the dithia[8]paracyclophanes 76 a, d and C1-symmetry of the oxa[7]paracyclophane 77 a and its carbocyclic analog 77 b in the solid state and in solution.

These cyclophanes were separated into their enantiomers and showed similar CD spectra with typical Cotton effects for electron transitions of the aromatic rings with significant contributions of the sulfur atoms. According to calculations the first oxa[7]paracyclophane 77 a [84] showed a remarkably short oxygen–benzene distance compared with the carbocyclic analog 77 b [85] which is indicated by a strong charge transfer band in the CD spectrum (see also Chapter 12). The chiral cyclophane 77 a was synthesized [86] via the dithia-route and sulfone pyrolysis [87] developed earlier by us.

2.2 Heteraphanes

The pathway to dithia[n]metacyclophanes 78 by thiol/bromide cyclizations [86] and subsequent sulfone pyrolysis allowed the first systematic studies of the previously poorly examined planar chiral [n]metacyclophanes. The motion of the cyclophane bridges was examined by dynamic NMR methods. It turned out that the cyclophane bridges undergo rather complex pseudorotation motions which can be frozen at low temperatures whereas a flip-process [87] is prevented by the intra-annular substituents. The distances between bridge and benzene ring obtained from Xray crystal structures correlate well with the observed NMR shifts [78 b].

2.2.1.5 Dioxa [2.2]Phanes and Oxaza [2.2]Phanes

These types of very narrowly bridged and therefore highly strained [2.2]cyclophanes were addressed several times through the past 20 years, but could so far not be synthesized in preparative amounts [78 a]. This goal remains a challenge for future.

2.2.1.6 Enantiomer Separations

The separation of enantiomers was carried out by HPLC on cellulose-tris(3,5-dimethylphenylcarbamate) columns on analytical scale, and repeated several times to yield enough material for CD measurements. Successful use of Okamoto’s separation resins [88] led to base-line separation of nearly all the chiral cyclophanes described above. For separation on a preparative scale we used semipreparative columns to get up to gram amounts of pure enantiomers for further examinations like gas phase CD measurements. Calculations of CD spectra with the AM1/MRD-CI and DFT/RPA methods were carried out on 1-thia- 68 and 1-oxa [2.2]metacyclophane 70 [89]. The spectra of

57

58

2 Hetera (Cyclo-)phanes

both compounds are remarkably similar showing three lowest lying negative bands followed by an intense positive peak at 200 nm and a negative band around 180 nm. These experimental features are well reproduced by the calculations. The corresponding thia [2.2]metacyclophane 68 shows a higher calculated CD intensity (by a factor of two) than the oxa-cyclophane 70 due to the stronger deviation of the geometry from C2h-symmetry.

Calculations of the CD spectra of [n]cyclophanes on first sight seem to be more difficult due to the inherent flexibility of the relatively long aliphatic chains (n ³ 6). In favorable cases, however the conformers which result from chain rotation are identical, as shown for the recently resynthesized 9,12-dimethyl-4-oxa[7]paracyclophane 77 a [86, 90] so that only one conformation has to be considered. Shortening of the aliphatic bridge in [n]paracyclophanes results in stronger deformations of the benzene ring towards a boat-shaped conformation. Strong interactions of the energetically high-lying lone-pair orbital at the oxygen atom with benzene ring orbitals can be expected from geometrical considerations. In the case of dithia [n]metacyclophanes 78 c, d [86 a] two energetically close lying conformers for each enantiomer must be taken into account to explain the experimental CD spectra. The best AM1/MRD-CI CD data were obtained for a 70 : 30% mixture of two conformers of 8,12-dimethyl-2,5-dithia[6]metacyclophane. This value is in good agreement with the ratio 70% derived from NMR data. Gas phase CD-measurements [91] and the extension of these measurements into the VUV i.e. to higher excited electronic states proved to be important for a better understanding of circular dichroism and for a reliable comparison with theoretical calculations.

2.2.1.7 Strongly Helical Heteraphanes

Owing to the elongated screw shape, the heteracyclophanes 83, 85 and the naphthalenophane 84 are expected to reveal high chirality comparable to the helicenes. Indeed, elongation of the screw (from biphenylo- to terphenylophanes) leads to an increased optical rotation [92].

2.2 Heteraphanes

This new helix type even surpasses some helicenes known for high rotatory strength and molar rotation. 2.2.2

Catenanes, Rotaxanes, and Knotanes of the Heteraphane Type 2.2.2.1 Template Synthesis of Rotaxanes Using Cyclophane Wheels

Some years ago, we reported on an efficient strategy with near to quantitative yields for the synthesis of a phenyl ether [2]rotaxane 90 based on the action of a preorganized supramolecular nucleophile 88, which is formed from the molecular

Fig. 2.12 Synthesis of ether rotaxanes with ether axles by the trapping method. The wheel is a heteraphane and it is only due to the hydrogen bonding abilities of its heteroatom that it is able to form the concave template for the trapping (threading) of the axle part

59

60

2 Hetera (Cyclo-)phanes

recognition of a phenolate anion stopper 86 by the tetralactam wheel 87 [93] which is a hetera-cyclophane. We termed this synthetic step “trapping” because the nucleophilic blocking group is caught by the wheel like a mouse in a trap. This supramolecular nucleophile 88 can now generate a rotaxane by an SN reaction with an electrophilic axle centre piece 89 as shown in Fig. 2.12. We have demonstrated that a variety of other anionic nucleophiles can be used as axle building blocks alternatively. New families of rotaxanes were synthesized upon formation of acetal, thioacetal, ester, thioester, sulfide, N-tosylamide, and phosphate bonds in their axles. The yields are high (23–81%) which underlines the versatility of this synthetic interlocking concept which makes use of the heteraphane 87 acting as concave template for anions [93].

2.2.2.2 Higher Order [n]Rotaxanes via Non-ionic Template Effect

We also developed a method for the synthesis of monodisperse higher order rotaxanes of the amide type based advantageously on an iterative reaction sequence [94]. Up to five heteraphane-type tetralactam wheels (87) could be threaded on one axle yielding the [6]rotaxane 93 (wheels and stoppers like in Fig. 2.12). To build up higher order rotaxanes in this way, semiaxles have to be synthesized that contain more amide groups, one for each wheel to be bound and then threaded. Therefore we elongated a bulky stopper group in a repetitive reaction sequence that led to the diamide 91, which is a semiaxle. Reaction of bis-wheeled 91 with the diacid dichloride 92 enables to bind one more wheel 87 on the axle on account of the additional amide group. Condensation with the second bis-wheeled 91 leads to the penta-wheeled [6]rotaxane 93 carrying five heteraphane units 87 topologically bound.

Fig. 2.13 Synthesis of oligo[n]rotaxanes 93 with heteracyclophane

wheels (87 in Fig. 2.12)

2.2 Heteraphanes

2.2.2.3 Combination of Anionic and Non-ionic Template

Applying the trapping (anionic template) synthesis in combination with the threading method (non-ionic template) for rotaxanes, we reacted an axle precursor with the phenolate 86 (Fig. 2.12) in presence of the heteraphane wheels 87 [95]. Besides small amounts of the free axle and the corresponding [2]rotaxanes (up to 10%) the [3]rotaxanes 94–96 were obtained in yields up to 29%. In all cases the isolation of the [3]rotaxanes 94–96 could easily be carried out by column chromatography.

2.2.2.4 Molecular Knots and Similar Macrocycles of the Heteraphane Type

A “molecular eight” type topologically chiral molecule 97 was described [96]. [1]Rotaxanes of the type 98 were obtained, too, which are also topologically chiral [97]. An analogous reaction of a double sulfonamide-substituted catenane [98] yielded “pretzelanes” [98]. A most simple synthesis of a molecular trefoil knot of the heteraphane type was achieved, in up to 20% yield, which proceeds under self-organization without any added auxiliary component [99] (Fig. 2.16).

Fig. 2.14 “Supramolecular eight” 97

61

62

2 Hetera (Cyclo-)phanes

Fig. 2.15 Sulfonamide-based [1]rotaxanes 98

As byproducts monocyclic heteraphanes like 101, 102 and in some cases higher oligomers including catenanes have been isolated as well [99]. 2.2.3

Further Heteraphanes, Metallaphanes and Supramolecular Phanes

In order to allow a more general view and comparison of heteraphane structures, more recent phanes with heteroatoms in the bridges are put together in the following Figs 2.17 – 2.19. These comprise usual heteroatoms like oxygen, nitrogen, but also boron, metal-carbonyl-complexed multiple bonds (Fig. 2.17), but also metal atoms (Fig. 2.18) and finally non-covalently connected bridges (Fig. 2.19).

2.3

Conclusions

This survey of cyclophanes and phanes (heteraphanes) substituted by heteroatoms in the bridge sets them in the context of aliphanes, heterophanes, metallo- and supramolecular phanes and certainly shows that a surprising variety of fascinating macrocyclic structures and even architectures have been synthesized and investigated during the past few years. All in all, this is a chapter of heteromacrocyclic chemistry, even if the heterophanes, the pyridinophanes, thiophenophanes etc. had to be more or less neglected. Heteraphanes are interesting because their preparation in many cases is easier than that of the corresponding hydrocarbons. On the other hand, hydrocarbons

2.3 Conclusions

Fig. 2.16 Synthesis of trefoil knot 103, a heteracyclic paracyclo(2,6)pyridinophane

cannot easily offer hydrogen bonding or p–p interactions or charges for host– guest interactions, whereas the heteraphanes can form supramolecular complexes and therewith can be considered as supramolecular phanes. The synthetic methods used to form C–X bonds (X = heteroatom) are additionally valuable. It is not surprising that gigantocyclic compounds, molecular knots, catenanes, rotaxanes have been tried with C–heteroatom connection rather than C–C-bond connection and so the chemistry of molecular ribbons and tubes has developed, too. The implementation of heteroatoms in the bridges of cyclophanes and phanes leads to many more possibilities than with the hydrocarbons, not only for intramolecular interactions, but also for host–guest interactions and with regard to symmetry breaking leading to chirality.

63

64

2 Hetera (Cyclo-)phanes

Fig. 2.17 Some selected further heteraphanes from more recent literature

2.3 Conclusions

Fig. 2.17 (cont.)

65

66

2 Hetera (Cyclo-)phanes

Fig. 2.17 (cont.)

2.3 Conclusions

Fig. 2.17 (cont.)

67

68

2 Hetera (Cyclo-)phanes

Fig. 2.17 (cont.)

2.3 Conclusions

Fig. 2.18 Some selected metalla-

phanes

69

70

2 Hetera (Cyclo-)phanes

Fig. 2.18 (cont.)

2.3 Conclusions

Fig. 2.18 (cont.)

As this area of research will grow with higher speed during the next years due to better synthetic and analytical methods, it will increase in importance and produce new complicated structures beyond molecular knots and beyond molecular tubes, which are of interest for applications. For the first time, the term supramolecular phanes is used here to collect and put together this new family of compounds under this topic. The fact that heteroatom containing macrocycles have at least presently some advantages compared to carbocyclic compounds, is demonstrated easily by the existence of heterocyclic knot molecules of the Sauvage and Vögtle type and nucleic acid type. So far, no carbocyclic knot molecule is known at all as the synthesis cannot be easily done via template synthesis. The authors hope that this collection of heteraphanes leads the reader to new ideas and the design of novel skeletons with implemented heteroatoms and also to new carbocyclic architectures.

71

72

2 Hetera (Cyclo-)phanes

Fig. 2.19 Selected su-

pramolecular phanes

2.3 Conclusions

Fig. 2.19 (cont.)

73

74

2 Hetera (Cyclo-)phanes

2.4

Acknowledgement

We are grateful to CERC3 (“Topological Stereochemistry, Topological Chirality”) and for the continuing collaboration in connection with the joint project entitled, “Topological chiral and cycloenantiomeric catenanes, rotaxanes and pretzelanes” (cooperation partners: Prof. Dr. A. v. Zelewsky (Fribourg), Prof. Dr. J.-P. Sauvage (Strasbourg), Dr. A. M. Brouwer (Amsterdam) including our group (F. Schelhase, R. Henkel).

2.5

References 1 2 3

4

5

6

7

8 9

10

11

12

F. Vögtle, Cyclophane Chemistry, WileyVCH, Chichester 1993. F. Diederich, Cyclophanes, Royal Society of Chemistry, Cambridge 1991. P. M. Keehn, S. M. Rosenfeld, Cyclophanes, Vol. I and II, Academic Press, New York 1983. F. Vögtle (ed.) Cyclophanes II, Top. Curr. Chem. 115, 1983; F. Vögtle, E. Weber, Macrocycles, Top. Curr. Chem. 161, 1992. N. V. Gerbeleu, V. B. Arion, J. Burgess, Template Synthesis of Macrocyclic Compounds, Wiley-VCH, Weinheim 1999. F. Diederich, P. J. Stang (eds.), Templated Organic Synthesis, Wiley-VCH, Weinheim 2000. B. Dietrich, P. Viout, J.-M. Lehn, Macrocyclic Chemistry. Aspects of Organic and Inorganic Supramolecular Chemistry, Wiley-VCH, Weinheim 1993. H. Schwierz, F. Vögtle, J. Incl. Phenom. Macrocycl. Chem. 2000, 37, 309–329. J. Guillard, O. Meth-Cohn, C. W. Rees, A. J. P. White, D. J. Williams, Chem. Commun. 2002, 232–233. G. Cafeo, F. H. Kohnke, G. L. La Torre, M. F. Parisi, R. Pistone Nascone, A. J. P. White, D. J. Williams, Chem. Eur. J. 2002, 8, 3148–3156. E. Mena-Osteritz, Adv. Mater. 2002, 14, 609–616; E. Mena-Osteritz, P. Bäuerle, ibid. 2001, 14, 243–246. Z. Hu, M. P. Cava, Tetrahedron Lett. 1994, 35, 3493–3496; E. Vogel, Pure & Appl. Chem. 1993, 65, 143–152; E. Vogel, P. Rohrig, M. Sichen, B. Knipp, A. Herrmann, M. Pohl, H. Schmickler, J. Lex, Angew. Chem. 1989, 101, 1683–1687; Angew. Chem. Int. Ed. Engl. 1989, 28, 1651–1655.

13

14

15

16

17 18

19

20

M. Iyoda, K. Nakao, T. Kondo, Y. Kuwatani, M. Yoshida, H. Matsuyama, K. Fukami, S. Nagase, Tetrahedron Lett. 2001, 42, 6869–6872. S. Tsuchiya, Y. Nakatani, R. Ibrahim, S. Ogawa, J. Am. Chem. Soc. 2002, 124, 4936– 4937. O. Henze, D. Lentz, A. Schäfer, P. Franke, A. D. Schlüter, Chem. Eur. J. 2002, 8, 357–365. F. Vögtle, S. Ibach, M. Nieger, C. Chartroux, T. Krüger, H. Stephan, K. Gloe, Chem. Commun. 1997, 1809–1810. T. Gerkensmeier, J. Mattay, C. Näther, Chem. Eur. J. 2001, 7, 465–474. G. Cafeo, G. Gattuso, F. H. Kohnke, A. Notti, S. Occhipinti, S. Pappalardo, M. F. Parisi, Angew. Chem. 2002, 114, 2226–2230; Angew. Chem. Int. Ed. 2002, 41, 2122–2126; see also: B. Turner, A. Shterenberg, M. Kapon, K. Suwinska, Y. Eichen, Chem. Commun. 2002, 404–405. D. Seidel, V. Lynch, J. L. Sessler, Angew. Chem. 2002, 114, 1480–1483; Angew. Chem. Int. Ed. 2002, 41, 1422–1425; see also: C. Bucher, R. S. Zimmerman, V. Lynch, J. L. Sessler, J. Am. Chem. Soc. 2001, 123, 2099– 2100; J. L. Sessler, D. Seidel, A. E. Vivian, V. Lynch, B. L. Scott, D. W. Keogh, Angew. Chem. 2001, 113, 611–614; Angew. Chem. Int. Ed. 2001, 40, 591–594; S. V. Shevchuk, J. M. Davis, J. L. Sessler, Tetrahedron Lett. 2001, 42, 2447–2450; further calixpyrrols and porphyrin analogues: J. L. Sessler, W.S. Cho, V. Lynch, V. Kral, Chem. Eur. J. 2002, 8, 1134–1143. N. Kobayashi, H. Miwa, V. N. Nemykin, J. Am. Chem. Soc. 2002, 124, 8007–8020.

2.5 References 21

22

23

24

25

26

27

28 29 30

31

32

E. Alcalde, N. Mesquida, L. Pérez-Garcia, S. Ramos, M. Alemany, M. L. Rodriguez, Chem. Eur. J. 2002, 8, 474–484; J.-Y. Shin, H. Furuta, A. Osuka, Angew. Chem. 2001, 113, 639–641; Angew. Chem. Int. Ed. 2001, 40, 619–621; M. Stepien, L. Latos-Grazynski, Chem. Eur. J. 2001, 7, 5113–5117; S. R. Graham, G. M. Ferrence, T. D. Lash, Chem. Commun. 2002, 894–895; A. Srinavasan, H. Furuta, A. Osuka, Chem. Commun. 2001, 1666–1667; H. Furuta, H. Maeda, A. Osuka, J. Am. Chem. Soc. 2001, 123, 6435–6436; S. R. Graham, D. A. Colby, T. D. Lash, Angew. Chem. 2002, 114, 1429–1432; Angew. Chem. Int. Ed. 2002, 41, 1371–1374; f. overview: J. L. Sessler, J. M. Davis, Acc. Chem. Res. 2001, 34, 989–997. J. R. Nitschke, T. D. Tilley, Angew. Chem. 2001, 113, 2200–2203; Angew. Chem. Int. Ed. 2001, 40, 2142–2145. E. Buncel, N. L. Mailloux, R. S. Brown, P. M. Kazmaier, J. Dust, Tetrahedron Lett. 2001, 42, 3559–3562. C. Pasquini, V. Desvergnes-Breuil, J. J. Jodry, A. Dalla Cort, J. Lacour, Tetrahedron Lett. 2002, 43, 423–426; see also: F. Vögtle, H. Förster, Chemiker-Ztg. 1973, 97, 386–387. F. Vögtle, S. Ibach, M. Nieger, C. Chartroux, T. Krüger, H. Stephan, K. Gloe, Chem. Commun. 1997, 1809–1810. C. Bucher, R. S. Zimmerman, V. Lynch, J. L. Sessler, J. Am. Chem. Soc. 2001, 123, 9716–9717. R. J. Schaller, R. Gleiter, J. Hofmann, F. Rominger, Angew. Chem. 2002, 114, 1231– 1234; Angew. Chem. Int. Ed. 2002, 41, 1181– 1183. S. Dahmen, S. Bräse, J. Am. Chem. Soc. 2002, 124, 5940–5941. G. P. Bartholomew, G. C. Bazan, J. Am. Chem. Soc 2002, 124, 5183–5196. C. Reuter, G. Pawlitzki, U. Wörsdörfer, M. Plevoets, A. Mohry, T. Kubota, Y. Okamoto, F. Vögtle, Eur. J. Org. Chem. 2000, 3059–3067; see also: P. Imming, M. Graf, S. Tries, R. Hirschelmann, E. Krause, G. Pawlitzki, Inflamm. Res. 2001, 50, 371–374. M. A. L. Sheepwash, R. H. Mitchell, C. Bohne, J. Am. Chem. Soc. 2002, 124, 4693– 4700; see also: M. Takeshita, T. Yamato, Angew. Chem. 2002, 114, 2260–2261; Angew. Chem. Int. Ed. 2002, 41, 2156–2157. T. Kawase, N. Ueda, K. Tanaka, Y. Seirai, M. Oda, Tetrahedron Lett. 2001, 42, 5509– 5511.

33

34 35 36 37

38

39 40 41

42

43

44

45

46

47

D. L. An, T. Nakano, A. Orita, J. Otera, Angew. Chem. 2002, 114, 179–181; Angew. Chem. Int. Ed. 2002, 41, 171–173. M. Ohkita, K. Ando, T. Tsuji, Chem. Commun. 2001, 2570–2571. Y. Hosokawa, T. Kawase, M. Oda, Chem. Commun. 2001, 1948–1949. M. Mayor, J.-M. Lehn, J. Am. Chem. Soc. 1999, 121, 11231–11232. Y. Tobe, R. Furukawa, M. Sonoda, T. Wakabayashi, Angew. Chem. 2001, 113, 4196– 4198; Angew. Chem. Int. Ed. 2001, 40, 4072– 4074. Z. Dominguez, H. Dang, M. J. Strouse, M. A. Garcia-Garibay, J. Am. Chem. Soc. 2002, 124, 2398–2399. Y. Rio, J.-F. Nierengarten, Tetrahedron Lett. 2002, 43, 4321–4324. F. Vögtle, K. Kadei, Chem. Ber. 1991, 124, 903–907. S. Ibach, V. Prautzsch, F. Vögtle, C. Chartroux, K. Gloe, Acc. Chem. Res. 1999, 32, 729–740. C. Chartroux, T. Krüger-Rambusch, K. Gloe, F. Bozkurt, M. Horn, S. Ibach, F. Vögtle, Czech. J. Phys. 1999, 49, 709–714; J. Gross, G. Harder, A. Siepen, J. Harren, F. Vögtle, H. Stephan, K. Gloe, B. Ahlers, K. Cammann, K. Rissanen, Chem. Eur. J. 1996, 2, 1585–1595; J. Groß, G. Harder, F. Vögtle, H. Stephan, K. Gloe, Angew. Chem. 1995, 107, 523–526; Angew. Chem. Int. Ed. Engl. 1995, 34, 481–484. C. Thilgen, F. Vögtle, Chem. Ber. 1991, 124, 671–672; C. Thilgen, F. Vögtle, Chem. Ber. 1990, 123, 2445–2447; F. Vögtle, C. Thilgen, Angew. Chem. 1990, 102, 1176–1178; Angew. Chem. Int. Ed. Engl. 1990, 29, 1162–1164. F. Vögtle, K. Saitmacher, S. Peyerimhoff, D. Hippe, H. Puff, P. Büllesbach, Angew. Chem. 1987, 99, 459–460; Angew. Chem. Int. Ed. Engl. 1987, 26, 470–471; K. Saitmacher, J. E. Schulz, M. Nieger, F. Vögtle, J. Chem. Soc., Chem. Commun. 1992, 175–176. Y. S. Park, E. J. Lee, Y. S. Chun, Y. D. Yoon, K. B. Yoon, J. Am. Chem. Soc. 2002, 124, 7123–7135. H.-J. Kwon, W. C. Smith, L. Xiang, B. Shen, J. Am. Chem. Soc. 2001, 123, 3385– 3386. E. La Porta, U. Piarulli, F. Cardullo, A. Paio, S. Provera, P. Seneci, C. Gennari, Tetrahedron Lett. 2002, 43, 761–766.

75

76

2 Hetera (Cyclo-)phanes 48 49

50 51

52

53 54

55 56

57

58

L. A. Paquette, J. Mendez-Andino, Tetrahedron Lett. 2001, 42, 967–970. D. R. Williams, L. Mi, R. J. Mullins, R. E. Stites, Tetrahedron Lett. 2002, 43, 4841– 4844. G.-W. Chen, A. Kirschning, Chem. Eur. J. 2002, 8, 2717–2729. D. Cao, H. Meier, Angew. Chem. 2001, 113, 193–195; Angew. Chem. Int. Ed. 2001, 40, 186–188. A. Fürstner, F. Jeanjean, P. Razon, Angew. Chem. 2002, 114, 2203–2206; Angew. Chem. Int. Ed. 2002, 41, 2097–2101. S. Hanessian, M. Angiolini, Chem. Eur. J. 2002, 8, 111–117. J. R. Dias, R. A. Pascal, Jr., J. Morrill, A. J. Holder, H. Gao, C. Barnes, J. Am. Chem. Soc. 2002, 124, 4647–4652. F. Yokokawa, A. Inaizumi, T. Shioiri, Tetrahedron Lett. 2001, 42, 5903–5908. H. J. Martin, M. Drescher, H. Kählig, S. Schneider, J. Mulzer, Angew. Chem. 2001, 113, 3287–3289; Angew. Chem. Int. Ed. 2001, 40, 3186–3188. Overviews: J. W. Steed, J. L. Adwood, Supramolecular Chemistry, Wiley-VCH, Chichester, 2000; Host–guest chemistry: C. A. Schalley, Angew. Chem. 2002, 114, 1583–1586; Angew. Chem. Int. Ed. 2002, 41, 1513–1515; see also: A. Day, R. J. Blanch, A. P. Arnold, S. Lorenzo, G. R. Lewis, I. Dance, Angew. Chem. 2002, 114, 285–287; Angew. Chem. Int. Ed. 2002, 41, 275–277; S.-Y. Kim, I.-S. Jung, E. Lee, J. Kim, S. Sakamoto, K. Yamaguchi, K. Kim, Angew. Chem. 2001, 113, 2177–2179; Angew. Chem. Int. Ed. 2001, 40, 2119–2121; H.-J. Kim, J. Heo, W. S. Jeon, E. Lee, J. Kim, S. Sakamoto, K. Yamaguchi, K. Kim, Angew. Chem. 2001, 113, 1574–1577; Angew. Chem. Int. Ed. 2001, 40, 1526–1529; K. A. Kellersberger, J. D. Anderson, S. M. Ward, K. E. Krakowiak, D. V. Dearden, J. Am. Chem. Soc. 2001, 123, 11316–11317; Cucurbituril: Rotaxanes: D. Tuncel, J. H. G. Steinke, Chem. Commun. 2002, 496–497; K.-M. Park, D. Whang, E. Lee, J. Heo, K. Kim, Chem. Eur. J. 2002, 8, 498–508; K. Park, S.-Y. Kim, J. Heo, D. Whang, S. Sakamoto, K. Yamaguchi, K. Kim, J. Am. Chem. Soc. 2002, 124, 2140–2147. J. Li, X. Chen, A. W. G. Burgett, P. G. Harran, Angew. Chem. 2001, 113, 2754– 2757; Angew. Chem. Int. Ed. 2001, 40, 2682– 2685; T. Ritter, E. M. Carreira, ibid. 2002, 14, 2601–2606, Int. Ed. 2002, 41, 2489– 2495.

59 60 61 62

63

64

65 66 67

68

69

70

71 72 73

74

T. Nagata, A. Nishida, M. Nakagawa, Tetrahedron Lett. 2001, 42, 8345–8349. F. Temal-Laib, J. Chastanet, J. Zhu, J. Am. Chem. Soc. 2002, 124, 583–590. P. J. Krenitsky, D. L. Boger, Tetrahedron Lett. 2002, 43, 407–410. W. R. Roush, L. Hertel, M. J. Schnaderbeck, N. A. Yakelis, Tetrahedron Lett. 2002, 43, 4885–4887. B. J. Carroll, S. J. Moss, L. Bai, Y. Kato, S. Toelzer, T.-W. Yu, H. G. Floss, J. Am. Chem. Soc. 2002, 124, 4176–4177. S. H. Kang, J. W. Jeong, Y. S. Hwang, S. B. Lee, Angew. Chem. 2002, 114, 1450–1453; Angew. Chem. Int. Ed. 2002, 41, 1392–1395. H. Schwierz, F. Vögtle, J. Incl. Phen. 2000, 37, 309–329. K. P. Meurer, F. Vögtle, Top. Curr. Chem. 1985, 127, 1–76. Overview: F. Vögtle, S. Grimme, K. H. Dötz, N. Krause, J. Hormes, Strain and Chirality, in S. Peyerimhoff, (ed.), Interactions in Molecules, Wiley-VCH, Weinheim, 2003, 15–109. P. Knops, P.-M. Windscheif, F. Vögtle, A. Roloff, M. Jansen, M. Nieger, E. Niecke, Y. Okamoto, Chem. Ber. 1991, 124, 1585–1590; D. Müller, M. Böhme, M. Nieger, K. Rissanen, F. Vögtle, J. Chem. Soc. Perkin Trans. 1, 1996, 2937–2943. F. Vögtle, J. Struck, H. Puff, P. Woller, H. Reuter, J. Chem. Soc., Chem. Commun. 1986, 1248–1250; F. Vögtle, K. Mittelbach, J. Struck, M. Nieger, J. Chem. Soc., Chem. Commun. 1989, 65–67. K. J. Przybilla, F. Vögtle, Chem. Ber. 1989, 122, 347–355; K. J. Przybilla, F. Vögtle, M. Nieger, S. Franken, Angew. Chem. 1988, 100, 987–989; Angew. Chem. Int. Ed. Engl. 1988, 27, 976–978. M. M. Pellegrin, Rec. Trav. Chim. Pays-Bas 1899, 18, 457. F. Vögtle, Tetrahedron Lett. 1968, 3623– 3626. F. Vögtle, Chem. Ind. (London) 1972, 346; P. Knops, N. Sendhoff, H.-B. Mekelburger, F. Vögtle, Top. Curr. Chem. 1991, 161, 1–67; L. Rossa, F. Vögtle, Top. Curr. Chem. 1983, 113, 1–86; see also: Katalog Labortechnik, Fa. Normag, Hofheim, 1992. A. Ostrowicki, E. Koepp, F. Vögtle, Top. Curr. Chem. 1992, 161, 37–67; The improvement of the synthetic procedure for 1-oxa[2.2]cyclophanes by use of the “cesium effect” is described in ref 70.

2.5 References 75

76

77

78

79

80 81 82

83 84 85

86

87

88

D. Wortmann-Saleh, S. Grimme, B. Engels, D. Müller, F. Vögtle, J. Chem. Soc. Perkin Trans. 2, 1995, 1185–1189. F. Vögtle, K. J. Przybilla, A. Mannschreck, N. Pustet, P. Büllesbach, H. Reuter, H. Puff, Chem. Ber. 1988, 121, 823–828. D. Müller, M. Nieger, F. Vögtle, J. Chem. Soc., Chem. Commun. 1994, 1361–1362; D. Müller, F. Vögtle, Synthesis 1995, 759– 760. a) The yield of the less stable cyclophane 13a is low (0.5 %): M. Habel, C. Niederalt, S. Grimme, M. Nieger, F. Vögtle, Eur. J. Org. Chem. 1998, 1471–1477; b) General survey on NMR spectra of phanes: L. Ernst, Progress Nucl. Magn. Res. Spectr. 2000, 37, 47–190. K. Schlögl, Top. Curr. Chem. 1984, 125, 29– 58; K. P. Meurer, F. Luppertz, F. Vögtle, Chem. Ber. 1985, 118, 4433–4438. S. Grimme, Chem. Phys. Lett. 1993, 201, 67– 74. I. Pischel, M. Nieger, A. Archut, F. Vögtle, Tetrahedron 1996, 52, 10043–10052 W. Offermann, F. Vögtle, Synthesis 1977, 272–273; W. Offermann, F. Vögtle, Angew. Chem. 1980, 92, 471–472; Angew. Chem. Int. Ed. Engl. 1980, 19, 464–465. B. Müller, I. Pischel, M. Nieger, F. Vögtle, Ber. 1994, 127, 759–765. S. Grimme, I. Pischel, S. Laufenberg, F. Vögtle, Chirality 1998, 10, 147–153. A. D. Wolf, V. V. Kane, R. H. Levin, M. Jones, Jr, J. Am. Chem. Soc. 1973, 95, 1680; L. W. Jenneskens, W. H. De Wolf, F. Bickelhaupt, Tetrahedron 1986, 42, 1571–1574. a) S. Grimme, I. Pischel, F. Vögtle, M. Nieger, J. Am. Chem. Soc. 1995, 117, 157– 162; b) S. Laufenberg, N. Feuerbacher, I. Pischel, O. Börsch, M. Nieger, F. Vögtle, Liebigs Ann. 1997, 1901–1906; c) W. Bieber, E. Hammerschmidt, F. Vögtle, Chem. Ber. 1978, 111, 2445–2447. F. Vögtle, L. Rossa, Angew. Chem. 1979, 91, 534–549; Angew. Chem. Int. Ed. Engl. 1979, 18, 514–529; J. Dohm, F. Vögtle, Top. Curr. Chem. 1991, 161, 69–106; F. Vögtle, P. Fornell, W. Löhr, Chem. Ind. (London) 1979, 416–418. N. Enamoto, S. Furukawa, Y. Ogasawara, H. Akano, Y. Kawamura, E. Yashima, Y. Okamoto, Anal. Chem. 1996, 68, 2798– 2804; E. Yashima, Y. Okamoto, Bull. Chem. Soc. Jpn. 1995, 68, 3289–3307; see also: Y. Okamoto, E. Yashima, Angew. Chem. 1998,

89

90

91 92

93

94 95

96

97 98

99

100 101

110, 1072–1095; Angew. Chem. Int. Ed. 1998, 37, 1020–1043; Chiralcel OD: cellulosetris[3,5-(dimethylphenyl)carbamate]; Chiralpak AD: amylosetris[3,5-(dimethylphenyl)carbamate]. S. Grimme, S. D. Peyerimhoff, S. Bartram, F. Vögtle, A. Breest, J. Hormes, J. Chem. Phys. Lett. 1993, 213, 32–40; S. Grimme, S. D. Peyerimhoff, Chem. Phys. 1996, 204, 411–417. H. Falk, K. Schlögl, Angew. Chem. 1968, 80, 405; Angew. Chem. Int. Ed. Engl. 1968, 8, 383. F. Pulm, J. Schramm, H. Lagier, J. Hormes, Enantiomer 1998, 3, 315–322. B. Kiupel, C. Niederalt, M. Nieger, S. Grimme, F. Vögtle, Angew. Chem. 1998, 110, 3206–3209; Angew. Chem. Int. Ed. 1998, 37, 3031–3034. C. Reuter, W. Wienand, G. M. Hübner, C. Seel, F. Vögtle, Chem. Eur. J. 1999, 5, 2692– 2697; G. M. Hübner, C. Reuter, C. Seel, F. Vögtle, Synthesis 1999, 103–108. A. H. Parham, R. Schmieder, F. Vögtle, Synlett 1999, 1887–1890. R. Schmieder, G. M. Hübner, C. Seel, F. Vögtle, Angew. Chem. 1999, 111, 3741– 3743; Angew. Chem. Int. Ed. 1999, 38, 3528– 3530. C. Reuter, W. Wienand, C. Schmuck, F. Vögtle, Chem. Eur. J. 2001, 7, 1728–1733; see also: Z.-L. Zhong, A. Ikeda, S. Shinkai, J. Am. Chem. Soc. 1999, 121, 11906– 11908. C. Reuter, C. Seel, M. Nieger, F. Vögtle, Helv. Chim. Acta 2000, 83, 630–640. A. Mohry, M. Nieger, F. Vögtle, Chirality 2000, 12, 76–83; W. Clegg, C. Gimenez-Saiz, D. A. Leigh, A. Murphy, A. M. Z. Slawin, S. J. Teat, J. Am. Chem. Soc. 1999, 121, 4124– 4129; D. A. Leigh, A. Murphy, J. P. Smart, A. M. Z. Slawin, Angew. Chem. 1997, 109, 752–756; Angew. Chem. Int. Ed. Engl. 1997, 36, 728–732; S. Ottens-Hildebrandt, M. Nieger, K. Rissanen, J. Rouvinen, S. Meier, G. Harder, F. Vögtle, J. Chem. Soc., Chem. Commun. 1995, 777–778. O. Safarowsky, M. Nieger, R. Fröhlich, F. Vögtle, Angew. Chem. 2000, 112, 1699– 1701; Angew. Chem. Int. Ed. 2000, 39, 1616– 1618. R. A. Bartsch, P. Kus, N. K. Dalley, X. Kou, Tetrahedron Lett. 2002, 43, 5017–5019. Y. Yamaguchi, S. Kobayashi, N. Amita, Tetrahedron Lett. 2002, 43, 3277–3280.

77

78

2 Hetera (Cyclo-)phanes 102 N. Komatsu, Org. Biomol. Chem. 2003, 1, 103 104 105 106 107

108 109

110

111 112

113 114 115

116 117

118 119

120

121

204–209. R. Breslow, Z. Fang, Tetrahedron Lett. 2002, 43, 5197–5200. S. Akine, T. Taniguchi, T. Nabeshima, Tetrahedron Lett. 2001, 42, 8861–8864. T. Bhattacharyya, U. J. Nilsson, Tetrahedron Lett. 2001, 42, 2873–2875. I. Bauer, W. D. Habicher, Tetrahedron Lett. 2002, 43, 5245–5248. M. Sánchez, H. Höpfl, M. E. Ochoa, N. Farfán, R. Santillan, S. Rojas-Lima, Chem. Eur. J. 2002, 8, 612–621. A. Lützen, O. Haß, T. Bruhn, Tetrahedron Lett. 2002, 43, 1807–1811. H. Akdas, E. Graf, M. W. Hosseini, A. De Cian, A. Bilyk, B. W. Skelton, G. A. Koutsantonis, I. Murray, J. M. Harrowfield, A. H. White, Chem. Commun. 2002, 1042– 1043; see also: E. Shokova, V. Tafeenko, V. Kovalev, Tetrahedron Lett. 2002, 43, 5153– 5156. A. Cattani-Scholz, C. Renner, C. Cabrele, R. Behrendt, D. Oesterhelt, L. Moroder, Angew. Chem. 2002, 114, 299– 302; Angew. Chem. Int. Ed. 2002, 41, 289– 292. R. Gibe, R. Green, Chem. Commun. 2002, 1550–1551. J. M. Ready, E. N. Jacobsen, Angew. Chem. 2002, 114, 1432–1435; Angew. Chem. Int. Ed. 2002, 41, 1374–1377. K. Tashiro, T. Aida, J. Incl. Phen. 2001, 41, 215–217. M. Herm, O. Molt, T. Schrader, Chem. Eur. J. 2002, 8, 1485–1499. J. Becerril, M. I. Burguete, B. Escuder, S. V. Luis, J. F. Miravet, M. Querol, Chem. Commun. 2002, 738–739. G. Rodríguez, M. Lutz, A. L. Spek, G. van Koten, Chem. Eur. J. 2002, 8, 45–57. T. Rosenau, A. Potthast, A. Hofinger, P. Kosma, Angew. Chem. 2002, 114, 1219– 1221; Angew. Chem. Int. Ed. 2002, 41, 1171– 1173. P. Rajakumar, M. Srisailas, Tetrahedron Lett. 2002, 43, 1909–1913. Overview: P. D. Beer, P. A. Gale, Angew. Chem. 2001, 113, 502–532; Angew. Chem. Int. Ed. 2001, 40, 486–516. L. Fabbrizzi, A. Leone, A. Taglietti, Angew. Chem. 2001, 113, 3156–3159; Angew. Chem. Int. Ed. 2001, 40, 3066–3069. O. D. Fox, T. D. Rolls, M. G. B. Drew, P. D. Beer, Chem. Commun. 2001, 1632–1633.

122 D. Ricard, M. L’Her, P. Richard, B. Boi-

trol, Chem. Eur. J. 2001, 7, 3291–3297. 123 R. D. Ionescu, T. Frejd, Chem. Commun.

2001, 1088–1089. 124 J. S. Kim, S. H. Yang, J. A. Rim, J. Y. Kim, J.

125 126

127 128 129 130

131 132 133

134

135

136 137

138

139

Vicens, S. Shinkai, Tetrahedron Lett. 2001, 42, 8047–8050; see also: K. Kavallieratos, B. A. Moyer, Chem. Commun. 2001, 1620– 1621. H. M. Colquhoun, F. Arico, D. J. Williams, Chem. Commun. 2001, 2574–2575. D.-W. Yoon, H. Hwang, C.-H. Lee, Angew. Chem. 2002, 114, 1835–1837; Angew. Chem. Int. Ed. 2002, 41, 1757–1759. A. Dondoni, X. Hu, A. Marra, H. D. Banks, Tetrahedron Lett. 2001, 42, 3295–3298. J. M. Mahoney, A. M. Beatty, B. D. Smith, J. Am. Chem. Soc. 2001, 123, 5847–5848. G. Pattenden, T. Thompson, Tetrahedron Lett. 2002, 43, 2459–2462. B. Escuder, A. E. Rowan, M. C. Feiters, R. J. M. Nolte, Tetrahedron Lett. 2001, 42, 2751–2753. K. Paek, J. Cho, Tetrahedron Lett. 2001, 42, 1927–1930. M. Darzac, T. Brotin, D. Bouchu, J.-P. Dutasta, Chem. Commun. 2002, 48–49. Overview: R. Warmuth, J. Yoon, Acc. Chem. Res. 2001, 95–105; see also: R. Warmuth, J.-L. Kerdelhué, S. Sánchez Carrera, K. J. Langenwalter, N. Brown, Angew. Chem. 2002, 114, 102–105; Angew. Chem. Int. Ed. 2002, 41, 96–99; C. Naumann, S. Place, J. C. Sherman, J. Am. Chem. Soc. 2002, 124, 16–17. H.-B. Mekelburger, J. Groß, J. Schmitz, M. Nieger, F. Vögtle, Chem. Ber. 1993, 126, 1713–1721. G. Brodesser, R. Güther, R. Hoss, S. Meier, S. Ottens-Hildebrandt, J. Schmitz, F. Vögtle, Pure & Appl. Chem. 1993, 65, 2325– 2328. F. Ebmeyer, F. Vögtle, Chem. Ber. 1989, 122, 1725–1727. A. Wallon, J. Peter-Katalinic´, U. Werner, W. M. Müller, F. Vögtle, Chem. Ber. 1990, 123, 375–379. A. Wallon, U. Werner, W. M. Müller, M. Nieger, F. Vögtle, Chem. Ber. 1990, 123, 859–867. C. Chartroux, K. Wichmann, G. Goretzki, T. Rambusch, K. Gloe, U. Müller, W. M. Müller, F. Vögtle, Ind. Eng. Chem. Res. 2000, 39, 3616–3624.

2.5 References 140 S. Breidenbach, S. Ohren, R. Herbst-Ir-

141

142

143 144

145

146

147

148

149

150 151 152

mer, S. Kotila, M. Nieger, F. Vögtle, Liebigs Ann. 1996, 2115–2121. M. Bauer, W. M. Müller, U. Müller, K. Rissanen, F. Vögtle, Liebigs Ann. 1995, 649–656. J.-C. Chambron, J.-P. Sauvage, K. Mislow, A. De Cian, J. Fischer, Chem. Eur. J. 2001, 7, 4085–4096. P. Samorí, F. Jäckel, Ö. Ünsal, A. Godt, J. P. Rabe, ChemPhysChem 2001, 7, 461–464. B. Korybut-Daszkiewicz, A. Wic¸kowska, R. Bilewicz, S. Domatala, K. Woz´niak, J. Am. Chem. Soc. 2001, 123, 9356–9366; for further heteracyclic Catenanes, Rotaxanes and Knotanes see: J.-P. Sauvage, C. O. DietrichBuchecker (eds.), Catenanes, Rotaxanes and Molecular Knots, Wiley-VCH, 1999; New Catenane Type: K. Hiratani, J. Suga, Y. Nagawa H. Houjou, H. Tokuhisa, M. Numata, K. Watanabe, Tetrahedron Lett. 2002, 43, 5747–5750. G. Dong, H. Gang, D. Chun-ying, P. Keliang, M. Qing-jin, Chem. Commun. 2002, 1096–1097. I. Fernandez, R. Ruiz, J. Faus, M. Julve, F. Lloret, J. Cano, X. Ottenwaelder, Y. Jornaux, M. C. Muñoz, Angew. Chem. 2001, 113, 3129–3132; Angew. Chem. Int. Ed. 2001, 40, 3039–3042. T. Nabeshima, A. Hashiguchi, T. Saiki, S. Akine, Angew. Chem. 2002, 114, 499–502; Angew. Chem. Int. Ed. 2002, 41, 481–484; see also: T. Nabeshima, A. Hashiguchi, Tetrahedron Lett. 2002, 43, 1457–1460. D. A. McMorran, P. J. Steel, J. Incl. Phen. 2001, 79–85; A. Lützen, M. Hapke, J. Griep-Raming, D. Haase, W. Saak, Angew. Chem. 2002, 114, 2190–2194; Angew. Chem. Int. Ed. 2002, 41, 2086–2089; G. A. Mines, B.-C. Tzeng, K. J. Stevenson, J. Li, J. T. Hupp, Angew. Chem. 2002, 114, 162–165; Angew. Chem. Int. Ed. 2002, 41, 154–157. B. Dominguez, N. S. Hodnett, G. C. Lloyd-Jones, Angew. Chem. 2001, 113, 4419–4421; Angew. Chem. Int. Ed. 2001, 40, 4289–4291. M. Albrecht, O. Spieß, M. Schneider, P. Weis, Chem. Commun. 2002, 786–787. Z. Luo, Q. Liu, L. Gong, X. Cui, A. Mi, Y. Jiang, Chem. Commun. 2002, 914–915. D. Monchaud, J. J. Jodry, D. Pomeranc, V. Heitz, J.-C. Chambron, J.-P. Sauvage, J. Lacour, Angew. Chem. 2002, 114, 2423– 2425; Angew. Chem. Int. Ed. 2002, 41, 2317– 2319.

153 M. E. Bluhm, S. S. Kim, E. A. Dertz, K. N.

154

155

156

157

158 159

160 161

162

163

Raymond, J. Am. Chem. Soc. 2002, 124, 2436–2437. G. R. Newkome, T. J. Cho, C. N. Moorefield, R. Cush, P. S. Russo, L. A. Godínez, M. J. Saunders, P. Mohapatra, Chem. Eur. J. 2002, 8, 2946–2954. S. J. Lee, W. Lin, J. Am. Chem. Soc. 2002, 124, 4554–4555; see also: Y. Yamamoto, Y. Nakagai, N. Ohkoshi, K. Itoh, J. Am. Chem. Soc. 2001, 123, 6372–6380; C. J. Kuehl, S. D. Huang, P. J. Stang, J. Am. Chem. Soc. 2001, 123, 9634–9641; C. J. Kuehl, S. D. Huang, P. J. Stang, J. Am. Chem. Soc. 2001, 123, 9634–9641. Y. Yamanoi, Y. Sakamoto, T. Kusukawa, M. Fujita, S. Sakamoto, K. Yamaguchi, J. Am. Chem. Soc. 2001, 123, 980–981; M. Yoshizawa, T. Kusukawa, M. Fujita, S. Sakamoto, K. Yamaguchi, J. Am. Chem. Soc. 2001, 123, 10454–10459. S. C. Johannessen, R. G. Brisbois, J. P. Fischer, P. A. Grieco, A. E. Counterman, D. E. Clemmer, J. Am. Chem. Soc. 2001, 123, 3818–3819; see also: Z. Qin, M. C. Jennings, R. J. Puddephatt, Chem. Commun. 2001, 2676–2677. D. K. Chand, K. Biradha, M. Fujita, Chem. Commun. 2001, 1652–1653. E. Stulz, Y.-F. Ng, S. M. Scott, J. K. M. Sanders, Chem. Commun. 2002, 524–525; see also: D. M. Guldi, C. Luo, A. Swartz, M. Scheloske, A. Hirsch, Chem. Commun. 2001, 1066–1067; C. A. Hunter, C. M. R. Low, M. J. Packer, S. E. Spey, J. G. Vinter, M. O. Vysotsky, C. Zonta, Angew. Chem. 2001, 113, 2750–2754; Angew. Chem. Int. Ed. 2001, 40, 2678–2682. S.-Y. Chang, J. S. Shoi, K.-S. Jeong, Chem. Eur. J. 2001, 7, 2687–2697. A. Hori, K. Kumazawa, T. Kusukawa, D. K. Chand, M. Fujita, S. Sakamoto, K. Yamaguchi, Chem. Eur. J. 2001, 7, 4142–4149; Au as a bridging element in Catenanes: C. P. McArdle, S. Van, M. C. Jennings, R. J. Puddephatt, J. Am. Chem. Soc. 2002, 124, 3959–3969; see also: C. O. DietrichBuchecker, N. Geum, A. Hori, M. Fujita, S. Sakamoto, K. Yamaguchi, J.-P. Sauvage, Chem. Commun. 2001, 1182–1183. C. P. McArdle, M. J. Irwin, M. C. Jennings, J. J. Vittal, R. J. Puddephatt, Chem. Eur. J. 2002, 8, 723–734. A. Aoki, M. Shiro, E. Kimura, Chem. Eur. J. 2002, 8, 929–939; see also: R. W. Saalfrank, H. Glaser, B. Demleitner, F. Ham-

79

80

2 Hetera (Cyclo-)phanes

164 165

166 167

168 169

170

171

172

pel, M. M. Chowdry, V. Schünemann, A. X. Trautwein, G. B. M. Vaughan, R. Yeh, A. V. Davis, K. N. Raymond, Chem. Eur. J. 2002, 8, 493–497; R. W. Saalfrank, B. Demleitner, H. Glaser, H. Maid, D. Bathelt, F. Hampel, W. Bauer, M. Teichert, Chem. Eur. J. 2002, 8, 2679–2683. H. Tsukube, H. Fukui, S. Shinoda, Tetrahedron Lett. 2001, 42, 7583–7586. Y. Miyake, A. Hosoda, M. Takagaki, E. Nomura, H. Taniguchi, Chem. Commun. 2002, 132–133. H. Shinmori, H. Furuta, A. Osuka, Tetrahedron Lett. 2002, 43, 4881–4884. V. Berl, M. Schmutz, M. J. Prische, R. G. Khoury, J.-M. Lehn, Chem. Eur. J. 2002, 8, 1227–1244. J. Lin, Q.-S. Hu, M.-H. Xu, L. Pu, J. Am. Chem. Soc. 2002, 124, 2088–2089. J. Brettar, J.-P. Gisselbrecht, M. Gross, N. Solladié, Chem. Commun. 2001, 733– 734. A. G. Martinez, J. O. Barcina, A. de Fresno Cerezo, Chem. Eur. J. 2001, 7, 1171– 1175; F. J. Carter, C. A. Hunter, D. J. Livingstone, J. F. McCabe, E. M. Seward, Chem. Eur. J. 2002, 8, 2848–2859. M. Wehner, T. Schrader, Angew. Chem. 2002, 114, 1827–1831; Angew. Chem. Int. Ed. 2002, 41, 1751–1754. X. Shi, J. C. Fettinger, J. T. Davis, Angew. Chem. 2001, 113, 2909–2913; Angew. Chem. Int. Ed. 2001, 40, 2827–2831; see also: R. K. O. Sigel, S. M. Thompson, E. Freisinger, F. Glahé, B. Lippert, Chem. Eur. J. 2001, 7, 1968–1980.

173 C. J. Wilds, M. A. Maier, V. Tereshko, M.

174

175

176

177 178

179 180

Manoharan, M. Egli, Angew. Chem. 2002, 114, 123–125; Angew. Chem. Int. Ed. 2002, 41, 115–117. H. Miyali, M. Didic, J. H. R. Tucker, I. Prokes, M. E. Light, M. B. Hursthouse, I. Stibor, P. Lhoták, Tetrahedron Lett. 2002, 43, 873–878. L. J. Prins, D. N. Reinhoudt, P. Timmerman, Angew. Chem. 2001, 113, 2446–2492; Angew. Chem. Int. Ed. 2001, 40, 2382–2426; L. J. Prins, R. Hulst, P. Timmerman, D. N. Reinhoudt, Chem. Eur. J. 2002, 8, 2288– 2301; T. Ishi-i, M. Crego-Calama, P. Timmerman, D. N. Reinhoudt, S. Shinkai, Angew. Chem. 2002, 114, 2004–2009; Angew. Chem. Int. Ed. 2002, 41, 1924–1929; A. G. Bielejewska, C. E. Marjo, L. J. Prins, P. Timmerman, F. de Jong, D. N. Reinhoudt, J. Am. Chem. Soc. 2001, 123, 7518– 7533. C. A. Schalley, Chem. uns. Zeit 2001, 35, 166–174; see also: A. Ikeda, H. Udzu, Z. Zhong, S. Shinkai, S. Sakamoto, K. Yamaguchi, J. Am. Chem. Soc. 2001, 123, 3872–3877. A. Bagno, G. Saielli, G. Scorrano, Chem. Eur. J. 2002, 8, 2047–2056. T. Mizuno, W.-H. Wei, L. R. Eller, J. L. Sessler, J. Am. Chem. Soc. 2002, 124, 1134– 1135. S. J. Park, J.-I. Hong, Chem. Commun. 2001, 1554–1555. K. Campbell, C. J. Kuehl, M. J. Ferguson, P. J. Stang, R. R. Tykwinski, J. Am. Chem. Soc. 2002, 124, 7266–7267.

81

3

Highly Strained Cyclophanes Takashi Tsuji

3.1

Introduction

When bridged with a side chain at the meta or para positions, the planarity-preferring benzene ring is forced to bend into a boat form, unless the bridging chain is sufficiently long. In the past 30 years, research on the preparation of bridged benzene derivatives, that is, metacyclophanes and paracyclophanes, with ever shorter bridges has made remarkable progress, driven by the interest in the properties of strained molecules as well as the challenges inherent in their synthesis and also in the pursuit of improved understanding of aromaticity [1, 2]. In this review, recent developments in the chemistry of strained cyclophanes including pyrenophanes are described. The reactivity of strained [n]meta- and [n]paracyclophanes (n £ 6) is greatly enhanced as the number of methylene groups in the bridging chain is decreased from six to five and then to four. Their enhanced reactivity is a consequence of the high steric strain present both in the bent benzene rings and in the distorted bridges, which is released in transition states upon reaction with a variety of reagents. Bending of benzene rings causes a rise of HOMO as well as a fall of LUMO in energy as a consequence of the torsion of p–bonds in the aromatic p–electron system. This also leads to activation of the benzene ring of strained cyclophane. The electron distribution in the benzene ring and the orbital coefficients of frontier orbitals are also perturbed. For the same bridge length a metacyclophane is less strained and, thus, significantly more stable in general than is its para analog.

3.2

[n]Metacyclophanes 3.2.1

Synthesis

As expected, the thermal stability of [n]metacyclophanes decreases with decreasing bridge length. Compared with shelf-stable [6]metacyclophanes 1 and the higher homologs, [5]metacyclophanes 2 exhibit markedly enhanced reactivity and the parModern Cyclophane Chemistry. Edited by Rolf Gleiter, Henning Hopf Copyright © 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 3-527-30713-3

82

3 Highly Strained Cyclophanes

Scheme 3.1

ent compound slowly polymerizes on standing at room temperature. Highly strained [4]metacyclophane 3 a has so far defied direct observation, but its generation as a fleeting intermediate has been demonstrated by interception reactions. Synthesis of 2,6-bridged p-nitrophenols 4 b, that is, substituted [n]metacyclophane derivatives (n = 6, 7, 9, 10 . . .), by base-catalyzed condensation of cycloalkanones with nitromalonaldehyde was reported as early as 1947 by Prelog et al. [3] (Scheme 3.1). Several useful routes to [6]metacyclophane and its derivatives have subsequently been devised [1]. They include (1) thermal rearrangement-aromatization of substituted [6.2.1]propellanes; (2) acid-catalyzed rearrangement of [6]paracyclophanes; (3) intramolecular diene-yne Diels-Alder reaction followed by dehydrogenation; and (4) cycloaddition of an m-xylylene diradical with 1,3-dienes. Of those methodologies, only the thermal rearrangement of propellanes has been successfully extended to the preparation of [5]metacyclophanes by Bickelhaupt et al., who developed a novel synthetic route which required relatively mild reaction conditions that sensitive [5]paracyclophanes could tolerate (Scheme 3.2). In a related approach, Reese et al. have accomplished the preparation of [5](1,3)naphthalenophane [4]. Condensation of cyclooctanone with nitromalonaldehyde by Prelog et al. did furnish a bridged compound, but the product was found to be 5 a (n = 5), a tautomer of substituted [5]metacyclophane 4 a (n = 5). Apparently, the relief of steric strain on going from 4 a to 5 a overcompensated the loss of aromatic stabilization energy. Bickelhaupt et al. found that, while treatment of syn-6 with t-BuOK in DMSO in an attempt to generate [4]metacyclophane 3 a only led to the formation of polymeric material, the corresponding reaction of anti-6 yielded 7, the Dewar isomer of 3 a (Scheme 3.3) [5]. Upon thermolysis at about 150 8C, 7 underwent rearrangement to give 3 a which was too unstable to be isolated, but instead was intercepted as Diels-Alder adducts as described later. Upon photolysis, however, 7 was quantitatively converted to the corresponding prismane, dampening the prospect of generating 3 a at low temperature [5 b].

Scheme 3.2

3.2 [n]Metacyclophanes

Scheme 3.3

3.2.2

Structures and Physical Properties

The most prominent geometrical feature of a strained metacyclophane is the outof-plane bending of the aromatic ring into an unsymmetrical boat conformation, represented by the projected deviation angles a and c (Scheme 3.4). The deformation of the aromatic ring is also recognized in the projected angle b of the benzylic carbon–carbon bonds, which reflects the degree of out-of-plane bending of the latter with respect to the aromatic ring and pyramidalization at the bridgehead carbon atoms. As expected, those deformation angles increase in the order of [4]- > [5]- > [6]metacyclophane (Tab. 3.1). Tab. 3.2 lists the strain energy (SE) evaluated for 1 a and 2 a using homodesmotic reactions and the results of their partitioning into the components, that is, the strain energies due to the distortion of benzene ring SE(C6H6), the distorted bridge SE(bridge), and the repulsive interactions between them SE(rep) [6]. The hexamethylene chain of 1 a resides at one side of the aromatic ring as illustrated in Scheme 3.4. Ring flip to the other side is slow at room temperature, but readily takes place at higher temperatures (DG{ = 17.4 kcal mol–1 at 76.5 8C). In 2 a, the bridge is too short to swing around to the other side. [6]Metacyclophanes un-

Scheme 3.4

Tab. 3.1 Deformation angles for strained [n]meta- and [n]paracyclophanes

Compound

Method

a (8)

b (8)

c (8)

(a + c)

(a + b)

1d 2d 3a 30 31 a 32 a 38 a

experimental experimental MNDO experimental SCF/DZ SCF/TZ2P B3LYP/6-31G*

19.4 26.8 40.6 19–21 23.7 29.7 28.0

39.7 48.0 46.1 18–21 28.6 38.0 46.4

8.4 12.0 12.0 – – – –

27.8 38.8 52.6 – – – –

– – – 39–40 52.3 67.7 74.5

83

84

3 Highly Strained Cyclophanes Tab. 3.2 Partition of the strain energy in strained [n]meta- and [n]paracyclophanes (kcal mol–1)

Compound

Method

SE(C6H6)

SE(bridge)

SE(sum)

SE(rep) a)

1a 2a 32 a 38 a

DFT DFT MP2/DZP MP2/6-31G*

12.1 28.4 78.2 90.5

12.8 10.7 12.2 9.1

24.9 39.2 90.4 99.6

8.4 12.0

a) SE(rep) = SE(HD)–SE(sum). SE(HD) is the strain energy obtained from a homodesmotic reaction. SE(rep) is ascribed to the repulsion between the bridge and the benzene ring.

dergo a second conformational movement, a pseudorotation of the hexamethylene bridge on the same side of the ring, with an activation barrier of DG{ = 11–13 kcal mol–1. A related conformational motion, the exo–endo equilibration, occurs in 2 with comparable ease (Scheme 3.4). The position of equilibrium depends on the structure of the bridge as well as on the substituents on the aromatic ring. The parent [5]metacyclophane exclusively exists in the exo conformer. The chlorine substituents at C2 and C5 tend to stabilize the endo conformer more than the exo conformer and the endo conformer is observed in a minor quantity in 2 d. 3.2.3

Reactions of Strained [n]Metacyclophanes

Bickelhaupt and coworkers investigated the chemistry of [5]metacyclophane and its derivatives, the smallest metacyclophane viable at room temperature, in great detail [7]. They have revealed not only the remarkable enhancement of reactivity, but also several reaction patterns that do not find an analogy in conventional aromatic behavior. They have also found that the reactivity of substituted [5]metacyclophanes is affected by some subtle factors including substituent effects and the hybridization state of the atoms in the bridge.

3.2.3.1 Thermal and Photochemical Reactions

Under flash vacuum thermolysis (FVT) conditions [5]metacyclophane 2 a undergoes homolytic cleavage at one of the benzylic bonds to generate a diradical, which cyclizes again, ultimately giving a mixture of 5- and 6-methyltetralins. FVT of 7, in contrast, mainly provided tetralin, together with lesser amounts of methylindans. The generation of [4]metacyclophane 3 a from 7 followed by rearrangement via a benzvalene intermediate has been proposed [5 d]. The preferred process on irradiation of 2 is rearrangement to the ortho isomers by way of benzvalenes, whose intermediacy is supported by labeling experiments [7 b]. It is noteworthy that higher homologs of [5]metacyclophane do not undergo an analogous rearrangement. Thus, the benzvalene formation seems to be facilitated by the distorted geometry of the benzene ring.

3.2 [n]Metacyclophanes

3.2.3.2 Addition Reactions

Simple alkylbenzenes resist the addition of dienophiles unless highly reactive dienophiles and forcing reaction conditions are employed. Strained metacyclophanes, however, react with dienophiles in Diels-Alder reactions with a rate that approaches and sometimes surpasses that of ordinary 1,3-dienes. The addition occurs across the aromatic ring at C2 and C5, because only this regiochemistry effectively removes the steric strain in the metacyclophanes. While 3 a has defied direct observation so far, it was intercepted by its Dewar benzene precursor under the condition of its generation at about 150 8C to furnish 8 a and 8 b as the primary products (Scheme 3.5). Note that these dimers resulted from the Diels-Alder reactions of 3 a as diene with the unactivated double bond of 7 as dienophile [5 e, 7 c]. In the presence of maleic anhydride or hexafluoro-2-butyne, the corresponding adduct has been obtained instead. Compared with 3 a, 2 a exhibits significantly reduced Diels-Alder reactivity, but in comparison with simple alkylbenzenes its reactivity is still spectacular [7 b, d]. With reactive dienophiles such as dimethyl acetylenedicarboxylate (DMAD) and acrylonitrile, it readily reacts at room temperature to give the corresponding adducts, respectively (Scheme 3.6). As anticipated, the reactivity of 1 is strongly reduced compared with that of 2. The parent compound 1 a still forms adducts with strong dienophiles, but chlorine substitution at the aromatic ring blocks the Diels-Alder reactions. Dichlorocarbene does not react with benzene itself. It does, however, react with aromatics in which the benzene ring is incorporated in a strained cyclophane system. Upon reaction with 2 a, dichlorocarbene preferentially adds in a 1,2-fashion to the formal “anti-Bredt” type double bond of the benzene ring of 2 a to afford the norcaradiene 9, which immediately rearranges to 11 via electrocyclic ring opening followed by a [1,5]sigmatropic chlorine migration (Scheme 3.7). A com-

Scheme 3.5

Scheme 3.6

85

86

3 Highly Strained Cyclophanes

Scheme 3.7

Scheme 3.8

putational study supports that the [1,5]chlorine migration in the rearrangement of 10 to 11 is essentially concerned with minor charge separation [8]. Reaction of 2 a with the precursor 12 of the phosphinidene complex 13 with CuCl as catalyst gave the adduct 14 as the only product (Scheme 3.8) [9]. This is the first addition of a phosphinidene complex to a benzene ring. Contrary to the 1,2-addition of dichlorocarbene, 13 adds to 2 a in a 1,4-fashion. The difference in behavior between these divalent species is rationalized in terms of a difference in size of the central atoms. As the C(sp3)–C(sp3) bond (1.54 Å) is considerably shorter than the analogous C–P bond (1.87 Å), the carbon atom is apparently too small to achieve sufficient overlap between the frontier orbitals of 2 a and dichlorocarbene in the transition state for the 1,4-addition. Theoretical calculations also support the mechanism of the reaction of 13 with 2 a as being a direct 1,4-addition. In sharp contrast to 12, 14 remains unchanged even under prolonged heating in the presence of CuCl. The decomplexation of 14 with iodine and methylimidazole afforded 15, the first uncoordinated 7k3-phosphanorbornadiene, which is moderately stable at room temperature toward loss of the phosphinidene bridge. Previous attempts to obtain a simple, unbridged 7k3-phosphanorbornadiene had failed, presumably because it rapidly excluded the phosphorus bridge. Compounds 14 and 15 undoubtedly owe their kinetic stabilization to the overall increase in strain that results on dissociation into highly strained 2 a and the phosphorus species.

3.2.3.3 Reactions with Electrophiles

A characteristic reaction of short-bridged metacyclophanes is the acid-catalyzed rearrangement to their corresponding ortho isomers. While 1 a is relatively stable towards acids, [5]metacyclophanes are particularly sensitive and undergo the rear-

3.2 [n]Metacyclophanes

Scheme 3.9

rangement under surprisingly mild conditions [7 b–d]. The mechanism illustrated in Scheme 3.9 is supported by labeling experiments and product analysis. The key step is protonation of one of the bridgehead positions, which is followed by a 1,2-alkyl shift. It should be noted that each step is facilitated by a significant reduction in steric strain. In the reaction of 2 d, a 1,2-chloride shift precedes the final deprotonation to afford 16 a. The replacement of the chlorine atom at C2 in 2 d with a methoxyl group as in 2 e leads to the formation of 19. Owing to the low migratory aptitude of the methoxyl group, 17 undergoes a further 1,2shift of the alkyl bridge followed by loss of the methyl group via attack by a nucleophile, eventually to furnish 19. The bromination of 2 a occurs without a catalyst at –75 8C, but again proceeds with rearrangement to afford 16 b.

3.2.3.4 Reactions with Nucleophiles

Haloaromatics are normally unreactive towards nucleophilic substitution unless the aromatic ring is activated by strongly electron-withdrawing groups [7 a]. Here, [5]metacyclophanes again constitute an exception, as they are unusually reactive in nucleophilic substitutions without the need for activation by electron-withdrawing groups. Thus, 2 c, d readily undergo attack by alkoxide ions at the ipso carbon atom to give 2 f via the Meisenheimer complex 20 (the SN2Ar mechanism: Scheme 3.10). The reaction is slower when the chlorine at C5 is missing, in line

Scheme 3.10

87

88

3 Highly Strained Cyclophanes

Scheme 3.11

with the SN2Ar mechanism. The reaction of 2 b with alkoxide ions similarly occurs at the ipso position, but the subsequent course of the reaction is different and the intermediate adds a proton to give 21 (Scheme 3.11). Fluorobenzenes are normally better substrates than the corresponding chlorides for substitution by the SN2Ar mechanism. The distinctive behavior of 2 b compared with that of 2 c, d has been ascribed to the shift of the rate-determining step from the initial addition of the nucleophile, which is normally rate-determining, to the next step, the extrusion of the halide ion, in the reaction of 2 b. In the reactions of 2, the attack by the nucleophile is accompanied by the release of a considerable amount of strain because the intermediates, such as 20, do not have “antiBredt” double bonds involving C2 whereas the extrusion of the halide ion is counteracted by the reintroduction of the strain. Under these specific circumstances, the poor nucleofugality of the fluorine substituent retards the fluoride ion extrusion to the extent that the alternative pathway of protonation can prevail, resulting in the formation of 21. When 2 b is treated with a large excess of MeONa, cyclohexadienone 22 is formed via the SN2 substitution with concomitant fluoride extrusion. Intuitively, one would have expected that the reaction of 2 c, d with the hydroxide ion should proceed by the SN2Ar mechanism similar to the attack by alkoxides. However, the reaction takes a completely different course, giving 24 instead of the substitution products [7 d]. This has been interpreted as outlined in Scheme 3.12. Nucleophilic attack at the bridgehead position is in general the kinetically preferred primary mode of attack, even in the alkoxide reactions. In these reactions, however,

Scheme 3.12

3.3 [n]Paracyclophanes

Scheme 3.13

the primary intermediates analogous to 23 do not have a rapid follow-up reaction available, but an unproductive reversion to the reactants, and the alternative, kinetically less favorable attack at C2 eventually opens the pathway leading to 2 f. In its reaction with hydroxide ion, 2 b again exhibits behavior quite different from that of 2 c, d, giving 22 instead of 24 (X = H). The kinetically preferred adduct, corresponding to 23, being incapable of undergoing a ring contraction because of the poor nucleofugality of the fluorine substituent, the attack at C2 provides a chance to furnish 25, from which 22 is derived. Treatment of 26 with t-BuOK in DMSO, in an attempt to synthesize dichloro[1.1]metacyclophane 27, led to the formation of 28 and 29 in 9 and 10% yield, respectively (Scheme 3.13) [10]. The formation of 28 has been rationalized in terms of the attack on 27 by an adventitious hydroxide ion at one of the bridgehead positions followed by ring opening. The yield of 28 increased to 40% when 4 equivalents of water were purposely added: in this case 29 was absent.

3.3

[n]Paracyclophanes

In the [n]paracyclophane series, the borderline for stability at ambient temperature has been found between [6]paracyclophane 30, which is an isolable, chemically fairly stable compound, and [5]paracyclophane 31, which has so far been observed only in solution below room temperature [1, 2]. [4]Paracyclophane 32 is observable only under matrix isolation at low temperature and rapidly disappears in fluid solution even below –100 8C. The instability of 31 and 32 most probably arises from their high propensity for undergoing addition at the bridgehead positions, whereby the steric strain inherent in the ring systems is largely released: their half-lives in solution are concentration-dependent, indicating a bimolecular or chain mechanism of polymerization. Thus, the kinetic stability of 32 is markedly ameliorated by introducing substituents that sterically hinder access to the bridgehead carbon atoms by other reagents. [4]Paracyclophane is so strained that its energy is comparable to or slightly higher than that of the Dewar isomer 36

89

90

3 Highly Strained Cyclophanes

and its kinetic stability is also limited by the intrinsic reactivity to rearrange thermally to the latter. 3.3.1

Synthesis

The first synthesis of [6]paracyclophane was accomplished by Jones and coworkers in 1974, using an ingenious methodology, a rearrangement of highly energetic carbene intermediate 33 (Scheme 3.14) [11]. This reaction, however, provided 30 a only in low yield and more efficient routes to [6]paracyclophane and its derivatives have subsequently been explored, that is, the thermal, photochemical or silver-ion-catalyzed aromatization of corresponding 1,4-bridged Dewar benzenes by Jones, Bickelhaupt, Tobe, and Gleiter, and the ring contraction of 3,6-bridged oxepin derivative by Tochtermann [1 c]. The preparation of 1,4-bridged Dewar benzenes has been achieved in turn by the methods including (1) silver-ion-catalyzed isomerization of 3,3'-hexamethylene bicyclopropenyl; (2) ring contraction of [6.3.2]propellane derivatives; and (3) AlCl3-catalyzed intramolecular dimerization of 1,10-dodecadiyne followed by cycloaddition of a third alkyne. [6]Paracyclophane is stable enough to permit structural modification and, by annelating benzene ring(s) to its aromatic core, [6](1,4)naphthalenophanes as well as [6](1,4)- and [6](9,10)anthracenophanes have been prepared [12]. Of these strategies, only the photochemical Dewar benzene route has successfully been applied for the generation of [5]- and [4]paracyclophanes (Schemes 3.15 and 3.16). While 30 is obtained in high yield by mildly heating the corresponding Dewar isomer, the lower homologs have been prepared solely by photolysis of their Dewar isomers. The thermal aromatization of 1,4-bridged Dewar benzene requires increasingly high temperature as the length of bridging chain is shortened. Strained 31 cannot survive the thermolysis conditions required to generate it from relatively stable 35. Photolysis, however, can be carried out regardless of the reaction temperature, allowing the preparation of sensitive 31 [13] and 32 [14, 15] at low temperature. Since both 31 and 32 possess more intense, far-reaching (in wavelength) chromophores than the Dewar precursors and are susceptible to the photochemical reversion to give the latter, the isomerization of the latter into the former by irradiation can not be brought to completion, but reaches a photostationary state usually consisting mainly of the Dewar isomer. The photochemical aromatization of the Dewar benzenes seems to take place in their singlet excited

Scheme 3.14

3.3 [n]Paracyclophanes

Scheme 3.15

Scheme 3.16

states and appears to be hindered by substituents which facilitate intersystem crossing to the triplet excited states [16 c]. 3.3.2

Structures and Physical Properties

The most prominent feature in the structure of strained paracyclophane is the bending of its aromatic ring into a boat form, accompanied by the out-of-plane bending of the benzylic carbon–carbon bonds. The degrees of these distortions are given by the critical angles a and b, defined respectively as in Scheme 3.15. Tab. 3.1 displays experimental geometrical parameters for 30, together with those calculated for 31 a, 32 a, and 38 a. The good agreement between the experimental and calculated values for 30 suggests that the values computed for the smaller analogs will also be sufficiently reliable. The calculated structure of 38 a, the most highly strained paracyclophane ever experimentally accessed, is instructive; 38 a has C2v symmetry and the bridge bisects the bent benzene ring [15 b]. The extent of bond alternation in the benzene ring is surprisingly small for the severe deformation of the ring, less than 0.02 Å, irrespective of the methods employed (SCF and MP2/6-31G*, B3LYP/6-31G*). Thus, the severely twisted bonds between the bridgehead and the adjoining carbon atoms are only slightly longer than the neighboring non-twisted bonds, suggesting the sustenance of cyclic delocalization of p-electrons in the benzene ring. The predicted deformation angles a and b are in the ranges of 27–29 8 and 45–48 8, respectively, and the sum (a + b) amounts to 74–75 8. As a result, the bridgehead carbon atoms are markedly pyramidalized, decreasing the sum of the bond angles around each of them to 339.3 8 (B3LYP/631G*) from 360 8 for a planar trivalent carbon atom. The residual aromatic carbon atoms are also pyramidalized (the sum of the bond angles is 355.6 8) as illustrated

91

92

3 Highly Strained Cyclophanes Tab. 3.3 Longest wavelength absorption bands of strained [n]paracyclophanes

kmax (nm)

30 a

30 b

31 a

31 b

32 a

32 b

296

329

330

360

370–380

425

in Scheme 3.16, so as better to maintain p-orbital overlap with adjacent bridgehead carbon atom. The strain energy due to the distortion of the benzene ring and that due to the distorted bridge are evaluated separately for 32 a and 38 a, and listed in Tab. 3.2 [2 b, 17]. The strain is evidently localized largely in the bent benzene moiety in the [4]paracyclophanes. The strain energies of 30 a and 31 a have been estimated to be 40 and 63 kcal mol–1, respectively. It is of interest that 38 is a fully unsaturated species and the p-bond isomerism in 38 gives rise to bicyclo[4.2.2]decapentaene 39 as an alternative structure. All the available data support the adoption of the former structure in preference to that of the latter; namely, the similarity in shape of the UV/Vis absorption spectra of 38 a, b to those of 32 a, b, respectively, and the structures of chemically intercepted products (see below), in addition to computational analysis. According to the theoretical calculations, however, the latter is only slightly less stable, by 3.9 kcal mol–1 at the B3LYP/6-31G* level, than the former, suggesting that the structural preference may be reversed by introducing suitable substituents [15]. The 1H NMR spectra of [5]- and [6]paracyclophanes exhibit temperature-dependent NMR behavior caused by flipping of the oligomethylene bridge. The barrier of flipping is in the range of 10–15 kcal mol–1. Protons in the central portion of the bridge of 31 are upfield shifted in their spectra and resonate at high field up to d –0.62, indicating that, despite the severe distortion, the benzene ring retains its aromatic character with remarkable tenacity [13]. This notion is further strengthened by the examination of the magnetic property of [4]paracyclophane as described below. Absorption bands in the UV/Vis spectra of paracyclophanes are increasingly shifted bathochromically with decreasing length of the bridging chain, in agreement with the predicted narrowing of the HOMO–LUMO energy gap in a bent benzene ring. The representative results are listed in Tab. 3.3. 3.3.3

Reactions of Strained [n]Paracyclophanes 3.3.3.1 Thermal and Photochemical Reactions

When heated, [6]paracyclophane undergoes homolysis at one of the benzylic carbon-carbon bonds to generate diradical 41 b, which then collapses to afford the spiro derivative 42 b (Scheme 3.17) [18 a]. This type of thermal transformation is reversible for 40 a/42 a, suggesting that 40 a and 42 a are nearly balanced in energy while the interconversion is in favor of the spiro form in the more strained [6]paracyclophane system. The strain of paracyclo-

3.3 [n]Paracyclophanes

Scheme 3.17

Scheme 3.18

phane is rapidly accumulated as the number of bridging carbon atoms decreases from eight to four, whereas the strain of the corresponding Dewar isomers remain virtually unaffected. Thus, the relative energy of a paracyclophane to its Dewar isomer is expected to reverse at a certain chain length. Theoretical calculations suggest that, when the bridging chain is shortened to tetramethylene, the Dewar form becomes isoenergetic to or slightly lower in energy than the cyclophane form [17 a, c]. This reversal of stability order has been corroborated for the pair of 43 and 44 : 44 thermally reverts to 43 with a half-life 15 ± 5 min at –20 8C (DG{ = 18.3 ± 0.3 kcal mol–1: Scheme 3.18) [16 b, c]. The efficient valence isomerization of benzene ring into the Dewar form upon irradiation is a characteristic reactivity common to strained [n]paracyclophanes (n £ 6). Although photochemical conversions of aromatic compounds into Dewar isomers have been well documented, they are inefficient, minor processes in most of the cases. This type of transformation is thus greatly facilitated by the enforced distortion of benzene ring into the boat form by a short bridge in the strained paracyclophanes. The activation of [7]- and [8]paracyclophanes toward the analogous isomerization appears to be insufficient and their photolysis merely produces polymeric material. While [6](1,4)- and [5](1,4)naphthalenophanes photochemically behave similarly, giving corresponding 1,4-bridged benzo Dewar isomers upon irradiation, [6](1,4)anthracenophane preferentially undergoes photo-dimerization to give a mixture of [2 + 2] stereoisomers conjoined at C1 and C2 [18 c]. Irradiation of a mixture of 32 a and 36 a in a quasi-photoequilibrium, in a glass matrix at 77 K, leads to the formation of p-quinodimethane 45, presumably via photochemical extrusion of ethylene from 32 a in competition with the cycloreversion to give 36 a (Scheme 3.19) [14].

93

94

3 Highly Strained Cyclophanes

Scheme 3.19

3.3.3.2 Reactions with Electrophiles

As is the case with strained metacyclophanes, 30 a is sensitive to acid and readily undergoes rearrangement, upon exposure to trifluoroacetric acid, to give a mixture of less strained 1 a and the ortho isomer. Since the reaction proceeds under the conditions under which 1 a is inert, the latter is formed directly from 30 a via sequential Wagner-Meerwein shifts in the protonated intermediate. In contrast, 32 a undergoes 1,4-addition at the bridgehead positions in preference to the rearrangement leading to tetralin in the reactions with protic solvent molecules (Scheme 3.20) [14, 19]. Due to the extreme steric strain, 32 a behaves as an unusually basic hydrocarbon and reacts with methanol even under neutral conditions. [5]Paracyclophane shows an intermediate behavior and affords both 1,4-adducts and the rearranged product. The observed distinctive behaviors of the strained paracyclophanes in the reaction with electrophiles are presumably dictated by factors including the charge distribution in the cation intermediate 47, the degree of strain release in the transition state leading to each product from 47, and the nucleophilicity of the reactant which adds to 47. When exposed to bromine, 30 a suffers addition reaction rather than substitution reaction to give the 1,4-adduct almost quantitatively [18 b]. The benzo-annelated analogs of 30 a undergo acid-induced oligomerization in preference to the rearrangement to the less strained isomers. Their dimers and trimers were isolated and their structures were elucidated by X-ray crystallography [18 b, d]. [6](9,10)Anthracenophanes are highly sensitive to acid and readily isomerize to the 9-methylene-10-dihydro derivatives, as is predicted by theoretical calculations [20].

3.3.3.3 Diels-Alder and Other Reactions

[6]Paracyclophane 30 a exhibits enhanced reactivity in the reaction with dienophiles, compared with the less strained homologs, and affords adducts at the non-

Scheme 3.20

3.3 [n]Paracyclophanes

Scheme 3.21

Scheme 3.22

bridgehead positions. The reactions of the benzo-annelated analogs of 30 a with dienophiles take different courses which are initiated by the addition to the bridgehead position and proceed stepwise via zwitterion and/or diradical intermediates depending on the dienophile and the polarity of the solvent [18 d]. The extremely bent benzene ring of [4]paracyclophadiene 38 a behaves as an activated alkene in the reaction with cyclopentadiene and the transient 38 a has been efficiently intercepted by the latter to afford a mixture of twofold adducts 48 a and 48 b; a rare, if not unprecedented, case in which an aromatic ring participates as a dienophile in a Diels-Alder reaction (Scheme 3.21) [15]. This result provided strong support for the preference of structure 38 over the p–bond shifted structure 39. A bridged cyclooctatetraene derivative has been obtained in the reaction of a substituted [6]paracyclophane with dichlorocarbene, while the major product in the reaction with dibromocarbene is a bridged cycloheptatriene [21]. Oxidation of 30 a with mCPBA leads to the dimeric oxidation product 50, for which epoxidation followed by rearrangement to give 49 has been proposed as a mechanism (Scheme 3.22) [18 a, d]. Butyllithium and s-BuLi add to 30 a in 1,2- or 1,4-fashion, while with t-BuLi or tBuOK/BuLi its aromatic ring is metalated, allowing the functionalization of the aromatic ring by subsequent treatment with electrophiles [22].

3.3.3.4 Kinetic Stabilization of [4]Paracyclophane Systems

The high lability of [4]paracyclophane evidently arises from its propensity for undergoing addition at the bridgehead positions, as exemplified by the formation of adducts such as 46 and 48. Thus, it is suggested that the ring system may be kinetically stabilized by introducing substituents that specifically shield the bridgehead sites from access by attacking reagents. Among a number of derivatives designed and prepared on the basis of such consideration, the best result has so far been obtained with 44 whose lifetime in solution was successfully extended to 12 min at –20 8C [16 a, c]. The measurement of the 1H NMR spectrum of [4]paracyclophane ought to provide not only definitive evidence for its formation

95

96

3 Highly Strained Cyclophanes

but also valuable information concerning the aromaticity of its severely bent benzene ring, but has been thwarted because of the thermal instability. The improved thermal stability of 44 is high enough to permit the measurement of the 1H NMR spectrum, which revealed the sustenance of a substantial diamagnetic current in the benzene ring, despite its extreme bending (a + b = 72.5 8). Thus, the isomerization of 43 to 44 is attended by the pronounced upfield shift of the Ha signal and the marked downfield shift of the Hb signal as depicted in Scheme 3.18. Theoretical calculations also support the retention of substantial aromaticity in the bent benzene moiety of 44 as judged by the magnetic criteria of aromaticity, namely diamagnetic susceptibility exaltation and nucleus-independent chemical shift (NICS).

3.4

Aromaticity of Bent Benzene Rings

The strain energy of [4]paracyclophane (90–100 kcal mol–1), which mainly arises from the distortion of the benzene ring as described above, far exceeds the resonance stabilization energy of benzene (20–30 kcal mol–1). Thus, its aromatic stabilization energy compensates only a fraction of the strain energy at best. Moreover, [4]paracyclophane and its derivatives behave chemically like activated alkenes and readily undergo addition reactions rather than substitution reactions. Accordingly, they cannot be regarded as aromatics so far as their chemical reactivities are concerned. How can the energy of [4]paracyclophane and its chemical properties be reconciled with the observed diatropicity, which is indicative of the sustenance of cyclic delocalization of p-electrons in the benzene ring? In benzene, distortion into a nonplanar geometry unavoidably gives rise to torsion of p-bonds, which is not suppressible by localization of p-bonds. Thus, the carbon atoms of a benzene ring undergo extensive pyramidalization–rehybridization upon severe bending, as theoretical calculations reveal, to avoid the rupture of the p-bonds leading to degeneration into a biradical(oid), at the expense of energy even greater than its aromatic stabilization energy, and, consequently, the cyclic conjugation seems to be retained. The sustenance of aromatic conjugation in the severely bent benzene rings, as found in [4]paracyclophane, is a consequence of accommodation of the geometrical constraint imposed on the ring and basically not attributable to the aromatic stabilization effect. Accordingly, strained cyclophanes such as [5]metaand [4]paracyclophanes are kinetically highly reactive, high energy species, yet they seem to sustain rather strong diatropicity and only a small degree of bond alternation [2, 16 c, 23].

3.5

Cyclophanes containing Polycyclic Aromatic Rings: (2,7)Pyrenophanes

Aiming at the construction of bowl-shaped molecules, Bodwell and coworkers have developed a new synthetic route to strained 2,7-bridged pyrenes in which the

3.5 Cyclophanes containing Polycyclic Aromatic Rings: (2,7)Pyrenophanes

Scheme 3.23

polycyclic aromatic hydrocarbon framework is bent over its entire aromatic surface [24]. Pyrene is a repeating subunit around the equator of D5h C70, D5d C80, and D6h C84. The synthetic approach is based upon the valence isomerization of [2.2]metacyclophane-1,9-dienes to 10b,10c-dihydropyrenes followed by dehydrogenation, and severe nonplanarity is imparted to the pyrene system by the incorporation of a tether between the 2 and 7 positions as outlined in Scheme 3.23. Among the known pyrenophanes including those prepared in this way, the pyrene moiety of 1,7-dioxa[7](2,7)pyrenophane 53 a suffers the highest degree of the overall bend (h = 109.1 8), as measured by the angle h between the C1-C2-C3 and C6-C7-C8 planes. It is even slightly more bent than the pyrene subunit in C70, though the local curvature is less as evaluated by the p-orbital axis vector (POAV) analysis [24 a]. The bending in 53 a is quite evenly distributed over the surface of the pyrene unit. Attempted synthesis of [6]- and 1,6-dioxa[6](2,7)pyrenophanes from the corresponding metacyclophane-1,9-dienes has resulted in the almost quantitative recovery of the latter. The way in which the distortion from planarity of the pyrene system influences its p-electron delocalization was investigated by using two quantitative measures of aromaticity based on geometry and magnetism. Both methods suggest that the p-electron structure of the pyrene moiety is almost untouched and its aromaticity is diminished only slightly upon increasing the bend angle h from 0 8 to 109.2 8 [24 e]. The [n](2,7)pyrenophanes exhibit enhanced reactivity with decreasing length of the bridging chain [24 b]. Thus, 53 a reacts with TCNE in benzene at room temperature to give 55, while 53 b does not react even at reflux (Scheme 3.24). Addition of PTAD to 53 a rapidly proceeds at room temperature to afford a 2 : 1 PTADpyrenophane adduct 56 a. In contrast to TCNE, PTAD did react with 53 b to furnish an analogous 2 : 1 adduct, but the reaction was clearly slower than that of

Scheme 3.24

97

98

3 Highly Strained Cyclophanes

Scheme 3.25

53 a. The higher homolog 53 c showed only traces of reaction with PTAD even after prolonged reaction time. AM1 calculations predict that the difference in strain energy between 53 a and 53 b is of the order of 15 kcal mol–1. Attempted bromination of 54 a and 54 b, and Friedel-Crafts acetylation of the former all met with difficulty and have failed so far to furnish the functionalized pyrenophanes. Treatment of 53 a or 53 b with t-BuLi, in an attempt to metalate the pyrenophane regioselectively, has resulted in the cleavage of the aryl-oxygen bond. Reduction of 54 a, d with lithium metal gives dimeric dianions 57 a, b, respectively, via the coupling of their radical anion intermediates at C6 (Scheme 3.25). Upon further reduction with lithium metal, these species undergo concomitant bond cleavage/intramolecular bond formation to afford monomeric dianions 58 a, b [25]. In contrast, the two electron reduction of pyrene with alkali metals converts it into an antiaromatic dianion (pyrene)2–. In the two-electron reduction of the strained pyrenophanes 54 a, d, the system dodges getting into an unfavorable strained antiaromatic state by undergoing the intramolecular bond formation which introduces a pronounced fold into what was a pyrene moiety.

3.6

[1.1]Paracyclophanes

The construction of [1.1]paracyclophane 61, that is the connection of two benzene rings at the para positions with two methylene bridges in a cyclic array, apparently requires severe bending of the rings and bonds. Despite its fascinating structure, to our knowledge, no report concerning 61 had been published when we embarked on its preparation. The successful generation of [4]paracyclophanes from the corresponding Dewar benzene precursors, however, suggested that 61 might well be accessible via valence isomerization of the bis(Dewar benzene) isomers 59. The first step of this isomerization from 59 to 60 is nothing else but the generation of a [4]paracyclophane skeleton, if a stepwise mechanism is assumed, and the second step corresponds to the formation of much less strained, six-carbonbridged paracyclophane from the respective Dewar benzene precursors (Scheme 3.26). Thus, the second step should be much easier than the first. This expectation is supported by computational analysis on 59 a–61 a which reveals, as described later, that the initial step is a slightly endothermic process whereas the second step is highly exothermic. The degree of ring deformation in 61 a is comparable to that calculated for [5]paracyclophane and much less than that for

3.6 [1.1]Paracyclophanes

Scheme 3.26

[4]paracyclophane. Although the parent [1.1]paracyclophane lacks the stability to persist in solution at ambient temperature, the ring system has been stabilized kinetically by introducing suitable substituents to such an extent that isolation as crystals is allowed. 3.6.1

Synthesis

Bis(Dewar benzene) precursor 59 is obtained via a synthetic pathway including twofold cycloaddition of acetylene to bis(enone) 62 as a key step (Scheme 3.27) [26]. The subsequent transformation of 63 to 59 was carried out in essentially the same manner as reported for the preparation of [4.2.2]propellatetraene 37 [27]. Compound 62 is in turn prepared in five steps from diethyl dihydroterephthalate. The photocycloaddition of acetylene to 62 occurs stereoselectively to furnish the anti-bis(acetylene) adduct 63 a. The observed stereoselectivity probably results from the preferential bending of the central six-membered ring toward the cyclobutene ring in the mono-adduct as suggested by force field calculations. Irradiation of 59 a, b gives rise to 61 a, b, respectively, which are both photochemically labile and subsequently converted into the transannular addition products 62. Neither the less symmetrical products resulting from the aromatization in only one of the Dewar benzene moieties nor prismane derivatives have been detected in the photolysate. The secondary transannular addition within 61 to give 64, to our knowledge, represents the first direct formation of a benzene p,p'-dimer and is certainly a consequence of the face-to-face arrangement of the bent benzene rings in close proximity (Scheme 3.28). The thermal stabilities of 61 a, b are comparable to that reported for [5]paracyclophane and their UV/Vis absorption spectra remain unchanged for several hours at –20 8C but decay within 4 h at room temperature, suggesting that the reactivity of [1.1]paracyclophane is primarily determined by the extent of bending of

Scheme 3.27

99

100

3 Highly Strained Cyclophanes

Scheme 3.28

the aromatic rings and the ring system is neither particularly stabilized nor particularly destabilized by the close stacking of the benzene rings. 3.6.2

Kinetic Stabilization of [1.1]Paracyclophane Systems

The known reactivity of strained paracyclophanes suggests that the kinetic stabilization of [1.1]paracyclophane systems requires the protection of all four bridgehead carbon atoms from access by other reagents. Such consideration led to the design of a number of substituted derivatives, among which 61 c gave the best result [28]. The substituents in 61 c are conformationally flexible and the effective bulkiness is very much dependent on their adopted conformations, yet they are expected to protect all the bridgehead carbon atoms in the most preferred conformation. The UV/Vis absorption due to 61 c decays only by 8% after a solution in n-decane is heated at 100 8C for 2 h, demonstrating the greatly improved stability. Irradiation of 59 tends to afford mainly 64 rather than 61, except at very low conversion, because 61 is efficiently transformed into 64 upon excitation and, moreover, the chromophore of 61 is more intense and far-reaching in wavelength than that of 59. The transannular adduct 64 c, however, undergoes thermal cycloreversion to regenerate 61 c, rendering the preparation of fairly air-sensitive 61 c in pure form rather simple. Thus, air-stable 64 c is readily isolated as colorless crystals and it can be quantitatively converted to 61 c by mildly heating its solution in a de-aerated solvent. 3.6.3

Structures and Physical Properties

The geometrical as well as electronic structures of [1.1]paracyclophane, its strain energy, and thermodynamic stability relative to related compounds are of special interest. To gain insights into these points, theoretical analyses were carried out by ab initio and DFT quantum mechanical methods [26 b]. The geometry-optimized structure of 61 a is D2h symmetric and the nonbonding interatomic distance between the opposing bridgehead carbon atoms is in a range of 2.36– 2.40 Å, compared with 2.778 Å in [2.2]paracyclophane, suggesting strong electronic interactions between the p-bonds of the aromatic rings. The bending angles a and b are about 23 8 and 27 8, respectively. Relative energies calculated for 61 a and relevant related compounds are collected in Tab. 3.4.

3.6 [1.1]Paracyclophanes

Inspection of Tab. 3.4 reveals that 61 a is certainly the most stable of the isomers and that the isomerization of 59 a to 60 a is endothermic whereas that of 60 a to 61 a is strongly exothermic. Accordingly, it is not unreasonable that 60 is not detected during the photolysis of 59, though a stepwise mechanism is likely in operation. Although 61 a is the most stable of the calculated isomers, it is still a highly strained molecule. For the evaluation of strain energy in 61 a, 9,10-dihydroanthracene serves as an ideal reference compound. The total strain energy of 61 a thus evaluated is 128.1, 93.6, and 106.5 kcal mol–1 at the RHF/6-31G*, MP2/ 6-31G*, and B3LYP/6-31G* levels, respectively. The calculations confirm that the HOMO and LUMO of 61 a are significantly raised and lowered in energy, respectively, compared with those of p-xylene as a consequence of bending of the benzene rings and of their mutual electronic coupling. The reduced HOMO–LUMO energy gap in 61 a (8.83 eV), compared with that in p-xylene (12.33 eV) manifests itself in the UV/Vis absorption spectrum. [1.1]Paracyclophane 61 a exhibits pronounced bathochromic shifts of absorption bands (kmax 290, 377 nm), compared with [2.2]paracyclophane (kmax 284, 302 nm), reflecting the severer distortion and the stronger transannular electronic interactions in the former than in the latter. A broad, relatively weak band (kmax 377 nm) in the range of 330–450 nm in the spectrum of 61 a is a characteristic feature common to the spectra of [1.1]paracyclophanes: 370–480 nm (kmax 405 nm) for 61 b and 400–530 nm (kmax 461 nm) for 61 c. The molecular structure of 61 c in the crystalline state has Ci symmetry and the polycyclic core is slightly distorted from ideal D2h symmetry (Fig. 3.1) [28]. The transannular interatomic distance between the opposing bridgehead carbon atoms is 2.376(5) Å, less than the sum of the van der Waals radii (3.5–3.6 Å) by more than 1.0 Å. The averaged total bending angle (a + b ) is 49.8 8, which is the largest value ever observed for a paracyclophane and only slightly less than that calculated for [5]paracyclophane. Thus, the geometrical parameters of 61 a given by the theoretical calculations are in good agreement with the experimental data. The bridgehead carbon atoms are effectively protected by the aromatic substituents, also in good agreement with the results of molecular modeling: this is certainly the reason for its remarkable stability. The thermal isomerization of 64 c to 61 c follows first-order kinetics and activation parameters for the process are: DH{ = 21.1 ± 0.8 kcal mol–1 and DS{ = –10.5 ± 2.6 cal mol–1 K. The half-life of 64 c at 40 8C is 191 ± 2 min. The thermal instability of 64 a, b might be due to their intrinsic reactivity to revert thermally to 61 a, b, respectively, which would be rapidly consumed under the conditions.

Tab. 3.4 Calculated energies of 59 a, 60 a, and 64 a relative to [1.1]paracyclophane 61 a (kcal mol–1)

Method

59 a

60 a

61 a

64 a

SCF/6-31G* B3LYP/6-31G*

51.5 70.7

62.3 –

0 0

8.7 25.6

101

102

3 Highly Strained Cyclophanes

Fig. 3.1

Crystal structure of 61 c

3.7

References and Notes (a) F. Bickelhaupt, W. H. de Wolf, in Advances in Strain in Organic Chemistry, ed. by B. Halton, JAI Press, Greenwich, Ct. 1993, 3, 185–227; (b) Y. Tobe, in Topics in Current Chemistry, ed E. Weber, Springer, Berlin 1994, 172, 1–40; (c) V. V. Kane, W. H. de Wolf, F. Bickelhaupt, Tetrahedron, 1994, 50, 4575–4622; (d) H. Hopf, Classics in Hydrocarbon Chemistry, Wiley-VCH, Weinheim, 2000, 321–378. 2 (a) T. Tsuji, in Advances in Strained and Interesting Organic Molecules, ed. B. Halton, JAI Press, Stamford, Ct., 1999, 7, 103–152; (b) T. Tsuji, M. Ohkita, H. Kawai, Bull. Chem. Soc. Jpn., 2002, 57, 415– 433. 3 (a) V. Prelog, K. Wiesner, Helv. Chim. Acta 1947, 30, 1465–1471; (b) V. Prelog, K. Wiesner, W. Ingold, O. Häfliger, Helv. Chim. Acta, 1948, 31, 1325–1341. 4 P. Grice, C. B. Reese, J. Chem. Soc., Chem. Commun., 1980, 424–425. 1

(a) L. A. M. Turkenburg, J. W. van Straten, W. H. de Wolf, F. Bickelhaupt, J. Am. Chem. Soc., 1980, 102, 3256–3257; (b) G. B. M. Kostermans, M. Hogenbirk, L. A. M. Turkenburg, W. H. de Wolf, F. Bickelhaupt, J. Am. Chem. Soc., 1987, 109, 2855–2857; (c) G. B. M. Kostermans, P. van Dansik, W. H. de Wolf, F. Bickelhaupt, J. Am. Chem. Soc., 1987, 109, 7887– 7888; (d) G. B. M. Kostermans, P. van Dansik, W. H. de Wolf, F. Bickelhaupt, J. Org. Chem., 1988, 53, 4531–4534. 6 M. J. van Eis, W. H. de Wolf, F. Bickelhaupt, R. Boese, J. Chem. Soc., Perkin Trans. 2, 2000, 793–801. 7 (a) P. A. Kraakman, J.-M. Valk, H. A. G. Niederländer, D. B. E. Brouwer, F. M. Bickelhaupt, W. H. de Wolf, F. Bickelhaupt, C. H. Stam, J. Am. Chem. Soc., 1990, 112, 6638–6646; (b) L. W. Jenneskens, H. J. R. de Boer, W. H. de Wolf, F. Bickelhaupt, J. Am. Chem. Soc., 1990, 112, 8941–8949; (c) F. Bickelhaupt, 5

3.7 References and Notes

8

9

10

11 12

13

14

15

16

W. H. de Wolf, J. Phys. Org. Chem., 1998, 11, 362–376; (d) G. W. Wijsman, W. M. Boesveld, M. C. Beekman, M. S. Goedheijt, B. L. M. van Baar, F. J. J. de Kanter, W. H. de Wolf, F. Bickelhaupt, Eur. J. Org. Chem., 2002, 614– 629. M. J. van Eis, B. S. E. van der Linde, F. J. J. de Kanter, et al., J. Org. Chem., 2000, 65, 4348–4354. (a) M. J. van Eis, C. M. D. Komen, F. J. J. de Kanter, et al., Angew. Chem., 1998, 110, 1656–1658; Angew. Chem. Int. Ed., 1998, 37, 1547–1550; (b) M. J. van Eis, H. Zappey, F. J. J. de Kanter, W. H. de Wolf, F. Bickelhaupt, J. Am. Chem. Soc., 2000, 122, 3386–3390. M. J. van Eis, F. J. J. de Kanter, W. H. de Wolf, F. Bickelhaupt, J. Org. Chem., 1997, 62, 7090–7091. V. V. Kane, A. D. Wolf, M. Jones, Jr., J. Am. Chem. Soc., 1974, 96, 2643–2644. (a) Y. Tobe, S. Saiki, N. Utsumi, T. Kusumoto, H. Ishii, K. Kakiuchi, K. Kobiro, K. Naemura, J. Am. Chem. Soc., 1996, 118, 9488–9497; (b) Y. Tobe, S. Saiki, H. Minami, K. Naemura, Bull. Chem. Soc. Jpn., 1997, 70, 1935–1942. (a) L. W. Jenneskens, F. J. J. de Kanter, P. A. Kraakman, L. A. M. Turkenburg, W. E. Koolhaas, W. H. de Wolf, F. Bickelhaupt, Y. Tobe, K. Kakiuchi, Y. Odaira, et al., J. Am. Chem. Soc., 1985, 107, 3716–3717; (b) Y. Tobe, T. Kaneda, K. Kakiuchi, Y. Odaira, Chem. Lett., 1985, 1301–1304; (c) G. B. M. Kostermans, W. H. de Wolf, F. Bickelhaupt, Tetrahedron, 1987, 43, 2955–2966; (d) D. S. van Es, F. J. J. de Kanter, W. H. de Wolf, F. Bickelhaupt, Angew. Chem., 1995, 107, 2728–2730; Angew. Chem. Int. Ed., 1995, 34, 2553–2555. (a) T. Tsuji, S. Nishida, J. Chem. Soc., Chem. Commun., 1987, 1189–1190; (b) T. Tsuji, S. Nishida, J. Am. Chem. Soc., 1988, 110, 2157–2164. (a) T. Tsuji, S. Nishida, J. Am. Chem. Soc., 1989, 111, 368–369; (b) T. Tsuji, S. Nishida, M. Okuyama, E. Osawa, J. Am. Chem. Soc., 1995, 117, 9804–9813. (a) M. Okuyama, T. Tsuji, Angew. Chem., 1997, 109, 1157–1158; Angew. Chem. Int. Ed., 1997, 36, 1085–1087; (b) M. Okuya-

17

18

19

20 21

22

23

24

ma, M. Ohkita, T. Tsuji, J. Chem. Soc., Chem. Commun., 1997, 1277–1278; (c) T. Tsuji, M. Okuyama, M. Ohkita, H. Kawai, T. Suzuki, J. Am. Chem. Soc., 2003, 125, 951–961. (a) S. Grimme, J. Am. Chem. Soc., 1992, 114, 10542–10547; (b) I. Frank, S. Grimme, S. D. Peyerimhoff, J. Am. Chem. Soc., 1994, 116, 5949–5953; (c) B. Ma, H. M. Sulzbach, R. B. Remington, H. F. Schaefer III, J. Am. Chem. Soc., 1995, 117, 8392–8400. (a) Y. Tobe, K. Ueda, K. Kakiuchi, Y. Odaira, Y. Kai, N. Kasai, Tetrahedron, 1986, 42, 1851–1858; (b) Y. Tobe, M. Jimbo, K. Kobiro, K. Kakiuchi, J. Org. Chem., 1991, 56, 5241–5243; (c) Y. Tobe, T. Takahashi, K. Kobiro, K. Kakiuchi, J. Am. Chem. Soc., 1991, 113, 5804–5808; (d) Y. Tobe, A. Takemura, M. Jimbo, T. Takahashi, K. Kobiro, K. Kakiuchi, J. Am. Chem. Soc., 1992, 114, 3479–3491. G. B. M. Kostermans, M. Bobeldijk, W. H. de Wolf, F. Bickelhaupt, J. Am. Chem. Soc., 1987, 109, 2471–2475. S. Rosenfeld, J. Org. Chem., 1993, 58, 7572–7575. (a) V. Königstein, W. Tochtermann, Tetrahedron Lett., 1986, 27, 2961–2964; (b) V. Königstein, W. Tochtermann, E.-M. Peters, K. Peters, H. G. von Schnering, Tetrahedron Lett., 1987, 28, 3483– 3486. (a) Y. Tobe, M. Jimbo, S. Saiki, K. Kakiuchi, K. Naemura, J. Org. Chem., 1993, 58, 5883–5885; (b) Y. Tobe, M. Jimbo, H. Ishii, S. Saiki, K. Kakiuchi, K. Naemura, Tetrahedron Lett., 1993, 34, 4969–4970. By applying their newly proposed procedure for the evaluation of aromatic stabilization energy (ASE), Schleyer and Puhlhofer have shown that the ASE of 31 a is estimated to be 19.5 kcal mol–1, thus, the distortion in 31 a reduces the ASE of the benzene moiety by 12 kcal mol–1 P. von R. Schleyer, F. Puhlhofer, Org. Lett., 2002, 4, 2873–2876. The ASE of 38 a, evaluated by applying the same procedure, is 13–17 kcal mol–1: T. Tsuji, unpublished results. (a) G. J. Bodwell, J. N. Bridson, T. J. Houghton, J. W. J. Kennedy, M. R. Man-

103

104

3 Highly Strained Cyclophanes nion, Chem. Eur. J., 1999, 5, 1823–2827; (b) G. J. Bodwell, J. J. Fleming, M. R. Mannion, D. O. Miller, J. Org. Chem., 2000, 65, 5360–5370; (c) G. J. Bodwell, J. J. Fleming, D. O. Miller, Tetrahedron, 2001, 57, 3577–3585; (d) G. J. Bodwell, D. O. Miller, R. J. Vermeij, Org. Lett., 2001, 3, 2093–2096; (e) G. J. Bodwell, J. N. Bridson, M. K. Cyranski, J. W. J. Kennedy, T. M. Krygowski, M. R. Mannion, D. O. Miller, J. Org. Chem., 2003, 68, 2089–2098. 25 (a) I. Aprahamian, G. J. Bodwell, J. J. Fleming, G. P. Manning, M. R. Mannion, T. Sheradsky, R. J. Vermeij, M. Rabinovitz, J. Am. Chem. Soc., 2003, 125, 1720–1721; (b) I. Aprahamian, G. J. Bodwell, J. J. Fleming, G. P. Manning,

M. R. Mannion, T. Sheradsky, R. J. Vermeij, M. Rabinovitz, Angew. Chem., 2003, 115, 2651–2654; Angew. Chem. Int. Ed., 2003, 42, 2547–2550. 26 (a) T. Tsuji, M. Ohkita, S. Nishida, J. Am. Chem. Soc., 1993, 115, 5284–5285; (b) T. Tsuji, M. Ohkita, T. Konno, S. Nishida, J. Am. Chem. Soc., 1997, 119, 8425–8431. 27 (a) T. Tsuji, Z. Komiya, S. Nishida, Tetrahedron Lett., 1980, 21, 3583–3586; (b) T. Tsuji, S. Nishida, Tetrahedron Lett., 1983, 24, 3361–3364. 28 (a) H. Kawai, T. Suzuki, M. Ohkita, T. Tsuji, Angew. Chem., 1998, 110, 827–829; Angew. Chem. Int. Ed., 1998, 37, 817–819; (b) H. Kawai, T. Suzuki, M. Ohkita, T. Tsuji, Chem. Eur. J., 2000, 6, 4177–4187.

105

4

Superphanes Rolf Gleiter and Rolf Roers

4.1

Introduction

Chemists, especially those who are dealing with polycyclic hydrocarbons, prefer descriptive names for their highly symmetrical molecules [1] such as cubane and adamantane. The name cyclophane was coined by Cram and Steinberg [2] from cyclo, phenyl and alkane in 1951 for tricyclo[8.2.2.24,7]hexadeca-4,6,10,12(1),13,15hexane (1). With cyclophane chemistry becoming more popular a systematic set of rules for the cyclophanes was proposed [3]. The suffix “phane” thereby defines a compound containing at least one aromatic (or cyclic conjugated) moiety and at least one bridge. Since the synthesis of [2.2]paracyclophane various kinds of phanes have been prepared [4]. In connection with this review we mention the multibridged species [5] and the first superphane [26](1,2,3,4,5,6)cyclophane (2) which was prepared by Boekelheide et al. [6] and paved the way for superphane chemistry [5, 7]. The attribute “super” was suggested by Hopf [6 a] for 2. The structure of 2 can be described as two benzene rings clamped parallel on top of each other by six ethano bridges. The superphane terminology can easily be extended to various other cyclic conjugated p-systems [7]. In the following we will discuss the syntheses, structures and properties of various superphanes, ordered according to the ring size of the cyclic conjugated p-system.

Chart 4.1

Modern Cyclophane Chemistry. Edited by Rolf Gleiter, Henning Hopf Copyright © 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 3-527-30713-3

106

4 Superphanes

4.2

[n2]Cyclopropenonophanes 4.2.1

Synthesis

The first cyclopropenono superphanes were synthesized by a stepwise process [8]. The treatment of cyclodeca-1,6-diyne (3(3)), cyclododeca-1,7-diyne (3(4)) and cyclotetradeca-1,8-diyne (3(5)) with an equimolar amount of LiCCl3 in THF at –78 8C to –10 8C yielded the dichlorobicycloalkenynes (4(n)) [9]. Subsequent hydrolysis yielded the [32]-, [42]- and [52]cyclopropenonoacetylenophanes (5(n)) but also the ring-enlarged cyclic diynes 6(n) (Scheme 4.1) [9]. The efforts to prepare [n2]cyclopropenonophanes directly from the cycloalkadiynes 3(n) failed even with a large excess of LiCCl3. More successful was the cyclopropanation of the triple bond in 5(n) with one equivalent of sodium trichloroacetate followed by hydrolysis. This treatment afforded the desired [n2]cyclopropenonophanes 7(n) in low yields (3–5%) (Scheme 4.2). Higher yields were obtained by an alternative protocol which is outlined in Schemes 4.3 and 4.4 [8]. The key compounds for this reaction were the dichlorotricyclodienes 8(n) and 9(n). These species were converted into the corresponding dichloro[n2]cyclopropenyliophanes 10(n) (Scheme 4.3) by hydride abstraction. The cations were then converted without isolation into [n2]cyclopropenonophanes 7(n) by hydrolysis.

Scheme 4.1

4.2 [n2]Cyclopropenonophanes

Scheme 4.2

Scheme 4.3

The starting materials, the isomeric mixture of 8(n) and 9(n), were prepared in a four-step sequence (Scheme 4.4) by applying a modified form of methods developed in the chemistry of monofunctional dichlorocyclopropenes (Scheme 4.4) [9]. The cyclodiynes 3(n) (n = 3–5) were treated with Schwartz’s reagent and N-bromosuccinimide (NBS) to yield 11(n) and 12(n), respectively. Treating 11(n) or 12(n) with a large excess of sodium trichloroacetate provided the dibromotetrachlorotricycloalkanes 13(n) and 14(n) in about 50% yield. Dehalogenation of 13(n) and 14(n) with methyl lithium gave 8(n) and 9(n) in yields of 90%.

Scheme 4.4

107

108

4 Superphanes

Molecular structure of 7(3); top view (above), side view (bottom).

Fig. 4.1

The [n2]cyclopropenonophanes 7(n)(n = 3–5) proved to be fairly stable crystalline compounds. X-ray structure analysis of 7(3) revealed that there is an interplanar angle between the cyclopropenone moieties of 95.58. The ten-membered ring adopts a boat conformation (Fig. 4.1). Further cyclopropenonophanes were obtained from the tetrathiacyclodiynes 15(m,n = 3–6) (Scheme 4.5) [10]. The reaction of 15(m,n) with sodium trichloroacetate yielded the bicyclic and tricyclic tetrathiacyclopropenonophanes 16(m,n) and 17(m,n). The superphanes 17(3,5), 17(4,6), 17(5,5), 17(5,6) and 17(6,6) were isolated. For 17(3,5), 17(4,6) and 17(5,5) [11] structural details were obtained from X-ray studies on single crystals. These studies revealed a conjugation between the sulfur substituents and the cyclopropenone rings. The protocol for preparing tetrathiacyclopropenonophanes shown in Scheme 4.5 can be extended to the corresponding tetraselena congeners. This has been demonstrated by preparing 1,1',7,7' tetraselena[7,7']cyclopropenonophane [12] from 1,4,10,13-tetraselenacyclooctadeca-2,11diyne [13].

4.4 C4-Superphanes

a, b

+

Scheme 4.5

4.3

Superbridged Cyclopropenyliophanes

So far there is only one example, 20, known of a superbridged cyclopropenyliophane [14]. It was generated by the superbridged [52]cyclopropenonophane 18 (Scheme 4.6), a congener of 7(5) [9]. The geminal dimethyl groups were used to avoid side reactions when the O-alkylated cyclopropenone derivative 19 was reacted with 1,5-diaminopentane [14]. Xray investigations on 20 revealed that the two cyclopropenylium rings are nearly parallel to each other. In contrast, in the twofold bridged [52]cyclopropenyliophane 19 the interplanar angle between the two rings amounts to 418. The bond lengths within the cyclopropenylium rings vary between 1.34 Å and 1.38 Å [14].

Scheme 4.6

4.4

C4-Superphanes

In 1988, a new pathway for cyclobutadieno superphanes was established, which was based on cyclic precursors [15, 16]. The reaction of 1,6-cyclodecadiyne (3(3)) with dicarbonyl(g5-cyclopentadienyl)cobalt, (CpCo(CO)2) gave superphane 26 in

109

110

4 Superphanes

12% yield (Scheme 4.7). A second low molecular mass product isolated with 6% yield was the intramolecular complex 23. The product distribution can be rationalized as an interplay of steric and electronic factors. Key intermediates were cobaltacyclopentadienyl species such as 22 and the tricyclic cyclobutadiene complex 25, which could not be isolated under these reaction conditions [7, 17]. The X-ray structure of 26 is shown in Fig. 4.2. The central carbon backbone has a C4h symmetry and the distance between the cyclobutadiene moieties is 2.9 Å. Fig. 4.2 shows a projection from the top and indicates a pinwheel conformation of the four propane chains [15]. Obviously, the intermolecular pathway with the intermediates 24 and 25 is preferred to the intramolecular route (intermediate 21, cobaltacycle 22). We ascribe this partly to the relative low first ionization energy of 3 which was assigned to the ionization from the highest occupied molecular orbital, the “in-plane” p-orbital of 3(3). This favours a side attack of electrophilic metal fragments on the triple bonds. Furthermore, the formation of the highly strained 22 needs higher activation energy than that of its intermolecular analogue [16].

Scheme 4.7

4.4 C4-Superphanes

Fig. 4.2

Molecular structure of 26; side view (left), top view (right). The distances are given in Å.

The higher activation energy can be explained with the anti-Bredt structure of 22. This concept prompted the question, whether the ratio of intermolecular and intramolecular products is a function of the diyne ring size. The investigation of symmetric cycloalkadiynes 3(3–6) in reaction with CpCo complexes showed that diynes with an odd number of chain atoms yield superphanes whereas an even number of chain atoms lead to intramolecular complexes preferably [16, 18]. Strong steric effects of bulky a-alkylsilyl substituents of unsymmetric monoalkynes are well known to be the key of transition metal-induced [2+2]cycloadditions which lead to C2v symmetric main products [19]. We found that diynes 27 and 28 [20] can be reacted to form superphanes 29 and 30 in high yields (40–60%) (Scheme 4.8) [21]. Obviously, silyl bridges have strong directing effects in so far as they block intramolecular pathways and repress oligomerizations in favor of the dimerization as well [17]. In our search for higher oligomers we reasoned that small chains on both sides of the triple bonds or small chains on one side and a chain with bulky groups on the other might favor higher oligomers. Indeed, we found that 1,1,2,2-tetramethyl1,2-disilacycloocta-3,7-diyne can be reacted to the belt-like phanes 31 and 32 (Chart 4.2) [22]. The one-pot concept of cyclodiyne dimerization gave access to fairly simple syntheses of cyclobutadieno superphanes. However, a stepwise synthesis, i.e. the di-

111

112

4 Superphanes

a) CpCo(COD), 120 8C Scheme 4.8

Chart 4.2

rected synthesis of the tricyclic diyne 25, was targeted for three reasons: 1) the exclusion of competing oligomerizations; 2) the incorporation of different chain lengths in one superphane; 3) the combination of different metal fragments. Our successful stepwise pathway to superphane 26 started from a masked cyclodecadiyne, 1,6-cyclodecynol (33) which was dimerized with CpCo(COD) and oxidized to the tricyclic diketone 34 (Scheme 4.9) [23]. The Lalezari procedure leads to the tricyclic diyne 25 in good yields [24].

4.4 C4-Superphanes

a) CpCo(COD); b) acetone, Al(O-/Pr)3; c) H3N2CONH2HOAc; d) SeO2/CH3CO2H; e) 180 8C–2008C/Cu Scheme 4.9

Scheme 4.10 summarizes four reactions of 25. Refluxing of 25 with CpCo(CO)2 yielded 26 in high yields after only 2 h reaction time [23]. This explains why 25 could never be isolated during the one-pot reaction of 3(3) to 26. When 25 was refluxed with CpCo(CO)2 together with an excess of DMAD, the mixed superphane 35 could be isolated in poor yields. Nevertheless, this finding supports the idea of an intramolecular cobaltacyclopentadiene intermediate. This species might undergo not only a straight forward rearrangement to 26, but a “trimerization” is also possible by incorporating a third alkyne unit. The potential of a stepwise superphane synthesis was further demonstrated in the reaction of 25 with Fe(CO)5 in refluxing toluene. The mixed superphane 36 was isolated with 8% yield and combines different metals in one superphane. Finally, the reaction with (Me5Cp)Co(CO)2 yielded 37, the first superphane of this family, which displayed different substitution patterns on the CpCo units. 37 was subsequently used as starting point for a series of donor–acceptor-substituted superphanes [25]. Chart 4.3 lists the other superphanes 38–40 which were accessible by stepwise syntheses [26]. A systematic extension of the stepwise superphane synthesis led to belt-like macrocycles with four (46) and eight (47) CpCo-cyclobutadienyl-building blocks [27] respectively. The synthesis of these species starts with 41 (Scheme 4.11) which can be obtained by reacting 33 with CpCo(COD). Selective oxidation of one alcohol function of 41 yields 42. Using the Lalezari protocol (cf. Scheme 4.9) allows the introduction of one alkyne unit (43). The CpCo(COD)-supported dimerization of 43 and subsequent oxidation of the alcohol functions led to 44 which was converted into 45, the key species for the synthesis of the belt-like species 46 and 47.

113

114

4 Superphanes

Scheme 4.10

Chart 4.3

4.4 C4-Superphanes

Scheme 4.11

4.4.1

Properties of Cyclobutadieno Superphanes

Biscyclobutadieno superphanes turned out to be ideal model compounds for the study of interactions between two CpCo cyclobutadiene units with varying distances and varying electronic environments, respectively [17, 28]. Cyclic voltammetry of 26 and 23 a (Fig. 4.3) showed that the oxidation potential of 26 is significantly smaller than that of the intramolecular reference complex 23 a. This lower value corresponds to a higher HOMO of 26 as compared to 23 a. The second potential of 26, which is higher relative to that of 23 a, indicates that the positive charge generated during the first oxidation is partially delocalized. By increasing the chain lengths (48 and 49) the CpCo units get more independent. Through the

115

116

4 Superphanes

CV oxidation potentials of 26, 48, 49, and 23 a. The data are referred to the ferrocene/ferrocenium couple.

Fig. 4.3

investigation of a series of donor–acceptor-substituted superphanes a ground state p/p interaction was detectable [25] (see also chapter 14). 4.4.2

Oxidative Demetallations

When 26 is reacted with (NH4)2[Ce(NO3)6], the CpCo fragments are removed [29]. As main product (80%) the [34]-bridged syn-tricyclo[4.2.0.02.5]octa-3,7-diene (50) was isolated (Scheme 4.12). Irradiation of 50 yielded the propella [34]cubane (51). Scheme 4.12 summarizes the pathway from the cyclic alkyne to the bridged cubane. This sequence corresponds to a fourfold dimerization of two fixed alkyne units.

4.5 C5-Superphanes

3(3)

Scheme 4.13

4.5

C5-Superphanes

For the sake of completeness we mention only briefly the superphanes which are derived from ferrocene, because Chapter 5 in this book is devoted to bridged ferrocenes. In the early 1980s, Hisatome et al. achieved the synthesis of a pentabridged ferrocenophane [30]. In Scheme 4.13 we show an abbreviated way how the first C5-superphane was achieved [31]. Following the earlier work of Schlögl et al. [32] ferrocene was bridged via Friedel-Crafts acylation and alkylation in a stepwise manner. The synthesis of [24](2,3,4,5)thiophenophane (59), which was seen as the first “ultimate” heterophane, was achieved in a five-step synthesis starting from 3,4bis(chloromethyl)-2,5-dimethylthiophene (55) (Scheme 4.14) [33]. It turned out, that the superthiophenophane shows remarkable bathochromic shifts when the UV spectrum was compared with those of doubly bridged thiophenophanes and tetramethylthiophene. This shift was considered as evidence for either a transannular interaction [34] of the two thiophene rings in 59, an increased distortion of the planes of the two rings, or both [35]. Analogous to the stepwise synthesis of 26, we developed a sequence for superphanes with CpCo-stabilized cyclopentadienone units, in which a [2+2+1]cycloaddition of two alkynes and one CO group is accomplished (Scheme 4.15) [35, 36]. 5-cyclodecynone (60) was refluxed with CpCo(CO)2 to yield the isomeric triketones 61. The Lalezari procedure led to tricyclic diyne 62.

Scheme 4.12

117

118

4 Superphanes

Scheme 4.14

a) CpCo(CO)2, 190 8C; b)H3N2CONH2HOAC; c)SeO2/CH3CO2H; d) 180 8C–2008C/Cu Scheme 4.15

Scheme 4.16

4.5 C5-Superphanes

Scheme 4.17

Starting from 62, both the mixed superphane 63 as well as the biscyclopentadienono superphane 64 could be synthesized (Scheme 4.16). The chemoselectivity of tricyclic diyne 62 was controlled by the reaction conditions. At 190 8C, the mixed superphane 63 was the main low molecular weight product, whereas irradiation of 62 led to 64 in which the CO units are exclusively syn oriented [36]. Oxidation of CpCo-stabilized cyclopentadienono superphanes turned out to be irreversible under CV standard conditions [35]. Protonation [37 a, b, c], alkylation [37 d, e] and acylation [37 f ] of cyclopentadienone metal complexes are well-known reactions. Treatment of CpCo cyclopentadienone complexes with an excess of triethyloxonium tetrafluoroborate yields yellow colored O-ethylcobalticinium salts. In the case of 64 the same reaction afforded the mono- and bis-O-ethylcobalticinium salts [38]. Protonation of 64 yields the diprotonated superphane [38]. Recently, a stepwise synthesis of a mixed thiopheno-cyclobutadieno superphane was accomplished, based on the zirconocene dichloride mediated thiophene formation [39]. Starting from a silaprotected 1,6-cyclodecynol 65 the tricyclic thiophene derivative 66 was prepared by applying a protocol suggested by Negishi et al. [39 a] and Nugent et al. [39 b]. The silyl-protected diol 66 was transferred in two steps into the dione 67. The synthesis of the tricyclic diyne 68 was accomplished via the Lalezari protocol (Scheme 4.17). The intramolecular ring closure to the mixed superphane 69 was achieved by heating 68 with CpCo(COD).

119

120

4 Superphanes

4.6

Superbridged Benzene Rings

The first superphane was synthesized by Boekelheide et al. via the application of an ingenious and efficient protocol [6]. The synthesis of 2 was based on the o-xylylene dimerization reaction [40] which was applied three times [6]. Gas phase pyrolysis of 2,4,5-trimethylbenzyl chloride (70) at 710 8C and 10–2 torr gave 71 (Scheme 4.18). Heating of 71 at 300 8C yielded the dimer 72. For the next dimerization step 72 was transformed to the chloride 75 by conventional means. Repeating the dimerization step a second time yielded the tetrabridged cyclophane 76. After conversion to 79 the stage was set to generate 2 (Scheme 4.18). A second route to 2 was published by Hopf et al. [41]. This six-step route commences with a modified Rieche formylation of 80 to afford a mixture of the dialdehydes 81 and 82 (Scheme 4.19). The next step, the transannular bond formation, was accomplished by a Bamford-Stevens reaction. In the cases of 81 and 82 two new transannular C–C bonds were formed to yield the fourfold bridged isomers 83 and 84. Repeating the

710 Scheme 4.18

4.6 Superbridged Benzene Rings

Scheme 4.19

Rieche formylation and Bamford-Stevens procedure yields the fivefold bridged cyclophane 87 which was converted via 88 to 2. The bottleneck of this elegant procedure is the low yield of 81 and 82 (16%). The molecular geometry of superphane 2 determined by X-ray crystallographic analysis [6 b, 42] is shown schematically in Fig. 4.4. As anticipated, the molecule adopts D6h symmetry with the ethano bridges in an eclipsed conformation. The benzene rings are separated by 2.624 Å. The most surprising feature is the distortion of the bond angle of the sp2–sp3 carbon–carbon bonds out of planarity with the benzene ring by 20.38. Calculations by Lindner predict [43] a strain energy for 2 between 60 and 79 kcal mol–1. The photoelectron spectroscopic investigation of 2 reveals an ionization energy of 7.55 eV [44]. This energy seems rather low, however due to symmetry reasons

121

122

4 Superphanes

Fig. 4.4

Perspective view of 2. The distances are given

in Å.

the HOMO of [2n]cyclophanes is effected only weakly by the ethano bridges and thus its electron donor capability is less than expected [45]. This behavior is also found in the UV/Vis absorption spectrum of the charge transfer complex of 2 with TCNQ [6 b]. The value recorded for the maximum of the long wavelength band (kmax = 572 nm) is considerably shorter than expected if the shortening of the distance between the decks and the alkyl effect of the bridges were independent and additive [6 b]. With respect to its high strain energy, the superphane 2 is relatively stable. Heating of 2 in the presence of trapping agents for benzylic radicals (diisopropyl benzene) at 350 8C for 24 h or dimethyl maleate at 200 8C (for 24 h) led only to the complete recovery of unchanged 2. Birch reduction of superphane 2 gave the expected dihydro product in only poor yield [6 b] whereas [2,2]paracyclophane and [24](1,2,4,5)cyclophane were reduced easily. A better way to achieve a dihydro derivative was found by treating 2 with the strong electrophile dimethoxycarbonium tetrafluoroborate [6 b]. This treatment gave a deep red solution whose 1H NMR spectrum is in accord with structure 89 (Scheme 4.20). The reduction of the red solution with NaBH4 led to the dihydro derivative 90. Reaction of 2 with an excess of NBS and a small amount of benzoyl peroxide yielded a monobromide, accompanied by traces of dibromides. Treatment of the resulting monobromide with DBN yielded [26](1,2,3,4,5,6)cyclophan-1-ene (91) (Scheme 4.20) [6 b]. The reaction of 2 with ethyl diazoacetate in presence of CuSO4 led to an enlargement of one ring in 2 (Scheme 4.21) [6 b] to yield a mixed phane with one cycloheptatriene ring 92. This product was transferred to the corresponding tropylium phane 94 by reduction and treatment with BF3. It is interesting to note that 94, when exposed to water, was transferred back to superphane 2 [6 b]. When a solution of 2 in dichloromethane was treated with an excess of dicyanoacetylene in the presence of aluminum trichloride the adduct 96 (X-ray analysis) was isolated after aqueous work-up (Scheme 4.22). A logical interpretation of structure 96 is that to each deck of 2 one equivalent of dicyanoacetylene was added to yield the intermediate 95 which leads via HCl addition and intramolecu-

4.6 Superbridged Benzene Rings

Scheme 4.20

; Scheme 4.21

lar Diels-Alder addition to 96. The unusual structure of 2 is matched by its unusual chemistry. A strong motivation for the synthesis of [36](1,2,3,4,5,6)cyclophane 97 was the prediction that it can be isomerized by photochemical means to the sixfold bridged hexaprismane 98 [46]. Shinmyozu et al. [47, 48] described two closely related routes to 97 in which the propano bridges between the benzene rings were introduced step by step via a transannular aldol condensation reaction between an acetyl group and a formyl group. The difference between the routes results from the starting material. The first route is summarized in Scheme 4.24. It commences with [33](1,3,5)cyclophane (99) [49] which was obtained by reaction of 1,3,5-tris(bromomethyl)benzene and (p-tolylsulfonyl)methyl isocyanide (TosMIC) [50]. Acylation of 99, followed by formylation produces 100. Intramolecular aldol type condensation between an acetyl group and a pseudogeminally sub-

123

124

4 Superphanes

Scheme 4.22

Scheme 4.23

Scheme 4.24

4.6 Superbridged Benzene Rings

Scheme 4.25

stituted formyl group and subsequent hydrogenation leads to the ketone 101 which was reduced to the fourfold bridged cyclophane 102 [47]. Twofold repetition of the five steps just described leads to 97 via the fivefold cyclophane 103 [51]. In a preceding variant of this synthesis [47] the key ring-closing step was carried out between an acetyl group and a pseudogeminally substituted CH2Cl group (100 in which CHO was replaced by CH2Cl). This was continued until 97 was obtained. A second, more efficient way started from [32](1,3)cyclophane (104) [52] which was converted into the tetrakis(bromomethyl)[32]cyclophane 105. Twofold ring closure between 105 and TosMIC (Scheme 4.25) in the presence of NaH in DMF yielded 106. The fourfold bridged cyclophane 107 was obtained from 106 by reduction with Li/NH3 in the presence of ethanol. By a sequence of acetylation/formylation, 107 was transferred to 103 and finally to 97 [48]. The multibridged cyclophanes with propano bridges such as 97, 99, 102, 103 and 107 are flexible in solution. The energy barrier (DG=) for the flipping of the propano bridges of one pinwheel conformation into the other (see Scheme 4.26) was estimated by NMR spectroscopy to be 10.9 kcal mol–1 (Tc = –40 8C) [48]. Ab initio calculations suggest a stepwise mechanism for this process. X-ray investigations were carried out on 97 and various complexes such as the complex 97 · TCNQ · C6H6 (1 : 1 : 1) at –190 8C, and 97 · TCNQF4 (1 : 1) [53]. In the first two crystals the cyclophane moiety is observed in a D6h conformation due to severe disorder of the bridges. No structural differences were observed in the cyclophane moieties of 97 and its TCNQ complex. Contrastly, in the complex of 97 · TCNQF4 the

Scheme 4.26

125

126

4 Superphanes Fig. 4.5 Perspective view of 97. The distances are given in Å.

cyclophane moiety shows approximately C6h symmetry in the solid state (Fig. 4.5). The distance between the two benzene rings in both 97 · TCNQ · C6H6 and 97 · TCNQF4 was found to be 2.94 Å. This value is larger than that reported for 2 by 0.32 Å, indicating that much less strain energy is present in 97. In line with the larger distance of the p-systems is the smaller angle (4.58) found for the bending of the C(sp2)–C(sp3) bond out of the plane of the benzene ring of 97 compared with 2 (208). The interplanar angles within the propano bridges vary between 1408 and 1428 (Fig. 4.5). A comparison between the longest wavelength band in the UV/Vis spectrum of the TCNE complexes of 2 (kmax = 572 nm) [6 b] and 97 (kmax = 728 nm) shows a considerable difference. This is in agreement with the observation that propano bridges influence the antisymmetric HOMO whereas ethano bridges influence a symmetric lower lying occupied MO [54].

4.7

Concluding Remarks

Since the first synthesis of 2, the superphane concept has been extended to cyclic p-systems of ring size three to five. Larger ring systems, e.g. tropylium have not been reported. A further extension, e.g. the replacement of all ethano bridges in 2 by double bonds (108) or even benzene rings to “hyperphanes” (109) is also missing. However, the latter case is not far away in view of the recent synthesis of a tetrabenzoannelated 1,2,4,5-cyclophane [55]. Although the superphanes provide excellent models for studying the interactions of rigid and sterically fixed p-systems by means of spectroscopy and electrochemistry, not many studies are known to date. This is mainly because the multistep syntheses required to prepare the various superphanes deter chemists from further investigations. Therefore more simple pathways are desired. By looking at the known superphanes one notices that most examples contain two p-systems of the same kind. However, donor–acceptor superphanes should also be of interest with respect to their properties.

4.9 References

Scheme 4.27

4.8

Acknowledgement

We are grateful to the Deutsche Forschungsgemeinschaft, the Fonds der Chemischen Industrie and the BASF Aktiengesellschaft, Ludwigshafen, for financial support.

4.9

References 1 2 3

4

5

6

7

A. Nickon, E. F. Silversmith, The Name Game, Pergamon Press, New York 1988. D. J. Cram, H. Steinberg, J. Am. Chem. Soc. 1951, 73, 5691–5704. a) W. H. Powell, Pure Appl. Chem. 1998, 70, 1513–1545; b) F. Vögtle, P. Neumann, Tetrahedron 1970, 26, 5847–5873. a) F. Vögtle, Cyclophane Chemistry: Synthesis, Structures and Reactions, Wiley, Chichester 1993; b) P. M. Keehn, S. M. Rosenfeld, Cyclophanes, Academic Press, New York 1983; c) F. Diederich, Cyclophanes, The Royal Society of Chemistry, London 1991; d) D. J. Cram, J. M. Cram, Container Molecules and their Guests, The Royal Society of Chemistry, London 1994. a) V. Boekelheide, Acc. Chem. Res. 1980, 13, 65–70; b) V. Boekelheide, Topic Curr. Chem. 1983, 113, 87–143; c) J. Kleinschroth, H. Hopf, Angew. Chem. 1982, 94, 485–496; Angew. Chem. Int. Ed. Engl. 1982, 21, 469–480. a) Y. Sekine, M. Brown, V. Boekelheide, J. Am. Chem. Soc. 1979, 101, 3126–3127; b) Y. Sekine, V. Boekelheide, J. Am. Chem. Soc. 1981, 103, 1777–1785. R. Gleiter, D. Kratz, Acc. Chem. Res. 1993, 26, 311–318.

8

9

10

11

12 13 14 15 16

17

18

R. Gleiter, M. Merger, A. Altreuther, H. Irngartinger, J. Org. Chem. 1996, 61, 1946–1953. a) R. Gleiter, M. Merger, Synthesis 1995, 969–972; b) R. Gleiter, M. Merger, A. Altreuther, H. Irngartinger, J. Org. Chem. 1995, 60, 4692–4696. C. Benisch, S. Bethke, R. Gleiter, T. Oeser, H. Pritzkow, F. Rominger, Eur. J. Org. Chem. 2000, 2479–2488. D. B. Werz, F. Rominger, I. Hyla-Kryspin, R. Gleiter, J. Org. Chem. 2001, 66, 3416–3422. D. B. Werz, R. Gleiter, F. Rominger, Eur. J. Org. Chem. 2003, 151–154. D. B. Werz, R. Gleiter, F. Rominger, J. Org. Chem. 2002, 67, 4290–4297. A. Schuster, F. Rominger, R. Gleiter, Tetrahedron Lett. 1999, 40, 7769–7772. R. Gleiter, M. Karcher, M. L. Ziegler, B. Nuber, Tetrahedron Lett. 1987, 28, 195–198. R. Gleiter, Angew. Chem. 1992, 104, 29– 46; Angew. Chem. Int. Ed. Engl. 1992, 31, 27–44. R. Gleiter, M. Merger, Angew. Chem. 1997, 109, 2533–2546; Angew. Chem. Int. Ed. Engl. 1997, 36, 2426–2439. R. Gleiter, G. Pflästerer, B. Nuber, J. Chem. Soc. Chem. Commun. 1993, 454– 456.

127

128

4 Superphanes 19

20

21 22

23

24

25

26

27

28

29

30

31

32 33

Y. Wakatsuki, O. Nomura, K. Kitaura, K. Morokuma, H. Yamazaki, J. Am. Chem. Soc. 1983, 105, 1907–1912. R. Gleiter, H. Stahr, F. Stadtmüller, H. Irngartinger, H. Pritzkow, Tetrahedron Lett. 1995, 36, 4603–4606. R. Gleiter, H. Stahr, B. Nuber, Tetrahedron Lett. 1995, 36, 4607–4610. G. Haberhauer, F. Rominger, R. Gleiter, Angew. Chem. 1998, 110, 3632–3634; Angew. Chem. Int. Ed. 1998, 37, 3376– 3377. R. Gleiter, H. Langer, V. Schehlmann, B. Nuber, Organometallics 1995, 14, 975– 986. a) I. Lalezari, A. Shafiee, M. Yalpani, Tetrahedron Lett. 1969, 10, 5105–5106. b) I. Lalezari, A Shafiee, M. Yalpani, Angew. Chem. 1970, 82, 484–485; Angew. Chem. Int. Ed. Engl. 1970, 9, 464–465. R. Gleiter, R. Roers, J. Classen, A. Jacobi, G. Huttner, T. Oeser, Organometallics 2000, 19, 147–151. a) For the four metal center superphane see R. Roers, J. Hofmann, J. Classen, R. Gleiter, Tetrahedron Lett. 2001, 42, 5871–5873. b) For the mixed cyclobutadieno ethyleno superphane see R. Gleiter, H. Langer, B. Nuber, Angew. Chem. 1994, 106, 1350–1352; Angew. Chem. Int. Ed. Engl. 1994, 33, 1272–1274. R. J. Schaller, R. Gleiter, J. Hofmann, F. Rominger, Angew. Chem. 2002, 114, 1231–1234; Angew. Chem. Int. Ed. 2002, 41, 1181–1183. M. E. Stoll, S. R. Lovelace, W. E. Geiger, H. Schimanke, I. Hyla-Kryspin, R. Gleiter, J. Am. Chem. Soc. 1999, 121, 9343–9351. R. Gleiter, M. Karcher, Angew. Chem. 1988, 100, 851–852; Angew. Chem. Int. Ed. Engl. 1988, 27, 840–841. M. Hisatome, J. Watanabe, K. Yamakawa, Y. Iitaka, J. Am. Chem. Soc. 1986, 108, 1333–1334. M. Hisatome, Y. Kawajiri, K. Yamakawa, Y. Harada, Y. Iitaka, Tetrahedron Lett. 1982, 23, 1713–1716. K. Schlögl, H. Seiler, Tetrahedron Lett. 1960, 7, 4–8. M. Takeshita, M. Koike, H. Tsuzuki, M. Tashiro, J. Org. Chem. 1992, 57, 4654–4658.

34

35 36

37

38

39

40

41 42 43 44

45 46 47

W. W. Paudler, M. D. Bezoari in Cyclophanes, P. M. Keehn, S. M. Rosenfeld; eds., Academic Press, New York 1983, Ch. 6. R. Roers, F. Rominger, R. Gleiter, Tetrahedron Lett. 1999, 40, 3141–3144. R. Roers, F. Rominger, B. Nuber, R. Gleiter, Organometallics 2000, 19, 1578– 1588. a) J. E. Sheats, M. D. Rausch, J. Org. Chem. 1970, 35, 3245–3249. b) R. E. Benson, R. V. Lindsey Jr., J. Am. Chem. Soc. 1957, 79, 5471–5474. c) R. Markby, H. W. Sternberg, I. Wender, Chem. Ind. (London) 1959, 1381–1382. d) P. Jutzi, U. Siemeling, A. Müller, H. Bögge, Organometallics 1989, 8, 1744–1750. e) R. Boese, D. Bläser, R. L. Haltermann, K. P. C. Vollhardt, Angew. Chem. 1988, 100, 592–594; Angew. Chem. Int. Ed. Engl. 1988, 27, 553–555. f) M. v. der Gruen, C. Schaefer, R. Gleiter, Organometallics 2003, 22, 2370–2372. R. Gleiter, R. Roers, F. Rominger, B. Nuber, I. Hyla-Kryspin, J. Organomet. Chem. 2000, 610, 80–87. a) E. Negishi, S. J. Holmes, J. M. Tour, J. A. Miller, F. E. Cederbaum, D. R. Swanson, T. Takahashi, J. Am. Chem. Soc. 1989, 111, 3336–3346; b) P. J. Fagan, W. A. Nugent, J. Am. Chem. Soc. 1988, 110, 2310–2312; c) S. Gath, R. J. Schaller, R. Gleiter, F. Rominger, Organometallics 2004, 116–1121. a) V. Boekelheide, G. Ewing, Tetrahedron Lett. 1978, 4245–4248; b) P. F. T. Schirch, V. Boekelheide, J. Am. Chem. Soc. 1979, 101, 3125–3126. S. El-Tamany, H. Hopf, Chem. Ber. 1983, 116, 1682–1685. A. W. Hanson, T. S. Cameron, J. Chem. Res. Synop. 1980, 336–337. H. J. Lindner, Tetrahedron 1976, 32, 753– 757. B. Kovac, M. Mohraz, E. Heilbronner, V. Boekelheide, H. Hopf, J. Am. Chem. Soc. 1980, 102, 4314–4324. E. Heilbronner, Z. Yang, Top. Curr. Chem. 1983, 115, 1–55. O. J. Cha, E. Osawa, S. Park, J. Mol. Struct. 1993, 300, 73–81. Y. Sakamoto, N. Miyoshi, M. Hirakida, S. Kusumoto, H. Kawase, J. M. Rudzins-

4.9 References

48

49

50 51

ki, T. Shinmyozu, J. Am. Chem. Soc. 1996, 118, 12267–12275. M. Yasutake, T. Koga, Y. Sakamoto, S. Komatsu, M. Zhou, K. Sako, H. Tatemitsu, S. Onaka, Y. Aso, S. Inoue, T. Shinmyozu, J. Am. Chem. Soc. 2002, 124, 10136–10145. K. Kurosawa, M. Suenaga, T. Inazu, T. Yoshino, Tetrahedron Lett. 1982, 23, 5335–5338. O. Possel, A. M. van Leusen, Tetrahedron Lett. 1977, 18, 4229–4232. T. Shinmyozu, M. Hirakida, S. Kusumoto, M. Tomonou, T. Inazu, J. M. Rudzinski, Chem. Lett. 1994, 669–672.

52

W. Sentou, T. Satou, M. Yasutake, C. Lim, Y. Sakamoto, T. Itoh, T. Shinmyozu, Eur. J. Org. Chem. 1999, 1223–1231. 53 M. Yasutake, Y. Sakamoto, S. Onaka, K. Sako, H. Tatemitsu, T. Shinmyozu, Tetrahedron Lett. 2000, 41, 7933–7938. 54 R. Gleiter, W. Schäfer, Acc. Chem. Res. 1990, 23, 369–375. 55 M. Brettreich, M. Bendikov, S. Chaffins, D. F. Perepichka, O. Dautel, H. Duong, R. Helgeson, F. Wudl, Angew. Chem. 2002, 114, 3840–3843; Angew. Chem. Int. Ed. 2002, 41, 3688–3691.

129

131

5

Carbon-Bridged Ferrocenophanes Joon-Seo Park and T. Randall Lee

5.1

Introduction

In the early 1950s, ferrocene was discovered independently by two research teams [1, 2]. Although Kealy and Pauson proposed a r-bonded structure in their report to Nature [1], Woodward soon afterward suggested the ‘sandwich structure’ for this new iron compound [3], which was quickly confirmed by X-ray diffraction studies [4, 5]. Despite intense study for more than 50 years, ferrocene is still attracting the interest of scientists working in a wide variety of disciplines. Ferrocene is the first known ‘sandwich complex’, in which two cyclopentadienyl (Cp) ligands lie in parallel planes with a separation of * 3.32 Å [6–8]. The distance between the two Cp rings is comparable to the van der Waals distance between two aromatic rings, which is * 3.40 Å. The Cp ligands in ferrocene are covalently bonded to the iron center, but they are almost free to rotate with respect to each other; the activation barrier to the rotation is about 4–8 kJ mol–1 [9– 12]. The staggered conformation of the rings with approximately 368 of ring twist angle is slightly preferred over the eclipsed conformation [13]. Ferrocene is a diamagnetic, 18-electron, d6 complex, and the aromatic character of ferrocene was demonstrated in 1952 [6]. Ferrocene acts as a weak base, and thus the Cp rings in ferrocene can undergo various electrophilic substitution reactions. Although the mechanism is still controversial, the reactions might start from the attack of the electrophile on the lone pairs of the metal center to give the bent intermediate followed by migration of the electrophile to the Cp ligand and then deprotonation. Ferrocene undergoes various aromatic reactions including Friedel-Crafts acylation and alkylation, Vilsmeier formylation, and mercuration [14, 15]. Ferrocene is 106 times more reactive than benzene in Friedel-Crafts acylation. Several useful reviews provide comprehensive summaries of the chemistry of ferrocene [16, 17]. Ferrocene and its derivatives have certain desirable characteristics that render them useful in a variety of applications. Unlike most organometallic compounds, ferrocene is soluble in most organic solvents and stable in air and at elevated temperatures [18]. Ferrocene, therefore, can be easily handled without decomposition. Ferrocene is reversibly oxidized by mild one-electron oxidizing agents to the more reactive deep green ferrocenium radical cation, which is an excellent radical trap Modern Cyclophane Chemistry. Edited by Rolf Gleiter, Henning Hopf Copyright © 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 3-527-30713-3

132

5 Carbon-Bridged Ferrocenophanes

[15]. This property renders ferrocene attractive for use in chemical sensing and other technologies [19]. The p-electrons of the Cp rings of ferrocene are delocalized through the d-electrons of iron, which makes ferrocene and its derivatives useful for constructing conducting and optoelectronic materials [20–22]. Ferrocenophanes are ferrocene derivatives in which the two Cp rings are connected by an atomic or molecular bridge. Ferrocenophanes also possess most of the desirable properties mentioned above for ferrocene. In ferrocenophane, however, the rotation of Cp rings can be restricted, and the dihedral angle of two Cp rings can be varied depending on the length, position, and structure of the bridges. The structural variation can affect the properties and thus the reactivities of ferrocenophanes. For example, the tilt of Cp rings can facilitate electrophilic attack on the metal center. Investigation of ferrocenophanes, therefore, includes synthesis and structural characterization of ferrocenophane derivatives and the study of their properties as a function of structure. Several useful reviews for this area of research have been published [23–25]. Recent development of various kinds of ferrocenophanes suggest that an updated review is timely. To this end, the focus of the present review is the synthesis and structure of ferrocenophanes having all carbon bridges. Descriptions of other ferrocenophanes and related metallocenes can be found in chapters 4, 6, and 16.

5.2

Nomenclature

In ferrocenophanes, the Cp rings of ferrocene are connected by either an atomic or a molecular bridge. The carbon-bridged ferrocenophanes can be categorized into two major groups depending on the number of ferrocene units in the compound. The first class consists of mononuclear ferrocenophanes in which the Cp rings of one ferrocene moiety are connected by one or more bridges. The mononuclear ferrocenophanes can be further divided into two subgroups depending on the number of bridges: singly-bridged ([m]) and multiply-bridged ferrocenophanes ([m]n). The second class can be described as multinuclear ferrocenophanes ([mn]) in which two or more ferrocene units are linked by one or more bridges (Fig. 5.1). This review adopts the most widely accepted nomenclature, which was proposed by Smith [26] and Vögtle and Neumann [27] for bridged organic aromatic cyclophanes and metallocenophanes. The mononuclear ferrocenophanes are denoted as [m][n][o][p][q]ferrocenophanes, where the number in square brackets indi-

Fig. 5.1

Structures of ferrocenophanes

5.3 Mononuclear Carbon-Bridged Ferrocenophanes

cates the length of the bridge. For multiply-bridged homonuclear ferrocenophanes, the relative location of the bridge is provided by the number in round brackets. The numbering of atoms begins at the end of the longest bridge. If multiple bridges have the same length, the substituted bridge is numbered first. According to this system of nomenclature, compound 1 is denoted as 1-hydroxy-2methyl[5]ferrocenophane, and the name of the multiply-bridged ferrocenophane 2 is 2,2-dimethyl[3](1,1')[3](3,3')ferrocenophane. Multinuclear ferrocenophanes are denoted as [m.n.o.p.q]ferrocenophanes. The ferrocenophane 3 is named as 1methyl[1.1](1,1'';1',1''')ferrocenophane, where the placement of the bridges is also provided by the number in round brackets. Because ferrocenophane structures are diverse, a different nomenclature can sometimes be more convenient to denote the structure.

5.3

Mononuclear Carbon-Bridged Ferrocenophanes 5.3.1

[1]Ferrocenophanes

[1]Ferrocenophane, with a single carbon atom in the bridge, is too strained to exist due to the extreme tilt of the Cp rings. A number of heteroatom-bridged [1]ferrocenophanes, however, have been reported; apparently, the larger atomic radii of the heteroatoms afford less tilted and thus more stable structures than the mono-carbon-bridged [1]ferrocenophane. The relative strain in heteroatom-bridged [1]ferrocenophanes is related to the ring tilt angle, which varies from 18 to 278 in known derivatives. The strain in these ferrocenophanes leads to thermal ringopening polymerization (ROP) to give ferrocene-containing polymers (Scheme 5.1). Since the first report by Manners and co-workers of the successful synthesis of poly(ferrocenylsilanes) by a thermal ROP [28], various heteroatom-bridged [1]ferrocenophanes, including those with Group 13 through Group 16 elements in their bridges, have been found to undergo ROP reactions [29–40]. Because heteroatom-bridged ferrocenophanes are beyond the scope of this chapter, we refrain from more detailed discussions and direct the reviewer to recent reviews on the topic [41, 42] as well as chapter 16.

133

134

5 Carbon-Bridged Ferrocenophanes

Scheme 5.1

5.3.2

[2]Ferrocenophanes

Analyses of ferrocenophanes bridged by two carbon atoms have shown that these compounds are highly strained because the length of the bridge is relatively short [43–45]. As mentioned in the previous section, the tilt angle of the Cp rings can be used to estimate the ring strain in ferrocenophanes. Regardless of their high strain, a number of interesting [2]ferrocenophanes have been reported. The reaction of 6,6-dimethylfulvene 6 with sodium followed by treatment with ferrous chloride gave the first carbon-bridged [2]ferrocenophane, 1,1,2,2-tetramethyl[2]ferrocenophane 8 (Scheme 5.2 and see chapter 6, section 6.4.2 for the analogous ruthenocenophane) [46]. Normally, the Cp rings of a [2]ferrocenophane are more tilted than those of ferrocene itself, and thus the [2]ferrocenophane is protonated more easily than ferrocene. The single crystal X-ray data for 8 showed a ring twist angle of 268 and a ring tilt angle of 238 [45]. The distance between carbon 1 of a Cp ring and carbon 1' of another Cp ring was found to be 2.70 Å, whereas the distance between the centers of two Cp rings was 3.32 Å. The nonsubstituted hydrocarbon-bridged analog exhibited a slightly smaller ring tilt angle of 21.68 [47]. The [2]ferrocenophanes bridged by two carbon atoms possess ring tilt angles comparable with those of highly strained silicon-bridged [1]ferrocenophanes (i.e. 20.88). Several interesting [2]ferrocenophanes having unsaturated bridges have been reported. In 1976, Yasufuku and coworkers described the reaction of transition metal ethynyl complexes with p-cyclopentadienyl(triphenylphosphine)dialkylacetylenecobalt [48]. The reaction of 1,1'-bis(phenylethynyl)ferrocene 9 with the diphenylacetylene cobalt complex afforded 1,2-(1,1'-ferrocenylene)-3,4,5,6-tetraphenylbenzene 10, which was the first ferrocenophane bridged with an aromatic ring [49]. When 9 was treated with p-cyclopentadienylbis(triphenylphosphine) cobalt, compound 11 was obtained in 10% yield [49]. The single crystal X-ray structure of 10

Scheme 5.2

5.3 Mononuclear Carbon-Bridged Ferrocenophanes

Scheme 5.3

exhibited a Cp ring tilt angle of 23.78 [50]. The dihedral angles between the central benzene ring and the Cp rings were found to be 90.68 and 94.98, and the two Cp rings were in a nearly eclipsed conformation. In compound 10, the length of bridging carbon–carbon bond (Csp2–Csp2) was found to be 1.388 Å, which is similar to the carbon–carbon length in benzene itself (1.397 Å). The trinuclear ruthenacycle [2]ferrocenophane 12 was prepared by the reaction of 1,1'-bis(phenylethynyl)ferrocene 9 with Ru3(CO)12 [51]. In the reaction, the dinuclear ruthenacycle complex 13 was also obtained as a minor product. The single crystal X-ray structure of 12 showed a ring tilt angle of 21.8; which is comparable to those of normal [2]ferrocenophanes even though two ruthenium atoms coordinate to the diene part of the complex.

The phosphorous-containing ferrocenophane, 1,1'-diphospha[2]ferrocenophane 14, was prepared by a four-step synthesis starting from 1-phenyl-3,4-dimethylphosphole-2-carboxaldehyde [52]. The aldehyde group of the 1-phenyl-3,4-dimethylphosphole-2-carboxaldehyde was converted to a bromomethyl group and then intermolecularly coupled via a Grignard reaction. After reduction and cleavage of the two Pphenyl bonds by lithium, treatment with ferrous chloride afforded the targeted compound 14. Because stabilized sp2 phosphorous derivatives are known as p-acceptor ligands rather than classical r-donor phosphines, the resulting compound promises an access to polymers containing unique sp2 phosphorous centers. The phosphorous-containing derivative 14 exhibited a Cp ring tilt angle of 208, which is similar to that of its all-carbon analog [52].

135

136

5 Carbon-Bridged Ferrocenophanes

Scheme 5.4

The structural data highlighted above indicate that [2]ferrocenophanes bridged by two saturated carbon atoms are highly strained. Like heteroatom-bridged ferrocenophanes, these highly strained ferrocenophanes can be used as monomers for ROP. As mentioned in the previous section, [1]ferrocenophanes bridged by single heteroatoms form organometallic polymers with high molecular weights due to their highly strained structures. The polymerization of a [2]ferrocenophane with two silicon atoms in the bridge, however, failed due to its minimal ring strain derived from a small tilt angle of 4.198 [53]. In contrast, the more strained carbonbridged [2]ferrocenophane with a tilt angle of 21.68 successfully polymerized to give poly(ferrocenylethylenes) [54]. Highly strained unsaturated [2]ferrocenophanes can also be utilized for polymerization. The intramolecular McMurry coupling of 1,1'-ferrocenedicarbaldehyde 15 afforded the vinylene-bridged ansa-ferrocene complex 16 (Scheme 5.4) [55]. The Xray crystallographic data of the complex, ansa-(vinylene)[2]ferrocenophane 16, showed a bent structure with a ring tilt angle of 238, which is comparable to those of saturated analogs. The ring-opening metathesis polymerization (ROMP) of this highly strained ferrocenophane with a molybdenum-based ROMP initiator [56] afforded insoluble poly(ferrocenylene vinylene) 17 (Scheme 5.4). The fully conjugated organometallic polymer is expected to facilitate electronic interactions between the metal centers due to its inherent electron delocalization. A pressed pellet of this polymer, however, exhibited weak conductivity after doping with I2. To overcome the poor solubility, block copolymers of ansa-(vinylene)[2]ferrocenophane with norbornene were prepared [55]. 5.3.3

[3]Ferrocenophanes

[3]Ferrocenophanes are the most readily accessible of all carbon-bridged ferrocenophanes because bridges that consist of three carbon atoms introduce relatively little structural strain in the system. The first [3]ferrocenophane was synthesized by Lüttringhaus and Kullick in the late 1950s. The reaction of sodium cyclopentadienide 18 with 1,3-dibromopropane gave the bis-cyclopentadienyl hydrocarbon 19, which was doubly deprotonated with sodium. Treatment of the disodium salt with ferrous chloride gave polymeric ferrocenes from which [3]ferrocenophane 20 was isolated in low yield (Scheme 5.5) [57]. A number of [3]ferrocenophanes have been prepared by intramolecular acylation reactions. The compound [3]ferrocenophane-1-one 22 was prepared by hetero-

5.3 Mononuclear Carbon-Bridged Ferrocenophanes

Scheme 5.5

annular Friedel-Crafts cyclization of b-ferrocenylpropionic acid 21 [58, 59]. Three reagents, trifluoroacetic anhydride, polyphosphoric acid, and aluminum chloride, were investigated for the self-condensation of b-ferrocenylpropionic acid 21, where trifluoroacetic anhydride afforded the desired product in the highest yield (87%). The starting propionic acid 21 was prepared by various methods, such as Willgerodt reaction of ferrocenyl ketones, carbethoxylation of acetylferrocene, and Friedel-Crafts acylation of ferrocene [60]. In the Friedel–Crafts acylation, the expected heteroannular cyclization might compete with the homoannular cyclization of acids. A systematic study of the effect of aliphatic chain length of x-ferrocenylaliphatic acids on cyclization showed that b-ferrocenylpropionic acid gave the heteroannularly-bridged [3]ferrocenophane-1-one 22, while x-ferrocenylaliphatic acids with longer chains afforded the homoannularly cyclized products or polymers [58]. The single crystal X-ray structure of [3]ferrocenophane-1-one 22 showed a tilt angle for the Cp rings of 8.88, which is smaller than those observed for [2]ferrocenophanes (> 20 8) [61]. The distance between carbon 1 of Cp and carbon 1' of Cp’ was found to be 3.10 Å, which is larger than those of [2]ferrocenophanes. These observations provide strong evidence that [3]ferrocenophanes possess less structural strain than [2]ferrocenophanes. Starting from 22, various derivatives of [3]ferrocenophane can be produced. Reduction of 22 by lithium aluminum hydride (LAH) afforded the corresponding alcohol 23. On the other hand, reduction with LAH-aluminum chloride gave the unsubstituted [3]ferrocenophane 20 in a nearly quantitative yield [58]. Furthermore, the modified McMurry coupling of [3]ferrocenophane-1-one 22 gave E,Zbis([3]ferrocenophane-1-ylidenes) 24 and 25 in 70% overall yield, in which the two [3]ferrocenophanes are separated by a rigid carbon–carbon double bond (Scheme 5.6) [62].

The intramolecular Friedel-Crafts cyclization can be extended to the asymmetric synthesis of ferrocenophanes [63]. A series of ferrocenyl ketones 26 underwent conversion to the corresponding alcohols 27 in high enantiomeric excesses via oxazaborolidine-catalyzed reduction. Subsequent methanolysis of the (R)-a-ferrocenyl

137

138

5 Carbon-Bridged Ferrocenophanes

Scheme 5.6

Scheme 5.7

alcohols 27 gave the corresponding methyl ethers 28, which were in turn converted to ethyl 3-ferrocenopropanoates 29 with retention of configuration on treatment with 1-ethoxy-1-(trimethylsilyloxy)ethene and BF3 · OEt2. After ester hydrolysis, heteroannular Friedel-Crafts cyclization of the resulting acids 30 afforded the enantiomerically enriched [3]ferrocenophanes 31. A number of [3]ferrocenophanes have been prepared by various condensation reactions. The [3]ferrocenophanes with functional groups at the 2-position of the bridge can be prepared by Dieckmann condensation of the dimethyl ester of ferrocene-1,1'-diacetic acid 32 followed by saponification and decarboxylation (Scheme 5.8) [64]. Like [3]ferrocenophane-1-one 22, [3]ferrocenophane-2-one 34 also can be reduced to the unsubstituted [3]ferrocenophane 20 and converted to the unsaturated [3]ferrocenophan-1-ene. Similarly, the intramolecular Claisen condensation of methyl 1'-acetylferrocenecarboxylate with potassium triphenylmethoxide in refluxing xylene afforded [3]ferrocenophane-1,3-dione 35 in high yield (78%) [65]. The single crystal X-ray structure of 35 showed that two Cp rings are tilted 9.88 with respect to each other [66]. The 2-substituted [3]ferrocenophane-1,3-dione 36 was prepared similarly. An alternative route to the 2-substituted compound, the alkylation of the carbanion

5.3 Mononuclear Carbon-Bridged Ferrocenophanes

Scheme 5.8

generated from [3]ferrocenophane-1,3-dione, however, afforded mainly the O-alkylation product 37, which was rationalized on the basis of steric hindrance of the 2-position by a-hydrogens of the Cp rings. On the other hand, condensation of [3]ferrocenophane-1,3-dione with benzaldehyde using triethylamine as catalyst gave 2-benzal[3]ferrocenophane-1,3-dione 38 [65]. The X-ray crystallographic data of 38 showed a little larger ring tilt angle (13.578) compared to that of 35 [67].

The compound 1-oxo-3-phenacyl[3]ferrocenophane 40 was prepared via internal Michael addition of 1-acetyl-1'-(3-oxo-3-phenyl-1-propenyl)ferrocene 39 using potassium hydroxide in boiling ethanol, where 1,14-dioxo[3.3]ferrocenophane-2,15-diene 41 was found as a co-product arising from the retro-Claisen-Schmidt reaction (Scheme 5.9) [68]. Subsequently, reaction with sodium methoxide in dimethylformamide was found to give only 40 in a slightly improved yield (21%). A series of 1,1'-dicinnamoylferrocenes 42 were converted to the corresponding [3]ferrocenophanes 43 by samarium diiodide-promoted reductive cyclization in a stereoselective manner (Scheme 5.10) [69]. In the single crystal X-ray structure, the Cp rings are in an eclipsed conformation with a ring twist angle of 2.78 and a ring tilt angle of 9.638. Unsaturated amino-substituted [3]ferrocenophanes 45 were prepared by an acidcatalyzed Mannich-type condensation of the corresponding enamines 44 [70, 71]. Catalytic hydrogenation of the resulting unsaturated [3]ferrocenophanes 45 in

Scheme 5.9

139

140

5 Carbon-Bridged Ferrocenophanes

Scheme 5.10

Scheme 5.11

THF at ambient temperature and pressure afforded the 1,3-trans-disubstituted [3]ferrocenophanes 46 (Scheme 5.11) [72]. 5.3.4

[4]Ferrocenophanes

[4]Ferrocenophanes can be prepared using many of the synthetic routes employed in the synthesis of [3]ferrocenophanes. The reaction of sodium cyclopentadienide 18 with 1,4-dibromobutane afforded the bis-cyclopentadienyl butane, which was then doubly deprotonated and treated with ferrous chloride to give [4]ferrocenophane 47 (Scheme 5.5) [57]. Another pathway to [4]ferrocenophanes involves the co-condensation of iron atoms with spirocyclic precursors. The co-condensation of iron atoms with a doubly vinylic cyclophane, [2.4]hepta-4,6-diene, yielded [4]ferrocenophane 47 as one of four different ferrocene derivatives [73]. 1-Methyl[4]ferrocenophane-4-one 48 was prepared by the treatment of methylenetriphenylphosphorane with 1,1'-diacetylferrocene and excess dimethylsulfinyl anion [74]. However, in contrast to the previous section, [4]ferrocenophane-1-one derivatives could not be prepared directly by intramolecular Friedel-Crafts cyclization of b-ferrocenylbutyric acid, which afforded the homoannularly cyclized compound 49 [58]. The alternative route, therefore, involves ring expansion of the trimethylene bridge of 22, which was prepared by heteroannular cyclization of b-ferrocenylpropionic acid. The bridge enlargement of 22 was achieved by the reaction with diazomethane to give two isomers, [4]ferrocenophane-1-one 50 and [4]ferrocenophane-2-one 51 [75]. When methanol was used as a catalyst, the formation of b-ketone 51 was exclusively dominant; in the reaction with BF3·OEt2 in benzene, however, the proportion of a-ketone 50 increased. In the latter case, bridge elonga-

5.3 Mononuclear Carbon-Bridged Ferrocenophanes

tion occurred even though the adjacent bridges sterically hindered the carbonyl group.

As mentioned in Section 5.3.2, ferrocenophanes possessing unsaturated carbon bridges are attractive for use as monomers in ROMP to generate conjugated organometallic polymers. Furthermore, heteroatom-bridged [1]ferrocenophanes can also be used to prepare conjugated organometallic polymers [28–42]. These latter materials, however, routinely exhibit lower conductivities than well-known p-conjugated organic polymers; the diminished conductivities have been attributed to the lack of p-overlap at the heteroatoms. On the other hand, ROMP of ferrocenophanes having unsaturated carbon bridges are expected to show enhanced conductivity because the unsaturated bonds are retained in the resulting polymers (Scheme 5.12). Several [4]ferrocenophanes having unsaturated carbon bridges have been prepared. The base-catalyzed heteroannular cyclization of 1,1'-bis((trimethylsilyl)ethynyl)ferrocene 54 afforded the fully conjugated 1,1'-(1-methoxy-1,3-butadienylene)ferrocene 55, where 54 was prepared by the coupling reaction of 1,1'-diiodoferrocene with two equivalents of (trimethylsilyl)acetylene catalyzed by (Ph3P)2PdCl2 and Cu(OAc)2· H2O [76, 77]. A variety of [4]ferrocenophanes can be produced from the 1,1'-(1-methoxy-1,3-butadienylene)ferrocene 55. Hydrolysis of 55 with aqueous acetic acid afforded the 1,1'-(4-oxo-1-butenylene)ferrocene 56 in 76% yield. The carbonyl group of 56 was reduced to the corresponding methylene group giving the [4]ferrocenophane 57 having a mono–olefinic bridge. On the other hand, the reduction of the 1,1'-(4-oxo-1-butenylene)ferrocene 56 with sodium borohydride gave the corresponding alcohol, 1,1'-(4-hydroxy-1-butenylene)ferrocene 58, which could be further converted to the diolefinic [4]ferrocenophane 52 by dehydration on activated alumina [77]. Analysis by single crystal X-ray crystallography showed that the ring tilt angles of the two Cp rings of 1,1'-(1-methoxy-1,3-butadienylene)ferrocene 55 were 7.168 and 11.078 for two independent molecules in the unit cell, which are comparable to that

Scheme 5.12

141

142

5 Carbon-Bridged Ferrocenophanes

Scheme 5.13

observed for [3]ferrocenophane-1-one 22 (8.88) [61] but significantly smaller than that of 1,1,2,2-tetramethyl[2]ferrocenophane 8 (23.28) [45]. The ring tilt angles suggest that the strain in 55 is comparable to that of [3]ferrocenophane-1-one 22; however, the longer bridge of the former complex allows a staggered conformation of the Cp rings [77]. The unsaturated butadiene-bridged ferrocenophanes 52 can also be prepared via reduction of the corresponding diketone 59 [78]. The starting compound, [4]ferrocenophane-1,4-dione 59, was prepared by the intramolecular coupling of commercially available 1,1'-diacetylferrocene [79]. Reduction of 59 with sodium borohydride in methanol gave a diastereomeric mixture of the corresponding diol 60, which upon treatment with p-toluenesulfonic acid in benzene afforded 1,1'(1,3-butadienylene)ferrocene 52. The single crystal X-ray structure of 52 exhibited a ring tilt angle of 10.28, which is slightly larger than that of [4]ferrocenophane1,4-dione 59 (6.78). Surprisingly, the sp2 carbons in the bridge of 52 showed bond angle distortions of * 108 (i.e., an average bond angle of 129.88 rather than 1208); these distortions were not observed in the diketone 59. The relief of strain is probably the major driving force for the ROMP of 52 [78]. In the early 1970s, Hisatome and his coworkers reported the preparation of various [4]ferrocenophanes by the acid-catalyzed reaction of 1,1'-bis(a-hydroxy-aphenylethyl)ferrocene 61. The mixture of products obtained from the reaction of a benzene solution of 61 with HCl included an insoluble peroxide derivative 62 and a soluble unsaturated derivative of [4]ferrocenophane 63 [80]. These compounds were also prepared by the oxidative reaction of 1,1'-bis(1-phenylvinyl)ferrocene

Scheme 5.14

5.3 Mononuclear Carbon-Bridged Ferrocenophanes

Scheme 5.15

with molecular oxygen in the presence of various diamagnetic or paramagnetic Lewis acids as two components of the product mixture [81]. As mentioned above, unsaturated butadiene-bridged ferrocenophanes are attractive candidates for the preparation of conjugated organometallic polymers via ROMP. Interest in ferrocene-based conjugated polymers arises from an expectation that the extended conjugation promises electrically conductive polymers. The development of processable high molecular weight polymers is desired for most microelectronics applications. It is anticipated that the ferrocene groups in ferrocene-containing polymers would be able to rotate, making the polymer soluble and thus processable. Several attempts to prepare such conjugated polymers, however, have led to materials having poor solubility, low conductivity, and low molecular weights. For example, the ROMP of [4]ferrocenophane-1,3-butadiene 52 afforded only low molecular weight oligomers/polymers that exhibited poor solubilities in common organic solvents (Scheme 5.12) [82]. The conductivity of the resulting low molecular weight materials after doping was ~10–4 X–1 cm–1, and interchain hopping was found to be the dominant mechanism for charge transport. In related work, the polymerization of 1,1'-(1-methoxy-1,3-butadienylene)ferrocene 55 afforded a soluble polymer in benzene [82]. The vinyl ether group in the resulting polymer, however, is known to undergo facile hydrolysis [76, 77]. To synthesize a soluble and thus processable polymer, 1,1'-(1-tert-butyl-1,3-butadienylene)ferrocene 65 was prepared [83]. ROMP of 65 with a tungsten-based metathesis catalyst afforded the deep red high molecular weight polymer 66, which was soluble in common organic solvents, including benzene, methylene chloride, and tetrahydrofuran. The UV/Vis spectrum of the resulting conjugated polymer 66 showed a bathochromic shift when compared with its monomer 65, suggesting extended conjugation for this new material [83].

Scheme 5.16

143

144

5 Carbon-Bridged Ferrocenophanes

Scheme 5.17

Scheme 5.18

Another example of an unsaturated [4]ferrocenophane having a benzoquinonelike structure 68 was prepared in 30% yield by the reaction of 1,1'-bis(a-bromoacetyl)ferrocene 67 with an excess of potassium butoxide in tetrahydrofuran [84]. Ring closing metathesis (RCM) of 1,1'-diallylferrocenes 69 and 70 with Grubbs’ catalyst afforded [4]ferrocenophane-2-enes 71 and 72, respectively. Starting with asubstituted derivatives (R = Ph, 4-MeOC6H4), RCM proceeded on the meso-diastereoisomers 69 and 70 to afford cis-disubstituted ferrocenophanes 71. For the methyl-substituted derivatives (R = Me), however, the trans-disubstituted ferrocenophanes 72 were also obtained. According to the results, large R groups appear to prevent RCM cyclization to the trans-disubstituted [4]ferrocenophane-2-ene 72 [85]. 5.3.5

[5]Ferrocenophanes

Like [3]ferrocenophanes and [4]ferrocenophanes, the most direct route to [5]ferrocenophanes involves treatment of the corresponding disodium salt of the bis-cyclopentadienyl hydrocarbon with ferrous chloride (Scheme 5.5) [59]. In 1960, Hauser and coworkers reported the base-catalyzed condensation of the readily available 1,1'-diacetylferrocene 73 with benzaldehyde, which afforded an unknown complex as well as the expected 1,1'-dicinnamoylferrocene 75 [86]. Later, Furdik suggested the 3-phenyl[5]ferrocenophane-1,5-dione 76 (R = Ph) structure for the unknown complex, and extended the reaction to the condensation with other aromatic [87–89] and aliphatic aldehydes [90]. Barr and Watts reported that the condensation of 73 with formaldehyde afforded the 2-ethoxymethyl[5]ferrocenophane1,5-dione 77 along with the [5]ferrocenophane-1,5-dione 76 (R = H) [91]. The mechanism for the formation of the [5]ferrocenophane-1,5-dione 76 (R = H) was proposed to consist of a Claisen-Schmidt reaction followed by internal Michael addi-

5.3 Mononuclear Carbon-Bridged Ferrocenophanes

Scheme 5.19

tion. Starting with these complexes, various [5]ferrocenophane derivatives could be prepared. The reduction of the diketone 76 by LAH–AlCl3 afforded the corresponding [5]ferrocenophane 78, which in turn was oxidized to the monoketone 79 by manganese dioxide. The resulting monoketone 79 was reduced by LAH to the alcohol 80, the dehydration of which introduced unsaturation into the bridge to give 81. On the other hand, the reduction of the diketone 76 by LAH afforded the tetrahydropyran derivative 83 instead of the expected diol 82. In addition to the preceding examples, a wide variety of [5]ferrocenophanes have been prepared using condensation reactions [91, 92].

The UV/Vis spectrum of [5]ferrocenophane-1-one 79 (R = H) exhibited a bathochromic shift of all absorption maxima, while the [3]ferrocenophane-1-one 22 failed to show such shifts. Furthermore, unlike the saturated analog 78, the unsaturated derivatives of [5]ferrocenophane-1-ene 81 showed well-defined maxima at 275 nm, indicating extended conjugation between the Cp rings and the adjacent double bonds. These observations indicate that overlap of the Cp p-electrons with those of adjacent double bonds in ferrocenophanes having bridges with five or more carbon atoms is conformationally possible due to their enhanced flexibility when compared to analogs bridged with shorter chains [91]. The compound 3-phenyl[5]ferrocenophane-1,5-dione 76 (R = Ph) can be utilized as a monomer for the preparation of nonconjugated conducting polymers. As shown in

145

146

5 Carbon-Bridged Ferrocenophanes

Scheme 5.20

the previous section, electronic interactions between ferrocenyl moieties can be facilitated by the preparation of conjugated polymers. Enhanced electronic communication also can be achieved by close co-facial alignment of the Cp rings of ferrocenyl moieties. From the dione 76 (R = Ph), 3-phenyl[5]ferrocenophane-1,5-dimethylene 84 was prepared via Wittig reaction in 72% yield. The resulting diene 84 was further radically cyclopolymerized to give the corresponding poly(ferrocenophane) 85, which allows close co-facial stacking of the Cp ligands [93]. 5.3.6

[m]Ferrocenophanes (m > 5)

Ferrocenophanes having bridges of more than five carbon atoms are expected to contain less strain than those having shorter bridges. The heteroannular cyclization of diester 86 can provide ferrocenophanes bridged by more than 5 carbon atoms 87 [94, 95]. By this condensation, [6]-, [8]-, [9]-, and [10]ferrocenophanes have been prepared in relatively good yields (50–75%), while [4]ferrocenophane was synthesized in 20% yield. An interesting class of ferrocenophanes having more than five carbon atoms in the bridge consists of ferrocenophanes having extended transannular p-electronic interactions between aromatic rings and double bonds. The compound [2]paracyclo[2]paracyclo[2]ferrocenophane-1,9,17-triene 89 was obtained as part of the product mixture from the intramolecular titanium-induced reductive coupling of dialdehyde 88, which was prepared by the palladium-catalyzed substitution of 1,1'diiodoferrocene with 4-vinylbenzaldehyde [96]. A similar intramolecular titaniuminduced reductive coupling of 1,1'-bis(p-formylphenylethynyl)ferrocene afforded [2]paracyclo[2]paracyclo[2]ferrocenophane-9-ene-1,17-diyne, which exhibited a bathochromic shift compared with 1-(p-methylphenylethynyl)ferrocene, indicating

Scheme 5.21

5.3 Mononuclear Carbon-Bridged Ferrocenophanes

Scheme 5.22

Scheme 5.23

a transannular p-electron interaction between the two benzene rings and the two triple bonds [96]. Various mononuclear and multinuclear ferrocenophanes having a conjugated chain of double bonds and para-substituted benzene rings in the bridge have been prepared by one-pot multiple Wittig reactions [97]. A method was developed for preparing macrocyclic ferrocenophanes via the cyclization of 1,1'-bis(x-cyanoalkyl)ferrocenes 90 promoted by methyl magnesium iodide or sodium N-methylanilide [98]. The solubilities of the ferrocenophanes 91 were found to increase with increasing bridge length. 5.3.7

Multiply-Bridged Mononuclear Ferrocenophanes ([m]nFerrocenophanes)

Schlögl and coworkers reported the synthesis of a di-bridged ferrocenophane via heteroannular Friedel-Crafts cyclization of b-ferrocenophanylpropionic acid 92 (see chapter 6, section 6.4.2 for the analogous ruthenocenophane) [99, 100]. The propionic acid 92 was prepared by formylation of [3]ferrocenophane 20 followed by condensation of the resulting aldehyde with malonic acid and then catalytic hydrogenation of the b-ferrcenophanylacrylic acid. Rinehart’s group also synthesized the multiply-bridged ferrocenophanes by Friedel-Crafts acylation [101]. They employed, however, a different synthetic strategy to prepare the propionic acid 92, which was synthesized by acetylation of [3]ferrocenophane 20, carboethoxylation of the acetyl derivative, catalytic hydrogenation of the b-carbonyl group, and saponification of the resulting ester. In both methods, [3]-a-oxo-[3]ferrocenophane 93 was reduced by LAH to afford the fully saturated [3][3]ferrocenophane 94. Rinehart’s method produced isomeric acetyl derivatives and thus [3][3]ferrocenophane isomers, [3][3](1,1';2,2')ferrocenophane, and [3][3](1,1';3,3')ferrocenophane. In contrast, Schlögl’s formylation method afforded only the (1,3) isomer. The tri-bridged ferrocenophane isomers 95 were prepared using Rinehart’s method, starting from [3][3]ferrocenophane [101].

147

148

5 Carbon-Bridged Ferrocenophanes

Scheme 5.24

Analysis of the multiply-bridged [3][3](1,1';3,3')ferrocenophane 94 by single crystal X-ray diffraction found that the average distance between the two cyclopentadienyl rings is 3.18 Å, which is smaller than that of unbridged ferrocene (3.32 Å) [102]. The reduced distance implies that the second trimethylene bridge introduces additional strain. The single crystal X-ray structure of [3][3][3]ferrocenophane 95 showed a shorter Cp–Cp' distance (3.15 Å) than that of [3][3]ferrocenophane 94 as well as that of ferrocene itself [103]. The ring tilt angle of 95 was 2.48, which is smaller than that of the di-bridged analog (98) [102]. Studies examining the effect of the number of trimethylene bridges on the structure of the corresponding ferrocenophanes found that the added bridges shorten the Cp–Cp’ distances and increase the electron density of the metal center [104]. The synthesis of [4]ferrocenophanes bridged with five tetramethylene groups has been reported (see chapter 4, section 4.5) [105, 106]. In contrast, the pentabridged [3]ferrocenophane has yet to be prepared due to its highly strained structure. Nevertheless, several tri-bridged [3]ferrocenophanes have been prepared [100, 101, 107, 108]; apparently, the structural strain introduced by an additional short trimethylene chain disturbs the preparation of [3]ferrocenophanes linked with more than three trimethylene bridges. In an effort to prepare the tetra-bridged [3][3][3][3]ferrocenophane 96, Schlögl and Peterlik applied their formylation method to [3][3]ferrocenophane [100]. Reinhart’s group, however, showed that the resulting compound was actually the di-bridged ferrocenophane 97 produced by homoannular cyclization [109]. Similarly, the attempt to prepare [3][3][3][3][3]ferrocenophane 98 yielded the [3][3][3]ferrocenophane derivative 99 [110]. In the single crystal X-ray structures of both 97 and 99, the Cp ligands reside in an eclipsed conformation.

Hisatome and coworkers assumed that the failure to prepare tetra- or pentabridged [3]ferrocenophanes from the propionic acid derivatives of [3][3][3]ferrocenophane 95 was due to the considerably reduced distance between the two Cp rings in the starting materials (3.15 Å), which prevents the acylium ion of a tribridged [3]ferrocenophane from attacking the corresponding carbon of the facing Cp ring [111]. Elongation of the distance between the two Cp rings and thus the

5.3 Mononuclear Carbon-Bridged Ferrocenophanes

Scheme 5.25

two carbon atoms to be linked was expected to facilitate heteroannular cyclization rather than homoannular cyclization (Scheme 5.25). According to structural studies, a distance longer than 3.20 Å between the two carbon atoms to be bridged to each other is required to form the expected tetra-bridged [3]ferrocenophane [112]. The elongation could be achieved by replacing one of the trimethylene groups of [3][3][3]ferrocenophane-3-propionic acid with a longer tetramethylene group. The introduction of a tetramethylene bridge into the position adjacent to that targeted for bridging was expected to give a distance of 3.35 Å between the Cp and Cp' carbons (e.g., the 2- and 2'-carbons adjacent to the tetramethylene bridge in the [4][3][3](1,1';3,3';4,4')ferrocenophane 101). Using this strategy, Hisatome and coworkers successfully prepared the tetra-bridged [4][3][3][3]ferrocenophane 103 and the penta-bridged [4][3][3][3][3]ferrocenophane 104 [109, 110]. The single crystal Xray structure of 104 showed that the two Cp rings reside in a nearly perfectly eclipsed conformation, almost parallel to each other, with a ring tilt angle of only 1.98. Recently, the heteroannular Friedel-Crafts cyclization of ferrocenylpropanoic acids was applied to the asymmetric synthesis of [3][3](1,1';3,3')ferrocenophane [63, 113]. The SnCl4-promoted intramolecular Friedel–Crafts cyclization of (R,R)1,1'-bis(2-chloroformyl-1-phenylethyl)ferrocene prepared from 105 afforded a

Scheme 5.26

149

150

5 Carbon-Bridged Ferrocenophanes

7 : 37 : 56 ratio of singly-bridged [3]ferrocenophane isomers. The major diastereomer was isolated after the mixture was converted to the corresponding methyl esters 106, 107, 108 due to the poor stabilities of the acids. After reduction and ester hydrolysis, TFAA-mediated intramolecular Friedel–Crafts cyclization of the acid 109 gave a ferrocenyl ketone, which was further reduced to a di-bridged C2-symmetric ferrocenophane 110. The single crystal X-ray structure of the ferrocenophane revealed a ring tilt angle of 6.58.

5.4

Multinuclear Ferrocenophanes

Multinuclear ferrocenophanes consist of more than two ferrocene units linked together. In these complexes, electronic communication between the metals may be transmitted “through bond” and/or “through space”. Multinuclear ferrocenophanes, therefore, have attracted many researchers due to their potential technical applications. The properties of multinuclear ferrocenophanes vary with the structure and the nature of the bridge. Depending on the degree of intramolecular interaction, multinuclear ferrocenophanes can exhibit properties of the individual components or something altogether different. 5.4.1

[0.0]Ferrocenophanes

The first example of a simple multinuclear ferrocenophane, [0.0]ferrocenophane 112, has been prepared via several synthetic strategies, including pyrolysis of polymercury ferrocenophane [114], modified Ullmann coupling of 1,1'-diiodoferrocene [115], coupling of 1,1'-dilithioferrocene using organocuprates [116], and the reaction of fulvalene dianion 111 with ferrous chloride [117]. The yields of these reactions are typically low due to the high probability of oligomerization and polymerization. The direct synthesis via the reaction of fulvalene dianion with FeCl2 gave the highest yield (40%) of all of these reactions (Scheme 5.27) [118]. The observed poor solubility of 112 in organic solvents was attributed to the rigidity of the molecule, which prevents adequate solvation. According to the single crystal X-ray structure, the two iron atoms of 112 are located in the center of the ferrocene units at a metal–metal distance of 3.98 Å [119]. The iron(II)-iron(III) mixed valence system of the 112 was obtained by benzoquinone oxidation. These complexes showed a broad near-infrared absorption, which can be assigned to intramolecular electron transfer. Com-

Scheme 5.27

5.4 Multinuclear Ferrocenophanes

pound 112 underwent facile oxidation by tetracyanobenzoquinodimethane (TCNQ) to give a mixed valence TCNQ salt, which was expected to show high conductivity. The room temperature bulk conductivity of compressed disks of the TCNQ salt was measured to be consistently above 10 X–1 cm–1, which indicates an unusually high conductivity along the main crystal axis [117]. Another known example of a [0.0]ferrocenophane is bis(indacenyl)diiron [120, 121]. This compound was prepared by the reaction of dilithium indacenide with ferrous chloride. The molecule exists as a mixture of syn 113 and anti 114 isomers.

5.4.2

[1n]Ferrocenophanes

Reduction of the bisfulvene 115 with lithium aluminum hydride or methyl lithium afforded the dianion 116, which in turn was treated with ferrous chloride to give [1.1]ferrocenophane 117 in yields ranging from 14 to 20% (Scheme 5.28) [122, 123]. The 1H NMR analysis of 117 suggested that the two methyl groups possess an exo configuration. Although a twisted anti conformer of the compound was proposed due to the steric strain in the syn conformer, the single crystal X-ray structure of the compound showed that the bridges exist in a syn conformation with the methyl groups in the exo position. The Cp rings of each ferrocene unit lie in a staggered conformation at an angle of 22.48 and with a ring tilt angle of 2.78 [124]. The exo,exo,anti-1,12-dimethyl[1.1]ferrocenophane, however, was later identified. The anti conformer showed large twist (36.18 and 34.08), rotation (3.48 and 53.98), and tilt angles (4.18 and 22.78), which suggests a high flexibility of the molecule, unlike the initial expectation [125, 126]. Through the intermediacy of 1,1'-bis(6-fulvenyl)ferrocene 119, the unsubstituted [1.1]ferrocenophane 118 was also prepared [127]. The reaction of 1,1'-dilithioferro-

Scheme 5.28

151

152

5 Carbon-Bridged Ferrocenophanes

cene with 6-dimethylaminofulvene followed by hydrolysis afforded 119 in better than 80% yield. Similarly, a mixed [1.1]metallocenophane was prepared [128] and its electrochemical properties were investigated by cyclovoltammetry [129]. The hetero-binuclear iron/ruthenium system 120 exhibited a reversible one-electron oxidation associated with ferrocene and an irreversible two-electron oxidation associated with ruthenocenophane, indicating a lack of interaction between the two metallocenophane units.

In related work, the aluminum chloride-catalyzed Friedel-Crafts reaction between ferrocene and 1,1'-bis(chlorocarbonyl)ferrocene 121 afforded the dinuclear [1.1]ferrocenophane-1,12-dione 122 (Scheme 5.29) [123]. The diketone was also synthesized by self-condensation of chlorocarbonylferrocene, where the yield was low due to polymerization.

Scheme 5.29

The reaction of bis(cyclopentadienyl)methane dianion with ferrous chloride or ferrous hexaamine thiocyanate afforded a mixture of four [1n]ferrocenophanes, where n = 2, 3, 4, and 5. The structures of the first two compounds 123 and 124 are shown below. An alternative route to the dimer, [1.1]ferrocenophane, involves the aluminum chloride-catalyzed reduction of the diketone 122 with LiAlH4 [120].

5.4 Multinuclear Ferrocenophanes

5.4.3

[mn]Ferrocenophanes

Multinuclear ferrocenophanes bridged by unsaturated carbon chains are interesting due to their possible electronic communication between metal centers. The intermolecular coupling of ferrocene-1,1'-dicarbaldehyde with TiCl4–LiAlH4 afforded [2.2]ferrocenophane-1,13-diene 125, which was further converted to the saturated analog 126 by catalytic hydrogenation [130]. An alternative route to the saturated [2.2]ferrocenophane involves the reductive coupling of 1,1'-bis(hydroxymethyl)ferrocene with TiCl4–LiAlH4. The UV/Vis spectrum of the [2.2]ferrocenophane-1,13diene 125 exhibited a bathochromic shift and an increase of intensity of the bands in the 230–320 nm region when compared with that of the mono-bridged analog 127. Another class of multinuclear ferrocenophanes having olefinic bridges consists of [3.3]ferrocenophane-1,15-diene-3,14-dione 128, [5.5]ferrocenophane-1,4,16,19-tetraene-3,18-dione 129, and [5.5]ferrocenophane-1,3,17,19-tetraene-5,16-dione 130, which were prepared by intermolecular base-catalyzed condensations [131]. The IR and NMR spectra of the [5.5]ferrocenophanes suggest trans olefinic bonds. The UV/Vis spectra showed bathochromic shifts and broadening compared with their mono-bridged analogs, indicating p-electronic interaction. The ferrocenoruthenocenophane analogs of these complexes have also been prepared [132].

The unsaturated [2.2]ferrocenophane-1,13-diyne 131 was prepared by the coppercatalyzed coupling of 1-ethynyl-1'-iodoferrocene [133]. This compound exhibited poor solubility and a high melting point due to its rigidity and high symmetry, respectively. The kinetically stabilized derivative of [4.4]ferrocenophane-1,3,15,17-tetrayne 132 was synthesized by oxidative Eglinton coupling of 2,2',4,4'-tetra-tert-butyl-1,1'-diethynylferrocene, which was prepared via deprotonation of 1,3-di-tert-butyl-5-vinylidenecyclopentadiene with lithium 2,2,6,6-tetramethylpiperidide (LiTMP) followed by subsequent treatment with ferrous chloride [134]. The UV/Vis spectrum of 132 exhibited a bathochromic shift compared with that of the correspond-

153

154

5 Carbon-Bridged Ferrocenophanes

ing monomer, which indicates delocalization through the conjugated system. Analysis by single crystal X-ray diffraction showed that the two ferrocene units are twisted by about 828 relative to each other. The cyclic voltammogram suggested a redox interaction between the two ferrocene units.

5.5

Summary

This review has focused on the synthesis and structure of ferrocenophanes having bridges composed only of carbon atoms. Despite this limitation, the ferrocenophanes described above have been observed to possess a wide variety of compositions and structures. These variations offer potentially useful applications for carbon-bridged ferrocenophanes, including their use in catalysis, sensor applications, and as stable lightweight electronic materials. Continued efforts to design and synthesize unique carbon-bridged ferrocenophanes will undoubtedly lead to their use in emerging devices and technologies.

5.6

References 1 2 3

4 5 6

7 8

T. J. Kealy, P. L. Pauson, Nature (London) 1951, 168, 1039–1040. S. A. Miller, J. A. Tebboth, J. F. Tremaine, J. Chem. Soc. 1952, 632–635. G. Wilkinson, M. Rosenblum, M. C. Whiting, R. B. Woodward, J. Am. Chem. Soc. 1952, 74, 2125–2126. E. O. Fischer, W. Pfab, Z. Naturforsch. Teil b 1952, 7, 377–379. P. F. Eiland, R. Pepinsky, J. Am. Chem. Soc. 1952, 74, 4971. R. B. Woodward, M. Rosenblum, M. C. Whiting, J. Am. Chem. Soc. 1952, 74, 3458–3459. J. D. Dunitz, L. E. Orgel, Nature 1953, 171, 121–122. P. Seiler, J. D. Dunitz, Acta Crystallogr. 1979, B35, 1068, 2020–2032.

9

10 11 12

13 14

15

A. J. Campbell, C. A. Fyfe, D. HaroldSmith, K. R. Jeffrey, Mol. Cryst. Liq. Cryst. 1976, 36, 1–23. A. Haaland, Acc. Chem. Res. 1979, 12, 415–422. S. Carter, J. N. Murrell, J. Organomet. Chem. 1980, 192, 399–408. M. C. Zerner, G. H. Loew, R. F. Kirchner, U. T. Mueller-Westerhoff, J. Am. Chem. Soc. 1980, 102, 589–599. J. D. Dunitz, L. E. Orgel, A. Rich, Acta Crystallogr. 1956, 9, 373–375. W. F. Little in Survey of Progress in Chemistry; A. Scott ed. Academic, New York, 1963, vol. 1, p. 133. J. P. Collman, L. S. Hegedus, J. R. Norton, R. G. Finke, Principles and Applica-

5.6 References

16

17

18

19 20 21 22 23 24 25 26 27 28 29

30

31

32

33 34

tions of Organotransition Metal Chemistry; University Science, Mill Valley, 1987. M. Rosenblum, Chemistry of the Iron Group Metallocenes: Ferrocene, Ruthenocene, Osmocene; Interscience Publishers, New York, 1965. Ferrocenes: Homogeneous, Catalysis, Organic Synthesis, Material Science, A. Togni, T. Hayashi, eds., VCH, New York, 1995. F. A. Cotton, G. Wilkinson, Advanced Inorganic Chemistry; Wiley, New York, 1980, p. 1164. R. M. Crooks, A. J. Ricco, Acc. Chem. Res. 1998, 31, 219–227. S. R. Marder, J. W. Perry, Science 1994, 263, 1706–1707. A. O. Patil, A. J. Heeger, F. Wudl, Chem. Rev. 1988, 88, 183–200. T. J. Marks, Science 1985, 227, 881–889. W. E. Watts, Organomet. Chem. Rev. 1967, 2, 231–254. U. T. Mueller-Westerhoff, Angew. Chem. Int. Ed. Engl. 1986, 25, 702–717. R. W. Heo, T. R. Lee, J. Organomet. Chem. 1999, 578, 31–42. B. H. Smith, Bridged Aromatic Compounds; Academic Press, New York, 1964. F. Vögtle, P. Neumann, Tetrahedron 1970, 26, 5847–5873. D. A. Foucher, B. Z. Tang, I. Manners, J. Am. Chem. Soc. 1992, 114, 6246–6248. A. Berenbaum, H. Braunschweig, R. Dirk, U. Englert, J. C. Green, F. Jäkle, A. J. Lough, I. Manners, J. Am. Chem. Soc. 2000, 122, 5765–5774. H. Braunschweig, R. Dirk, M. Müller, P. Nguyen, R. Resendes, D. P. Gates, I. Manners, Angew. Chem. Int. Ed. Engl. 1997, 36, 2338–2340. M. J. MacLachlan, J. Zheng, K. Thieme, A. J. Lough, I. Manners, C. Cordas, R. LeSuer, W. E. Geiger, L. M. LiableSands, A. L. Rheingold, Polyhedron 2000, 19, 275–289. J. K. Pudelski, D. A. Foucher, C. H. Honeyman, P. M. Macdonald, I. Manners, S. Barlow, D. O’Hare, et al., Macromolecules 1996, 29, 1894–1903. K. Temple, S. Dziadek, I. Manners, Organometallics 2002, 21, 4377–4384. A. Berenbaum, A. J. Lough, I. Manners, Organometallics 2002, 21, 4415–4424.

35

36

37

38 39

40

41 42 43 44 45 46

47 48 49 50 51

52 53

54

R. N. Kapoor, G. M. Crawford, J. Mahmoujd, V. V. Dementiev, M. T. Nguyen, A. F. Diaz, K. H. Pannell, Organometallics 1995, 14, 4944–4947. D. A. Foucher, M. Edwards, R. A. Burrow, A. J. Lough, I. Manners, Organometallics 1994, 13, 4959–4966. R. Rulkens, A. J. Lough, I. Manners, Angew. Chem. Int. Ed. Engl. 1996, 35, 1805–1807. T. Mizuta, M. Onishi, K. Miyoshi, Organometallics 2000, 19, 5005–5009. R. Rulkens, D. P. Gates, D. Balaishis, J. K. Pudelski, D. F. McIntosh, A. J. Lough, I. Manners, J. Am. Chem. Soc. 1997, 119, 10976–10986. J. K. Pudelski, D. P. Gates, R. Rulkens, A. J. Lough, I. Manners, Angew. Chem. Int. Ed. Engl. 1995, 35, 1506–1508. I. Manners, Angew. Chem. Int. Ed. Engl. 1996, 35, 1602–1621. R. D. Hudson, J. Organomet. Chem. 2001, 637–639, 47–69. H. L. Lentzner, W. E. Watts, Tetrahedron 1971, 27, 4343–4351. T. H. Barr, W. E. Watts, Tetrahedron 1968, 24, 6111–6118. M. B. Laing, K. N. Trueblood, Acta Crystallogr. 1965, 19, 373–381. K. L. Rinehart, Jr., A. K. Frerichs, P. A. Kittle, L. F. Westman, D. H. Gustafson, R. L. Pruett, J. E. McMahon, J. Am. Chem. Soc. 1960, 82, 4111–4112. J. M. Nelson, H. Rengel, I. Manners, J. Am. Chem. Soc. 1993, 115, 7035–7036. K. Yasufuku, H. Yamazaki, J. Organomet. Chem. 1976, 121, 405–411. K. Yasufuku, H. Yamazaki, J. Organomet. Chem. 1977, 127, 197–207. K. Yasufuku, K. Aoki, H. Yamazaki, Inorg. Chem. 1977, 16, 624–628. K. Onitsuka, K. Miyaji, T. Adachi, T. Yoshida, K. Sonogashira, Chem. Lett. 1994, 2279–2282. E. Deschamps, L. Ricard, F. Mathey, Organometallics 2001, 20, 1499–1500. W. Finckh, B. Z. Tang, D. A. Foucher, D. B. Zamble, A. Lough, I. Manners, Organometallics 1993, 12, 823–829. J. M. Nelson, P. Nguyen, R. Petersen, H. Rengel, P. M. Macdonald, A. J. Lough, I. Manners, N. P. Raju, J. E.

155

156

5 Carbon-Bridged Ferrocenophanes

55 56

57 58

59 60

61

62

63 64 65 66

67 68 69 70 71 72

73

74

Greedan, S. Barlow, D. O’Hare, Chemistry-A Europ. J. 1997, 3, 573–584. M. A. Buretea, T. D. Tilley, Organometallics 1997, 16, 1507–1510. R. R. Schrock, J. S. Murdzek, G. C. Bazan, J. Robbins, M. R. DiMare, M. O’Regan, J. Am. Chem. Soc. 1990, 112, 3875– 3886. A. Lüttringhaus, W. Kullick, Makromol. Chem. 1961, 44–46, 669–681. K. L. Rinehart, Jr., R. J. Curby, Jr., D. H. Gustafson, K. G. Harrison, R. E. Bozak, D. E. Bublitz, J. Am. Chem. Soc. 1962, 84, 3263–3269. K. L. Rinehart, Jr., R. J. Curby, Jr., J. Am. Chem. Soc. 1957, 79, 3290–3291. K. L. Rinehart, Jr., R. J. Curby, Jr., P. E. Sokol, J. Am. Chem. Soc. 1957, 79, 3420–3424. N. D. Jones, R. E. Marsh, J. H. Richards, Acta Crystallogr. 1965, 19, 330– 336. P. Härter, K. Latzel, M. Spiegler, E. Herdtweck, Polyhedron 1998, 17, 1141– 1148. A. J. Locke, C. J. Richards, Organometallics 1999, 18, 3750–3759. W. Mock, J. H. Richards, J. Org. Chem. 1962, 27, 4050–4051. M. Salisova, S. Toma, E. Solcaniova, J. Organomet. Chem. 1977, 132, 419–427. E. Gyepes, T. Glowiak, S. Toma, J. Soldanova, J. Organomet. Chem. 1984, 276, 209–214. E. Gyepes, T. Glowiak, S. Toma, J. Organomet. Chem. 1986, 316, 163–168. S. Toma, M. Salisova, J. Organomet. Chem. 1973, 57, 191–198. S.-J. Jong, J.-M. Fang, Org. Lett. 2000, 2, 1947–1949. S. Knüppel, R. Fröhlich, G. Erker, J. Organomet. Chem. 1999, 586, 218–222. S. Knüppel, R. Fröhlich, G. Erker, J. Organomet. Chem. 2000, 595, 308–312. P. Liptau, S. Knüppel, G. Kehr, O. Kataeva, R. Fröhlich, G. Erker, J. Organomet. Chem. 2001, 637–639, 621–630. J. A. Gladysz, J. G. Fulcher, R. C. Ugolick, A. J. Lee Hanlan, A. B. Bocarsly, J. Am. Chem. Soc. 1979, 101, 3388–3390. W. M. Horspool, R. G. Sutherland, Chem. Comm. 1967, 5, 240.

75

76 77 78

79

80

81

82

83 84

85

86 87 88 89 90

91 92 93

94

M. Hisatome, Y. Kawajiri, K. Yamakawa, J. Organomet. Chem. 1982, 226, 71– 84. J. K. Pudelski, M. R. Callstrom, Organometallics 1992, 11, 2757–2759. J. K. Pudelski, M. R. Callstrom, Organometallics 1994, 12, 3095–3109. M. S. Erickson, F. R. Fronczek, M. L. McLaughlin, Tetrahedron Lett. 1993, 34, 197–198. Y. Ito, T. Konoike, T. Harada, T. Saegusa, J. Am. Chem. Soc. 1977, 99, 1487– 1493. M. Hisatome, S. Minagawa, K. Yamakawa, J. Organomet. Chem. 1973, 55, C82– C84. M. Hisatome, T. Takayuki, K. Yamakawa, J. Organomet. Chem. 1976, 117, C23– C26. C. E. Stanton, T. R. Lee, R. H. Grubbs, N. S. Lewis, J. K. Pudelski, M. R. Callstrom, M. S. Erickson, M. L. McLaughlin, Macromolecules 1995, 28, 8713–8721. R. W. Heo, F. B. Somoza, T. R. Lee, J. Am. Chem. Soc. 1998, 120, 1621–1622. P. Molina, A. Tarraga, D. Curiel, M. D. Velasco, J. Organomet. Chem. 2001, 637–639, 258–265. A. J. Locke, C. Jones, C. J. Richards, J. Organomet. Chem. 2001, 637–639, 669– 676. T. A. Mashburn, C. E. Cain, C. R. Hauser, J. Org. Chem. 1960, 25, 1982–1986. M. Furdik, S. Toma, J. Suchy, et al. Chem. Zvesti 1961, 15, 45–62. M. Furdik, S. Toma, J. Suchy, Chem. Zvesti 1961, 15, 789. M. Furdik, S. Toma, J. Suchy, Chem. Zvesti 1962, 16, 449–457. M. Furdik, M. Dzurilla, S. Toma, J. Suchy, Acta Facultatis Rerum Naturalium Universitatis Comenianae, Chimia 1964, 8, 569–579. T. H. Barr, W. E. Watts, Tetrahedron 1968, 24, 3219–3235. T. H. Barr, W. E. Watts, Tetrahedron 1969, 25, 861–869. C. J. Neef, D. T. Glatzhofer, K. M. Nicholas, J. Polym. Sci. Part A: Polym. Chem. 1997, 35, 3365–3376. K. Schlögl, H. Seiler, Monatsh. 1960, 91, 79–91.

5.6 References 95 K. Schlögl, H. Seiler, Angew. Chem. 96 97 98

99 100 101

102 103 104

105

106

107 108

109 110

111 112

113

114

115

1960, 72, 38. K. Kasahara, T. Izumi, I. Shimizu, Chem. Lett. 1979, 1119–1122. D. Tanner, O. Wennerström, Acta Chem. Scand. B 1980, 34, 529–535. A. N. Nesmeyanov, M. I. Rybinskaya, G. B. Shul’Pin, A. A. Pogrebnyak, J. Organomet. Chem. 1975, 92, 341–367. K. Schlögl, H. Seiler, Tetrahedron Lett. 1960, 1, 4–8. K. Schlögl, M. Peterlik, Tetrahedron Lett. 1962, 3, 573–576. K. L. Rinehart, Jr., D. E. Bublitz, D. H. Gustafson, J. Am. Chem. Soc. 1963, 85, 970–982. I. C. Paul, Chem. Comm. 1966, 377–378. M. Hillman, E. Fujita, J. Organomet. Chem. 1978, 155, 87–98. M. Hillman, B. Gordon, A. J. Weiss, A. P. Guzikowski, J. Organomet. Chem. 1978, 155, 77–86. M. Hisatome, J. Watanabe, K. Yamakawa, Y. Iitaka, J. Am. Chem. Soc. 1986, 108, 1333–1334. M. Hisatome, J. Watanabe, Y. Kawajiri, K. Yamakawa, Y. Iitaka, Organometallics 1990, 9, 497–503. K. Schlögl, M. Peterlik, Monatsh. 1962, 93, 1328–1342. M. Hisatome, T. Sakamoto, K. Yamakawa, J. Organomet. Chem. 1976, 107, 87– 101. D. E. Bublitz, K. L. Reinhart, Jr., Tetrahedron Lett. 1964, 5, 827–833. L. D. Spaulding, M. Hillman, G. J. Williams, J. Organomet. Chem. 1978, 155, 109–116. J. Watanabe, M. Hisatome, K. Yamakawa, Tetrahedron Lett. 1987, 28, 1427–1430. M. Hisatome, J. Watanabe, K. Yamakawa, K. Kozawa, T. Uchida, Bull. Chem. Soc. Jpn. 1995, 68, 635–644. A. J. Locke, C. J. Richards, D. E. Hibbs, M. B. Hursthouse, Tetrahedron: Asymmetry 1997, 8, 3383–3386. M. D. Rausch, R. F. Kovar, C. S. Kraihanzel, J. Am. Chem. Soc. 1969, 91, 1259–1261. F. I. Hedberg, H. Rosenberg, J. Am. Chem. Soc. 1969, 91, 1258–1259.

116 A. Davison, A. W. Rudie, Synth. React.

Inorg. Met.-Org. Chem. 1980, 10, 391–395. 117 U. T. Mueller-Westerhoff, P. Eil-

118

119 120 121 122 123 124

125

126

127

128

129

130 131 132

133

134

bracht, J. Am. Chem. Soc. 1972, 94, 9272–9274. C. LeVanda, K. Bechgaard, D. O. Cowan, U. T. Mueller-Westerhoff, P. Eilbracht, G. A. Candelar, R. L. Collins, J. Am. Chem. Soc. 1976, 98, 3181–3187. M. R. Churchill, J. Wormwald, Inorg. Chem. 1969, 8, 1970–1974. T. J. Katz, N. Acton, G. Martin, J. Am. Chem. Soc. 1973, 95, 2934–2939. T. J. Katz, W. Slusarek, J. Am. Chem. Soc. 1980, 102, 1058–1063. W. E. Watts, J. Am. Chem. Soc. 1966, 88, 855–856. T. H. Barr, H. L. Lentzner, W. E. Watts, Tetrahedron 1969, 25, 6001–6013. J. S. McKechnie, B. Bersted, I. C. Paul, W. E. Watts, J. Organomet. Chem. 1967, 8, P29–P31. U. T. Mueller-Westerhoff, A. Nazzal, W. Prossdorf, J. Am. Chem. Soc. 1981, 103, 7678–7681. M. Löwendahl, Ö. Davidsson, P. Ahlberg, M. Håkansson, Organometallics 1993, 12, 2417–2419. A. Cassens, P. Eilbracht, A. Nazzal, W. Prössdorf, U. T. Mueller-Westerhoff, J. Am. Chem. Soc. 1981, 103, 6367–6372. U. T. Mueller-Westerhoff, A. Nazzal, M. Tanner, J. Organomet. Chem. 1982, 236, C41–C44. A. F. Diaz, U. T. Mueller-Westerhoff, A. Nazzal, M. Tanner, J. Organomet. Chem. 1982, 236, C45–C48. A. Kasahara, T. Izumi, Chem. Lett. 1978, 21–24. A. Kasahara, T. Izumi, I. Shimizu, Chem. Lett. 1979, 1317–1320. S.-I. Kamiyama, A. Kasahara, T. Izumi, I. Shimizu, H. Watabe, Bull. Chem. Soc. Jpn. 1981, 54, 2079–2082. M. Rosenblum, N. M. Brawn, D. Ciappenelli, J. Tancrede, J. Organomet. Chem. 1970, 24, 469–477. K. H. H. Fabian, H.-J. Lindner, N. Nimmerfroh, K. Hafner, Angew. Chem. Int. Ed. 2001, 40, 3402–3405.

157

159

6

Endohedral Metal Complexes of Cyclophanes Rolf Gleiter, Bernhard J. Rausch, and Rolf J. Schaller

6.1

Introduction

[m,n]Phanes can act as ligands towards metal fragments due to their p-electrons. Four different structural types are possible (A–D). In the cases of A and B either one or both p-systems act as ligands. In the latter case the formation of multidecker systems is also possible as indicated in C. In type D the metal atom is situated inside the cage built by the cyclophane. Following fullerene chemistry – in which fullerenes with a metal center situated inside the cage are called endohedral fullerenes – type D is referred to as endohedral metal complexes. Since examples for A–C are legion we mention some reviews [1]. This chapter is devoted to endohedral metal complexes of cyclophanes. Because they contain bridges, these systems have the capability to stabilize unusual oxidation states and, therefore, they deserve special interest. The following paragraphs summarize the syntheses and properties of the known endohedral metal p-complexes of type D and are ordered according to the position of the lightest caged metal of the group in the periodic table.

Scheme 6.1

Modern Cyclophane Chemistry. Edited by Rolf Gleiter, Henning Hopf Copyright © 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 3-527-30713-3

160

6 Endohedral Metal Complexes of Cyclophanes

6.2

Singly-Bridged Group IVB Metallocenes

If one speaks of “metallocenophanes” one does not usually include singly-bridged group IVB metallocenes. These compounds have great importance in applications as catalyst precursors for olefin polymerization [2] as well as reagents for organic synthesis [3], so their methods of preparation are briefly summarized in this chapter. While much experimental work on singly- and doubly-bridged ansa-metallocenes of zirconium, titanium and hafnium has been reported, most of it has been done on complexes with one or two bridging atoms. Additionally, the electronic structure of these compounds has been examined by physical methods, such as IR- [4] and ESCA-spectroscopy [5], as well as cyclic voltammetry [4, 6] and computational studies [7]. These investigations provided the theoretical background for the experimental results. This chapter is limited to describing the preparation of the group IVB metallocene dihalogenides, and focusses mainly on the most recent work (1998–2002) [8] in this field. 6.2.1

Singly-Carbon-Bridged Group IVB Metallocenes

Several strategies were utilized to generate the ligand systems for complexes incorporating one or two carbon atoms in their bridging moiety (Scheme 6.2) [9]. Recently, a quite frequent synthetic approach was used to generate metallocene 1 [10] with a single carbon bridge. Treatment of indenyl lithium with 6,6-dimethylfulvene generated the corresponding ligand core which was converted into the zirconocene 1 by deprotonation and metallation with zirconium tetrachloride. In order to tailor the framework of zirconocene 2 and titanocene 3 [11], Halterman used the palladium-mediated cross-coupling of 1,2-diiodobenzene with cyclopentadienyl lithium in the presence of ZnCl2. Frequently, the complexes were tethered at the cyclopentadienyl units of indenyl- or fluorenyl-based metallocenes (4) [12]. In addition, group IVB metallocenes were generated which were bridged at the benzene rings of the indenyl moiety (5–6) [13] as presented in Scheme 6.2. Furthermore, novel syntheses of ansa-metallocenes with long aliphatic bridging chains (9, 12–14) have been described in recent publications [14, 15]. Starting from titanocenes (7) with pendant alkenyl groups attached to the cyclopentadienyl

M = Zr 2 M = Ti 3 Scheme 6.2

M = Zr 5 M = Ti 6

6.2 Singly-Bridged Group IVB Metallocenes

Scheme 6.3

Scheme 6.4

ligands, bent metallocenes (9) were generated by magnesium-mediated reduction and opening of the corresponding titanacycles (8) with HCl (see Scheme 6.3) [14]. In an independent approach [15], the alkenyl groups in 10 underwent intramolecular olefin metathesis [16] yielding the alkenyl-bridged zirconocene 11, and after hydrogenation the corresponding ansa-metallocene with aliphatic bridging units. Additionally, the ligand framework was also demonstrated to be generated by treatment of an a,x-bis(methanesulfonyl chloride) with cyclopentadienyl or indenyl lithium [17] yielding metallocenes 12–14 (Scheme 6.4). 6.2.2

Singly-Silicon-Bridged Group IVB Metallocenes

Several silicon-bridged metallocenes of zirconium, titanium and hafnium have been synthesized [18]. Recently, ansa-metallocenes with the cyclopentadiene moiety annulated to pyrroles (15) and thiophenes (16) have been examined [19]. The silicon bridge was introduced by deprotonation of cyclopenteno[b]pyrrole or cyclopenteno[b]thiophene with BuLi and treatment with Me2SiCl2 to form the ligand

161

162

6 Endohedral Metal Complexes of Cyclophanes

Scheme 6.5

precursor. Metallation by standard procedure (BuLi, ZrCl4) formed zirconium dichlorides 15 and 16. Adopting a similar protocol created the Ph2Si- or (Me2Si)2linked metallocenophanes 17 [20] and 18 [21] shown in Scheme 6.5. 6.2.3

Singly-Boron- and Phosphorous-Bridged Group IVB Metallocenes

The novel class of boron- (19–20) [22] and phosphorous-bridged (21–25) [23] ansametallocenes has gained great interest among researchers. Even though the boron- and phosphorous-tethered complexes are not usually very sensitive to moisture and can be handled easily [22 a], the synthetic techniques to generate the highly reactive ligand precursors and to metallate them in good yields have only been developed recently. The boron-bridged ansa-zirconocene 19 was isolated by treatment of bis(trimethylstannyl)cyclopentadiene with PhBCl2 followed by zirconium tetrachloride [22 a]. In an analogous way the ansa system of 20 was prepared [22 b]. Similarly, metallocenes with phosphorous in the bridging moiety were generated by treatment of cyclopentadienyl or fluorenyl lithium with PhPCl2 to yield the corresponding zirconocenes 21–25 [23 b] presented in Scheme 6.6.

Scheme 6.6

6.3 Doubly-Bridged Group IVB Metallocenes

6.3

Doubly-Bridged Group IVB Metallocenes

In order to create conformationally well-defined complexes with beneficial dihedral angles promoting the efficiency and activity of metallocene-based catalysts for the olefin polymerization, the approaches to doubly-bridged zirconocenes, titanocenes and hafnocenes have been intensively studied. 6.3.1

Doubly-Carbon-Bridged Group IVB Metallocenes

The synthesis of achiral carbon-bridged [22]metallocenophanes of zirconium and titanium was anticipated in 1994. Starting from ethano-bridged bis(cyclopentadiene), 26, the tricyclic ligand framework was synthesized [24] by deprotonation with BuLi and cyclizing condensation with glyoxal sulfate to form 27. After reduction and aqueous workup the cyclopentadiene 28 was reacted with ZrCl4 to yield the [22]zirconocenophane 29 as outlined in Scheme 6.7. The major drawback of this approach was the very low overall yield (2%) of 29. An independent strategy [25] started from 1,2,5,6-tetramethylenecyclooctane (30) which was treated with tBuOK and bromoform to yield a mixture of dibromocyclopropanes 31 a and 31 b. The ligand framework was created by the Skattebøl rearrangement [26] induced by MeLi. The doubly-bridged bis(cyclopentadiene) 32 was deprotonated and converted into the corresponding zirconocene 29 and titanocene complexes 33 (see Scheme 6.8) with an overall yield of 13% and 20%, respectively. Utilizing a similar strategy, also the chiral [22]titanocenophane 34 [27] was synthesized starting from a chiral bicyclic diketone. In order to synthesize doubly-carbon-bridged indenyl-based group IVB complexes (39–40) [28] the Nazarov cyclization [29] was applied. Starting from 1,5-di-

f) ZrCl4 Scheme 6.7

163

164

6 Endohedral Metal Complexes of Cyclophanes

Scheme 6.8

methyl-2,6-cyclooctadiene (35) the diketone 36 was synthesized in three steps as presented in Scheme 6.9. Treatment with polyphosphoric acid (PPA) at 120 8C for 3 h yielded the Nazarov cyclization product 37 in 54% yield. The straight forward

a) SeO2, TBHP; b) PhMgBr; c) Swern oxidation; d) PPA; e) LiAlH4; f) p-TsOH; g) n-BuLi; h) TiCl3, [O] or Zr(NMe2)4, HNMe2 · HCl Scheme 6.9

6.3 Doubly-Bridged Group IVB Metallocenes

synthesis of the doubly-bridged indene ligand precursor 38 was accomplished by a reduction and dehydratization strategy. Metallation [30] yielded in both cases the racemic bis(indenyl) metal dichloride (39, 40) as the only diastereomers, since the formation of the meso-isomer was prevented through the constraints imposed by the bridging moiety. 6.3.2

Doubly-Silicon-Bridged Group IVB Metallocenes

Recently, the synthesis of several differently substituted bridged metallocenes with two bridging Me2Si-units was reported [31, 32]. The two adjacent silicon bridges had a great effect on the opening of the metallocene wedge and led to the formation of highly rigid complexes. The usual protocol to synthesize the doubly-tethered ligand framework utilized an iterative deprotonation of a substituted cyclopentadiene (41) [31 c] followed by treatment with Me2SiCl2. This first resulted in the formation of a singly-bridged bis(cyclopentadiene) (42) and finally yielded a doubly-tethered ligand (43) as presented in Scheme 6.10. The synthesis of the corresponding doubly-silicon-bridged metallocenophanes was achieved by standard methods generating the differently substituted complexes 44–50 [31] shown in Scheme 6.11. Additionally, synthesis of the twofold (SiMe2-O-SiMe2)-bridged [23]zirconocenophane 54 and [23]hafnocenophane 55 has been reported [32]. The singly-tethered bis(cyclopentadiene) 51 was converted into the corresponding C2v-symmetric metallocenes 54 and 55 with an overall yield of 40% and 30%, respectively, as presented in Scheme 6.12.

Scheme 6.10

165

166

6 Endohedral Metal Complexes of Cyclophanes

Scheme 6.11

a) O(SiMe2Cl)2; b) t-BuLi; c) SnMe3Cl, MCl4 Scheme 6.12

6.3.3

Structural Features of Doubly-Bridged Group IVB Metallocenes

Tailoring of the substitution pattern in twofold tethered group IVB metallocenes has a great effect on the geometry and structural parameters such as the dihedral angle a and the interplanar angle b (for definition see Fig. 6.1) of the two cyclopentadienyl ligands of the corresponding complexes. A survey of the reported geometrical data obtained by X-ray analysis is presented in Tab. 6.1.

6.3 Doubly-Bridged Group IVB Metallocenes Fig. 6.1

Definition of a, b, and c (Tab. 6.1) of bridged metallo-

cenes

Tab. 6.1 Structural parameters in doubly-bridged zirconocenes and titanocenes

Compound

a

b

c

Ref.

29 (Zr) 33 (Ti) 34 (Ti) 39 (Zr) 44 (Zr) 45 (Zr) 46 (Zr) 47 (Ti) 49 g (Zr) 50 (Zr) 54 (Zr)

120.0 124.5 124.2 120.2 121.8 122.5 120.6 126.0 122 128.1 137.3

62.5 57.8 – – 72.1 72.9 69.6 64.4 – 69.7 44.1

97.8 (Cl) 94.8 (Cl) 95.4 (Br) 93.9 (Cl) 98.7 (Cl) 99.7 (Cl) 99.6 (Cl) 96.7 (Cl) 105 (Cl) – 96.2 (Cl)

[24] [26] [26] [27] [30 a] [30 a] [30 b] [30 b] [30 e] [30 d] [31]

The dihedral angles a in doubly-ethano-bridged zirconocenophanes (29, 39) revealed an average value of about 1208, whereas a was found to be about 1228 in doubly-SiMe2-bridged derivatives (44–46, 49 g). The angle increased even further when sterically demanding substituents were introduced at the cyclopentadienyl rings, and this does not favor the bent conformation (50) [31 d]. [22]Titanocenophanes exhibited an average dihedral angle of 1248 in carbon-bridged (33, 34) and with slightly larger values of about 1268 in silicon-bridged derivatives (47). The c angles defined by the central atom and its two halogen ligands (see Fig. 6.1) varied in the range of 97 ± 3 8. The interplanar angle b of the two cyclopentadienyl rings was found to be about 718 in twofold Me2Si-bridged zirconocenes (44–46, 50), however, this angle decreased to about 628 in the doubly-ethano-bridged [22]zirconocenophane 29. A similar trend was discovered for the corresponding titanium derivatives where the b angle was revealed to be about 648 in doublySiMe2-bridged 47 and about 588 in twofold C2H4-bridged 33. The X-ray structure of the zirconocene 54 containing two adjacent disiloxane bridges exhibited an even smaller interplanar angle of about 448 due to its further enlarged tethering moiety. In general, the introduction of a second bridging unit at the ligand core of the group IV B complexes led to an increase of the interplanar angle b in comparison to the corresponding singly-tethered systems [4, 33, 34]. Only in the case of 54 this angle did decrease, by about 78 [35], since the two bulky disiloxane units led to an enlargement of the Cp–Zr distances.

167

168

6 Endohedral Metal Complexes of Cyclophanes

6.4

Endohedral Group VIB and VIIIB Metal p-Complexes 6.4.1

Bridged Bis[benzene]chromium Complexes

The first example of an endohedral cyclophane with chromium in the center was published in 1978, namely (g12-[2.2]paracyclophane)chromium(0) (57) [36]. The preparation of 57 was achieved by cocondensation of the ligand with the metal (Scheme 6.13). The prerequisites for this procedure were a thermally stable ligand and the fact that both the ligand and the metal could be vaporized. Using [2.2]paracyclophane (56) as ligand the yield of 57 was 5%. One year later the congener (g12-[3.3]paracyclophane)chromium(0) (59) was prepared (Scheme 6.14) via the same route [37], with purification of 59 achieved by conversion to the radical cation. The proposed structure of 59 was confirmed by investigating the triiodide of the radical cation 59+· [38]. In the solid state the bridges of 59+· adopted exclusively the chair conformation, whereas in solution a 1 : 1 ratio between chair and boat conformation of 59 (Scheme 6.14) was found by NMR [38]. The benzene rings revealed an eclipsed conformation where the transannular distance between the centers of the rings amounted to 3.22 Å. The silabridged bis[benzene]chromium 62 was synthesized from 60 in a stepwise fashion as shown in Scheme 6.15 [39]. This approach was also preferred in

Scheme 6.13

Scheme 6.14

6.4 Endohedral Group VIB and VIIIB Metal p-Complexes

Scheme 6.15

view of the low availability of 1,1,2,2,9,9,10,10-octamethyl-1,2,9,10-tetrasila[2.2]paracyclophane [40]. Starting from 60 the bis[benzene]chromium derivative 61 was synthesized by metal atom-ligand cocondensation. Treating 61 with lithium in naphthalene yielded the desired cyclophane 62 [39]. The structure of 62 was studied in detail by means of X-ray diffraction [39]. The distance between the planes of the rings was found to be 3.24 Å. All three complexes, 57, 59 and 62, are thermally stable solids. The metalla-bridged heterocyclophane 65 was also prepared in a stepwise approach analogous to 62 [41]. A nucleophilic substitution of the chromocene derivative 63 with LiP(C6H5)2 led to 64 which reacted with Ni(CO)4 to yield 65 (Scheme 6.16). Electrochemical studies revealed that the oxidation is only reversible for the first step (65 ? 65+·). The reduction to 65 occurred in a quasi-reversible way [41].

Scheme 6.16

169

170

6 Endohedral Metal Complexes of Cyclophanes

6.4.2

Bridged Metallocenophanes of Group VIIIB Metals

Many endohedral ferrocenophanes have been reported in the literature [42], therefore, chapter 5 of this book is devoted to the syntheses and properties of those species. In this chapter we restrict ourselves to ruthenocenes (see also chapters 14 and 16). The synthesis of ruthenocenophanes closely followed the route which was used by Schlögl et al. to synthesize ferrocenophanes [43]. In Scheme 6.17 the syntheses of two twofold bridged ruthenocenes are summarized [44]. The bridged species 66 a and 66 b were the starting points. They were first converted by formylation, Knoevenagel reaction and hydrogenation reactions into 67 a and 67 b. The second chain was then attached to the Cp ligand by intramolecular Friedel-Crafts acylation to yield 68 a and 68 b. From these key compounds 70 a and 70 b followed by reduction of the keto-group with LiAlH4/AlCl3 [44]. The ring expansion of 68 a and 68 b with diazomethane led to 69 a and 69 b from which 71 a and 71 b could be reached. Since the metallocene is rather stable, the overall yield in this multi-step sequence was 18% [44]. Single bridged [n]ruthenocenophanes have also been reported in the literature [45–47]. Their preparation proceeded either by starting with two tethered cyclopentadiene anions (72, Scheme 6.18) or with dilithiated ruthenocene, 75.

Scheme 6.17

6.4 Endohedral Group VIB and VIIIB Metal p-Complexes

Scheme 6.18

Both routes could easily be extended to derivatives substituted on the ring and chain as shown for 78–81 in Scheme 6.19 [45–47]. The disilane-bridged ferrocenophanes and ruthenophanes will be treated in chapter 16 of this book.

Scheme 6.19

To reach an 18-valence electron configuration for an endohedral Co(I)- or Rh(I)phane the ligands of choice should be a cyclophane containing a cyclopentadiene anion (or another 6e-equivalent) and a cyclobutadiene (or another 4e-equivalent). The problem of preparing such a phane as a ligand is the instability of the cyclic 4p system. This problem was circumvented by in situ generation of the cyclobutadiene moiety [48]. The feasibility of this concept was demonstrated by the reaction sequence shown in Scheme 6.20. The cobalt complexes with the pendant alkyne groups 83 a–c were synthesized by reaction of the corresponding cyclopentadiene complexes 82 a–c, first with nbutyl lithium to generate the corresponding cyclopentadiene anions and then with Co2(CO)8 and iodine (Scheme 6.20). By heating the complexes 83 a–c at 160 8C in cyclooctane the endohedral cobalt phanes 84 a–c were generated. In the case of 82 c (n = 5) two products were isolated, 84 c and 85. In the latter system the two tert-butyl groups are situated adjacent to each other on the four-membered ring. The parent system with propano bridges was also prepared by heating the bis(trimethylsilyl) congener of 83 a, 86. As a result 87 was generated (Scheme

171

172

6 Endohedral Metal Complexes of Cyclophanes a, b

84 a–c

Scheme 6.20

6.21) [49]. The trimethylsilyl groups from 87 were then removed by reaction with tetramethylammonium fluoride to generate 88. For 84 a and 84 b detailed structure determination was possible on the basis of X-ray diffraction studies on single crystals. These investigations revealed different distances between the two p-ligands in 84 a (3.24 Å) and 84 b (3.36 Å) (Fig. 6.2).

Scheme 6.21

6.4 Endohedral Group VIB and VIIIB Metal p-Complexes Fig. 6.2 Molecular structures of 84 a (top) and 84 b (bottom)

In 84 a the C(sp2)–C(sp3) bonds were bent towards the metal by 148 and 6.58, respectively. In the unbridged parent system (g5-cyclopentadienyl)(g4-cyclobutadien)cobalt (CpCoCb), the two p-ligands were separated by 3.34 Å [50]. When the complex 82 a with the pendant alkyne units was heated at 190 8C in decaline, only one cyclopentadienone complex (89 a or 90 a) was isolated in 10% yield. Treatment of 82 b under the same conditions gave an equimolar mixture of 89 b and 90 b in only 2% yield (Scheme 6.22) [51]. When the rhodium complex 91 was heated at 160 8C in cyclooctane a mixture of three diasteriomeric cyclopentadienone complexes was isolated in 10% overall yield (Scheme 6.23) [51]. The spectroscopic data showed the presence of one C1-symmetric complex, 92, and two CS-symmetric complexes 93 and 94 a or 94 b [51]. The structures of 92 and 93 were confirmed by X-ray diffraction studies (see Fig. 6.3).

173

174

6 Endohedral Metal Complexes of Cyclophanes

a) 190 8C, decaline Scheme 6.22

Scheme 6.23

Scheme 6.24

6.4 Endohedral Group VIB and VIIIB Metal p-Complexes Fig. 6.3 Molecular structures of 92 (top) and 93 (bottom)

The distances between the metal and the center of the Cp-ring amount to 1.78 Å (92) and 1.81 Å (93). The values obtained for the distance between the metal and the center of the cyclopentadienone ring were 1.74 Å (92) and 1.75 Å (93). A further example for the in situ generation of the rhodium stabilized cyclopentadienone ring by intramolecular [2+2+1]cycloaddition was achieved by warming the complex 95 at 50 8C for 8 h or by irradiation (Scheme 6.24) [52]. The structure of the air stable product 96 was confirmed by X-ray investigations.

175

176

6 Endohedral Metal Complexes of Cyclophanes

6.5

Cavities as Hosts for Cations of Group IB, IIIA and IVA Metals 6.5.1

Endohedral Silver Complexes with Cyclophanes

It has been known for quite some time that simple p-systems, such as olefins, alkynes and benzenes, form complexes with small metal cations [53]. For example complexes with Ag+ were quite stable if four double bonds, three triple bonds or three benzene rings were arranged in such a way as provided by [97 · Ag]+ [54], [98 · Ag]+ [55] or [99 · Ag]+ [56]. In the latter case the [2.2.2]paracyclophane was named “p-prismand” due to the rigid prismatic juxtaposition of the three benzene rings [56]. The [2.2.2](1,4)cyclophane (99) formed a 1 : 1 complex with silver triflate. The structure elucidation by X-ray diffraction on 99 · AgCF3SO3 [57] reveals that the silver ion was not situated in the center of the p-prismand but was situated close to the rim of the belt formed by the three benzene rings. The silver ion was bonded to one carbon double bond in each of the phenyl rings and to one oxygen of the triflate anion (Fig. 6.4). The silver atom was situated in an average distance of 2.51 Å from the center of the double bonds closest to the metal, furthermore, it was found to be external by 0.23 Å from the mean plane of the three closest double bonds (Fig. 6.4). As anticipated the structure reported for 99 · AgClO4 [58] was in close agreement to the triflate. In the AgClO4 complex the Ag+ shift from the plane containing the centers of the three nearest ring double bonds amounted to 0.24 Å. In [26](1,2,4,5)cyclophane (100) (deltaphane) [57] each of the three benzene rings was tied to the neighbors by four ethano bridges. This made 100 more rigid than 99. X-ray crystallographic investigations of single crystals of 100 and its silver triflate complex revealed that the geometry of the deltaphane moiety was essentially the same. The silver atom was situated on an average distance of 2.43 Å from the three nearest carbon atoms, and 0.23 Å from the plane of these three atoms on the external side away from the center of the deltaphane moiety. The Ag–C vectors were nearly perpendicular (858) to the benzene planes (see Scheme 6.27).

Scheme 6.25

6.5 Cavities as Hosts for Cations of Group IB, IIIA and IVA Metals

Fig. 6.4 Molecular structure of [99 · Ag]+[CF3SO3]–. The oxygen centers of the anion are hatched.

Boekelheide et al. also discovered an exchange of the silver ion according to Scheme 6.26 as suggested by NMR data [57]. Concentration studies showed that the exchange process shown in Scheme 6.26 is due to an intermolecular mechanism. The same holds for 99 · AgO3SCF3 [57]. The seminal work of Pierre et al. [56] and Boekelheide et al. [57] was extended by Rissanen and his group [59, 60] to the silver complexes of [2,2,2]-m,m,p-(101), [2.2.2]-p,p,m-(102), [2.2.1]-p,p,m-(103) and [2.2.1]-p,p,p-cyclophane (104) (see Scheme 6.27). The silver complexes of 101 and 102 were remarkably similar to that of the silver triflate of 99, although the cycles are smaller. The silver ion was sitting on one side of the belt in such a way that it interacted with one double bond of the three benzene rings in 101 and 102. The average distance of the silver ion from the center of the closest double bond of each benzene ring was found to be 2.51 Å for 101 and 2.59 Å for 102. For 103 the interaction between Ag+ and the cyclophane was accomplished by bonding with one of the double bonds in each of the two p-substituted phenyl rings and with one carbon atom of the m-substituted phenyl ring. The average distance of the silver ion from the center of the two closest double bonds was found

Scheme 6.26

177

178

6 Endohedral Metal Complexes of Cyclophanes

Scheme 6.27

to be 2.54 Å. In 104 the silver ion was coordinated to one double bond of each of the benzene rings. The average distance of the silver ion to the center of the three double bonds was found to be 2.50 Å. The coordination sphere in the complexes 101–104 was supplemented by one oxygen atom of the triflate anion with Ag–O bond lengths close to those of the Ag–C distance. The geometry around the silver ion could be described as a distorted tetrahedron. Quantum chemical calculations on 99 · Ag+ [61] and 101 · Ag+–104 · Ag+ [62] revealed that the bonds between Ag+ and the p-system were formed by r-donation and d-p* back-donation between the silver ion and the hydrocarbon skeleton resulting in a hexahapto (three times g2) overall p-bonding. One g2-bonding to one aromatic ring in the p-prismands related by strength to a single moderately strong hydrogen bond and varied between 25 and 50 kJ mol–1 [62]. Related to the silver complexes of 100–104 (Scheme 6.27) is that of 5,12-methano[2.2.2]paracyclophane (105) (Scheme 6.28), described by Hopf et al. [64]. X-ray

6.5 Cavities as Hosts for Cations of Group IB, IIIA and IVA Metals +

crystallographic analysis of 105 · Ag reveals three independent molecules in the unit cell. In two of them there is a direct contact between the anion (ClO–4) and silver, in a third one the silver atom is associated with one water molecule. It is found that the metal is bound somewhat closer to bonds C5–C6, C11–C12 and C24–C25 (see Scheme 6.28) than to the others. Further silver complexes were reported from (E)- and (Z)-2,3-bis(4'-[2.2.2]paracyclophanyl)but-2-ene (106, 107) [65, 66] and 1,4-bis(4'- [2.2.2]paracyclophanyl)buta1,3-diyne (108) [67]. The Ag


C interactions in 106–108 involve one C–C bond per ring as found in 99 · Ag+. Close contacts between the anion (SbF–6) and the metal (Ag


F = 2.545 Å) are observed in 108 · Ag+ [67]. Like in the other silver complexes with 99 as ligand, each metal atom in 108 · Ag+ is bonded to three pairs of C-atoms that correspond to one bond in each of the three rings (see Scheme 6.28). The distances between the centers of the C–C bonds and the silver ion vary between 2.5 Å and 2.7 Å. Furthermore, the (Z)-stilbene derivative 109 formed a 1 : 1 silver complex [67] in which the silver ion was located between the cleft formed by the two aromatic rings as shown in Scheme 6.29. The macrocycles 110 and 111 containing two (Z)stilbene units were available by McMurry coupling reaction [68]. The reaction of 110 and 111 with silver triflate led to the corresponding silver complexes 110 · Ag+ and 111 · Ag+. Single crystals of the silver complexes were not obtained, however, structural details were unraveled by NMR spectroscopy.

Scheme 6.28

179

180

6 Endohedral Metal Complexes of Cyclophanes

Scheme 6.29

The highest upfield shift in the 1H and 13C NMR spectra was found for the carbon and hydrogen atoms in the ortho position to the sp3 centers. This data pointed to a symmetrical geometry of 110 · Ag+ and 111 · Ag+ in which the metal ion was placed in the center of the cavity. From the temperature dependence of the NMR spectra of 111 · Ag+ a rapid hopping of the silver ion between two different binding sites of the belt-like host was concluded [68]. 1H NMR studies suggested that the hopping occurred between the (Z)-stilbene subunits and not between the two 7,7-diphenylnorbornane units. Further studies showed that the complex 111 · Ag+ is more stable by a factor of 102–104 than 110 · Ag+. 6.5.2

Silver Complexes with p-Prismands

The bicyclic systems 4,10,15-(1,4)tribenzena-1,7-diazabicyclo[5.5.5]heptadecaphane (112), 11,16-(1,4)dibenzena-1,8-diazabicyclo[6.5.5]octadecaphane-4-yne (113) and 17-(1,4)benzena-1,8-diazabicyclo[6.6.5]nonadecaphane-4,11-diyne (114) represent new types of prismands [69, 70]. Reaction with silver triflate yielded in all three cases the 1 : 1 complex in which the silver ion was situated inside the cavity. X-ray investigations of 112 · AgCF3SO3 showed that the silver ion did not reside exactly in the center of the cavity but was spread over two equivalent positions.

6.5 Cavities as Hosts for Cations of Group IB, IIIA and IVA Metals

Scheme 6.30

Each silver ion was displaced towards the bridgeheads. As a result the symmetry of 112 · Ag+ is reduced from C3h to C3 [69, 70]. In solution, however, NMR investigations revealed C3h symmetry. This data suggested a relatively fast intramolecular hopping process of the silver ion from one bridgehead position to the other. The proposed fluctuation of the silver ion could not be resolved down to –100 8C. Since the separation of the two silver positions in the crystal structure was small (1.05 Å) and the tunneling path is flanked by mobile p-systems on three sides, the activation energy of the intramolecular exchange process has been relatively low. For the 1 : 1 complexes between 113 and 114 with silver triflate, X-ray studies revealed that the metal ion is situated in the center between the two nitrogen atoms [70]. Endohedral metal complexes were also reported for 5-cylindrophane 115 [71]. In this molecule two benzene rings were connected with three 3-thiapentamethylene chains which provide a cavity with two parallel benzene rings and three sulfur centers [72]. Crystals of 115 · Cu+ and 115 · Ag+ with tetrafluoroborate as anion were obtained and investigated. It was shown [71] that the metal ions sit in the center and were coordinated by three sulfur atoms. NMR data indicated that the benzene rings are also coordinated to the metal [72]. 6.5.3

Group IIIA and IVA Complexes of Cyclophanes

The work by Schmidbaur et al. showed that Ga(I), In(I) and Tl(I) formed arene complexes in which these ions were g6-bonded to either one or two aromatic rings [73]. Examples included Ga(I) complexes with hexamethyl benzene [74] or benzene [75], In(I) and Tl(I) complexes with mesitylene [76, 77]. These results suggested that Ga(I) and its congeners In(I) and Tl(I) might also be complexed by cyclophanes and prismands. A solution of (C6H6)2Ga · GaBr4 reacted with [2.2.2]paracyclophane 99 to yield the complex [99 · Ga]+ [GaBr4]– [78]. A detailed structural analysis (Fig. 6.5) of this complex revealed that the gallium ion was shifted by 0.43 Å sideways from the center towards one Br center of the tetrabromogallate anion [78]. To achieve g6-bonding of all three benzene rings in 99, they were slightly tilted. Quantum chemical calculations on tris(g6-benzene)gallium(I) [79] assuming D6h symmetry revealed that the empty 5px and 5py (e') orbitals of the metal interacted stronger with the occupied p-orbitals (e') of the three ligands than the 5pz orbital (a'') due to better overlap (Fig. 6.6). The resulting 7 a'' orbital of the complex was

181

182

6 Endohedral Metal Complexes of Cyclophanes

Molecular structure of [99 · Ga]+[GaBr4]–. The gallium atom is hatched, the bromine atoms are dotted. Fig. 6.5

suited to interact with a neighboring ligand, such as the anion in [99 · Ga]+ [GaBr4]– or the nitrogens in [112 · Ga]+ (see Section 6.5.4). It is interesting to note that [2.2]paracyclophane and [3.3]paracyclophane yielded only exohedral complexes with Ga(I) [80]. The [2.2.2]paracyclophane cavity was also capable to host Ge(II) and Sn(II) complexes [81]. The reaction of 99 with SnCl2 and GeCl2 in presence of AlCl3 yielded the complexes [(99 · Sn)(AlCl4)]+AlCl–4 and (99 · GeCl)2(Al2O2Cl10). X-ray investigations on both complexes showed that the metal atom is placed inside the belt formed by the three benzene rings [81]. Analogous to the gallium (I) complex (99 · Ga)+ the Sn(II) center was shifted towards the opening of the ring in order to have close contact to a Cl atom of [(AlCl4)–, (Sn


Cl = 3.07 Å)] [81]. The second anion showed no close contact to the Sn center. In the case of the Ge(II) complex one Cl atom was strongly bound to Ge (Ge– Cl = 2.22 Å). The Ge center in (99 · GeCl)+ showed tetrahedral coordination. The distances of the Ge atom to the centers of the arene rings varied between 2.72 Å and 2.80 Å. They were found longer than the corresponding distances for [99 · Sn]+ (2.53 Å to 2.67 Å) [81]. 6.5.4

Group IIIA Complexes with p-Prismands

The p-prismand 112 provided an ideal cavity for group IIIA cations due to the three benzene rings and the two nitrogen donor atoms. The reaction of 112 with Ga2Cl4 in toluene at room temperature yielded [112 · Ga]+ [GaCl4]– [82]. X-ray investigations on single crystals of [112 · Ga]+ [GaCl4]– showed that the metal ion was located in the center of the cavity (Fig. 6.7) and the three benzene rings were oriented in such a way that they faced the metal ion to provide g6-coordination of

6.5 Cavities as Hosts for Cations of Group IB, IIIA and IVA Metals

Fig. 6.6

Frontier orbitals of (C6H6)3Ga+(D3h).

Fig. 6.7 Molecular structure of [112 · Ga]+[GaCl4]–. The chlorine atoms of the anion are hatched.

183

184

6 Endohedral Metal Complexes of Cyclophanes

the aromatic system. The distance between the metal and the center of the arene rings was determined to be 2.73 Å (mean value) which was somewhat longer than that of [99 · Ga]+ [GaBr4]– (2.64 Å). The distance between the nitrogen centers and the Ga+ amounted to 2.91 Å. From Fig. 6.7 it was evident that there is no interaction between the anion [GaCl4]– and the endohedral Ga+.

6.6

Concluding Remarks

Endohedral metal complexes with cyclophanes are expected for many more cyclophanes known so far. Therefore, we also expect endohedral species for those main group elements for which non-bridged metallocenes have been reported [83, 84]. For cyclophanes with large cavities we anticipate the inclusion of cations: one example is 2,7,9,14-tetraoxa-1,8(1,4)-dibenzenacyclo-tetradecaphane-4,11-diyne [85]. With respect to applications in further reactions only the singly- and doublybridged metallocenes of group IV B metals play a role as catalysts as well as reagents in synthesis. Much less attention has been devoted to the known representatives of the endohedral metal complexes of group VI B, VIII B, III A and IVA. This is probably due to their limited availability. Procedures for synthesizing larger quantities are therefore desirable.

6.7

Acknowledgments

We thank the Deutsche Forschungsgemeinschaft, the Fonds der Chemischen Industrie and the BASF AG, Ludwigshafen, for financial support. We are grateful to P. Kraemer for making the drawings. 6.8

References Reviews: a) U. T. Mueller-Westerhoff, Angew. Chem. 1986, 98, 700–716; Angew. Chem. Int. Ed. Engl. 1986, 25, 702–718; b) J. Schulz, F. Vögtle, Top. Curr. Chem. 1994, 172, 41–86. 2 a) P. C. Möhring, N. J. Coville, J. Organomet. Chem. 1994, 479, 1–29. b) H. H. Brintzinger, D. Fischer, R. Mülhaupt, B. Rieger, R. M. Waymouth, Angew. Chem. 1995, 107, 1255–1283; Angew. Chem. Int. Ed. Engl. 1995, 34, 1143–1170. c) R. H. Grubbs, G. W. Coates, Acc. Chem. Res. 1996, 29, 85–93. d) W. Kaminsky, J. Chem. Soc., Dalton Trans. 1998, 1413– 1

1418. e) J. C. Green, Chem. Soc. Rev. 1998, 27, 263–271. 3 a) E. Negishi, T. Takahashi, Synthesis 1988, 1–19. b) S. L. Buchwald, in Comprehensive Organometallic Chemistry II, E. W. Abel, F. G. A. Stone, G. Wilkinson (eds.), Pergamon, Elmsford, NY, Vol. 12, 771–784. c) A. H. Hoveyda, J. P. Morken, Angew. Chem. Int. Ed. Engl. 1996, 35, 1262–1284. d) Titanium and Zirconium in Organic Synthesis, V. Snieckus, I. Marek (eds.), VCH, Weinheim, 2002. 4 Review: C. E. Zachmanoglou, A. Docrat, B. M. Bridgewater, G. Parkin,

6.8 References

5

6

7

8

9

10

C. G. Brandow, J. E. Bercaw, C. N. Jardine, M. Lyall, J. C. Green, J. B. Keister, J. Am. Chem. Soc. 2002, 124, 9525– 9546. a) P. G. Gassman, D. W. Macomber, J. W. Hershberger, Organometallics 1983, 2, 1470–1472. b) P. G. Gassman, C. H. Winter, Organometallics 1991, 10, 1592–1598. c) P. G. Gassman, P. A. Deck, C. H. Winter, D. A. Dobbs, D. H. Cao, Organometallics 1992, 11, 959–960. For recent studies see: a) J. Langmaier, Z. Samec, V. Varga, M. Horácˇek, K. Mach, J. Organomet. Chem. 1999, 579, 348–355. b) J. Langmaier, Z. Samec, V. Varga, M. Horácˇek, R. Choukroun, K. Mach, J. Organomet. Chem. 1999, 584, 323–328. c) B. J. Rausch, R. Gleiter, F. Rominger, J. Organomet. Chem. 2002, 658, 242–250. a) L. Cavallo, P. Corradini, G. Guerra, L. Resconi, Organometallics 1996, 15, 2254–2263 (and references cited therein). b) G. V. Loukova, V. V. Strelets, J. Organomet. Chem. 2000, 606, 203–206. For reviews on earlier work see: a) R. Beckhaus, in Metallocenes, A. Togni, R. L. Halterman (eds.), VCH, Weinheim, 1998, 153–239. b) R. L. Halterman, in Metallocenes, A. Togni, R. L. Halterman (eds.), VCH, Weinheim, 1998, 455–544. For some recent work on carbon-bridged group IVB metallocenes see: a) I. E. Nifant’ev, P. V. Ivchenko, V. V. Bagrov, L. G. Kuz’mina, Organometallics 1998, 17, 4734–4738. b) H. G. Alt, M. Jung, W. Milius, J. Organomet. Chem. 1998, 558, 111–121. c) M. Ringwald, R. Stürmer, H. H. Brintzinger, J. Am. Chem. Soc. 1999, 121, 1524–1527. d) R. W. Baker, M. A. Foulkes, P. Turner, J. Chem. Soc., Dalton Trans. 2000, 431–433. e) C. J. Schaverien, R. Ernst, P. Schut, T. Dall’Occo, Organometallics 2001, 20, 3436–3452. f) S. A. Miller, J. E. Bercaw, Organometallics 2002, 21, 934–945. g) B. Y. Lee, Y. H. Kim, Y. C. Won, J. W. Han, W. H. Suh, I. S. Lee, Y. K. Chung, K. H. Song, Organometallics 2002, 21, 1500–1503. F. J. Gómez, R. M. Waymouth, Macromolecules 2002, 35, 3358–3368.

11

12

13

14

15

16

17

18

R. L. Halterman, A. Tretyakov, M. A. Khan, J. Organomet. Chem. 1998, 568, 41–51. a) U. Dietrich, M. Hackmann, B. Rieger, M. Klinga, M. Leskelä, J. Am. Chem. Soc. 1999, 121, 4348–4355. b) J. Kukral, P. Lehmus, T. Feifel, C. Troll, B. Rieger, Organometallics 2000, 19, 3767–3775. c) E. J. Thomas, M. D. Rausch, J. C. W. Chien, Organometallics 2000, 19, 4077–4083. a) R. L. Halterman, D. Combs, M. A. Khan, Organometallics 1998, 17, 3900– 3907. b) C. J. Schaverien, R. Ernst, P. Schut, W. M. Skiff, L. Resconi, E. Barbassa, D. Balboni, Y. A. Dubitsky, A. G. Orpen, P. Mercandelli, M. Moret, A. Sironi, J. Am. Chem. Soc. 1998, 120, 9945–9946. a) M. Horácˇek, P. Šteˇpnicˇka, R. Gyepes, I. Císarˇová, I. Tisˇlerová, J. Zemánek, J. Kubisˇta, K. Mach, Chem. Eur. J. 2000, 6, 2397–2408. b) M. Horácˇek, P. Šteˇpnicˇka, K. Fejfarová, R. Gyepes, I. Císarˇová, J. Kubisˇta, K. Mach, J. Organomet. Chem. 2002, 642, 148–155. c) L. Lukešová, P. Šteˇpnicˇka, K. Fejfarová, R. Gyepes, I. Císarˇová, M. Horácˇek, J. Kubisˇta, K. Mach, Organometallics 2002, 21, 2639– 2653. D. Hüerländer, N. Kleigrewe, G. Kehr, G. Erker, R. Fröhlich, Eur. J. Inorg. Chem. 2002, 2633–2642. a) P. Schwab, M. B. France, J. W. Ziller, R. H. Grubbs, Angew. Chem. 1995, 107, 2179–2181; Angew. Chem. Int. Ed. Engl. 1995, 34, 2039–2041. b) R. H. Grubbs, S. Chang, Tetrahedron 1998, 54, 4413–4450. a) G. Erker, C. Mollenkopf, M. Grehl, R. Fröhlich, C. Krüger, R. Noe, M. Riedel, Organometallics 1994, 13, 1950–1955. b) T. Jödicke, F. Menges, G. Kehr, G. Erker, U. Höweler, R. Fröhlich, Eur. J. Inorg. Chem. 2001, 2097–2106. For some recent work on silicon-bridged group IV B metallocenes see: a) M. Hüttenhofer, F. Schaper, H. H. Brintzinger, Angew. Chem. 1988, 110, 2378–2380. Angew. Chem. Int. Ed. 1998, 37, 2268– 2270. b) H. G. Alt, K. Föttinger, W. Milius, J. Organomet. Chem. 1998, 564, 109–114. c) O. W. Lofthus, C. Slebodnick, P. A. Deck, Organometallics 1999,

185

186

6 Endohedral Metal Complexes of Cyclophanes

19

20

21

22

23

18, 3702–3708. d) N. Suzuki, Y. Masubuchi, Y. Yamaguchi, T. Kase, T. K. Miyamoto, A. Horiuchi, T. Mise, Macromolecules 2000, 33, 754–759. e) X. Zhang, Q. Zhu, I. A. Guzei, R. F. Jordan, J. Am. Chem. Soc. 2000, 122, 8093–8094. f) P. Jutzi, C. Müller, B. Neumann, H. G. Stammler, J. Organomet. Chem. 2001, 625, 180–185. g) J. A. Pool, C. A. Bradley, P. J. Chirik, Organometallics 2002, 21, 1271–1277. a) J. A. Ewen, R. L. Jones, M. J. Elder, A. L. Rheingold, L. M. Liable-Sands, J. Am. Chem. Soc. 1998, 120, 10786–10787. b) J. A. Ewen, M. J. Elder, R. L. Jones, A. L. Rheingold, L. M. Liable-Sands, R. D. Sommer, J. Am. Chem. Soc. 2001, 123, 4763–4773. c) A. N. Ryabov, D. V. Gribkov, V. V. Izmer, A. Z. Voskoboynikov, Organometallics 2002, 21, 2842–2855. V. V. Izmer, A. Y. Agarkov, V. M. Nosova, L. G. Kuz’mina, J. A. K. Howard, I. P. Beletskaya, A. Z. Voskoboynikov, J. Chem. Soc., Dalton Trans. 2001, 1131–1136. K. H. Thiele, C. Schliessburg, K. Baumeister, K. Hassler, Z. Anorg. Allg. Chem. 1996, 622, 1806–1810. For some recent work on boron-bridged group IV B metallocenes see: a) K. A. Rufanov, V. V. Kotov, N. B. Kazennova, D. A. Lemenovskii, E. V. Avtomonov, J. Lorberth, J. Organomet. Chem. 1996, 525, 287–289. b) K. Rufanov, E. Avtomonov, N. Kazennova, V. Kotov, A. Khvorost, D. Lemenovskii, J. Lorberth, J. Organomet. Chem. 1997, 536/ 537, 361–373. c) D. S. Stelck, P. J. Shapiro, N. Basickes, A. L. Rheingold, Organometallics 1997, 16, 4546–4550. d) C. T. Burns, D. S. Stelck, P. J. Shapiro, A. Vij, K. Kunz, G. Kehr, T. Concolino, A. L. Rheingold, Organometallics 1999, 18, 5432–5434. e) H. Braunschweig, C. von Koblinski, M. Mamuti, U. Englert, R. Wang, Eur. J. Inorg. Chem. 1999, 1899–1904. f) M. T. Reetz, M. Willuhn, C. Psiorz, R. Goddard, J. Chem. Soc., Chem. Commun. 1999, 1105– 1106. g) S. J. Lancaster, M. Bochmann, Organometallics 2001, 20, 2093–2101. For some recent work on phosphorousbridged group IV B metallocenes see: a) G. K. Anderson, M. Lin, Organometallics

24

25

26 27

28

29 30 31

32

33

1988, 7, 2285–2288. b) H. G. Alt, M. Jung, J. Organomet. Chem. 1998, 568, 127–131. c) N. Leyser, K. Schmidt, H. H. Brintzinger, Organometallics 1998, 17, 2155–2161. d) J. H. Shin, T. Hascall, G. Parkin, Organometallics 1999, 18, 6–9. e) J. H. Shin, B. M. Bridgewater, G. Parkin, Organometallics 2000, 19, 5155–5159. a) K. Hafner, C. Mink, H. J. Lindner, Angew. Chem. 1994, 106, 1566–1568; Angew. Chem. Int. Ed. Engl. 1994, 33, 1479– 1480. b) C. Mink, K. Hafner, Tetrahedron Lett. 1994, 35, 4087–4090. B. Dorer, M. H. Prosenc, U. Rief, H. H. Brintzinger, Organometallics 1994, 13, 3868–3872. L. Skattebøl, Tetrahedron 1967, 23, 1107–1117. R. B. Grossman, J.-C. Tsai, W. M. Davis, A. Gutiérrez, S. L. Buchwald, Organometallics 1994, 13, 3892–3896. R. L. Halterman, A. Tretyakov, D. Combs, J. Chang, M. A. Khan, Organometallics 1997, 16, 3333–3339. K. L. Habermas, S. E. Denmark, T. K. Jones, Org. React. 1994, 45, 1–158. G. M. Diamond, S. Rodewald, R. F. Jordan, Organometallics 1995, 14, 5–7. a) W. Mengele, J. Diebold, C. Troll, W. Röll, H. H. Brintzinger, Organometallics 1993, 12, 1931–1935. b) A. Cano, T. Cuenca, P. Gómez-Sal, B. Royo, P. Royo, Organometallics 1994, 13, 1688–1694. c) T. A. Herzog, D. L. Zubris, J. E. Bercaw, J. Am. Chem. Soc. 1996, 118, 11988–11989. d) S. Miyake, L. M. Henling, J. E. Bercaw, Organometallics 1998, 17, 5528–5533. e) D. Veghini, L. M. Henling, T. J. Burkhardt, J. E. Bercaw, J. Am. Chem. Soc. 1999, 121, 564–573. f) T. J. Peckham, P. Nguyen, S. C. Bourke, Q. Wang, D. G. Harrison, P. Zoricak, C. Russel, L. M. Liable-Sands, A. L. Rheingold, A. J. Lough, I. Manners, Organometallics 2001, 20, 3035–3043. J. Jung, S. K. Noh, D.-H. Lee, S. K. Park, H. Kim, J. Organomet. Chem. 2000, 595, 147–152. a) H. Wiesenfeldt, A. Reinmuth, E. Barsties, K. Evertz, H. H. Brintzinger, J. Organomet. Chem. 1989, 369, 359–370. b) B. Dorer, J. Diebold, O. Weyand,

6.8 References

34

35

36

37

38

39

40

41

42 43 44 45

46 47 48

49

H. H. Brintzinger, J. Organomet. Chem. 1992, 427, 245–255. a) J. A. Smith, J. von Seyerl, G. Huttner, H. H. Brintzinger, J. Organomet. Chem. 1979, 173, 175–185. b) C. S. Bajgur, W. Tikkanen, J.L. Peterson, Inorg. Chem. 1985, 24, 2539–2546. a) M. D. Curtis, J. J. D’Errico, P. S. Epstein, D. N. Duffy, L. G. Bell, Organometallics 1983, 2, 1808–1814. b) S. Ciruelos, T. Cuenca, P. Gómez-Sal, A. Manzanero, P. Royo, Organometallics 1995, 14, 177–185. C. Elschenbroich, R. Möckel, U. Zenneck, Angew. Chem. 1978, 90, 560–561; Angew. Chem. Int. Ed. Engl. 1978, 17, 531–532. A. R. Koray, M. L. Ziegler, N. E. Blank, M. W. Haenel, Tetrahedron Lett. 1979, 26, 2465–2466. R. Benn, N. E. Blank, M. W. Haenel, J. Klein, A. R. Koray, K. Weidenhammer, M. L. Ziegler, Angew. Chem. 1980, 92, 45–46; Angew. Chem. Int. Ed. Engl. 1980, 19, 44–45. C. Elschenbroich, J. Hurley, W. Massa, G. Baum, Angew. Chem. 1988, 100, 727–729; Angew. Chem. Int. Ed. Engl. 1988, 27, 684–686. H. Sakurai, S. Hoshi, A. Kamiya, A. Hosomi, C. Kabuto, Chem. Lett. 1986, 1781–1784. C. Elschenbroich, J. Sebbach, B. Metz, G. Heikenfeld, J. Organomet. Chem. 1992, 426, 173–185. R. W. Heo, T. R. Lee, J. Organomet. Chem. 1999, 578, 31–42 and references therein. K. Schlögl, H. Seiler, Tetrahedron Lett. 1960, 7, 4–8. S. Kamiyama, A. Kasahara, Bull. Chem. Soc. Jpn. 1984, 57, 719–724. J. M. Nelson,. A. J. Lough, I. Manners, Angew. Chem. 1994, 106, 1019–1021; Angew. Chem. Int. Ed. Engl. 1994, 33, 989–991. J. M. Nelson, A. J. Lough, I. Manners, Organometallics 1994, 13, 3703–3710. M. Herberhold, T. Bärtl, Z. Naturforsch. 1995, 50 b, 1692–1698. G. Scholz, R. Gleiter, F. Rominger, Angew. Chem. 2001, 113, 2559–2561; Angew. Chem. Int. Ed. 2001, 40, 2477–2479. G. Scholz, R. Gleiter, F. Rominger, to be published.

50 51 52

53

54

55

56

57

58

59

60 61

62 63

64

65

66 67 68

P. E. Riley, R. E. Davis, J. Organomet. Chem. 1976, 113, 157– 166. G. Scholz, C. Schaefer, F. Rominger, R. Gleiter, Org. Lett. 2002, 4, 2889–2892. M. V. Baker, D. H. Brown, B. W. Skelton, A. H. White, J. Chem. Soc., Dalton Trans. 2002, 2595–2596. H. W. Quinn, J. H. Tsai, Adv. Inorg. Radiochem. 1969, 12, 217–273; M. Herberhold, Metal p-Complexes. Elsevier, Amsterdam 1972; Gmelins Handbook of Inorganic Chemistry, Silver, Vol. 5 b, 8th edn. Springer, Heidelberg 1975. J. E. McMurry, G. J. Haley, J. R. Matz, J. C. Clardy, J. Mitchell, J. Am. Chem. Soc. 1986, 108, 515–516. J. D. Ferrara, A. Djebli, C. TessierYoungs, W. J. Youngs, J. Am. Chem. Soc. 1988, 110, 647–649. J.-L. Pierre, P. Baret, P. Chautemps, M. Armand, J. Am. Chem. Soc. 1981, 103, 2986–2988. H. C. Kang, A. W. Hanson, B. Eaton, V. Boekelheide, J. Am. Chem. Soc. 1985, 107, 1979–1985. C. Cohen-Addad, P. Baret, P. Chautemps, J.-L. Pierre, Acta Cryst. 1983, C39, 1346–1349. T. Lahtinen, E. Wegelins, K. Airola, E. Kolehmainen, K. Rissanen, J. Prakt. Chem. 1999, 341, 237–244. T. Lahtinen, E. Wegelius, K. Rissanen, New J. Chem. 2001, 25, 905–911. P. Saarenketo, R. Suontamo, T. Jödicke, K. Rissanen, Organometallics 2000, 19, 2346–2353. P. Saarenketo, R. Suontamo, K. Rissanen, Organometallics 2002, 21, 5473–5485. F. R. Heirtzler, H. Hopf, P. G. Jones, P. Bubenitschek, V. Lehne, J. Org. Chem. 1993, 58, 2781–2784. F. R. Heirtzler, H. Hopf, P. G. Jones, P. Bubenitschek, Tetrahedron Lett. 1995, 36, 1239–1242. P. G. Jones, P. Bubenitschek, S. Heirtzler, et al., Acta Cryst. 1996, C52, 1380–1384. P. G. Jones, F. Heirtzler, H. Hopf, Acta Cryst. 1996, C52, 1384–1388. J. E. Gano, G. Subramaniam, R. Birnbaum, J. Org. Chem. 1990, 55, 4760–4763. A. Garcia Martinez, J. Osio Barcina, M. del Rosario Colorado Heras, A. de

187

188

6 Endohedral Metal Complexes of Cyclophanes

69 70 71

72

73

74

75

76

Fresno Cerezo, M. del Rosario Torres Salvador, Chem. Eur. J. 2003, 9, 1157– 1165. A. Kunze, S. Bethke, R. Gleiter, F. Rominger, Org. Lett. 2000, 2, 609–612. A. Kunze, R. Gleiter, F. Rominger, to be published. M. Mascal, J.-L. Kerdelhué, A. J. Blake, P. A. Cooke, Angew. Chem. 1999, 111, 2094–2096; Angew. Chem. Int. Ed. 1999, 1968–1971. a) M. Mascal, J. Hansen, A. J. Blake, W. S. Li, Chem. Commun. 1998, 355–356; b) J. Hansen, A. J. Blake, W.-S. Li, M. Mascal, J. Chem. Soc. Perkin Trans. 1998, 3371–3376. H. Schmidbaur, Angew. Chem. 1985, 97, 893–904; Angew. Chem. Int. Ed. Engl. 1985, 893–904. V. Thewalt, T. Zafiropoulos, H. Schmidbaur, Z. Naturforsch. 1984, 39 b, 1642–1646. M. Uson-Finkenzeller, W. Bublak, B. Huber, G. Müller, H. Schmidbaur, Z. Naturforsch. 1986, 41 b, 346–350. J. Ebenhöch, G. Müller, J. Riede, H. Schmidbaur, Angew. Chem. 1984, 96, 367–368; Angew. Chem. Int. Ed. Engl. 1984, 23, 386–387.

77

78

79 80

81

82 83 84 85

H. Schmidbaur, W. Bublak, J. Riede, G. Müller, Angew. Chem. 1985, 97, 402–403; Angew. Chem. Int. Ed. Engl. 1985, 24, 414– 415. H. Schmidbaur, R. Hager, B. Huber, G. Müller, Angew. Chem. 1987, 99, 354– 356; Angew. Chem. Int. Ed. Engl. 1987, 26, 338–339. A. Kunze, R. Gleiter, S. Bethke, unpublished results. H. Schmidbaur, W. Bublak, B. Huber, G. Müller, Organometallics 1986, 5, 1647–1651; H. Schmidbaur, W. Bublak, B. Huber, G. Müller, Helv. Chim. Acta 1986, 69, 1742–1747. T. Probst, O. Steigelmann, J. Riede, H. Schmidbaur, Angew. Chem. 1990, 102, 1471–1473; Angew. Chem. Int. Ed. Engl. 1990, 29, 1397–1398. A. Kunze, R. Gleiter, S. Bethke, F. Rominger, to be published. S. Harder, Coord. Chem. Rev. 1998, 176, 17–66. P. Jutzi, N. Burford, Chem. Rev. 1999, 99, 969–990. R. Behm, C. Gloeckner, M. A. Grayson, M. L. Gross, G. W. Gokel, Chem. Comm. 2000, 2377–2378.

211

8

Reactive Intermediates from Cyclophanes Wolfram Sander

Two basic types of reactive intermediates are formed from cyclophanes: (1) reactive centers (radicals, carbocations etc.) are formed in the aromatic core or the aliphatic bridges with retaining the cyclophane system or (2) ring-opening of strained cyclophanes results in diradicals or quinoid systems. Reactive intermediates of type (1) are not specific for cyclophanes, the reactive centers generated behave similarly to those in other aromatic or aliphatic compounds. An exception are cyclophanes with donor and acceptor functions, where electron transfer leads to a charge-separated system where the counterions are held in proximity by the cyclophane molecule. Electron transfer processes within cyclophanes will be described elsewhere in this book (Chapter 14). Type (2) intermediates are specific for cyclophanes. Thermolysis or photolysis of strained cyclophanes leads to the homolysis of one of the aliphatic bridges. The release of strain during ring-opening of the cyclophanes allows diradicals or quinoid compounds to be obtained which are not easily accessible by other methods. The systems which have been studied most intensely are [2.2]cyclophanes which balance the strain energy stored in these molecules with ease of synthetic availability. The focus of this chapter is thus the thermal or photochemical ring-opening of [2.2]cyclophanes and the characterization by spectroscopical and chemical methods of the intermediates obtained in these reactions. Methods to identify these reactive intermediates are classical trapping experiments in solution, kinetic studies both in solution and in the gas phase, low temperature spectroscopy in organic glasses and inert gas matrices, and time-resolved spectroscopy in solution. In combination with high-level ab initio calculations these studies provide a detailed picture of the bond cleavage in cyclophanes.

8.1

Thermolysis of [2.2]Paracyclophanes

[2.2]Paracyclophane 1 a is the cyclophane which has been studied most intensely and for which most physical data are available. The heat of formation of 1 a was determined as DH8f (g) = 58.8 ± 0.8 kcal mol–1 [1], and by comparison with Benson’s strain free increments [2] a strain energy of 30.1 kcal mol–1 is obtained [3]. Modern Cyclophane Chemistry. Edited by Rolf Gleiter, Henning Hopf Copyright © 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 3-527-30713-3

212

8 Reactive Intermediates from Cyclophanes

Scheme 8.1

The first synthesis of paracyclophane 1 a was achieved by Brown and Farthing in 1949 by gas phase pyrolysis of p-xylene 2 at temperatures above 750 8C (Scheme 8.1) [4]. The relation between 1 a and its precursor p-xylylene 3 a is already shown in the title of this early publication: “Preparation and structure of dip-xylylene”. Two years previously, Szwarc had shown that under similar conditions p-xylylene 3 a is produced as a biradical that is stable in the gas phase but in the condensed phase rapidly produces an insoluble polymer [5]. The synthesis of paracyclophane 1 a from xylene 2 requires the loss of two hydrogen atoms from each molecule of 2 via cleavage of benzylic CH bonds. It was speculated that either p-xylylene 3 a or the diradical 4 a are intermediates along the pathway to 1 a (Scheme 8.1). On the other hand, thermolysis of 1 a results in the formation of 3 a, and at temperatures above 200 8C 1 a and 3 a are in equilibrium [6]. This indicates that 3 a is indeed an intermediate in the synthesis of 1 a from 2. The reported heat of formation of 3 a varies between 50 kcal mol–1 [7] and 58 kcal mol–1 [6], but in any case the cleavage of 1 a to 3 a is highly endothermic (41– 49 kcal mol–1) despite the release of the strain energy of the cyclophane. The reverse reaction, formation of 1 a from 3 a, is thus exothermic, however, the cyclization has to compete with the entropically more favorable oligo- or polymerization. For p-xylylene 3 a both a diradicaloid and a quinoid (quinodimethane) resonance structure can be formulated to explain its high reactivity. Although the ground state of 3 a is singlet, the molecule partially shows the behavior of a diradical. At –80 8C in THF solution 3 a is stable enough to record the NMR spectrum [8, 9], however, at room temperature it oligomerizes, polymerizes, cyclizes back to 1 a, or is trapped by a solvent or substrate molecule. Thus, 1 a can be used as a stable storage form that on flash vacuum pyrolysis releases p-xylylene 3 a for synthetic applications. The polyparaxylylenes (parylenes) that form on polymerization of 3 a and its derivatives are of interest due to their excellent thermal stability and elec-

8.1 Thermolysis of [2.2]Paracyclophanes

Scheme 8.2

tric properties. Therefore p-xylylenes 3 have been used for the vapor deposition (CVD) of polymeric coatings [10–13]. The flash vacuum thermolysis (FVP) of 1 a with subsequent trapping of the products in solid argon at cryogenic temperatures has been used by several authors for the spectroscopic characterization of 3 a [14–16]. While FVP at 600 8C produces 3 a in a clean reaction [15], at higher temperatures (930 8C) a complex mixture of further products such as styrene (main product), p-xylene 2, toluene, benzene, and benzocyclobutene are formed [14]. Obviously, these processes require intermolecular reactions including the transfer of hydrogen atoms and methyl or methylene groups. A key step in this reaction sequence is the rearrangement of 3 a to p-tolylmethylene 5 (Scheme 8.2). A very clean reaction is the FVP (flash vacuum pyrolysis) of octafluoro[2.2]paracyclophane 1 b [17, 18] which quantitatively yields tetrafluoro-p-xylylene 3 b [19]. The formation of 3 b was followed both by mass spectrometry and matrix IR spectroscopy. Towards photolysis, even with the 248 nm light of an KrF excimer laser, the matrix-isolated cyclophane 1 b is completely stable, in contrast to many other cyclophanes. Xylylene 3 b, on the other hand, is photolabile and rearranges in a complex sequence to heptafulvene 6. The fluorinated parylene produced via CVD of 3 b is thermally stable up to 240 8C [20] and is of interest for coatings in electronic devices [21–23].

213

214

8 Reactive Intermediates from Cyclophanes

Among many other derivatives of p-xylylene that have been synthesized in preparative scale by FVP of the corresponding cyclophanes are the naphthalene- and the anthracene-p-quinodimethane 3 c and 3 d, respectively [8, 9]. The NMR as well as the UV/Vis and IR spectra of these species have been reported.

The question of whether the diradical 4 is an intermediate in the thermolysis of paracyclophanes 1 was investigated by Reich and Cram using chiral cyclophanes (Scheme 8.3) [24]. Upon heating of (–)-4-carbomethoxy[2.2]paracyclophane 1 e to approximately 200 8C racemization was observed with rates which were only slightly dependent on the solvent polarity. An activation barrier for the racemization of 38 kcal mol–1 was reported. This suggests that p,p’-dimethylenedibenzyl diradicals of type 4 are indeed formed as intermediates, and that the lifetime of these diradicals is long enough to allow rotation of the phenyl rings. The diradicals 4 can either be trapped, cleave to the corresponding p-xylylenes 3, or perform a ring-closure back to 1. Heating of cyclophane 1 a as a solution in p-diisopropyl-

benzene to 250 8C produced p,p’-dimethyldibenzyl as the only nonpolymeric material in 21% yield [24]. This clearly indicates that 4 a is formed as the primary intermediate and long-lived enough to abstract hydrogen atoms from the solvent.

Scheme 8.3

8.1 Thermolysis of [2.2]Paracyclophanes

Enthalpy profile for the ring-opening of paracyclophane a (reproduced with permission from ref. 6)

Fig. 8.1

Since simple alkyl and aryl radicals recombine without activation barrier, the question arises whether or not diradical 4 is a true intermediate lying in an energy well with an activation barrier separating 4 from 1. To answer this question a very detailed thermochemical study of the 1 a ? 3 a interconversion was reported by Roth et al. [6]. The thermolysis of 1 a was performed in the gas phase in the presence of NO, a very efficient radical scavenger, at temperatures between 168 and 238 8C. From the temperature dependence of the depletion of 1 a, an Arrhenius activation barrier for the ring opening of 37.7 kcal mol–1 was determined. Principally both the diradicals 4 a and 3 a could be formed and trapped by NO, and it is thus initially not clear to which of the two intermediates the activation barrier is leading. However, since the reaction enthalpy for the formation of p-xylylene 3 a from 1 a is, at 56.1 kcal mol–1, higher than the observed activation barrier, only 4 a can be the trapped intermediate. From the kinetic data the activation barrier for the cyclization of diradical 4 a to cyclophane 1 a was determined as 11.8 kcal mol–1 and that of the cleavage of 4 a to two molecules of p-xylylene 3 a as 30.2 kcal mol–1. Thus, 4 a is indeed a real intermediate lying in an energy well and separated by substantial activation barriers from 1 a and 3 a. Since the two benzyl radicals in 4 a are separated by an ethano bridge in p-position, the through-bond electronic interaction between the radical moieties should be negligible and the activation barrier arise mainly from the increasing steric repulsion upon bond formation. The energy of the four conformers of 4 a A–D (R = H, Scheme 8.3) should be therefore similar, as long as the steric energy of the cyclophane can be released. An interesting example of a diradical, where the steric energy is only partially released, is obtained by thermolysis of the triply bridged 1,3,5-[2.2.2]cyclophane 7

215

216

8 Reactive Intermediates from Cyclophanes

Scheme 8.4

Development of the steric and electronic energy during the formation of a C–C bond in diradicals of type 4. The bond distance varies between 3.5 and 1.5 Å (reproduced with permission from ref. 6) Fig. 8.2

(Scheme 8.4) [6]. In this case the thermochemistry of the ring-opening was determined by oxygen trapping in the gas phase. At 31.5 kcal mol–1, the barrier for breaking of one of the three ethano bridges to form the metacyclophane diradical 8 is considerably smaller than in the case of 1. The barrier for the radical recombination in 8 to give back 7 is now only 3.4 kcal mol–1. The 1,2,4-[2.2.2]cyclophane 9 is an intermediate case (Scheme 8.4) [25]. Thermolysis results in the formation of the less strained orthocyclophane diradical 10, which can be trapped in solution with maleic ester. Under preparative conditions the diradical abstracts hydrogen

8.2 Photolytical Cleavage of [2.2]Paracyclophanes

atoms from the solvent to give dibenzocyclooctane 11 [25]. The activation barrier for the 9 ? 10 reaction was determined by oxygen trapping as 34.4 kcal mol–1 and for the reverse reaction 6.8 kcal mol–1 [6]. These experiments nicely demonstrate that the energy well of diradicals of type 4 increases with increasing difference in strain energy between the diradical and the cyclophane. An activation barrier for the radical recombination arises if the increasing steric energy along the reaction pathway is not overcompensated by the gain in electronic stabilization during bond formation. This relation was confirmed for the ring closure of 4 a by calculations using the MMEVBH force field [26] which was parameterized to reproduce radical reactions of hydrocarbons. While the steric repulsion increases almost linearly when the distance between the radical centers is reduced from 3.5 Å to 1.5 Å, the electronic stabilization is significant only at distances below 2.5 Å. This results in an activation barrier with a maximum around 2.5 Å.

8.2

Photolytical Cleavage of [2.2]Paracyclophanes

The photochemical cleavage of cyclophane 1 a has also been thoroughly investigated. Irradiation of 1 a in the gas phase with an ArF excimer laser (193 nm, vacuum UV light) produces 3 a in a multiphoton hot molecule reaction [27, 28]. Excited states and radical cations formed via ionization of 1 a could be excluded. Under these conditions the electronic excitation is rapidly transformed to a hot ground state molecule which subsequently decomposes to two molecules of 3 a. In contrast, photolysis of 1 a under the conditions of matrix isolation at low temperature proceeds via a triplet mechanism (see below). The direct observation of diradical 4 a as an intermediate in the photolysis of 1 a at low temperature was described by Kaupp et al. [29, 30] and by Ishikawa [31] using organic glasses. Kaupp et al. photolyzed cyclophane 1 a in a organic glass (2-methyltetrahydrofuran) with 254 nm of a low pressure mercury arc lamp [30]. From the UV/Vis and fluorescence spectra it was concluded that at 83 K p-xylylene 3 a was formed, whereas at 77 K the diradical 4 a was produced. The UV/Vis spectrum of 4 a is similar to that of the benzyl radical which suggests that there is only minor interaction between the two benzyl radical moieties in 4 a. Photolysis of cyclophane 12 under similar conditions produced the diradical 13, again with a similar UV/Vis spectrum to that of the benzyl radical and 4 a. Continuous irradiation into the visible absorption of 13 results in the cleavage of the second ethano bridge and formation of the bisquinoid system 14. Photolysis in organic glasses of furanophane 15 produces 2,5-dimethylen-dihydrofuran 16 and of naphthalenophane 17 yields benzo-p-xylylene 18 [29]. The intermediate diradicals were not reported. The formation of the diradicals 4 as intermediates in the photolysis of paracyclophanes 1 to give p-xylylenes 3 clearly demonstrates that these photochemical cleavages are stepwise reactions rather than concerted [6 + 6]cyclorevisions.

217

218

8 Reactive Intermediates from Cyclophanes

The photochemical cleavage of 4 a to give 3 a was found by Nagakura et al. to be a biphotonic process proceeding via excited triplet states (Scheme 8.5) [31]. The diradical 4 a was characterized by its UV, fluorescence, and ESR spectra. Careful analysis of the ESR spectra in several organic glasses confirmed that indeed the diradical 4 a and not benzyl radicals were formed on photolysis of 1 a. However, the authors could not confirm the photochemical cleavage of 4 a to p-xylylene 3 a. The spectroscopic properties of 4 a was remarkably dependent on the viscosity of the glass indicating conformational changes in softer glasses. The direct observation of 4 a as a transient species in solution was reported by Scaiano et al. [32]. Photolysis of 19 with a 10-ns pulse of 266 nm laser light produced the radical 20 with an absorption maximum at 320 nm and a lifetime (in

Scheme 8.5

8.2 Photolytical Cleavage of [2.2]Paracyclophanes

Scheme 8.6

cyclohexane at room temperature) of 9 ls. Irradiation into the absorption maximum with a second laser (308 nm) resulted in only very small conversion of 20 and no useable transient species. A preparative scale run of the photolysis of 19, however, produced minor amounts of cyclophane 1 a which suggests that 4 a is produced as an intermediate (Scheme 8.6). A better precursor of 4 a is the cyclophaneketone 21, which on 254 nm irradiation in benzene produces 1 a as the only product. Irradiation with a pulsed laser at 266 nm produces with a small time delay a transient species with similar spectroscopic properties than 20 (kmax = 320 nm, lifetime 8.6 ls) which was assigned to 4 a. These experiments provided detailed insight into the reactivity of 4 a. As expected, it reacts with molecular oxygen close to the diffusion limit. While it thermally reacts to 1 a or cyclic oligomers, there is no indication that it thermally (at room temperature) or photochemically cleaves to p-xylylene 3 a. In the presence of a triplet quencher the photolysis of 1 a produces small amounts of 3 a besides 4 a. However, even under these conditions 4 a is not cleaved to 3 a which clearly shows that 3 a is directly formed from 1 a. The cleavage of 4 a to 3 a requires thermolysis at high temperatures (see above). Both the low temperature photolysis of 1 a to give 4 a via triplet states or the direct formation of 3 a via vacuum UV photolysis (193 nm) at room temperature are biphotonic processes.

219

220

8 Reactive Intermediates from Cyclophanes

8.3

Cleavage of [2.2]Paracyclophanes via Electron Transfer

The oxidative cleavage (via single electron transfer, SET) of the C–C bond of diarylethanes produces a benzyl radical and a benzyl cation [33]. The formation of a distonic (charge and radical site are separated) radical cation 23 by chemical or electrochemical oxidation of [2.2]paracyclophane 1 a was reported by several authors [34, 35]. The primarily formed radical cation 24 spontaneously opens to the distonic radical cation 23 (Scheme 8.7). Direct spectroscopic evidence for 23 is not available, however, trapping experiments provide convincing evidence for its formation as an intermediate in the oxidation of 1 a. A detailed study of the oxidative cleavage of [2.2]paracyclophanes was reported by Hopf et al. [35]. Oxidation of cyclophane 25 with either FeCl3 or NOBF4 produced the radical cation 26 which reacted with NO to the benzyl cation 27 (Scheme 8.8). The cation subsequently reacted in an intramolecular electrophilic addition via the oxime to 28.

Scheme 8.7

Scheme 8.8

8.4 Cleavage of Cyclophanes with Unsaturated Bridges

8.4

Cleavage of Cyclophanes with Unsaturated Bridges

The influence of conjugation was analyzed by Roth et al. investigating the ringopening of the unsaturated paracyclophanes 29 a–d [6]. Thermolysis of [2.2]paracyclophan-1-ene 29 a results in the formation of the diradical 30 a (Scheme 8.9). Due to the double bond in the bridge, the two benzyl radical moieties in 30 a are now conjugated and a quinoid structure can be formulated. Rotation at the C1–C2 bond results in the anti (or trans) configuration 31 of the diradical. The activation barrier for the ring opening of 29 a was determined as 34.4 kcal mol–1 and is therefore comparable to that of 1 a. The diradical 30 a, however, is more stabilized than 4 a, which results in a much higher equilibrium concentration. Thus, even with small concentrations of scavenger the diradical 30 a is completely trapped and the cyclization back to 29 a not observed. The barrier for the cyclization of 30 a was estimated as 23.7 kcal mol–1, which means that it lies in a deep thermodynamic well. The diradical 30 a could be directly characterized by Sander et al. using matrix isolation spectroscopy [36, 37]. Irradiation of cyclophene 29 a, matrix-isolated in argon at 10 K, with short-wavelength UV light (248 nm) results in a yellow-green coloring of the matrix and appearance of a series of absorptions in the visible region of the spectrum extending to 600 nm assigned to 30 a. Due to the rigidity of the argon matrix the isomerization of 30 a to the anti form 31 a was considered to

Scheme 8.9

221

222

8 Reactive Intermediates from Cyclophanes

be unlikely. In organic glasses the 600 nm absorption is missing and the other bands are slightly red-shifted. These shifts in the absorption spectrum were attributed to geometric relaxations in the softer organic glass. The IR spectrum of 30 a could also be recorded in solid argon and was compared to ab initio calculations at the MP2 level of theory. The good agreement between experimental data and theoretical predictions confirms the assignment of 30 a. The experimental UV/Vis spectrum of 30 a was later reassessed by Hohlneicher et al. on the basis of correlated ab initio calculations [38]. These authors come to the interesting conclusion, that in the rigid argon matrix two structures of 30 a coexist: the quinoid structure A and a higher energy biradical structure B. Thus, A and B are not resonance structures but represent distinct electronic structures which can interconvert.

The introduction of two methyl groups in the cyclophane 29 b leads to a twisting of the benzene rings in 30 b, resulting in a loss of most of the electronic stabilization of the diradical [6]. Consequently, diradical 30 b lies in a well of only 10.9 kcal mol–1 compared to 23.7 kcal mol–1 for 30 a. The benzene annulation in 30 c has a similar effect. The phenyl rings of the two benzyl radical moieties are strongly tilted towards the central benzene ring, and the interaction between the two unpaired electrons resulting in a quinoid resonance structure is therefore negligible. The activation barrier for the ring closure in 30 c is 11.9 kcal mol–1 and is thus basically identical to that of the parent system 4 a.

8.5 Cleavage of Cyclophanes with Carbonyl Groups in the Bridge

Another highly interesting system investigated by the same authors is 1,2-bismethylene[2.2]paracyclophane 29 d. In this case the ring-opening results in the formation of a non-stabilized “non-Kekulé” diradical 30 d which can be described as a derivative of tetramethylenethane. In addition, in the syn conformer of the diradical the steric interaction between the methylene groups results in a tilting of the benzene rings. As expected, the two unpaired electrons in 30 d do not interact and the activation barrier for the ring-closure to 29 d is with 11.6 kcal mol–1 again almost identical to that of 4 a. The experimental determination of thermodynamic data of a series of diradicals derived from cyclophanes allows a deep insight into the electronic structure of these species. On the basis of these data an empirical force field for the precise prediction of the enthalpy profile of organic radicals could be developed.

8.5

Cleavage of Cyclophanes with Carbonyl Groups in the Bridge

The photolytic ring-opening of the cyclophanes described so far requires short wavelength UV irradiation to cleave the aliphatic C–C bond. Introduction of a carbonyl group in the bridge provides a chromophore which absorbs light above 300 nm and which leads to ring-opening of the cyclophane via a-cleavage. An example of this is the cyclophaneketone 21 described above. The photolysis of cyclophaneketons often results in a clean photodecarbonylation. This has been used as a general synthetic route to [2.2]cyclophanes and [2.3]cyclophanes [39]. Thus, the bisketone 31 produces 21 in 57% yield after 4 h irradiation, which on continuous irradiation decarbonylates to give 95% yield of 1 a after 20 h.

Irradiation of cyclophanediketone 32 in argon at 10 K at 335 nm results in the destruction of the starting material and formation of a mixture of 2,5-cyclohexadiene1,4-bis(ylideneketene) 34 and p-xylylene 3 a (Scheme 8.10) [40]. The IR spectra gave no evidence for diradical 33 which is formed after the first a-cleavage. Presumably diradical 33 is more photolabile than cyclophane 32 and therefore produced in only very small concentrations during the photolysis of 32. Similar results were obtained when 32 was irradiated in organic glasses at 77 K. If irradiated in ethanol at room temperature, however, diradical 33 is efficiently trapped and the ester 35 and diethyl terephthalate 36 are formed. Prolonged irradiation transforms 35 completely to 36.

223

224

8 Reactive Intermediates from Cyclophanes

Scheme 8.10

In solution bisketene 34 is highly reactive, and the photolysis of cyclophane 32 is so far the only method for its synthesis. The photodecarbonylation of 34 should lead to p-benzyne 37, in analogy to the synthesis of o-benzyne 39 from the corresponding bisketene 38 [41]. However, very short wavelength photolysis was necessary to decarbonylate 34 and under these conditions p-benzyne 37 could not be observed [42].

The FVP of terephthaloyl diiodide 40, a thermal precursor of 34, directly produces iodine, CO, and hex-3-ene-1,4-diyne 41, the product of the ring-opening of p-benzyne 37. Bisketene 34 and p-benzyne 37 were not observed under these conditions. Thus, although 34 is a potential precursor of p-benzyne 37, the harsh conditions for the thermal or photochemical decarbonylation of 34 do not allow to isolate 37 generated from this precursor.

8.5 Cleavage of Cyclophanes with Carbonyl Groups in the Bridge

Scheme 8.11

More successful was the synthesis of m-benzyne 43 a and derivatives via cyclophane photolysis. During the last years metaparacyclophanes of type 42 have been used as photochemical precursors for m-benzynes 43 (Scheme 8.11) [43–47]. Irradiation (254 nm) of metaparacyclophane 42 a in argon at 10 K resulted in the formation of a new species with an IR absorption at 1833 cm–1, characteristic of acyl radicals. The most reasonable assignment of this new molecule is the diradical 44 formed by a-cleavage of 42 a. However, 45 or 46 are also in accordance with an IR band at 1833 cm–1. Aromatic acyl radicals are labile and are expected to thermally or photochemically cleave to CO and aryl radicals. Indeed, continuous irradiation produces p-xylylene 3 a and m-benzyne 43 a. The m-benzyne was identified by independent synthesis in argon matrices from several other precursors and by comparison of the experimental IR spectra with high level ab initio calculations. These experiments provided a complete set of IR absorptions of 43 a and finally settled the question whether m-benzyne is better described as an open shell diradical 43 a or as a closed shell bicyclic molecule 47. Neither conventionally trapping experiments in solution nor theoretical studies alone were able to definitely differentiate these two structures, and for both, evidence was presented in literature. By comparison of the IR spectra of matrix-isolated 43 a with spectra predicted at the coupled cluster level of theory it was clearly shown that structure 43 a is the most accurate description, although m-benzyne is not a regular hexagon but rather exhibits a trapezoid distortion with a shortening of the distance between the radical centers to about 2 Å. This structure balances the gain in energy by formation of a bond between the radical centers (through bond and through space interaction) and the rise in energy due to the geometric distortion of the benzene skeleton.

225

226

8 Reactive Intermediates from Cyclophanes

The cleavage of metaparacyclophanes 42 b–d was used to synthesize the corresponding m-benzynes 43 b–d with substituents in the 5-position such as fluorine or methoxy [48]. This allowed the influence of electron withdrawing or electron donating groups on the spectroscopic properties of 43 to be studied.

The cleavage of strained cyclophanes provides a unique method for the synthesis of aromatic diradicals and quinoid compounds. The formation of these reactive species is assisted by the release of strain upon ring-opening. A further advantage of this method, compared to the more conventional stepwise cleavage of two labile groups in a precursor molecule, is that the two radical centers are formed instantaneously by braking only one C–C bond. Thus, using cyclophanes as precursors for diradicals excludes a possible stepwise formation of the two radical centers which might interfere in kinetic and trapping studies. Due to the large variability of cyclophanes this method provides a general access to diradicals which has not yet been used systematically.

8.6

References 1

2

3

4 5 6

7

K. Nishiyama, N. Sakiyama, S. Seki, H. Horita, T. Otsubo, S. Misumi, Bull. Chem. Soc. Jpn, 1980, 53, 869–877. S. W. Benson, Thermochemical Kinetics: Methods for the Estimation of Thermochemical Data and Rate Parameters, 2nd edn, 1976. Wiley, New York. A. de Meijere, S. I. Kozhuskov, K. Rauch, H. Schill, S. P. Verevkin, M. Kümmerlin, H.-D. Beckhaus, C. Rüchardt, D. S. Yufit, J. Am. Chem. Soc. 2003, 125, 15 110–15 113. C. J. Brown, A. C. Farthing, Nature (London) 1949, 164, 915–916. M. Szwarcz, Discuss. Faraday Soc., 1947, 46–49. W. R. Roth, H. Hopf, A. de Meijere, F. Hunold, S. Boerner, M. Neumann, T. Wasser, J. Szurowski, C. Mlynek, Liebigs Ann. Chem., 1996, 2141–2154. S. K. Pollack, B. C. Raine, W. J. Hehre, J. Am. Chem. Soc., 1981, 103, 6308–6313.

8 9

10

11

12

13 14 15

D. J. Williams, J. M. Pearson, M. Levy, J. Am. Chem. Soc., 1970, 92, 1436–1438. J. M. Pearson, H. A. Six, D. J. Williams, M. Levy, J. Am. Chem. Soc., 1971, 93, 5034–5036. V. Massa, A. Cicuta, W. Cavigiolo, in Eur. Pat. Appl. (Ausimont S.p.A., Italy), Ep, 1988, p. 9. T. Nakagawa, N. Niku, in Jpn. Kokai Tokkyo Koho (Kimura Yoshio, Japan), Jp, 1993, p. 3. L. I. Trakhtenberg, E. Axelrod, G. N. Gerasimov, A. E. Grigoriev, E. I. Grigoriev, S. A. Zav’yalov, Y. Feldman, Scientific Israel–Technol. Advantages 1999, 1, 34–42. S. Iwatsuki, Adv. Polym. Sci., 1984, 58, 93– 120. O. L. Chapman, U. P. E. Tsou, J. Am. Chem. Soc. 1984, 106, 7974–7976. Y. Yamakita, Y. Furukawa, M. Tasumi, Chem. Lett., 1993, 2, 311–314.

8.6 References 16 17

18

19

20 21

22

23 24 25

26

27 28

29 30 31 32

33

Y. Yamakita, M. Tasumi, J. Phys. Chem., 1995, 99, 8524–8534. W. R. Dolbier, Jr., M. A. Asghar, H. Q. Pan, L. Celewicz, J. Org. Chem., 1993, 58, 1827–1830. S.-z. Zhu, Y.-y. Mao, G.-f. Jin, C.-y. Qin, Q.-l. Chu, C.-m. Hu, Tetrahedron Lett., 2002, 43, 669–671. H. H. Wenk, W. Sander, A. Leonov, A. De Meijere, Eur. J. Org. Chem. 1999, 3287–3290. A. O. Oyewale, R. A. Aitken, J. Therm. Anal., 1995, 45, 1393–1401. C. J. Lee, H. Wang, G. A. Foggiato, in PCT Int. Appl. (Quester Technology, Inc., USA), Wo, 1999, p. 69. H. Maruyama, in Jpn. Kokai Tokkyo Koho (Daisan Kasei K. K., Japan), Jp, 1998, p. 6. J. J. Senkevich, S. B. Desu, Appl. Phys. Lett., 1998, 72, 258–260. H. J. Reich, D. J. Cram, J. Am. Chem. Soc. 1967, 89, 3078–3080. H. Hopf, J. Kleinschroth, A. E.-F. ElSayed Murad, Is. J. Chem., 1980, 20, 291– 293. W. R. Roth, V. Staemmler, M. Neumann, C. Schmuch, Liebigs Ann. Chem., 1995, 1061–1118. S. Shimizu, N. Nakashima, Y. Sakata, Chem. Phys. Lett. 1998, 284, 396–400. Y. Hosoi, T. Yatsuhashi, K. Ohtakeyama, S. Shimizu, Y. Sakata, N. Nakashima, J. Phys. Chem. A 2002, 106, 2014– 2019. G. Kaupp, Angew. Chem. 1976, 88, 482– 484. G. Kaupp, E. Teufel, H. Hopf, Angew. Chem. 1979, 91, 232–234. S. Ishikawa, J. Nakamura, S. Nagakura, Bull. Chem. Soc. Jpn. 1980, 53, 2476–2480. M. A. Miranda, E. Font-Sanchis, J. Perez-Prieto, J. C. Scaiano, J. Org. Chem., 2001, 66, 2717–2721. L. W. Reichel, G. W. Griffin, A. J. Muller, P. K. Das, S. N. Ege, Can. J. Chem., 1984, 62, 424–436.

34

35

36

37

38 39

40

41

42

43

44 45 46

47

48

W. Adam, M. A. Miranda, F. Mojarrad, H. Sheikh, Chem. Ber., 1994, 127, 875– 879. S. Sankararaman, H. Hopf, I. Dix, P. G. Jones, Eur. J. Org. Chem. 2000, 2711–2716. R. Marquardt, W. Sander, T. Laue, H. Hopf, Liebigs Ann. Chem., 1996, 2039– 2043. W. Sander, R. Marquardt, G. Bucher, H. Wandel, Pure Appl. Chem., 1996, 68, 353–356. D. Henseler, G. Hohlneicher, J. Mol. Struct., 1999, 480/481, 515–518. H. Isaji, M. Yasutake, H. Takemura, K. Sako, H. Tatemitsu, T. Inazu, T. Shinmyozu, Eur. J. Org. Chem., 2001, 2487– 2499. R. Marquardt, W. Sander, T. Laue, H. Hopf, Liebigs Ann. Chem., 1995, 1643– 1648. O. L. Chapman, C. C. Chang, J. Kolc, N. R. Rosenquist, H. Tomioka, J. Am. Chem. Soc., 1975, 97, 6586–6588. R. Marquardt, W. Sander, T. Laue, H. Hopf, Liebigs Ann. Chem., 1995, 1643– 1648. R. Marquardt, W. Sander, E. Kraka, Angew. Chem. 1996, 108, 825–877; Angew. Chem. Int. Ed. Engl., 1996, 35, 746–748. W. Sander, M. Exner, J. Chem. Soc., Perkin Trans. 2, 1999, 2285–2290. W. Sander, Acc. Chem. Res., 1999, 32, 669–676. W. Sander, M. Exner, M. Winkler, A. Balster, A. Hjerpe, E. Kraka, D. Cremer, J. Am. Chem. Soc., 2002, 124, 13072–13079. H. Wenk Hans, M. Winkler, W. Sander, Angew. Chem. 2003, 115, 519–546; Angew. Chem. Int. Ed. Engl. 2003, 42, 502–528. W. Sander, M. Exner, J. Chem. Soc., Perkin Trans. 2, 1999, 2285.

227

189

7

Intramolecular Reactions in Cyclophanes Henning Hopf

7.1

Introduction

There are not many organic compounds in which p-systems that are not directly connected come as close in space as the benzene rings of [2.2]paracyclophane (1), the layered organic molecule par excellence (Scheme 7.1). In fact its intra-annular distance (ca. 3.1 Å for the non-bridgehead carbon atoms) may be equated with the “length” of a p-orbital and is comparable with the distance between the planes of two other important layered structures: graphite (3.4 Å) and the intrastrand distance between the bases of DNA (3.4 Å). The electronic interaction between the (bent) benzene rings in 1 was noted and studied by Cram in his pioneering investigations on cyclophanes [1]. That there could also be a chemical interaction, i.e. bond formation, between these rings was originally less obvious. After all, intramolecular (or intradeck) ring formation would not only entail destruction of aromatic subunits [2], but also generate cage structures of presumably high strain. In the meantime many reactions have been reported between the benzene rings of [2.2]paracyclophanes as well as between the aromatic nuclei of multibridged cyclophanes and of isomers of 1, in particular of [2.2]metacyclophane. These reactions will be discussed in Section 7.2. The rigidity of 1 not only holds the atoms of this molecular framework “in position” [3] but also extends to the vicinity of the molecule. For example, many derivatives, such as the ketones 2, prefer the conformation shown in Scheme 7.1 in which the carbonyl group points towards the ethano bridge [4]. These functionalized cyclophanes often show a remarkable substituent effect in electrophilic aromatic substitution reactions: the substituents induce an incoming electrophile (E+ in Scheme 7.1) to attack the unsubstituted ring of the substrate just opposite the original functional group in the so-called pseudo-geminal (pseudo-gem) position. This intramolecular directing effect, also first detected by Cram and coworkers (see below), has been of crucial importance for the preparation of multibridged cyclophanes as well as numerous chiral cyclophanes, which are of growing importance in stereoselective synthesis, as discussed in Chapter 17 of this volume. These pseudo-gem substitutions will be discussed in Section 7.3.

Modern Cyclophane Chemistry. Edited by Rolf Gleiter, Henning Hopf Copyright © 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 3-527-30713-3

190

7 Intramolecular Reactions in Cyclophanes

Scheme 7.1

Intramolecular reactions in [2.2]paracyclophanes

If the layered nature of 1 causes preferred orientations in the immediate vicinity of the cyclophane core, the question may be asked whether this “ordering effect” can also be extended to reactive groups further away from the aromatic decks, but connected to them by suitable spacers, as shown in general form in 3 in Scheme 7.1. If, after an appropriate intramolecular reaction, the phane system can be detached from the reaction product, e.g. by hydrolysis, we might expect that the inherent order, the layered geometry of the cyclophane nucleus, could be transferred to the cleavage product. This effect which we have called “topochemical reaction control in solution” (see below) is addressed in Section 7.4. Most examples presented here have been reported during the last 15 years; I thus hope to bridge the gap between cutting edge cyclophane research and the older cyclophane review literature [1, 5].

7.2

Reactions between the Benzene Rings of Cyclophanes

Possibly the first reaction between the aromatic decks of a cyclophane was the photoisomerization of anti-[2.2](1,4)-naphthalenophane (4) to dibenzoequinene (5) reported by Wasserman and Keehn in 1967 (Scheme 7.2) [6]. More recent studies by Gleiter and coworkers [7] have shown that this process, which results in a complete breakdown of the central cyclophane core, is general, and that derivates such as 6 a (20%), 6 b (27%), and 6 c (58%) can also be prepared from the appropriate precursor cyclophanes, as can the less symmetrical derivative 6 d (9%), all isomerizations being carried out in benzene under the influence of 350 nm light. Interestingly, when two more benzene rings are annulated to 4, such as in the anthracenophane 7 (Scheme 7.3), the reaction takes a completely different course, yielding the hydrocarbon 8, in which the two decks are now held together by cyclobutane rings [8]. Whereas in the 4 ? 5-isomerization the number of “complete”

7.2 Reactions between the Benzene Rings of Cyclophanes

Scheme 7.2

Photocyclizations of [2.2]paracyclophanes to equinenes

Scheme 7.3

Photocyclization of anthracenophanes

191

192

7 Intramolecular Reactions in Cyclophanes

Scheme 7.4

From cyclophanes to hexaprismanes?

aromatic rings in substrate and product has been maintained, it has increased by two for the 7 ? 8-isomerization, possibly accounting for the driving force of the process. A comparable process has also been observed for the 2,11-diaza[3.3](9,10)anthracenoparacyclophane 9 [9], indicating that the bridge length can be extended without jeopardizing the intra-annular photoaddition, although the limit of this extension has not been determined. The formation of single bonds between the bridgeheads in 7 suggests the possibility of preparing a derivative of the still unknown hexaprismane from a suitable cyclophane precursor. For example, it has been proposed that the [36]superphane 10 could be photoisomerized to the fully-bridged hexaprismane derivative 11 (Scheme 7.4) [10]. Beginning with [33](1,3,5)cyclophane (12) Shinmyozu and coworkers in a comprehensive effort have attempted to accomplish such face-to-face cycloadditions [11]. In none of their numerous experiments could a hexaprismane derivative be isolated. Hydrocarbon 12, for example, yielded the chloride 13 a when irradiated in dry dichloromethane. When the experiment was repeated in the same solvent containing water, the adducts 13 b and 14 were obtained. Changing to methanol/ hydrochloric acid leads to still other cage hydrocarbon systems. It could well be that the initial photoproduct is the expected hexaprismane. But this is apparently protonated under the reaction conditions, the resulting carbocations equilibrate and the most stable cation is finally intercepted by the nucleophile present (chloride, water, methanol). The corresponding fourfold bridged cyclophanes [34](1,2,4,5)cyclophane [12] and its isomer [34](1,2,3,5)cyclophane [13] led to similar results. The structures of these often bizarre rearrangement/addition products were determined by X-ray structural analysis whenever possible, so there can be no doubt as to the accuracy of the assignments. In no case did the isolated products contain remaining double bonds. In an interesting proposal Prinzbach has suggested using the [43](1,2,4,5)cyclophane 15 as a precursor and subject it to

7.2 Reactions between the Benzene Rings of Cyclophanes

Scheme 7.5

Another possible route to hexaprismanes

Scheme 7.6

Intramolecular Diels-Alder additions in [2.2]paracyclophanes

photolysis. According to B3LYP/6-31G* calculations this could undergo a [2 + 2]cycloaddition and the resulting 16 could ring-close to the prismane 17 [14]. This intriguing idea has not been put to the test experimentally. If a benzene ring of a cyclophane can be converted into a reactive subsystem this in principle could attack the facing aromatic nucleus in, for example, DielsAlder fashion. A case in point is provided by the cyclobutenophane 18 [15]. When heated, it opens up to the o-xylylene intermediate 19, the diene part of which is positioned exactly above the dienophilic double bond in the other half of the molecule. Diels-Alder addition takes place and furnishes the internal adduct 20. The four-membered ring of 20 could in principle also open up to provide a tetramethylenethane intermediate (known from allene dimerization), which could add to the opposing aromatic ring. But this process is not observed, presumably because of the high strain increase associated with it. It has been known for some time that the benzene rings of cyclophanes can function as dienes [16]; here we see the rare case in which they play the role of the dienophilic partner. In another Diels-Alder process, the aryne intermediate 22 was generated by treating the bromide 21 with potassium tert-butoxide in tert-butylbenzene at 175– 180 8C [17]. Under these conditions 22 is either intercepted by tert-butoxide to produce the ether 23 or it undergoes an intramolecular Diels-Alder addition to provide the bridged barrelene derivative 25, for whose formation the aryne evidently has to be long-lived enough to undergo a rotational process to conformation 24 before it can cycloadd (Scheme 7.7). Interestingly, the same cycloaddition product is produced when the isomer 26 of the substrate 21 is dehydrobrominated, indicating that the aryne 27 can also adopt the perpendicular conformation 24. In this latter elimination the endo-ether

193

194

7 Intramolecular Reactions in Cyclophanes

Scheme 7.7

Dehydrobenzene (aryne) intermediates as intramolecular addition partners

28 is also formed, in agreement with the overall mechanistic picture. The corresponding phenols are produced as well as the ethers 23 and 28. Although other intramolecular aryne additions of this type have been reported [18, 19], 4,5-dehydro[2.2]paracyclophane, one of the simplest cyclophynes (see Chapter 1), generated from the corresponding bromide [20] did not undergo intramolecular trapping, presumably because the strain increase in the cycloaddition step is too steep. Still, this parent hydrocarbon of the whole series, compound 33, could be prepared by the alternate route summarized in Scheme 7.8 [21]. When the bis-epoxide 29, readily prepared from the mono Birch reduction product of [2.2]paracyclophane (1), was treated with trifluoroacetic acid in dichloromethane, the diol 31 was obtained in acceptable yield. Formally this derivative is produced by ring opening of the oxirane rings, rotation of the still intact benzene ring to an “upright position”, and double intramolecular Friedel-Crafts alkylation as symbolized by structure 30. Treatment of 31 with borontribromide in dichloromethane converts it in good yield to the dibromide 32, which undergoes twofold dehydrobromination with potassium tert-butoxide in THF to furnish the doubly bridged benzobarrelene 33. That epoxides are interesting and reactive intermediates in intramolecular cycloadditions of paracyclophanes, is also illustrated by the example reproduced in

7.2 Reactions between the Benzene Rings of Cyclophanes

Scheme 7.8

The synthesis of the simplest cyclophane barrelene derivative

Scheme 7.9

Intramolecular cycloaddition initiated by epoxidation of a [2.2]paracyclophane

Scheme 7.9. Here the tetramethyl[2.2]paracyclophane 34 was oxidized under very mild conditions with p-nitroperbenzoic acid in dichloromethane in the presence of sodium bicarbonate as a neutralizing reagent [22]. Rather than the expected primary product 35, the epoxidation afforded an isomer, 36, the structure of which was established by X-ray structural analysis. Clearly, an intramolecular [2 + 4]cycloaddition had occurred again. Whereas for these examples practical applications are presently hardly apparent, this is different for the unusual heterocyclic cyclophane 37 (Scheme 7.10), recently prepared by Bodwell and coworkers [23]. When this indolophane was heated in N,N-diethylaniline for two days it underwent an intramolecular cycloaddition – accompanied by nitrogen loss – to provide the pentacyclic product 38 in excellent yield (90%) [24]. The resemblance of 38 to the so-called A–E rings, shown in structure 39, of strychnine is striking. Before turning to the [2.2]metacyclophanes as the other large group of cyclophanes for which intramolecular reactions have been reported, we note that cyclophanes have been used as complexing reagents which can encapsulate guests of very different character and size, depending on the size of the cyclophane host. In these systems no “internal” bonds are generated between the aromatic decks, but noncovalent interactions established between them via the guest atoms, ions or whole molecules (see also Chapter 20) [25, 26].

195

196

7 Intramolecular Reactions in Cyclophanes

Scheme 7.10

Application of cyclophane chemistry in natural product synthesis

The prototypical intramolecular reaction of a [2.2]metacyclophane is its conversion to a (hydrogenated) pyrene derivative, a process that was discovered first for metacyclophandienes and investigated very extensively by Boekelheide and later Mitchell and their groups [1, 27]. Among the reported examples by far the most prominent is the isomerization of the anti-[2.2]metacyclophandiene 40 to trans-10 b,10 c-dimethyl-10 b,10 c-dihydropyrene (41) by irradiation with UV light. This valence isomerization, which converts a layered molecule into a more or less flat structure – 41 is a [14]annulene with a central substituted ethano bridge – occurs stereospecifically, and since the central single bond can be opened again by irradiation with visible light, the whole system functions as a light-driven molecular switch (Scheme 7.11). This photochromic behavior has been investigated particularly by Mitchell and coworkers, who have also been able to incorporate the 40/41-switch into more complex p-systems such as the hydrocarbon 42, which not only photochemically opens to the “half-open” structure 43, but also to the fully-opened hydrocarbon 44 [28]. Whether these interesting stretched cyclophanes can indeed be practically employed as three-way molecular p-switches remains to be seen. Since the syn-[2.2]metacyclophandienes are more difficult to prepare than their anti-isomers [29] the chemistry of the corresponding cis-dihydropyrenes is not very well explored experimentally. One possible way to overcome this limitation consists of the incorporation of a molecular bridge into appropriate precursor molecules which force the paracyclophandienes and hence the dihydropyrenes into the orientation that is not normally favored. The first experiments toward these goals have recently been reported [30]. The parent hydrocarbon, [2.2]metacyclophane (45), also undergoes an intra-annular ring closure reaction as shown by Sato and coworkers who obtained the tetrahydropyrene derivative 46 by treating 45 with bromine in the presence of iron powder [31] presumably by way of an addition/elimination mechanism (Scheme 7.12). That the process can become quite complex – and strongly depends on the reaction conditions – has in particular been demonstrated by Yamato and his coworkers

7.2 Reactions between the Benzene Rings of Cyclophanes

Vis

Scheme 7.11

Photoisomerization of [2.2]metacyclophandienes and dihydropyrenes

Scheme 7.12

Ring closure in [2.2]metacyclophanes by electrophilic reagents

197

198

7 Intramolecular Reactions in Cyclophanes

Scheme 7.13

Further electrophilic ring closure reactions of [2.2]metacyclophanes

who have investigated a large number of substituted [2.2]metacyclophanes. For example, the bromination of 8,16-dimethyl[2.2]metacyclophane (47) under similar conditions as for the parent hydrocarbon afforded the isomerization – transannular cyclization – polysubstitution product 2,7-dimethyl-1,3,6,8-tetrabromo-4,5,9,10-tetrahydropyrene (48, Scheme 7.13) [32]. That especially the latter two processes are supported by the iron powder (or the Lewis acid derived therefrom) is borne out by the analogous reaction of the 8,16-diethyl derivative 49, which in the absence of the metal (salt) leads to 50 in good yield, whereas in its presence the cyclized bromides 51 and 52 are generated (Scheme 7.13) [33, 34].

7.3

The Pseudo-gem Effect

The pseudo-gem effect is one of the most important and significant intramolecular effects in paracyclophane chemistry. It was first described by Cram and Reich in 1969 in a series of legendary publications which still belong to the highlights of cyclophane literature [35]. It will be presented here for the bromination of the monoester 53, which on treatment with bromine/iron powder in dichloromethane furnishes the pseudo-gem bromide 54 in excellent yield (82%) practically exclusively, the isomeric esters being produced in < 2% yield. The authors explained

7.3 The Pseudo-gem Effect

this high regioselectivity by postulating the formation of the r-complex 55 in which the externally attacking bromine has displaced the originally aromatic hydrogen atom towards the most basic oxygen atom of the ester group. This is oriented in optimal position for a slow intramolecular proton transfer step. The deprotonation of the thus generated 56 does not play a role in the rate-limiting step anymore. Of course, other r-complexes than 55 could also be produced, but none can exploit the advantage of the interdeck “delivery process”. Generalization then showed that the bromination of methoxycarbonyl, carboxy, acetyl and nitro-[2.2]paracyclophanes gave the pseudo-gem derivatives in all but the last case, essentially exclusively. But even with the nitro group the other isomers were produced in the % range only: pseudo-ortho: 3%, pseudo-para: 6%, and pseudo-

Scheme 7.14

The pseudo-gem effect

Scheme 7.15

Synthesis of [2.2.2](1,2,4)cyclophane

199

200

7 Intramolecular Reactions in Cyclophanes

meta: 8%, whereas the geminally-substituted product was isolated in 70% yield. When, in contrast, 4-bromo-[2.2]paracyclophane was nitrated or the cyano compound brominated little transannular directive influence was visible [35]. Since, furthermore, the pseudo-gem derivatives can be converted into the pseudo-meta-isomers by simple heating to 200 8C in good yield this substitution pattern is also available readily. The importance of the pseudo-gem directing effect is threefold. It can first be used to introduce additional bridges into the [2.2]paracyclophane nucleus. This is shown in exemplary fashion for the preparation of [23](1,2,4)cyclophane (Scheme 7.15) [36, 37].

Scheme 7.16

The synthesis of various pseudo-geminally substituted [2.2]paracyclophanes

7.4 Intramolecular Reaction between Functional Groups in Cyclophanes

Chloromethylation of the ketone 57 leads to the pseudo-geminally substituted derivative 58 exclusively. When this is subjected to the routine steps shown in Scheme 7.15 it is transformed to the triply-bridged hydrocarbon 59. Comparable transformations also played a role in the synthesis of other multibridged cyclophanes, the preparation of which has been reviewed several times and must not be repeated here [38–40]. Secondly, this method of stereocontrol can be employed for the directed synthesis of cyclophanes otherwise difficult to prepare, as demonstrated for the pseudogem diamine 64 in Scheme 7.16. Rieche formylation of 53 first yielded the pseudo-gem aldehyde ester 60, which was subsequently oxidized to the diacid 61. When this was subjected to Curtius degradation, the stable bis-isocyanate 63 was obtained via the azide 62. The versatile intermediate 63 furnished the desired diamine 64 on hydrolysis and could, inter alia, be trapped as the crown ether 65 [41]. Thirdly the pseudo-gem effect plays a crucial role in the synthesis of multifunctionalized chiral [2.2]paracyclophane derivatives, compounds that are of increasing importance as ligands in stereoselective synthesis. Since a whole chapter of this monograph is devoted to these novel chiral inductors (see Chapter 17) only a few of the leading references are listed here [42].

7.4

Intramolecular Reaction between Functional Groups in Cyclophanes

Cyclophanes are ideal model systems to explore the question of whether and how functionalized groups can interact and eventually react with each other. As illustrated in Scheme 7.17 in structure 66 for a pseudo-gem arrangement, the distance between the two functional groups F1 and F2 can be deliberately adjusted by altering the intramolecular distances d1 and d2 by changes in the length and type of the molecular bridges. Furthermore, by selecting different anchor points for F1 and F2, as in the pseudo-ortho situation 67, the dihedral angle between them can be varied. Since one is not restricted to the use of benzene rings as aromatic subunits – naphthaleno, phenanthreno, pyreno and many other aromatic and heteroaromatic cyclophanes are known [5] – in principle F1 and F2 can be positioned in three-dimensional space in nearly any arrangement with respect to each other. Although other core

Scheme 7.17

Cyclophanes as molecular workbenches

201

202

7 Intramolecular Reactions in Cyclophanes

Scheme 7.18

Topochemical reaction control in solution

units, e.g. cage hydrocarbons such as cubane or adamantane, could also be used for such purposes, none of them is so readily available and can be varied over such a broad range as the cyclophanes, making them unique as “molecular work benches” [43]. In fact, the potential of this system for stereoselection and stereocontrol has hardly been tapped so far, and most reported reactions deal with pseudo-gem-[2.2]paracyclophanes. Just two examples will be discussed here. Treatment of the already mentioned diamine 64 with cinnamoyl chloride provided the bis-amide 68 (Scheme 7.18). When this was irradiated in acetone with a 150 W medium-pressure mercury lamp it underwent an intramolecular [2 + 2]photoaddition to provide the cyclobutane derivative 69 in 76% yield [41] saponification of which yielded b-truxinic acid 70 in practically quantitative yield (98%), the spacer molecule 64 also being recovered quantitatively. It should be noted that photolysis of trans-cinnamic acid in solution largely leads to trans/cis-photoisomerization. Good yields of the photodimer 70 are only produced when the so-called b-modification of trans-cinnamic acid, in which two molecules of the substrate are hold in head-to-head orientation at an intermolecular distance of 3.6 Å by the crystal lattice, is photolyzed in the solid state [44]. Clearly, in 68 the paracyclophane linker is functioning as a substitute for the crystal lattice and hence the description of the process as “topochemical reaction control in solution” [45] is justified. That the controlling influence of the “cyclophane lattice” can be extended beyond the first double bond was demonstrated with the series of vinylogous diesters 71 a–c (Scheme 7.19) [45–47]. With n = 1 72 a was not only produced in quantitative yield, but also with the highest quantum yield ever observed (0.8) for a photodimerization of a cinnamic

7.4 Intramolecular Reaction between Functional Groups in Cyclophanes

Scheme 7.19

Ladderanes from pseudo-geminally substituted [2.2]paracyclophanes

acid derivative [46]. Whereas both 71 b and c photo-close readily – in the latter case the ladderane 73 [48] with its characteristic step-like structure (as shown by X-ray structural analysis) is produced – the control effect of the [2.2]paracyclophane unit seems to break down from n = 3 on. Under no conditions could the expected ladderane 72 d be isolated. According to NMR spectral evidence the photoproducts obtained still contained olefinic double bonds, indicating that either the fully saturated ladderane had not been generated or that it is reverting to only partially closed structures under the reaction/work-up conditions [49]. Since 72 d would contain seven annulated cyclobutane rings it would be highly strained and hence show a tendency to ring-open again. Similar photochemical ring closure reactions have recently been described by Nishimura and coworkers for various dimethoxy-[2.n]metacyclophanes [50]. Other cases in which two functional groups in a [2.2]para- or [2.2]metacyclophane are close enough in space to undergo an intramolecular reaction include the diketone 75 which is obtained when the diol precursor 74 is subjected to PCC oxidation (Scheme 7.20) [51]. McMurry coupling readily induces bridge formation to 76, and since this is a cis-stilbene derivative its oxidative photocyclization to the phenanthrenophane 77 comes as no surprise. In the metacyclophane series Vögtle and coworkers have achieved a simultaneous ring contraction/bridge-formation process for the [3.3]metacyclophane derivative 78. This loses carbon monoxide and a formal fragment C2H6O under flash vacuum pyrolysis conditions and provides the triple-bridged cyclophane 79 (15%) [52]. The construction of [n.1]- or even [1.1]paracyclophanes has been difficult and has required multistep approaches during which the extremely high strain is introduced in small portions [53]. It appears that a cyclophane unit can cope better with an oxa than with a methano bridge. Other examples supporting this view are

203

204

7 Intramolecular Reactions in Cyclophanes

Scheme 7.20

Intramolecular bridge formation

provided by the ether 81 which is produced from the bis-phenol 80 in good yield on treatment with AlCl3/CH3NO2 in benzene (Scheme 7.21) [54]. Another example reported by the same group [55] involves the nitration of 5,13di-tert-butyl-6,8-dimethoxy[2.2]metacyclophane (82) which with various nitration reagents yielded the 8,16-epoxy[2.2]metacyclophane 83, for whose formation not only an intramolecular condensation must have taken place but also an anti ? syn-ring inversion process. The main products of the reaction are formed by ipso-substitution, not surprising in view of the presence of tert-butyl substituents in the substrate. In principle a very attractive route to incorporate an additional interdeck bridge into a cyclophane precursor consists in its reaction with a bifunctional reagent. In practice this approach has rarely been employed, one successful example being the formation of the [2.2.2]paracyclophanketone 86 by treatment of the parent hydrocarbon 84 with oxalyl chloride in the presence of aluminum trichloride. The latter two reagents are known to generate phosgene in situ which then attacks 84 to produce the intermediate acid chloride 85 (Scheme 7.22) [56]. That the bridge building could be completed in this case presumably has to do with the prismoid structure of 84 (and hence 85) which brings an unsubstituted benzene ring into close vicinity to the acid chloride function.

7.4 Intramolecular Reaction between Functional Groups in Cyclophanes

Scheme 7.21

Oxabridged cyclophanes

Scheme 7.22

Bridge formation with a bifunctional reagent

A fascinating interplay of inter- and intramolecular [2 + 2]photoaddition reactions in different di- and oligo-vinylarenes has been uncovered by Nishimura and his students [57]. As far as cyclophane chemistry is concerned the examples summarized in Scheme 7.23 are particularly instructive. Irradiation of 1,3,5-trivinylbenzene (87) under high dilution conditions in benzene through a Pyrex filter for 63 h furnished the cyclobutane-bridged cyclophanes 88 and 89 in 1% yield in a ratio of 24 : 76, i.e. a nearly statistical yield [58]. Although different routes to these paddlanes are obviously conceivable, the ones shown in the scheme certainly belong to the network connecting 87 and its dimers. In a first step the cyclobutane derivative 90 is produced which subsequently could close to 91 by a second [2 + 2]cycloaddition. From here on one is dealing with intramolecular cyclophane chemistry and from conformation 91 the triplybridged hydrocarbon 88 could be produced. Alternatively, if the vinyl groups in 91

205

206

7 Intramolecular Reactions in Cyclophanes

Scheme 7.23

From vinylarenes to multibridged cyclophanes

rotate into conformation 92 this could cycloadd to the less symmetrical dimerization product 89. Although the process so far could not be extended to hexavinylbenzene (which in principle could dimerize to a new superphane, see Chapter 4), 1,2,4,5-tetravinylbenzene (93) was dimerized to the tetrabridged cyclophane 94, admittedly in very low yield (0.6%) but under generation of a very high degree of structural complexity in a one-pot process from a very simple precursor [59]. That this approach to novel cyclophane frameworks is general was proven by the synthesis of many other polybridged cyclophanes incorporating a large variety of aromatic subunits [60]. If, finally, the vinyl arene is incorporated into the ring structure of a [18]annulene, as was done by Meier and coworkers for a large number of benzannulated derivatives of these non-benzenoid aromatics [61], a most remarkable photodimerization could be induced. As shown in Scheme 7.24, derivative 95 (the alkoxy groups are required to increase the solubility) on irradiation with a medium pressure mercury lamp dimerized in excellent yield to the belt-like cyclophane 96 (see Chapter 13). Clearly, this must be a stepwise process as well, with one cyclobutane ring being closed after another. However, as can be concluded from the high yield, there must be attractive forces between the phenanthrene units of 95 right from the very beginning.

7.5 Conclusions

Scheme 7.24

From [18]annulenes to belt-structered cyclophanes

7.5

Conclusions

Intramolecular reactions in cyclophanes – especially in [2.2]para- and [2.2]metacyclophanes – can take place directly between the aromatic subunits. The resulting highly-strained cage compounds are of interest for mechanistic and structural reasons and often cannot be prepared by other routes. The most important intramolecular effect of a functional group in a [2.2]paracyclophane is the pseudo-gem effect. This stereo-controlling influence has been employed for the preparation of countless cyclophane derivatives, compounds that are now becoming of increasing importance in stereo-selective synthesis. Furthermore, the pseudo-gem effect has been applied for the introduction of additional ethano bridges into the [2.2]paracyclophane core. Finally, the rigid, layered cyclophane unit can influence the reactivity of functional groups that are relatively far away from this control ele-

207

208

7 Intramolecular Reactions in Cyclophanes

ment as demonstrated by the phenomenon of topochemical reaction control in solution. As frequent reference of intramolecular effects to other areas of cyclophane chemistry in this chapter shows, these proximity effects indeed play an important role in modern cyclophane chemistry.

7.5

References 1

2

3

4

5

6 7

8

For a review and leading references see H. Hopf, Classics in Hydrocarbon Chemistry, Wiley-VCH, Weinheim, 2000, Ch. 12.3, 337–378. Although the benzene rings are deformed (boat-like) their spectral data and many of their chemical properties still indicate their aromatic nature. In fact, 1 is not completely rigid, its benzene rings rotate disrotatorily about the axis that passes through their respective centers even at very low temperatures, the rotational barrier having been calculated as 2.5 kJ mol–1 D. Henseler, G. Hohlneicher, J. Phys. Chem. A 1998, 102, 10828–10833. (a) P. G. Jones, P. Bubenitschek, H. Hopf, Z. Pechlivanidis, Z. Kristallogr. 1993, 208, 136–138; (b) P. G. Jones, P. Bubenitschek, H. Hopf, B. Kaiser, Z. Kristallogr. 1995, 210, 548–549; (c) P. G. Jones, P. Bubenitschek, H. Hopf, B. Kaiser, Z. Kristallogr. 1995, 210, 550–551; (d) P. G. Jones, J. Hillmer, H. Hopf, Acta Cryst. 2002, C58, 301–304. A selection from the cyclophane review literature: (a) P. M. Keehn, S. M. Rosenfeld (eds.), Cyclophanes I, II, Academic Press, New York, 1983; (b) F. Diederich, Cyclophanes, Royal Society of Chemistry, London, 1991; (c) F. Vögtle, Cyclophane Chemistry, J. Wiley and Sons, Chichester, 1993; (d) G. J. Bodwell, Angew. Chem., 1996, 108, 2221–2224; Angew. Chem. Int. Ed., 1996, 35, 2085–2088. H. H. Wasserman, P. M. Keehn, J. Am. Chem. Soc. 1967, 89, 2770–2772. R. Gleiter, K. Staub, H. Irngartinger, T. Oeser, J. Org. Chem. 1997, 62, 7644– 7649. J. H. Golden, J. Chem. Soc. 1961, 3741– 3748.

9

10

11

12

13

14

15

(a) H. Okamoto, H. P. J. M. Dekkers, K. Satake, M. Kimura, J. Chem. Soc. Chem. Commun. 1998, 1049–1050; (b) M. Usui, T. Nishiwaki, K. Anda, M. Hida, Chem. Lett. 1984, 1561–1564. Y. Sakamoto, T. Kumagai, K. Matohara, C. Lim, T. Shinmyozu, Tetrahedron Lett. 1999, 40, 919–922. K. Matohara, C. Lim, M. Yasutake, R. Nogita, T. Koga, Y. Sakamto, T. Shinmyozu, Tetrahedron Lett., 2000, 41, 6803– 6807. Ionic additions to multibridged [2n]cyclophanes have already been carried out by Boekelheide and co-workers, who also were unable to isolate hexaprismane derivatives. However, by reducing the monochloride obtained by AlCl3/HCltreatment of [23](1,3,5)cyclophane with Li/tert-BuOH they were able to prepare a cage hydrocarbon in which two cyclohexane rings are bridged by ethano bridges in the former 1-, 3-, and 5-positions, whereas the remaining, originally unsubstituted benzene positions are now connected directly (by “zero-bridges”): V. Boekelheide, R. A. Hollins, J. Am. Chem. Soc. 1973, 95, 3201–3208. C. Lim, M. Yasutake, T. Shinmyozu, Angew. Chem., 2000, 112, 593–594; Angew. Chem. Int. Ed. 2000, 39, 578–580. (a) C. Lim, M. Yasutake, T. Shinmyozu, Tetrahedron Lett. 1999, 40, 6781–6784; (b) W. Senton, T. Saton, M. Yasutake, C. Lim, T. Shinmyozu, Eur. J. Org. Chem. 1999, 1223–1231. M. Wollenweber, M. Etzkorn, J. Reinbold, T. Wahl, T. Voss, J.-P. Melder, C. Grund, R. Pinkos, D. Hunkler, M. Keller, J. Woerth, L. Knothe, H. Prinzbach, Eur. J. Org. Chem. 2000, 3855–3886. H. Hopf, P. Blickle, Tetrahedron Lett. 1978, 449 – 452 and unpublished work.

7.5 References 16

17

18

19 20

21 22

23 24

25

26

27

H. Hopf, J. Kleinschroth, A. E. Mourad, Angew. Chem. 1980, 92, 388–389; Angew. Chem. Int. Ed. Engl. 1980, 19, 389–390. (a) N. Mori, M. Horiki, H. Akimoto, J. Am. Chem. Soc. 1992, 114, 7927–7928; (b) N. Mori, T. Takemura, U. Tsuchiya, J. Chem. Soc. Chem. Commun., 1988, 575–576. H. Matsuzawa, K. Kozawa, T. Uchida, H. Akimoto, N. Mori, Acta Crystallogr. Sect. C 1980, 46, 479–481. Y. Fukazawa, M. Kikuchi, O. Kajita, S. Ito, Tetrahedron Lett. 1976, 17, 4559–4562. (a) D. T. Longone, G. R. Chipman, J. Chem. Soc. Chem. Commun. 1969, 1358– 1359; (b) D. J. Cram, A. C. Day, J. Org. Chem. 1966, 31, 1227–1232. R. Savinsky, H. Hopf, I. Dix, P. G. Jones, Eur. J. Org. Chem. 2001, 4595–4606. H. Hopf, C. Marquard in Strain and Its Implications in Organic Chemistry (A. de Meijere, S. Blechert, eds.), Kluwer Academic Publishers, Dordrecht, 1989, 297– 332.; cf. D. Wullbrandt, PhD Thesis, Braunschweig, 1982. G. J. Bodwell, J. Li, Org. Letters 2002, 4, 127 – 130. G. J. Bodwell, J. Li, Angew. Chem. 2002, 114, 3395–3396; Angew. Chem. Int. Ed. 2002, 41, 3261–3262. D. J. Cram, J. M. Cram, Container Molecules and Their Guests, Royal Society of Chemistry, London, 1992. The following references are only a small selection from a long list of examples: (a) Lithium inclusion compounds: C. Mink, K. Hafner, Tetrahedron Lett. 1984, 39, 4087–4090; (b) Inclusion of inorganic anions: A. C. Benniston, A. Harriman, D. Philp, J. Fraser Stoddart, J. Am. Chem. Soc., 1993, 115, 5298–5299; (c) Inclusion of neutral organic molecules, e.g. pyrene: D. B. Smithrud, T. B. Wyman, F. Diederich, J. Am. Chem. Soc. 1991, 113, 5420–5426; (d) Increase of internal electron density between the rings by radical anion formation: J. Bruhin, U, Buser, F. Gerson, T. Wellauer, Helv. Chim. Acta 1990, 73, 2058–2069. (a) V. Boekelheide, J. B. Phillips, J. Am. Chem. Soc. 1963, 85, 1545–1546; (b) R. H. Mitchell in Advances in Theoretically Interesting Molecules, R. P. Thummel (ed.), JAI Press, Greenwich, 1989, 1, 35–199.

28

29

30 31

32 33

34

35 36

37

R. H. Mitchell, T. R. Ward, Y. Wang, P. W. Dibble, J. Am. Chem. Soc. 1999, 121, 2601–2602 and refs. cited therein. (a) R. H. Mitchell, V. Boekelheide, J. Am. Chem. Soc. 1970, 92, 3510–3512; (b) V. Boekelheide, R. H. Mitchell, J. Am. Chem. Soc. 1974, 96, 1547–1557. G. J. Bodwell, J. N. Bridson, S.-L. Chen, J. Li, Eur. J. Org. Chem. 2002, 243–249. (a) T. Sato, M. Wakabayashi, Y. Okamura, T. Amada, K. Hata, Bull. Chem. Soc. Jpn. 1967, 40, 2363 – 2365; (b) T. Sato, K. Nishiyama, J. Org. Chem. 1972, 37, 3254–3260. M. Tashiro, T. Yamato, J. Org. Chem. 1981, 46, 1543–1552. (a) T. Yamato, A. Miyazawa, M. Tashiro, J. Org. Chem. 1992, 57, 266–270; (b) T. Yamato, S. Ide, K. Takuhisa, M. Tashiro, J. Org. Chem. 1992, 57, 271–275. For the reaction of various tert-butylated derivatives see M. Tashiro, T. Yamato, J. Am. Chem. Soc. 1982, 104, 3701–3707. When these derivatives carry additional functional groups such as methoxy or hydroxy (phenols) sometimes highly unexpected intramolecular products are formed, both with the benzene rings intact and also completely substituted: (a) T. Yamato, J. Matsumota, K. Tokuhisa, K. Suehiro, S. Horie, M. Tashiro, J. Org. Chem. 1992, 57, 6368–6371; (b) T. Yamato, J. Matsumoto, K. Tokuhisa, M. Kojihara, K. Suehiro, M. Tashiro, Chem. Ber. 1992, 125, 2443–2454; (c) T. Yamato, J. Matsumoto, K. Tokuhisa, K. Tsuji, K. Suehira, M. Tashirto, J. Chem. Soc. Perkin I, 1992, 2675–2682; (d) T. Yamato, A. Miyazawa, M. Tashiro, J. Chem. Soc. Perkin I, 1993, 3129–3137. H. J. Reich, D. J. Cram, J. Am. Chem. Soc. 1969, 91, 3505–3516. (a) D. J. Cram, E. A. Truesdale, J. Am. Chem. Soc., 1973, 95, 5825–5827; (b) D. J. Cram, E. A. Truesdale, J. Org. Chem. 1980, 45, 3974–3981; (c) D. J. Cram, R. B. Hornby, E. A. Truesdale, H. J. Reich, M. H. Delton, J. M. Cram, Tetrahedron 1974, 30, 1757–1768. (a) H. Hopf, S. Trampe, K. Menke, Chem. Ber., 1977, 110, 371–372; (b) A. E. Mourad, H. Hopf, Chem. Ber. 1980, 113, 2358–2371.

209

210

7 Intramolecular Reactions in Cyclophanes 38 39

40

41 42

43

44

45

46

47

H. Hopf, Naturwiss., 1983, 70, 348–358. V. Boekelheide in Strategies and Tactics in Organic Synthesis, Th. Lindberg (ed.), Academic Press, New York, 1984, 1–19. H. Hopf in P. M. Keehn, S. Rosenfeld (eds.), The Cyclophanes II, Academic Press, New York, 1983, 521–572. H. Zitt, I. Dix, H. Hopf, P. G. Jones, Eur. J. Org. Chem. 2002, 2298–2307. (a) H. J. Reich, K. E. Yelm, J. Org. Chem. 1991, 56, 5672–5679; (b) A. Pelter, R. A. N. C. Crump, H. Kidwell, Tetrahedron Lett., 1996, 37, 1273–1276; (c) A. Pelter, B. Mootoo, A. Maxwell, A. Reid, Tetrahedron Lett., 2001, 42, 8391– 8394; (d) X.-W. Wu, X.-L. Hou, L.-X. Dai, J. Tao, B.-X. Cao, J. Sun, Tetrahedron: Asymmetry, 2001, 12, 529–532; (e) V. I. Rozenberg, D. Yu Antonov, E. V. Sergeeva, E. V. Vorontsov, Z. A. Starikova, I. V. Fedyanin, C. Schulz, H. Hopf, Eur. J. Org. Chem., 2003, 2056–2061. (a) H. Hopf, K.-L. Noble, L. Ernst, Chem. Ber., 1984, 117, 474 – 488; (b) J. Mulzer, K. Schein, J.-W. Bats, J. Buschmann, P. Luger, Angew. Chem., 1998, 110, 1625–1628; Angew. Chem. Int. Ed. 1998, 37, 1566–1569; (c) J. Mulzer, I. Böhm, J.-W. Bats, Tetrahedron Lett. 1998, 39, 9643–9646. (a) M. D. Cohen, G. M. J. Schmidt, J. Chem. Soc., 1964, 1996–2000; (b) G. M. J. Schmidt, J. Chem. Soc., 1964, 2014–2021; (c) For a collection of papers by G. M. J. Schmidt and his collaborators see Solid State Photochemistry (D. Ginsburg, ed.), Verlag Chemie, Weinheim, 1976. H. Hopf, H. Greiving, P. G. Jones, P. Bubenitschek, Angew. Chem. 1995, 107, 742–744; Angew. Chem. Int. Ed. 1995, 34, 685–687. H. Greiving, H. Hopf, P. G. Jones, P. Bubenitschek, J. P. Desvergne, H. Bouas-Laurent, Liebigs Ann. 1995, 1949– 1956. H. Greiving, H. Hopf, P. G. Jones, J.-P. Desvergne, H. Bouas-Laurent, J. Chem. Soc. Chem. Commun., 1994, 1075–1076.

48

49

50 51 52

53

54

55 56

57

58 59

60

61

H. Hopf, Angew. Chem. 2003, 115, 2928– 2931; Angew. Chem. Int. Ed. 2003, 42, 2822–2825. H. Hopf, L. Bondarenko, unpublished work; cf. Chr. Beck, PhD dissertation, Braunschweig, 1999. Y. Okada, M. Kaneko, J. Nishumura, Tetrahedron Lett. 2001, 42, 1919–1921. H. Hopf, C. Mlynek, J. Org. Chem. 1990, 55, 1361–1363. J. Breitenbach, F. Ott, M. Nieger, F. Vögtle, Chem. Ber. 1992, 125, 1283– 1285. T. Tsuji, M. Ohkita, T. Konno, S. Nishida, J. Am. Chem. Soc. 1997, 119, 8425–8431. T. Yamato, J. Matsumota, K. Tokuhisa, M. Kajihara, K. Suehiro, M. Tashiro, Chem. Ber. 1992, 125, 2443–2454. T. Yamato, H. Kamimura, T. Furukawa, J. Org. Chem. 1997, 62, 7560–7564. F. R. Heirtzler, H. Hopf, P. G. Jones, P. Bubenitschek, V. Lehne, J. Org. Chem. 1993, 58, 2781–2784. For a review see: J. Nishimura, Y. Nakamura, Y. Hayashida, T. Kudo, Acc. Chem. Res., 2000, 33, 679–686. Y. Wada, T. Ishimura, J. Nishimura, Chem. Ber. 1992, 125, 2155–2157. Y. Nakamura, Y. Hayashida, Y. Wada, J. Nishimura, Tetrahedron, 1997, 53, 4593– 4600. (a) K. Nakanishi, K. Mizuno, Y. Otsuji, J. Chem. Soc., Perkin I, 1990, 3362–3363; (b) Y. Nakamura, T. Tsuihiji, T. Mita, T. Minowa, S. Tobita, H. Shizuka, J. Nishimura, J. Am. Chem. Soc., 1996, 118, 1006–1012; (c) Y. Nakamura, M. Kaneko, K. Tani, T. Shinmyozu, J. Nishimura, J. Org. Chem. 2002, 67, 8706– 8709. (a) H. Meier, K. Mueller, Angew. Chem. 1995, 107, 1598–1600; Angew. Chem. Int. Ed. 1995, 34, 1437–1439; (b) K. Mueller, H. Meier, H. Bouas-Laurent, J. P. Desvergne, J. Org. Chem. 1996, 61, 5474– 5480.

229

9

X-ray Crystal Structures of Porphyrinophanes as Model Compounds for Photoinduced Electron Transfer Hermann Irngartinger and Thomas Oeser

9.1

Introduction

Heme proteins containing the porphyrin system have a variety of natural functions: electron transport, oxygen transport and storage, oxygen reduction, hydrocarbon oxidation, and catalytic processes. Therefore tremendous efforts have been undertaken to investigate these processes in nature and in artificial model systems [1–5]. Photoinduced electron transfer reactions of porphyrin–quinone systems are of fundamental importance in the initial steps of biological photosynthesis in plants and bacteria [6–10]. The structure of the natural reaction center for photosynthesis has been determined [11]. Many molecular systems have been synthesized in which a porphyrin and a quinone or a similar electron acceptor group is linked together by a flexible chain [1–5, 12–25]. If the two components are connected together to a cyclophane type molecule the possibilities of motion are more restricted. The distance between the two groups and their mutual orientation are very important quantities for the electron transfer rate. They can be modeled in cyclophane-type molecules very easily. Many porphyrinophanes with quinone or aromatic acceptor groups have been synthesized in recent years [12–37]. In this article we give a review of the X-ray structure determinations of porphyrin–quinone cyclophanes and their Zn complexes, which were synthesized as biomimetic models to investigate photoinduced electron transfer processes. Because of their very close structural similarity we also included the porphyrinophanes which have planar aromatic groups as electron acceptor components instead of quinones. We have omitted iron and other metal complexes which are constructed specifically to model the other functions of the porphyrin system mentioned above.

Modern Cyclophane Chemistry. Edited by Rolf Gleiter, Henning Hopf Copyright © 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 3-527-30713-3

230

9 X-ray Crystal Structures of Porphyrinophanes as Model Compounds for Photoinduced Electron Transfer

9.2

Porphyrin–Quinone Cyclophanes 9.2.1

Double-Bridged Porphyrin–Quinone Cyclophanes

Porphyrin and quinone units should be linked in a rigid face-to-face arrangement so that the vertical parameters on which the electron transfer depends can be studied. According to molecular models the quinone–porphyrin–quinone units in the model cyclophane 1 are expected to have a rigid parallel face-to-face arrangement.

1

Staab et al. [38–42] synthesized compound 1, determined the electron transferrelated properties, and investigated the X-ray crystal structure analysis [38, 40]. According to the molecular S4 symmetry in the crystal, both quinone rings and the least squares plane of the porphyrin system are in fact exactly parallel to each other (Fig. 9.1). As can be seen from the top-view of the molecule (Fig. 9.1A) the centers of the three rings are exactly superimposed along the S4 axis. The interplanar distance of the porphyrin system to the quinone rings on both sides is 342 pm. As the side view (Fig. 9.1B) indicates the porphyrin shows considerable deviations from planarity. The four pyrrole rings have an inclination angle of 9 8 against the porphyrin plane. The pyrrole rings themselves are nearly planar. The maximum deviation of the five-membered rings is in the order of 1 pm. The phenyl rings substituted in the meso-position of the porphyrin system are twisted by 67 8 against this plane. The tetramethylene links have energetically favorable antiperiplanar conformations in the crystal, which are also assumed for the molecules in solution as is indicated by the 1H NMR data [38]. Nevertheless a significant flexibility of the molecule has to be anticipated. Therefore the rigid S4 molecular symmetry in the solid state is determined by packing effects. The molecules of 1 are stacked in the crystal along the fourfold rotoinversion axis (c-axis) which is

9.2 Porphyrin–Quinone Cyclophanes

1 Fig. 9.1

Molecular structure of 1: A top-view and B side view [111]

Fig. 9.2

Crystal packing of 1

1

perpendicular to the ring planes (Fig. 9.2). As a result of this symmetry the quinone rings of two neighboring molecules facing each other are rotated by 90 8. The intermolecular distance of 337 pm between the quinone units is only slightly shorter than the intramolecular distance between the quinone planes to the porphyrin plane.

231

232

9 X-ray Crystal Structures of Porphyrinophanes as Model Compounds for Photoinduced Electron Transfer

9.2.2

Single-Bridged Porphyrin–Quinone Cyclophanes 9.2.2.1 Phenyl-Spacered Porphyrin–Quinone Cyclophanes

In the course of the program of controlled variation of structural parameters affecting the electron transfer rate, the singly-bridged porphyrin–quinone cyclophanes 2 and 3 were synthesized and investigated according to their spectroscopic

[43, 44] and structural properties [45–47]. The quinone ring in compound 2 is substituted by two methoxy groups to vary the acceptor strength of this unit.

2 Molecular structure of 2: A top view onto the mean plane of the porphyrin and B side view

Fig. 9.3

9.2 Porphyrin–Quinone Cyclophanes

In contrast to the symmetrical structure of 1 the quinone rings in the cyclophanes 2 and 3 are now tilted by an angle of 26 8 and 44 8 respectively against the porphyrin plane (Figs. 9.3 and 9.4). The larger inclination angle in cyclophane 3 is obviously a result of the steric requirements of the two phenyl substituents in the remaining meso-positions of the porphyrin. In the projection onto the porphyrin plane the center of the quinone unit is shifted from the center of the porphyrin system by 155 and 160 pm for compound 2 and 3 respectively (Figs. 9.3 A and 9.4 A). The center-to-center distances from the quinone to the porphyrin ring system, at 474 and 432 pm in 2 and 3 respectively, are considerably longer than the interplanar distance of 342 pm between the parallel porphyrin and quinone planes in 1. The porphyrin system of derivative 3 is nearly planar. The maximum deviations from the mean porphyrin plane are 0.4 pm for the nitrogen atoms and 28 pm for the carbon atoms [46]. In compound 2 however the pyrrole subunits are twisted by 9 8 from the mean porphyrin plane not far from D2 symmetry (Fig. 9.3 B). The phenyl rings in cyclophane 2 deviate only by a small amount of 10 8 from a perpendicular orientation to the porphyrin plane. The corresponding deviations in 3 are 9 8 and 21 8 resp. The two phenyl rings in 3 not involved in the bridge are twisted by 69 8 and 62 8 respectively against the porphyrin plane. The characteristic feature of the packing arrangement of both porphyrinophanes 2

3

Fig. 9.4 Molecular structure of 3: A top view onto the porphyrin plane and B side view

233

234

9 X-ray Crystal Structures of Porphyrinophanes as Model Compounds for Photoinduced Electron Transfer Fig. 9.5 Packing arrangement of compound 3

3

and 3 is the face-to-face orientation of the porphyrin planes of two neighboring molecules at a distance of 360 and 352 pm, respectively, related by a center of symmetry. A second aspect determining the packing arrangement of 3 is the antiparallel orientation of two quinoid carbonyl groups at an intermolecular C···O' distance of 306.3 pm (Fig. 9.5) [46]. At first glance the molecular structure of the porphyrin–quinone cyclophane 4 [48] (Fig. 9.6) which is now a Zn-complex, looks similar to those of the corresponding metal-free cyclophanes 2 and 3.

4

The quinone ring is neither parallel nor centered with regard to the porphyrin system. However the inclination angle of 40 8 between the porphyrin and the quinone plane is found to be considerably larger than in the analogous metal-free porphyrin derivative 2 (26 8). The reason for this strong tilting is found in the orientation of one quinoid carbonyl group into the direction of the central Zn atom of compound 4. The oxygen atom of this group is placed exactly above the zinc with a distance of only 253.2(2) pm. The comparable distance of this oxygen atom to the center of the porphyrin system of 2 is 429 pm. The close Zn···O distance in 4 is an indication for attractive coordination which is supported by the fact that the concerned C=O group is tilted by 6 8 from the quinone plane into the direction to Zn, and Zn itself deviates for 4 pm towards the oxygen, out of the mean plane through the four nitrogen atoms of the porphyrin. This remarkable structural difference between the metal-free porphyrinophanes 2 and 3 compared with the Zn complex 4 should have consequences for the mechanism of the electron transfer process (see Chapter 9.4).

9.2 Porphyrin–Quinone Cyclophanes

4 Molecular structure of 4: A side view and B top view onto the mean plane of the porphyrin

Fig. 9.6

9.2.2.2 Naphthalene-Spacered Porphyrin–Quinone Cyclophanes

Linked by Covalent Bonds In order to investigate the dependence of the electron transfer rates on donor–acceptor distances the length of the bridge in the cyclophanes was systematically varied. By replacing the phenyl groups in compounds like 2–4 by 1-naphthyl groups the quinones should be shifted to longer porphyrin–quinone distances. Therefore Staab et al. [49] synthesized naphthalene-spacered porphyrin–quinone cyclophanes and determined the structure of the corresponding derivative 5 by Xray diffraction (Fig. 9.7) [50].

5

235

236

9 X-ray Crystal Structures of Porphyrinophanes as Model Compounds for Photoinduced Electron Transfer

5 Fig. 9.7

Molecular structure of 5: side view

As found for the phenyl-spacered analogous porphyrinophane 2, of the two tetramethylene chains on both sides of the central quinone ring, one chain is inclined downwards towards the porphyrin, whereas the other is directed upwards away from the porphyrin plane. This results in an inclination of the quinone ring to the porphyrin plane of 5 by 56 8 (Fig. 9.7). The corresponding inclination in the phenyl-spacered analog 2 amounts to less than a half of that value. The vertical distance from the center of the quinone ring to the mean plane of the porphyrin is 503 pm in compound 5, whereas the corresponding vertical distance in the phenyl-spacered cyclophane 2 is only 448 pm [45]. Thus, in the naphthalene-spacered analog an increased porphyrin– quinone distance is indeed achieved, although this is not as much as had been anticipated for a parallel orientation of the quinone and porphyrin planes. Host–Guest Complex The two components of the porphyrin–quinone cyclophanes described so far are linked by covalent bonds. Hayashi, Ogoshi et al. [51] determined the structure of a host–guest complex 6 of meso-a,a,a,a-tetrakis(2-hydroxy-1-naphthyl)-porphyrin and tetramethoxyquinone joined together by hydrogen bonds and noncovalent weak interactions.

6

9.2 Porphyrin–Quinone Cyclophanes

6 Molecular structure of host-guest complex 6: A top view and B side view Fig. 9.8

The quinone fits exactly into the pocket of the porphyrin. All the hydroxy groups at the meso-naphthyl substituents of the porphyrin core are hydrogenbonded to the carbonyl and methoxy groups of the quinone (Fig. 9.8). The crystallographic twofold rotation axis C2 has a perpendicular orientation to the ring planes of both complex components and passes through their centers. Consequently the two ring systems are coplanar with a separation distance of 335 pm, which enables effective p–p-interactions. The porphyrin ring is planar within 10–12 pm. Two of the four methoxy groups on the quinone system are oriented upward to avoid steric hindrance with the porphyrin molecules, and the other two MeO groups are coplanar with the quinone ring (Fig. 9.8).

237

238

9 X-ray Crystal Structures of Porphyrinophanes as Model Compounds for Photoinduced Electron Transfer

9.3

Porphyrin-Aromatic-Ring Cyclophanes 9.3.1

Single Bridge from Opposite meso-Positions 9.3.1.1 Substituted Phenyl Rings as Cyclophane Components

Porphyrinophanes which have an aromatic ring system, attached by donor and acceptor substituents, opposite the porphyrin plane instead of a quinone group, are structurally very similar to the model compounds discussed so far. The structures of such pyridinophanes have been determined by Staab et al. for the phenyl-spacered derivatives 7 [45, 47] and 8 [46] respectively and for the naphthyl-spacered derivative 9 [50, 52].

7: X = Y = OMe 8: X = NMe2; Y = H

9

As was found for the corresponding porphyrin–quinone cyclophanes (Chapter 9.2), in these analogous porphyrin–aryl cyclophanes 7–9 one chain is inclined downwards towards the porphyrin whereas the other is directed upwards away from the porphyrin plane. This results in an inclination of the central phenyl ring to the porphyrin plane of the same order as was found for the quinone analogs (Fig. 9.9, Tab. 9.1). In the phenyl-spacered cyclophanes 7 and 8 this inclination angle amounts to less than half the value of that of the naphthalene-spacered cyclophane 9 (Tab. 9.1). In the perpendicular projection onto the porphyrin plane, the center of the phenyl group is shifted towards the center of the porphyrin system by a large amount for the three compounds (Tab. 9.1). These molecular conformations result in center-to-center distances of the two ring systems far beyond van der Waals contacts (Tab. 9.1). Therefore the p–p-interactions are considerably reduced. The atoms of the porphyrin system of 7–9 deviate from planarity by only small amounts. The spacering phenyl and naphthyl groups in the meso-position are almost perpendicular to the porphyrin plane (Tab. 9.1). Two methyl groups of the methoxy substituents in 7 and of the dimethylamino groups in 8 point inwards to the porphyrin plane and the other two groups outwards (Fig. 9.9 A and C). The methoxy groups in 9 are almost coplanar with the phenyl ring (Fig. 9.9 D).

9.3 Porphyrin-Aromatic-Ring Cyclophanes

7

7

9

8 Fig. 9.9

Molecular structures of 7–9: A side view of 7; B top view of 7; C side view of 8; D side

view of 9

Tab. 9.1 Molecular dimensions of 7–9

Compound

Dihedral angle porphyrin/ aromatic plane

Distance (pm) central points porphyrin ··· phenyl ring

Shift (pm) center of porphyrin/center phenyl ring in projection to porphyrin plane

Av. dihedral angle porphyrin/ pyrrole

Av. deviation from perpendicular orientation porphyrin/meso-phenyl or mesonaphthyl plane

7 8 9

24 8 26 8 56 8

502 450 559

225 150 215

58 – 58

58 98 1 8, 9 8

In the molecular structures of 10 and 11 determined by Gunter et al. [53] a hydroquinol ring is linked by crown ether chains to the phenyl-spacered porphyrin. Since the 7- and 10-atom chains of 10 and 11 respectively are considerably longer than the corresponding 4-atom chains in 7–9 there is more space for rotation of the hydroquinol group. Actually in both cases the hydroquinol ring adopts an essentially orthogonal orientation relative to the porphyrin plane with phenyl–porphyrin dihedral angles of 86.7 8 and 83.3 8 respectively (Figs. 9.10 and 9.11).

239

240

9 X-ray Crystal Structures of Porphyrinophanes as Model Compounds for Photoinduced Electron Transfer

10: n = 1 11: n = 2

This edge-to-face geometry has been observed with many other aromatic systems [54–57]. There are close contacts between the hydroquinol C–H and the pelectrons of the porphyrin system in each case. In compound 10 the hydroquinol ring is located almost centrally over the porphyrin and is aligned along an N···N axis (Fig. 9.10), whereas in compound 11 with the longer bridge, the hydroquinol ring is shifted from its central position along the N···N axis (Fig. 9.11). On the other hand an exactly parallel orientation of the phenyl ring and porphyrin plane was observed by Cammidge et al. [58] in the Zn complex of the naphthyl-space-

Fig. 9.10 Molecular structure of 10: A side view; B top view

10

9.3 Porphyrin-Aromatic-Ring Cyclophanes

11 Fig. 9.11 Molecular structure of 11: side view

12

red cyclophane 12 with very short 2-atom chains (Fig. 9.12). In the crystal a vertical twofold rotation axis passes through the centers of both ring systems. Pyromellitic diimide is an effective electron acceptor unit, having a first reduction potential of E1red = –1.37 V. The typical absorption of the anion of the pyromellitic diimide at k&715 nm should make it possible to indicate a charge separation spectroscopically [59–64]. A corresponding sharp absorption is missing for the quinone groups. Therefore electron transfer interactions have been studied with por-

12

Fig. 9.12 Molecular structure of 12: top view

241

242

9 X-ray Crystal Structures of Porphyrinophanes as Model Compounds for Photoinduced Electron Transfer

13

phyrins linked to the pyromellitic diimide unit by several groups [61–64]. Since the components were joined by single conformational flexible chains in those derivatives, Staab et al. [59, 60] synthesized pyromellitic diimide–porphyrin cyclophanes and determined the structure of the derivative 13 by X-ray diffraction. The molecular structure of crystalline 13 differs considerably from the corresponding structures of the porphyrin cyclophanes discussed in this article. The molecule lies on a crystallographic mirror plane vertical to the porphyrin plane. Consequently the tetramethylene chains show mirror symmetry and do not show the one-sided inward bending that was found for many porphyrinophanes. Both tetramethylene chains are bent away from the center of the porphyrin unit resulting in a tilt of the pyromellitic diimide group by an dihedral angle of 69 8 against the porphyrin plane (Fig. 9.13). The vertical projection from the acceptor center meets the donor plane outside the porphyrin cycle at a distance of 497 pm, so that there is negligible intramolecular interaction between the two groups. The reason for this orientation may arise from the long extension of 670 pm of the pyromellitic diimide group which is only 53 pm shorter than the corresponding extension of the porphyrin donor. The corresponding extension of approx. 300 pm for the quinone groups is much shorter. The porphyrin group is planar, the fivemembered rings of the acceptor group are tilted by 7.2 8 against the central ring. Both phenyl rings are oriented almost perpendicular (92 8) to the porphyrin plane.

13

Fig. 9.13 Molecular structure of 13: side view

9.3 Porphyrin-Aromatic-Ring Cyclophanes Fig. 9.14 Packing arrangement of 13: projection of (001)-mirror plane

13

In contrast to the minor intramolecular contact of the donor and acceptor group in 13 the pyromellitic diimide unit is arranged in an almost parallel orientation to a porphyrin plane of a neighboring molecule in the crystal packing of 13 (Fig. 9.14). At an intermolecular distance of about 330 pm between both planes and a partial overlap there is considerable p–p-interaction between the neighboring molecules in the crystal.

9.3.1.2 Polycyclic Aromatic Ring Systems as Cyclophane Components

The size of pyrene compares well with that of the pyromellitic diimide unit. However both amido bridges to the phenyl-spacered porphyrin system in compound 14 are shorter than the bridges in the previous molecule. In addition cyclophane 14 is strapped by a second non-aromatic chain on the other side of the porphyrin plane. As the structure determination conducted by Weiss et al. [65] reveals, the porphyrin core shows a small saddle-shaped distortion, probably because of the double-strapped geometry. The averaged dihedral angles of the pyrrole rings with the porphyrin plane are only 5.8 8. The pyrene plane is displaced slightly off-center and is tilted with respect to the mean porphyrin plane by 27.4 8 (Fig. 9.15). The distance between the centers of the planes is 493 pm.

14

243

244

9 X-ray Crystal Structures of Porphyrinophanes as Model Compounds for Photoinduced Electron Transfer Fig. 9.15 Molecular structure of 14: side

view

14

In the Zn-porphyrin complex 15, the structure of which was determined by Gunter et al. [66], a naphthalene unit is linked by 10-atom polyether chains to the phenyl-spacered porphyrin system. Both long polyether chains leave space for the inclusion of bipyridinium cation (paraquat) generating a triple-decker p-stacking arrangement (Fig. 9.16). The two pyridinium rings in the paraquat unit are twisted by 26 8. The N···N axis of the paraquat deviates by only 2.9 8 from a corresponding N···N axis of the porphyrin system. One of the pyridinium rings of the included paraquat unit forms a dihedral angle of 1 8 with the plane of one of the four pyrrole rings below, and a corresponding angle of 7 8 with one of the sixmembered rings of the naphthalene unit above, forming a p-stacking arrangement with distances of the order of 360 pm (Fig. 9.16).

9.3 Porphyrin-Aromatic-Ring Cyclophanes Fig. 9.16 Molecular structure of the inclusion complex of 15 and paraquat

15

9.3.1.3 Aromatic Heterocycles as Cyclophane Components

Weiss et al. [67] combined heterocyclic aromatic units with the porphyrin system to give the cyclophanes 16 and 17, and determined their molecular structures. In spite of the linking of the pyridine ring in the meta-position, molecule 16 adopts a conformation in which both the porphyrin and the pyridine rings have quasi-parallel orientation. The pyridine ring is only slightly tilted with respect to the por-

16

17

phyrin mean plane, the dihedral angle being 8.2 8. The center-to-center distance of the two planes is 353.3 pm. This short intramolecular separation indicates that the pyridine ring is in van der Waals p–p-contact with the porphyrin ring. In addition to this optimal orientation of the two ring systems, the centroid of the pyridine unit lies approximately on the normal to the porphyrin plane, passing through the centroid of the nitrogen atoms of this unit (Fig. 9.17).

16

Fig. 9.17 Molecular structure of molecule 16: side view

245

246

9 X-ray Crystal Structures of Porphyrinophanes as Model Compounds for Photoinduced Electron Transfer

On the other hand the pyridine rings forming the bis-pyridine strap of 17 are inclined by dihedral angles of 65.4 8 and 108.6 8, respectively, to the porphyrin plane with the nitrogen atoms oriented to the porphyrin core. The distances between the centroids of the pyridine rings and that of the porphyrin system are 629.9 and 673.5 pm, respectively. Weiss et al. investigated the very rigid porphyrinophanes 18 and 18 a which are strapped by a phenanthroline, linked by p-phenylene rings on both sides to the

phenyl-spacered porphyrin system. The structures of the free base 18 [68] and the Zn complexes are shown; the latter have included in the pocket of the molecule: water 18-Zn-W [68], imidazole 18-Zn-I [69], 2-methyl-imidazol 18-Zn-MI [69], 18 aZn-MI [70], 2-phenyl-imidazol 18-Zn-PhI [70], 18 a-Zn-PhI [70] and 2-methyl-benzimidazol 18 a-Zn-MBI [70]. In the free base structure 18 a methanol guest molecule resides in the pocket, fixed by hydrogen bonds to the phenanthroline nitro-

18-Zn-W

Fig. 9.18 Molecular structure of 18-Zn-W

9.3 Porphyrin-Aromatic-Ring Cyclophanes

gen atoms and further C–H···O-interactions. Inside the 18-Zn-W system one H2O molecule is coordinated to Zn and a second is integrated by a hydrogen bond network (Fig. 9.18). In the remaining inclusion complexes, one molecule of imidazole or imidazole derivative is included inside the porphyrinophanes 18 and 18 a, respectively, and coordinated to Zn. In all the structures the phenanthroline system is tilted against the porphyrin unit by 61 8–78 8. The porphyrin subunit of 18 a-Zn-MBI is considerably distorted to a puckered shape. Efficient p–p- and CH···p-interactions contribute to the stabilization of the phenyl–imidazol and 2-methyl–benzimidazol inclusion Zn-complexes of 18-Zn-PhI, 18 a-Zn-PhI and 18 a-Zn-MBI. The molecular structures of 2,2’-substituted biphenyl-strapped meso-diphenylporphyrins 19 a–d have been determined by Weiss et al. [71]. In all these biphenyl-

strapped porphyrins the 2,2’-substituents are oriented toward the porphyrin core. The doming observed for the porphyrin systems in these molecules increases from 19 a to 19 d. The four porphyrin-nitrogen atoms are displaced from the porphyrin plane by 3 to 12 pm away from the straps (Fig. 9.19). The dihedral angles between the phenyl rings forming the biphenyl groups have very different values: 69.3 8 to 128.5 8. The cavity heights of the porphyrinophanes 19, which are defined as the distances between the center of the C–C bond joining both phenyl rings of

19 c

Fig. 9.19 Molecular structure of 19 c

247

248

9 X-ray Crystal Structures of Porphyrinophanes as Model Compounds for Photoinduced Electron Transfer

the biphenyl group and the intersection of the normal to the porphyrin mean plane, are 525 pm on average. In the cavity of a diphenylmethane-capped porphyrin system space is available for an included guest molecule. The Zn-complex 20, the structure of which was determined by Diederich et al. [72], has CH2Cl2 included (Fig. 9.20). The dihedral angles between the phenyl rings of the diphenylmethane group and the plane defined by C(sp2)–C(sp3)–C(sp2) of this central group are 44 8 and 78 8 respectively.

20

9.3.2

Single Bridge from Opposite non-meso-Positions

Most model compounds of porphyrinophanes are bridged from opposite meso-positions (methin carbon atoms) of the porphyrin system. But there are also examples with the bridges fixed at carbon atoms (C2, C12) of two opposite pyrrole

20

Fig. 9.20 Molecular structure of 20

9.3 Porphyrin-Aromatic-Ring Cyclophanes

rings in the porphyrin system. Battersby et al. [73] determined the molecular structures of porphyrinophanes with anthracene 21 and pyridine 22, respectively, linked by two 8-atom bridges to opposite pyrrole rings at C2 and C12 of the porphyrin core. The anthracene and pyridine groups are displaced from symmetric positions and are inclined by 18.4 8 and 6.0 8, respectively, to the porphyrin plane (Fig. 9.21). Carbon atoms of the overlapping part of the anthracene unit in 21 have contacts to the porphyrin plane as close as 307 pm. The corresponding distances in 22 range from 320 to 350 pm.

22

21

Fig. 9.21 Molecular structures of 22 (left) and 21 (right)

9.3.3

Capped Porphyrins

Ibers et al. investigated many capped porphyrinophanes which have a benzene ring linked by four chains to phenyl rings substituted in all meso-positions of a porphyrin. These porphyrinophanes and their Fe and Ru complexes served as model compounds for hemoglobin and myoglobin for probing the structure–function relationship they exhibit [74, 75]. The free bases, however, have the same structural pattern as the model compounds for electron transfer dealt with in this article. Therefore their structure characteristics will be discussed in the following. In the porphyrinophanes 23 a to 23 g [74–79] tetraphenylporphyrin is capped by a benzene ring connected with 4- to 7-atom chains of equal length for the four linking chains in each case (Tab. 9.2). The benzene cap is, to a good approxima-

249

250

9 X-ray Crystal Structures of Porphyrinophanes as Model Compounds for Photoinduced Electron Transfer

Tab. 9.2 Molecular dimensions of “capped” porphyrinophane 23 and “pocket” porphyrinophane

24 Compound Number of atoms in the chain

Center to center separation porphyrin ··· “cap” (pm)

Lateral displacement of the cap from the center of porphyrin (pm)

Dihedral angle porphyrin/ “cap”

Av. displacement from the mean 24 atoms porphyrin plane (pm)

23 a 23 b 23 c 23 d 23 e 23 f 23 g 24

402 381 390 396 474 349 728 421

121 54 110 12 54 20 69 186

7 88 11 8 08 14.3 8 1.7 8 68 12.7 8

7.4 10.2 6.1 13.4 7 27 8 10

4 4 4 5 5 6 7 3

9.3 Porphyrin-Aromatic-Ring Cyclophanes

A 23 a

C 23 g

B 23 d

Fig. 9.22 Molecular structure of 23 a: A side view; 23 d: B top view; 23 g: C side view with included CHCl3

tion, parallel to the porphyrin plane in all cases. The dihedral angle between the benzene ring and the mean plane of the porphyrin varies only from 0 8 to 14.3 8 (Tab. 9.2, Fig. 9.22 A). The separation between these planes is given by the centerto-center distances of these planes from 349 to 474 pm for 23 a–f, resulting in weak p–p-interactions, if any exist at all. Compound 23 g has an exceptionally long separation of 728 pm, since it has a chloroform guest included under the cap occupying two disordered positions (Tab. 9.2, Fig. 9.22 C). In general the benzene caps are only slightly off-centered from the centroid of the porphyrin unit (Tab. 9.2, Fig. 9.22 B). The deviations from planarity of the 24-atom plane of the porphyrin system have only small amounts for compounds 23 (Tab. 9.2). A similar geometry has been found for the “pocket” porphyrinophane 24 [74] in which only three 3-atom chains connect the benzene ring with the porphyrin system. The ortho-position of the fourth phenyl ring is substituted by a pivalamido group extending toward the cap. Compared with the four arms of “capped” por-

251

252

9 X-ray Crystal Structures of Porphyrinophanes as Model Compounds for Photoinduced Electron Transfer

24

phyrinophanes 23, the benzene cap in the “pocket” analog 24 is shifted laterally by a considerable amount from the porphyrin centroid towards the middle arm of the phenyl group 2 (Fig. 9.23, Tab. 9.2).

Fig. 9.23 Molecular structure of the “pocket” porphyrinophane 24: side view

24

The average deviations of the atoms from the 24-atom plane of the porphyrin system in compound 25 [80] at about 29 pm, are somewhat larger than in the previous analogs 23 and 24 (Fig. 9.24). The distance from the center of the four N atoms to the center of the central C–C bond in the strap was calculated to be 928 pm as a rough measure of the height of the cavity of 25.

25

9.4 Porphyrinophanes with Fullerene Hosts Fig. 9.24 Molecular structure of 25

25

9.4

Porphyrinophanes with Fullerene Hosts

Several porphyrinophanes strapped by fullerene groups, as model compounds with back electron transfer in the “Marcus-inverted” regions have been synthesized [81, 82]. Unfortunately no crystal structure analysis has been reported up to now. The structure of the Zn complex 26 of a dimeric porphyrinophane with C60 as included host molecule has been published recently by Saigo et al. [83]. Short distances of 276.5 and 291.8 pm from the Zn to two neighboring C-atoms of C60 suggest the presence of p interactions (Fig. 9.25).

26

253

254

9 X-ray Crystal Structures of Porphyrinophanes as Model Compounds for Photoinduced Electron Transfer Fig. 9.25 Molecular structure of 26 with C60 as inclusion host (side view)

26

9.5

Concluding Remarks

Solid structural parameters are obtained from X-ray structural analysis for molecular model systems. On the basis of these experimental results, conclusions and extrapolations, it is possible to estimate the structure of further model compounds for which no structural analyses exist. The results from X-ray diffraction meet the structural requirements for the analysis of spectroscopic measurements for the porphyrinophanes of the type 1–5, 13 and further derivatives: porphyrinophanes with varied acceptor strength of the quinone subunit [43, 84], Zn complexes of the porphyrinophanes [48], vertically stacked porphyrin–quinone(1)–quinone(2) [85] and porphyrin(1)–porphyrin(2)–quinone triple-decker cyclophanes [86], porphyrin–quinone cyclophanes with varied donor–acceptor distances (benzene-, naphthalene-, biphenylene-, anthracene-spacered cyclophanes) [49, 50, 87, 88], fourfold-bridged porphyrin–quinone “capped” cyclophanes [89], and porphyrinophanes with the strong electron-acceptors 7,7,8,8-tetracyanoquinodimethane [90] and pyromellitic diimide [59, 60] instead of quinones. Time-resolved absorption and emission measurements for those porphyrinophanes were undertaken to determine the photoinduced intramolecular electron transfer rates. These spectroscopic data could be analyzed with the structural results given in this article [91– 98]. The field of intramolecular photoinduced electron transfer problems is still being actively investigated. Fullerenes have recently been introduced as electron acceptor component in porphyrinophanes [81, 82, 99–110]. Corresponding structural analyses are anxiously awaited.

9.6 References

9.6

References 1

2 3 4 5 6

7 8

9

10

11

12

13

14

15

16 17

For review see: B. Morgan, D. Dolphin, in J. W. Buchler (ed.) Structure and Bonding, Vol. 64, Springer, Berlin Heidelberg 1987, p. 113–203. T. G. Traylor, Acc. Chem. Res. 1981, 14, 102–109. J. P. Collman, Acc. Chem. Res. 1977, 10, 265–272. R. D. Jones, D. A. Summerville, F. Basolo, Chem. Rev. 1979, 79, 139–179. P. G. Jene, J. A. Ibers, Inorg. Chem. 2000, 39, 5796–5802 and ref. therein. Review and survey articles: in V. Balzani (ed.) Electron Transfer in Chemistry Vol. 1– 5, Wiley-VCH, Weinheim 2001. Textbook: D.-P. Häder (ed.) Photosynthese, G. Thieme, Stuttgart 1999. Textbook: D. Hall, R. Krishna, Photosynthesis, Cambridge Univ. Press, Cambridge (UK) 1999. Reviews: Photoinduced Electron Transfer I–IV in J. Mattay (ed.) Topics in Current Chemistry Vols. 156, 158, 159 and 163, Springer, Heidelberg 1990–1992. Reviews: in M. A. Fox, M. Chanon (eds.) Photoinduced electron transfer Vol. Part A– D, Elsevier Science Publisher B.V. 1988. J. Deisenhofer, O. Epp, K. Miki, R. Huber, H. Michel, Nature 1986, 318, 618– 624. J. Springer, G. Kodis, L. de la Garza, A. L. Moore, T. A. Moore, D. Gust, J. Phys. Chem. A 2003, 107, 3567–3575 and references therein. D. Gust, T. A. Moore, A. L. Moore, in V. Balzani (ed.) Electron Transfer in Chemistry, Vol. 3, Wiley-VCH, Weinheim 2001, p. 272–336. Porphyrin Handbook: D. Gust, T. A. Moore, Vol. 8, Academic Press, San Diego 2000, p. 153–190. M. O. Senge, M. Speck, A. Wiehe, H. Dieks, S. Aguirre, H. Kurreck, Photochem. Photobiol. 1999, 70, 206–216 and references therein. T. Hayashi, H. Ogoshi, Chem. Soc. Rev. 1997, 26, 355–364. M. D. Ward, Chem. Soc. Rev. 1997, 26, 365–375.

18

19 20 21 22 23

24

25

26

27 28

29

30 31

32

33 34 35

H. Kurreck, M. Huber, Angew. Chem. 1995, 107, 929–947; Angew. Chem. Int. Ed. Engl. 1995, 34, 849–866. D. L. Atkins, C. Guo, Adv. Mat. 1994, 6, 512–516. G. Q. Li, R. Govind, Ind. Eng. Chem. Res. 1994, 33, 755–783. D. Gust, T. A. Moore, A. L. Moore, Acc. Chem. Res. 1993, 26, 198–205. M. R. Wasielewski, Chem. Rev. 1992, 92, 435–461. D. Gust, T. A. Moore, in J. Mattay (ed.), Top. Curr. Chem., Vol. 159, Springer, Berlin Heidelberg 1991, p. 103–151. S. Connolly, J. R. Bolton, in M. A. Fox, M. Chanon (eds.) Photoinduced electron transfer, Vol. Part D, Chap. 6.2, Elsevier Science Publisher B.V. 1988, p. 303–393. M. R. Wasielewski, in M. A. Fox, M. Chanon (eds.) Photoinduced electron transfer, Vol. Part A, Chap. 1.4, Elsevier Science Publisher B.V. 1988, p. 161–206. L. Flamigni, A. M. Talarico, S. Serroni, F. Puntoriero, M. J. Gunter, M. R. Johnston, T. P. Jeynes, Chem. Eur. J. 2003, 2649–2659 and references therein. J. P. Collman, L. Fu, Acc. Chem. Res. 1999, 32, 455–463. E. Kaganer, E. Joselevich, I. Willner, Z. Chen, M. J. Gunter, T. P. Gayners, M. R. Johnson, J. Phys. Chem. B 1998, 102, 1159–1165 and references therein. S. Anderson, H. L. Anderson, J. K. M. Sanders, Acc. Chem. Res. 1993, 26, 469– 475 J. S. Lindsey, T. Chaudhary, B. T. Chait, Anal. Chem. 1992, 64, 2804–2814. M. Gubelmann, A. Harriman, J.-M. Lehn, J. L. Sessler, J. Chem. Phys. 1990, 94, 308–315. J. S. Lindsey, J. K. Delaney, D. C. Mauzerall, H. Linschitz, J. Am. Chem. Soc. 1988, 110, 3610–3612. A. Hamilton, J.-M. Lehn, J. L. Sessler, J. Am. Chem. Soc. 1986, 108, 5158–5167. A. Osuka, H. Furuta, K. Maruyama, Chem. Lett. 1986, 479–482. B. Morgan, D. Dolphin, Angew. Chem. 1985, 97, 1000–1002; Angew. Chem. Int. Ed. Engl. 1985, 24, 1003–1004.

255

256

9 X-ray Crystal Structures of Porphyrinophanes as Model Compounds for Photoinduced Electron Transfer 36 37 38 39 40 41 42

43 44

45

46 47 48

49

50

51

52 53

54

55

56 57

J. E. Baldwin, P. Perlmutter, Topics Curr. Res. 1984, 121, 181–220. K. N. Ganesh, J. K. M. Sanders, J. Chem. Soc. Perkin Trans. 1 1982, 1611–1615. H. A. Staab, J. Weiser, E. Baumann, Chem. Ber. 1992, 125, 2275–2283. D. Mauzerall, J. Weiser, H. A. Staab, Tetrahedron 1989, 45, 4807–4814. C. Krieger, J. Weiser, H. A. Staab, Tetrahedron Lett. 1985, 26, 6055–6058. J. Weiser, H. A. Staab, Tetrahedron Lett. 1985, 26, 6059–6062. J. Weiser, H. A. Staab, Angew. Chem. 1984, 96, 602–603; Angew. Chem. Int. Ed. Engl. 1984, 23, 623–625. H. A. Staab, T. Carell, A. Döhling, Chem. Ber. 1994, 127, 223–229. H. A. Staab, J. Weiser, M. Futscher, G. Voit, A. Rückemann, C. Anders, Chem. Ber. 1992, 125, 2285–2301. C. Krieger, M. Dernbach, G. Voit, T. Carell, H. A. Staab, Chem. Ber. 1993, 126, 811–821. T. Carell, Dissertation, Heidelberg 1993. G. Voit, Dissertation, Heidelberg 1991. H. A. Staab, C. Krieger, C. Anders, A. Rückemann, Chem. Ber. 1994, 127, 231– 236. H. A. Staab, A. Feurer, R. Hauck, Angew. Chem. 1994, 106, 2542–2545; Angew. Chem. Int. Ed. Engl. 1994, 33, 2428–2431. H. A. Staab, A. Feurer, C. Krieger, A. S. Kumar, Liebigs Ann./Recueil 1997, 2321– 2336. T. Hayashi, T. Miyahara, N. Koide, J. Kato, H. Masuda, H. Ogoshi, J. Am. Chem. Soc. 1997, 119, 7281–7290. A. Feurer, Dissertation, Heidelberg 1993. M. J. Gunter, D. C. R. Hockless, M. R. Johnston, B. W. Skelton, A. H. White, J. Am. Chem. Soc. 1994, 116, 4810–4823. J. E. Cochran, T. J. Parrott, B. J. Whitlock, H. W. Whitlock, J. Am. Chem. Soc. 1992, 114, 2269–2270. P. L. Anelli, P. R. Ashton, N. Spencer, A. M. Z. Slawin, J. F. Stoddart, D. J. Williams, Angew. Chem. 1991, 103, 1052–1054; Angew. Chem. Int. Ed. Engl. 1991, 30, 1036–1039. S. K. Burley, G. A. Petsko, J. Am. Chem. Soc. 1986, 108, 7995–8001. S. K. Burley, G. A. Petsko, Science 1985, 229, 23–28.

58 59 60 61

62

63

64 65 66

67 68

69

70

71

72

73

74

A. N. Cammidge, O. Öztürk, J. Org. Chem. 2002, 67, 7457–7464. H. A. Staab, S. Nikolic, C. Krieger, Eur. J. Org. Chem. 1999, 1459–1470. S. Nikolic, Dissertation, Heidelberg 1997. A. Osuka, S. Nakajima, T. Okada, S. Taniguchi, K. Nozaki, T. Ohno, I. Yamazaki, Y. Nishimura, N. Mataga, Angew. Chem. 1996, 108, 98–101; Angew. Chem. Int. Ed. Engl. 1996, 35, 92–95; further references therein. H. L. Anderson, C. A. Hunter, M. Nafees Meah, J. K. M. Sanders, J. Am. Chem. Soc. 1990, 112, 5780–5789; further references therein. R. J. Harrison, B. Pearce, G. S. Beddard, J. A. Cowan, J. K. M. Sanders, Chem. Phys. 1987, 116, 429–448. J. A. Cowan, J. K. M. Sanders, J. Chem. Soc. Perkin Trans. 1 1985, 2435–2437. P. Schmitt, D. Mandon, J. Fischer, R. Weiss, New J. Chem. 1992, 16, 763–765. M. J. Gunter, T. P. Jeynes, M. R. Johnston, P. Turner, Z. Chen, J. Chem. Soc. Perkin Trans. 1 1998, 1945–1957. A. Diebold, J. Fischer, R. Weiss, New J. Chem. 1996, 20, 959–970. P. Ochsenbein, M. Bonin, K. Schenk, J. Froidevaux, J. Wytko, E. Graf, J. Weiss, Eur. J. Inorg. Chem. 1999, 1175–1179. J. Froidevaux, P. Ochsenbein, M. Bonin, K. Schenk, P. Maltese, J.-P. Gisselbrecht, J. Weiss, J. Am. Chem. Soc. 1997, 119, 12362–12363. D. Paul, F. Melin, C. Hirtz, J. Wytko, P. Ochsenbein, M. Bonin, K. Schhenk, P. Maltese, J. Weiss, Inorg. Chem. 2003, 42, 3779–3787. L. Jaquinod, N. Kyritsakas, J. Fischer, R. Weiss, New J. Chem. 1995, 19, 453– 460. D. R. Benson, R. Valentekovich, C. B. Knobler, F. Diederich, Tetrahedron 1991, 47, 2401–2422. W. B. Cruse, O. Kennard, G. M. Sheldrick, A. D. Hamilton, S. G. Hartley, A. R. Battersby, J. Chem. Soc. Chem. Commun. 1980, 700–701. For a compilation of Fe complexes see: C. Slebodnick, J. C. Fettinger, H. B. Peterson, J. A. Ibers, J. Am. Chem. Soc. 1996, 118, 3216–3224 and references therein.

9.6 References 75

76 77

78 79 80

81

82

83

84

85

86

87 88 89

90

91

For a compilation of Ru complexes see: C. Slebodnick, K. Kim, J. A. Ibers, Inorg. Chem. 1993, 32, 5338–5342 and references therein. C. Slebodnick, M. L. Duval, J. A. Ibers, Inorg. Chem. 1996, 35, 3607–3613. M. R. Johnson, W. K. Seok, W. Ma, C. Slebodnick, K. W. Wilcoxen, J. A. Ibers, J. Org. Chem. 1996, 61, 3298–3303. M. R. Johnson, W. K. Soek, J. A. Ibers, J. Am. Chem. Soc. 1991, 113, 3998–4000. G. B. Jameson, J. A. Ibers, J. Am. Chem. Soc. 1980, 102, 2823–2831. J. P. Collman, V. J. Lee, C. J. KellenYuen, X. Zhang, J. A. Ibers, J. I. Brauman, J. Am. Chem. Soc. 1995, 117, 692– 703. D. M. Guldi, C. Luo, M. Prato, E. Dietel, A. Hirsch, Chem. Commun. 2000, 373–374. N. Armaroli, G. Marconi, L. Echegoyen, J.-P. Bourgeois, F. Diederich, Chem. Eur. J. 2000, 6, 1629–1645 and references therein. J.-Y. Zheng, K. Tashiro, J. Hirabayashi, K. Kinbara, K. Saigo, T. Aida, S. Sakamoto, K. Yamaguchi, Angew. Chem. 2001, 113, 1910–1913; Angew. Chem. Int. Ed. Engl. 2001, 40, 1857–1861. H. A. Staab, G. Voigt, J. Weiser, M. Futscher, Chem. Ber. 1992, 125, 2303– 2310. H. A. Staab, M. Tercel, R. Fischer, C. Krieger, Angew. Chem. 1994, 106, 1531– 1534; Angew. Chem. Int. Ed. Engl. 1994, 33, 1463–1466. H. A. Staab, T. Carell, Angew. Chem. 1994, 106, 1534–1536; Angew. Chem. Int. Ed. Engl. 1994, 33, 1466–1468. H. A. Staab, R. Hauck, B. Popp, Eur. J. Org. Chem. 1998, 631–642. H. A. Staab, B. Kratzer, S. Quazzotti, Eur. J. Org. Chem. 1998, 2149–2160. H. A. Staab, A. Döhling, P. Voit, M. Dernbach, Tetrahedron Lett. 1984, 35, 7617. H. A. Staab, J. Weikard, A. Rückemann, A. Schwögler, Eur. J. Org. Chem. 1998, 2703–2712. F. Pöllinger, C. Musewald, H. Heitele, M. E. Michel-Beyerle, C. Anders, M. Futscher, G. Voit, H. A. Staab, Ber.

92

93

94

95

96

97

98

99

100

101

102

103 104 105 106

Bunsen-Ges. Phys. Chem. 1996, 100, 2076– 2080. T. Häberle, J. Hirsch, F. Pöllinger, H. Heitele, M. E. Michel-Beyerle, C. Anders, A. Döhling, C. Krieger, A. Rückemann, H. A. Staab, J. Phys. Chem. 1996, 100, 18269–18274. H. Heitele, F. Pöllinger, T. Häberle, M. E. Michel-Beyerle, H. A. Staab, J. Chem. Phys. 1994, 98, 7402–7410. F. Pöllinger, H. Heitele, M. E. MichelBeyerle, M. Tercel, H. A. Staab, Chem. Phys. Lett. 1993, 209, 251–257. H. Heitele, Angew. Chem. 1993, 105, 378–398; Angew. Chem. Int. Ed. Engl. 1993, 32, 359–377. H. Heitele, F. Pöllinger, K. Kremer, M. E. Michel-Beyerle, M. Futscher, G. Voit, J. Weiser, H. A. Staab, Chem. Phys. Lett. 1992, 188, 270–278. W. Frey, R. Klann, F. Laermer. T. Elsaesser, E. Baumann, M. Futscher, H. A. Staab, Chem. Phys. Lett. 1992, 190, 567–573. F. Pöllinger, H. Heitele, M. E. Michel-Beyerle, C. Anders, M. Futscher, H. A. Staab, Chem. Phys. Lett. 1992, 198, 645–652. S. N. Smirnov, P. A. Liddell, I. V. Vlassiouk, A. Teslja, D. Kuciauskas, C. L. Braun, A. L. Moore, T. A. Moore, D. Gust, J. Phys. Chem. A 2003, 107, 7567– 7573. T. Da Ros, M. Prato, D. M. Guldi, M. Ruzzi, L. Pasimeni, Chemistry Eur. J. 2001, 7, 1597–1605. S. Fukuzumi, D. M. Guldi, in V. Balzani (ed.) Electron Transfer in Chemistry, Vol. 2, Wiley-VCH, Weinheim 2001, p. 270–337. D. M. Guldi, C. Luo, M. Prato, A. Troisi, F. Zerbetto, M. Scheloske, E. Dietel, W. Bauer, A. Hirsch, J. Am. Chem. Soc. 2001, 123, 9166–9167. D. Gust, T. A. Moore, A. L. Moore, Acc. Chem. Res. 2001, 34, 40–48. D. Gust, T. A. Moore, A. L. Moore, J. Photochem. Photobiol. B 2000, 58, 63–71. F. Diederich, M. Gomez-Lopez, Chem. Soc. Rev. 1999, 28, 263–277. F. Diederich, R. Kessinger, Acc. Chem. Res. 1999, 32, 537–545.

257

258

9 X-ray Crystal Structures of Porphyrinophanes as Model Compounds for Photoinduced Electron Transfer 107 H. Imahori, Y. Sakata, Eur. J. Org. Chem.

111 All the Figs were drawn according to the

1999, 2445–2457. 108 J.-P. Bourgeois, F. Diederich, L. Echegoien, J.-F. Nierengarten, Helv. Chim. Acta 1998, 81, 1835–1844. 109 N. Martin, L. Sanchez, B. Illescas, I. Perez, Chem. Rev. 1998, 98, 2527–2547. 110 H. Imahori, Y. Sakata, Adv. Mater. 1997, 9, 537–546.

atomic coordinates given in: F. H. Allen, The Cambridge Structural Database: a quarter of a million crystal structures and rising. Acta Crystallogr. 2002, B58, 380–388.

259

10

Ultraviolet Photoelectron Spectra of Cyclophanes Heidi M. Muchall

10.1

Introduction

The interest in electronic interactions between the two decks in [2.2]cyclophane systems because their separation is small (ranging from ca. 3 Å down to 2.6 Å in the fully bridged superphane, as compared to the distance of 3.4 Å between layers in graphite) has been one of the driving forces in the photoelectron (PE) spectroscopic investigation of cyclophanes. In particular the matter of through-space and through-bond interactions between the decks in cyclophanes has been the focus of early studies [1]. An early review of these and other highlights in the PE spectroscopy of cyclophanes was given by Heilbronner and Yang in 1983 [1]. PE spectroscopic data of cyclophanes have since appeared in a databank [2] and in books and reviews [3–6]. In this review, we present several aspects that have appeared in the literature from 1983 onwards and only include older literature where it helps the discussion. Additional information for the interpretation of the PE spectrum of [2.2]paracyclophane is gained from the use of a different ionization technique (Section 10.2) as well as from chemical modification of the bridges (Section 10.3). The rigidity of the [2.2]cyclophane system is exploited in conformational studies with conjugating substituents (Section 10.4) and in studies on donor–acceptor interactions between the two decks (Section 10.5). Through-bond interactions are revealed in heterocyclophanes with modified bridges (Section 10.6) and, finally, phanes with only one bridge (Section 10.7) are presented with respect to strain, pinteractions and homoconjugation. The PE spectroscopic data of cyclophanes more recent than 1983 are presented in Tab. 10.1. This review does not strive to be exhaustive, for example the PE spectra of metal-capped cyclophanes or superphanes have not been included.

Modern Cyclophane Chemistry. Edited by Rolf Gleiter, Henning Hopf Copyright © 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 3-527-30713-3

260

10 Ultraviolet Photoelectron Spectra of Cyclophanes Tab. 10.1 Vertical ionization energies IPv (eV) and their assignment for cyclophanes

Compound

IPv

Ref.

7.8 (b3u), 8.3 (b2g), 8.3 (b3g), 9.2 (b2u), 9.8 (ag)

15

8.0 (p), 8.0 (p), 8.4 (p), 9.2 (p), 10.1 (r), 10.5 (p)

20

8.0, 8.0, 8.4, 9.0, 10.0, 10.3

20

8.0, 8.0, 8.4, 9.0, 10.0, 10.2

20

8.0, 8.0, 8.4, 9.1, 9.7, 10.3

20

7.9, 7.9, 8.3, 8.8, 9.5, 10.1

20

7.7 (sh), 7.9, 8.3, 8.9, 10.0, 10.3

20

7.9, 7.9, 8.3, 9.0, 10.2, 10.2

20

10.1 Introduction Tab. 10.1

(continued)

Compound

R = OCH3 R = CN

D = OCH3, A = COOCH3 D = OCH3, A = NO2

IPv

Ref.

7.8, 7.8, 8.3, 9.0, 9.7, 10.4

20

8.1, 8.1, 8.4, 9.3, 9.9, 10.3

20

7.35 (b2g), 8.00 (b3u), 8.25 (b3g), 9.05 (b2u), 9.65 (n) 8.90 (b2g), 8.90 (b3u), 9.25 (b3g), 10.12 (b2u)

23 23

7.60 (b2g), 8.50 (b3g), 8.75 (b3u), 9.65 (n), 9.90 (b2u)

23

7.20 (b2g), 8.00 (b3u), 8.20 (b3g), 9.10 (b2u), 9.70 (n), 9.80, 10.0 7.60 (b2g), 8.50 (b3g), 9.10 (b3u), 9.80 (n), 10.00 (b2u)

23

8.60, 9.10, 9.20, 9.62, 9.79, 10.05

23

23

261

262

10 Ultraviolet Photoelectron Spectra of Cyclophanes Tab. 10.1

(continued)

Compound

n=1 n=2 n=3

IPv

Ref.

7.42, 8.31, 8.76, 9.20, 9.30, 9.45, 10.03

23

7.54, 8.66, 8.95, 9.35, 9.41, 9.85, 10.10

23

7.4 (au), 8.0 (ag), 8.4 (bg), 8.7 (bu), 9.4 (ag)

15

7.4 (au), 8.1 (bg), 8.9 (ag), 9.4 (bu), 9.4 (ag)

15

7.33, 8.44, 9.20, 9.8

27

6.95, 8.28, 8.5, 9.7

27

8.42 11.4 8.45 8.48

32

(no), 9.33 (pD–pt,i), 9.64 (pt,o), 10.14 (pD+pt,i), (w) (no), 9.53–9.8 (pD–pt,i, pt,o, pD+pt,i) (no), 9.47–9.69 (pD–pt,i, pt,o, pD+pt,i)

32 32

10.2 [2.2]Paracyclophane Tab. 10.1

(continued)

Compound

IPv

Ref.

8.5 (par), 8.8 (par), 9.2–9.5 (p+o , p–i , p–o, p+i )

38

8.5 (par), 8.7 (sh) (par), 9.2–9.7 (pd, p–o, p+o , p–i ), 9.9 (p+i )

38

8.6 (par), 8.7 (par), 9.4–9.5 (p–i , p–o, p+o , p+i )

38

8.5 (par), 8.6 (par), 8.9 (pd), 9.2–9.5 (p–i , p+o , p–o), 9.7 (p+i )

38

10.2

[2.2]Paracyclophane

1

The PE spectrum of the parent cyclophane 1 (Fig. 10.1) was recorded as early as 1971 by Pignataro and coworkers [7] and its electronic structure has been studied extensively and reviewed [1, 8, 9]. Two problems with the interpretation of the PE spectrum of 1 addressed early were the number of ionizations in the low energy region and their assignment to molecular orbitals (Fig. 10.1). Now accepted, a prediction by Gleiter [10], from a correlation based on molecular orbital energies from extended Hückel calculations, attributed the first band to three ionizations as given on the second line below the spectrum in Fig. 10.1. More support for this interpretation is supplied in a more recent ab initio study by Canuto and Zerner [8]. Using the Dunning-Hay split valence double zeta basis set [11], they analyzed the first five ionization events in the PE spectrum of 1. Based on experimental geometry from an X-ray analysis [12], treatments from both molecular orbital energies in Koopmans’ approximation (which relates the vertical ionization energy

263

264

10 Ultraviolet Photoelectron Spectra of Cyclophanes

Fig. 10.1 PE spectrum of 1 and suggested assignments (reprinted with permission from ref. 1. © Copyright 1990 Springer, modified)

with the orbital energy, Iv,j&–ej) [13] and vertical ionization energies calculated directly from the radical cation states, reproduce three ionizations close in energy, followed by two separate events [8]. As for the second problem, in the orbital sequence four p-orbitals (Fig. 10.2) are followed by a r-orbital, and thus band 5 at 10.3 eV is now attributed to an ionization from a r-orbital (Fig. 10.1). This assignment is supported by comparison with the PE spectrum of 2,3',5,6'-tetrahydro[2.2]paracyclophane (2) [14], in which the four p ionizations are found between 8.3 and 8.8 eV, and the next band at 10.5 eV necessarily belongs to the first r ionization. The subject of through-bond interaction in 1 is covered in Section 10.3.

The most recent PE spectroscopic study of 1, a Penning ionization (PI) through collision with metastable He*(23S) atoms, was performed by Ohno and coworkers [9] and sheds more light on the r/p issue for the band at 10.3 eV. For a series of polycyclic aromatic hydrocarbons such as naphthalene and anthracene, the PI electron spectra reveal that ionizations from p-orbitals give more intense bands than ionizations from r-orbitals [16]. Because a PI electron spectrum is comparable to a UV PE spectrum with a similar ionization energy, observed intensities can give valuable information on orbital assignments to the ionization bands. In the PI electron spectrum of 1 the band at 10.4 eV is less intense than the band at 9.7 eV, and this is further support for the assignment given in Fig. 10.1 (bottom).

10.3 Modified bridges in [2.2]paracyclophane

Fig. 10.2 The four highest occupied molecular orbitals of 1 (redrawn and modified from ref. 15)

The simulated PI electron spectrum, with ionization energies determined from OVGF (outer-valence Green’s function) calculations and intensities from the EED (exterior electron density) model [17] at HF/6-31 + G*//HF/4-31G, also allowed for the location of the remaining p ionizations. p2 lies at 11.4 eV, p1 at 13.5 eV, corresponding to bands 10 and 19, respectively [9].

10.3

Modified Bridges in [2.2]Paracyclophane

The third ionization in the spectrum of 1 is attributed to a p-orbital that shows a considerable amount of r/p mixing with a r-orbital from the bridges (b3u in Fig. 10.2) [1, 3, 15]. The extent of this through-bond interaction of the two p systems, the amount of mixing, can be controlled by the energy of the corresponding r-orbital. A higher (less negative) orbital energy permits increased mixing and therefore the energy of the resulting orbital will be higher; the corresponding ionization band will be found at a lower energy. On the other hand, lowering the energy of the r-orbital will eventually lead to a loss of mixing. Heilbronner and Maier [18] found that the inductive effect of the fluorine atoms causes the through-bond interaction to be turned off in the octafluoro derivative 3, and the third ionization at 10.5 eV is found well separated from the first two at 9.3 eV. A similar but less pronounced effect is seen upon introduction of the unsaturated bridges in 4 [1]. According to Gleiter et al. [19] and in contrast to 3, the higher base energy of the r-orbital in cyclopropane allows for more efficient mixing in the cyclopropane derivative 5, and the first three ionizations are found between 7.9 and 8.1 eV, the strong overlap of the bands rendering impossible any further analysis. An even greater interaction between the ring p-orbitals and the r-orbitals of the bridges is found in the disilane 6 [15].

265

266

10 Ultraviolet Photoelectron Spectra of Cyclophanes

Fig. 10.3 Correlation diagram for the first four ionization bands in the PE spectra of 1 and 3–6. The filled bars depict those p-orbitals that show a large r contribution from the bridges. p-orbitals from the bridges in 4 are labeled pbr (reprinted with permission from ref. 3. © Copyright 1990 American Chemical Society, modified)

Whereas the first ionization band (8.1 eV) in the PE spectrum of 1 shows a steep onset (Fig. 10.1), the first ionization band (8.3 eV) in the PE spectrum of 6 exhibits a distinct shoulder on its low-energy side (7.8 eV), corresponding to one ionization event. Orbital energies for the unsubstituted disilane (SiH2–SiH2 bridges) from an FMO approach and those from an MNDO calculation using the optimized geometry support this assignment. The energy difference between HOMO and HOMO-1 is larger than that between HOMO-1 and HOMO-2, and the calculations show a HOMO with r/p interaction [15], analogous to HOMO-2 given in Fig. 10.2. A correlation diagram for the first four ionization bands in the spectra of 1, 5 and 6 reveals that the through-bond interaction becomes more important along this series, while the through-space interaction is drastically decreased in 6, probably because of the larger inter-deck distance. The correlation diagram also allows for the assignment of the first ionization in 5, which as for 6 should be ascribed to a HOMO with significant r/p interaction. The orbital correlation diagram [3] for all species discussed here (1, 3–6) is given in Fig. 10.3.

10.4

Conjugating Substituents in [2.2]Paracyclophane

As is common practice when discussing the PE spectra of cyclophanes [1], the carbon centers in the p system of the upper and lower deck are numbered as in the formula.

10.4 Conjugating Substituents in [2.2]Paracyclophane

1A

Heilbronner and coworkers [20] have studied a possible influence of p–p conjugation of the aromatic rings with substituents in position 2 of the [2.2]paracyclophane unit. In analogy to PE spectroscopic studies on the conformations of substituted styrenes [21, 22], the series of substituted vinyl cyclophanes 7–14 was investigated. With calculated PE band positions from a ZDO treatment, bands 1–4 and 6 are identified as arising from p ionizations. In these conformationally flexible systems, the energies of the p-orbitals are dependent on the torsion angle. For 7, the estimated shifts of the PE bands as a function of the torsion angle h give good agreement with the experimental band positions for 0 8< h < 45 8, which suggests coplanarity of the aromatic ring and its attached vinyl group. This is confirmed by the similarity of the ionization energies of bands 1–5 in the spectrum of 15, whose orbital energies are not a function of a torsion angle. On this basis the spectrum of 13 shows good agreement with calculated ionization energies for h = 0 8. Estimated band shifts for the methyl-substituted compounds 8–11 reveal band 6 to be the most sensitive to changes in h. A comparison with the experimental ionization energies suggests a torsion angle close to 0 8 in 8, 10 and 11, whereas that in 9 could be as large as 45 8, in agreement with an inhibition of coplanarity on purely steric grounds. Even though there is good agreement between the estimated and observed data, the authors conclude that in contrast to styrenes [21, 22], PE spectroscopy does not give worthwhile information about the torsion angle in the vinyl cyclophanes 7–11, part of the reason lying in the unique nodal properties of the molecular orbitals of 7 [20].

267

268

10 Ultraviolet Photoelectron Spectra of Cyclophanes

10.5

Donor–Acceptor Cyclophanes

[2.2]Paracyclophanes with their rigid arrangement of parallel, eclipsed benzene rings should be ideal candidates for the study of donor–acceptor interactions with respect to the relative orientation of the donor (D) and acceptor (A) functional groups. Gleiter et al. [23] have interpreted the PE spectra of 16–23 to determine the extent to which the ground-state donor and acceptor p systems interact. For pairs of compounds such as 22/23, UV/Vis spectroscopic studies show the pseudogeminal derivative to exhibit a more intense charge-transfer band at a longer wavelength than the pseudo-ortho derivative; charge-transfer is found to occur predominantly in the excited state, cf. Chapter 11 [24]. Assignments of the observed ionization bands in the PE spectra of 16–23 have been undertaken through comparison with the spectra of the underlying parent compounds 24–29, and with calculated ionization energies from a ZDO treatment [23]. The interpretation of the spectra of the cyclophanes 16–20 reveals a through-space interaction between the upper and the lower deck of 0.3 eV. This value is smaller than that for other [2.2]cyclophanes [1] and is attributed to a reduced inter-deck p overlap due to the strongly electron withdrawing substituents. The PE spectra of 21–23 can be constructed from superpositions of the spectra of the parent systems 24–29, which leads to the conclusion that additional stabilizing donor–acceptor interactions that would lead to larger negative ionization energies are not found in the spectra of 16–23.

10.5 Donor–Aacceptor Cyclophanes

269

270

10 Ultraviolet Photoelectron Spectra of Cyclophanes

10.6

Heterocyclophanes

In analogy to the analysis of the disilane 6, a strong interaction between the ring p-orbitals and the r-orbitals of the bridge Si–Si bonds was found in the disilanes 30 and 31 [15]. For the assignment of the ionization bands in the PE spectra of 30 and 31, orbital energies were calculated at the HF/STO-3G level for the unsubstituted disilanes (SiH2–SiH2 bridges) and the first four ionization bands were correlated with those of 32 and 33 [25, 26]. As for 6, the HOMO in 30–33 shows r/p mixing (Fig. 10.4), which is in contrast to the change in HOMO-character found between 1 and 6. Further support for this assignment is drawn from the nodal properties of the HOMO in 30–33. It does not possess a coefficient on the heteroatom (Fig. 10.4), and so its energy and hence the first ionization energy for 30 and 31 is comparable.

10.7

Miscellaneous Compounds 10.7.1

[6]Phanes with Higher Aromatic Systems

Tobe et al. [27] reported the PE spectra of the naphthalenophane 34 and anthracenophane 35, the “smallest-bridged isolable cyclophanes with bent acene nuclei”. The X-ray crystal structure reveals the bridged aromatic ring in 35 [27] to be even

Fig. 10.4 The four highest occupied p molecular orbitals of 30 (X = S) and 31 (X = O) (redrawn and modified from ref. 15)

10.7 Miscellaneous Compounds

more deformed than those in the substituted [6]paracyclophanes 36 [28] and 37 [29]. Yet a comparison of the PE spectra of 34 and 35 with those of 1,4-diethylnaphthalene (38) and 1,4-diethylanthracene (39), both strain-free molecules, reveals great similarity. The shifts in the ionization energies in the pairs 34/38 and 35/39 is only on the order of 0.1 to 0.3 eV, and from this it is concluded that the bridge has mainly an inductive effect on the p ionizations of 34 and 35 [27]. This finding is not completely unexpected. Earlier studies by Gleiter et al. [30] on [8]paracyclophane, a molecule with negligible bending of the aromatic system, revealed a split of the first two ionization bands of 0.65 eV. As this value is also found in the PE spectra of several 1,4-dialkylbenzenes it is concluded that the split is due to inductive and hyperconjugative effects [30]. Furthermore, splits of a similar size are found in the bent [6]- (0.8 eV) and [7]paracyclophane (0.65 eV) [31]. While this was initially attributed to the deformation of the benzene rings and loss of aromatic character [31], the results on [8]paracyclophane led to a reevaluation and rather suggest the split to be due to electron releasing effects of the bridges for [6]- and [7]paracyclophane [30].

10.7.2

Cyclopropenophanes

In general, strong interactions with large splits of the corresponding ionization bands are expected for orbitals with similar energies and large resonance integrals. Gleiter et al. investigated p interactions in the series of cyclopropenophanes

271

272

10 Ultraviolet Photoelectron Spectra of Cyclophanes

40–42 [32, 33] because the p ionization energy for the C=C double bond (pD) in the PE spectrum of the cyclopropenone 43 (9.67 eV) is similar to those of the triple bond in cyclooctyne (44) (9.18 and 9.30 eV) [34] and therefore measurable interactions can be expected. The in-plane p-orbitals (pt,i) of the two alkyne units in cyclodeca-1,6-diyne (45), for example, are split by 1.5 eV, the out-of-plane p-orbitals (pt,o) by 0.3 eV [35]. The PE spectra of 40–42 exhibit a band at 8.4–8.5 eV, which is attributed to the oxygen lone pair (n0) as in 43 (8.56 eV) [34]; the p ionizations are found between 9 and 10 eV. While the larger systems 41 and 42 show one band corresponding to three ionizations, the spectrum of 40 shows three distinct peaks in this region indicating considerable p interaction. The split of about 0.5 eV for the p ionizations in 40 is larger than that for the out-of-plane combination in 45 and this is found to arise from the boat conformation of the ten-membered ring in 40, in which the cyclopropene C=C double bond mimics an in-plane interaction with the triple bond. The split is reproduced by MINDO/3 and HF/6-31G* calculations [32].

10.7.3

Metacyclophanediynes

Gleiter and coworkers investigated dienediynes such as 46 and found considerable homoconjugation between the p systems [36, 37]. In the metacyclophanediynes 47–50 the aromatic system is separated from the alkyne units through methylene groups, which also allows for possible homoconjugation [38]. In general, the PE spectra of 47–50 can be readily understood from the spectra of the parent compounds, m-xylene and a medium-sized cyclic diyne, as a simple addition of spectral features, and this assignment is supported by calculated orbital energies at the semi-empirical AM1 level. No evidence for homoconjugation between the p systems of the aromatic ring (par, first band at 8.6 eV, two ionizations) and the triple bonds (second band between 9.2 and 9.7 eV, four ionizations) is found. In contrast, the second band in the spectra of 48 and 50 exhibits a larger half-width than

10.8 References

that in the spectra of 47 and 49, and the authors attribute this to a possible homoconjugation between the alkene (pd) and alkyne units similar to that found in dienediynes [36, 37]. Yet a detailed analysis is hindered by the strong overlap of the ionization bands.

10.8

References 1 2

3 4 5 6 7

8 9 10 11

12 13

E. Heilbronner, Z.-Z. Yang, Top. Curr. Chem. 1983, 115, 1–55. Z.-Z. Yang, B. Kovacˇ, E. Heilbronner, J. Lecoultre, C. W. Chan, H. N. C. Wong, H. Hopf, F. Vögtle, Helv. Chim. Acta 1987, 70, 299–307. R. Gleiter, W. Schäfer, Acc. Chem. Res. 1990, 23, 369–375. F. Vögtle, Cyclophane Chemistry, Wiley, Chichester, 1993. R. Gleiter, D. Kratz, Acc. Chem. Res. 1993, 26, 311–318. P. Rademacher, Chem. Rev. 2003, 103, 933–975. J. N. A. Ridyard, S. Pignataro, V. Mancini, H. J. Lempka, J. Chem. Soc., Chem. Comm. 1971, 3, 142–143. S. Canuto, M. C. Zerner, J. Am. Chem. Soc. 1990, 112, 2114–2120. Y. Yamakita, M. Yamauchi, K. Ohno, Chem. Phys. Lett. 2000, 322, 189–198. R. Gleiter, Tetrahedron Lett. 1969, 4453– 4456. T. H. Dunning, Jr., P. J. Hay, in Methods of Electronic Structure Theory, H. F. Schaefer III (Ed.), 1, 1977. H. Hope, I. Bernstein, K. N. Trueblood, Acta Cryst. 1972, B28, 1733–1743. T. Koopmans, Physica 1934, 1, 104–113.

14

15

16

17

18 19

20

21 22

23 24

R. Gleiter, J. Spanget-Larsen, H. Hopf, C. Mlynek, Chem. Ber. 1984, 117, 1987– 1990. R. Gleiter, W. Schäfer, G. Krennrich, H. Sakurai, J. Am. Chem. Soc. 1988, 110, 4117–4120. M. Yamauchi, Y. Yamakita, H. Yamakado, K. Ohno, J. Electron Spectrosc. Relat. Phenom. 1998, 88/91, 155–161. K. Ohno, Y. Harada, in Theoretical Models of Chemical Bonding, Part 3, Z. B. Maksicˇ (ed.), Springer, Berlin, 1991. E. Heilbronner, J. P. Maier, Helv. Chim. Acta 1974, 57, 151–159. R. Gleiter, M. Eckert-Maksicˇ, W. Schäfer, E. A. Truesdale, Chem. Ber. 1982, 115, 2009–2011. Z.-Z. Yang, E. Heilbronner, H. Hopf, S. Ehrhardt, S. Hentschel, J. Phys. Chem. 1988, 92, 914–917. J. P. Maier, D. W. Turner, J. Chem. Soc. Faraday Trans. 2, 1973, 196–206. T. Kobayashi, K. Yokota, S. Nagakura, J. Electron Spectrosc. Rel. Phenom. 1973, 3, 449–454. R. Gleiter, W. Schäfer, H. A. Staab, Chem. Ber. 1988, 121, 1257–1264. H. A. Staab, C. P. Herz, C. Krieger, M. Rentea, Chem. Ber. 1983, 116, 3813– 3830.

273

274

10 Ultraviolet Photoelectron Spectra of Cyclophanes 25

26

27

28

29

30

F. Bernardi, A. Bottoni, F. P. Colonna, G. Distefano, U. Folli, P. ViVarelli, Z. Naturforsch. A 1978, 33 a, 959–963. B. Kovacˇ, M. Allan, E. Heilbronner, J. P. Maier, R. Gleiter, M. W. Haenel, P. M. Keehn, J. A. Reiss, J. Electron Spectrosc. Relat. Phenom. 1980, 19, 167–177. Y. Tobe, T. Takahashi, T. Ishikawa, M. Yoshimura, M. Suwa, K. Kobiro, K. Kakiuchi, R. Gleiter, J. Am. Chem. Soc. 1990, 112, 8889–8894. J. Liebe, C. Wolff, C. Krieger, J. Weiss, W. Tochtermann, Chem. Ber. 1985, 118, 4144–4178. Y. Tobe, A. Nakayama, K. Kakiuchi, Y. Odaira, Y. Kai, N. Kasai, J. Org. Chem. 1987, 52, 2639–2644. R. Gleiter, H. Hopf, M. Eckert-Maksicˇ, K.-L. Noble, Chem. Ber. 1980, 113, 3401–3403.

31 32

33 34

35

36 37

38

H. Schmidt, A. Schweig, W. Thiel, Chem. Ber. 1978, 111, 1958–1961. R. Gleiter, M. Merger, A. Altreuther, H. Irngartinger, J. Org. Chem. 1995, 60, 4692–4696. R. Gleiter, M. Merger, Angew. Chem. Int. Ed. Engl. 1997, 36, 2426–2439. G. Bieri, E. Heilbronner, E. KlosterJensen, A. Schmelzer, J. Wirz, Helv. Chim. Acta 1974, 57, 1265–1283. R. Gleiter, M. Karcher, R. Jahn, H. Irngartinger, Chem. Ber. 1988, 121, 735–740. R. Gleiter, R. Merger, B. Nuber, J. Am. Chem. Soc. 1992, 114, 8921–8927. R. Gleiter, R. Merger, H. Irngartinger, J. Am. Chem. Soc. 1992, 114, 8927– 8932. M. Ramming, R. Gleiter, J. Org. Chem. 1997, 62, 5821–5829.

275

11

UV/Vis Spectra of Cyclophanes Paul Rademacher

11.1

Introduction

The electronic structure of cyclophanes has been a subject of continuous research ever since this field of aromatic chemistry was established with Cram’s studies on [m.n]paracyclophanes in the early 1950s [1]. In the past 50 years, cyclophane chemistry has emerged as a central research discipline with numerous links to other areas without losing its fascination [2]. One reason is certainly connected with the unusual structural features of many of these compounds, which coincide with novel electronic properties, especially electronic interactions between the aromatic rings. Ideal spectroscopic tools to probe these effects are methods related to uptake or release of electrons. Photoelectron (PE) spectroscopy (see Chapter 10), UV/Vis and ESR spectroscopy, and cyclovoltammetry (see Chapter 14) have been employed extensively to determine such effects in cyclophanes [3]. UV/Vis absorption and emission spectroscopy is still one of the most efficient experimental methods in the field of cyclophane chemistry. Several excellent reviews on cyclophane chemistry have been published. Only a few of the more recent are cited here which include comments on electronic and spectroscopic properties [4–8]. The paper entitled “The Spectral Consequences of Bringing Two Benzene Rings Face to Face” of Cram, Allinger and Steinberg [9] from 1954, can be cited as the first systematic interpretation and discussion of spectral properties of [m.n]paracyclophanes, in particular those based on UV spectroscopy. In the following sections, no comprehensive account on electronic spectra of cyclophanes can be supplied, but rather a few important, characteristic and interesting aspects will be highlighted.

11.2

Characteristic Properties of Cyclophanes with Implications for their Electronic Spectra

Sterically fixing two p systems into close proximity has numerous spectroscopic implications. The p molecular orbitals (MOs) are split into bonding and antibonding sets by transannular through-space interaction and – depending on symmetry Modern Cyclophane Chemistry. Edited by Rolf Gleiter, Henning Hopf Copyright © 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 3-527-30713-3

276

11 UV/Vis Spectra of Cyclophanes

– are further perturbed by through-bond interaction with the r-orbitals of the connecting bridges [10–12]. The normal MO pattern of the p system is therefore modified and exhibits stabilization or destabilization of HOMO and LUMO as well as of other MO energies. Deformation of the p system will twist the p orbitals which may enhance or diminish overlap and further contribute to changes in MO energies. Strain is also exerted on the r bonds of the bridges causing r- and r*-orbitals with unusually high or low energy, respectively. Furthermore, the p systems may be either electron-excessive or electron-deficient, and the dependence of donor–donor, donor–acceptor and acceptor–acceptor interactions can be modified by their transannular distances and orientations. These special features of the electronic structure are reflected in “cyclophanespecific” UV/Vis spectra. The spectra are often characterized by excimer formation which is accompanied by typical fluorescence and charge transfer (CT) absorptions. UV/Vis absorption spectra of arenes have been classified by Clar [13], and this terminology, by which three groups of bands are distinguished according to their intensity and other properties as a, para (or p) and b, is occasionally used for cyclophanes. Alternatively, sometimes the classification introduced by Platt [14] is used, which may be considered more rational since it relates to excited states. In the Platt system Clar’s a bands are called 1Lb, the para bands are termed 1La, while the b bands are labelled 1Bb. The a band, which can be masked by the others, is characterized by low intensity (lg e = 2–3) and a complex vibrational fine structure. The para band is of medium intensity (lg e & 4) and displays a regular splitting pattern. The b band is very intensive (lg e > 5) and usually featureless. In a cyclophane, in which two aromatic rings are arranged in a rigid, geometrically defined orientation, electronic interactions favor the formation of excited dimers or complexes, which are called excimers or exciplexes, respectively. Excimers are excited homodimers stabilized by exciton resonance (MM* « M*M) and by charge-transfer (CT) configurations (M+M– « M–M+) [15–18]. While for a system of two separated molecules (M + M) the interaction in the ground state (except for a van der Waals minimum and perhaps other nonbonded attractions) is repulsing, the interaction can be stabilizing if one molecule is in an excited state, so that in the complex

Fig. 11.1 MO scheme for the formation of an excimer

11.3 [n]Cyclophanes Fig. 11.2 Potential energy hypersurfaces for excimer formation and fluorescence of monomer M and excimer 1(MM)*

HOMO and LUMO are both singly occupied (Fig. 11.1). This leads to a relative minimum of the excimer 1(MM)* in the potential energy of the excited state. Excimers are identified by their emission spectra: a singlet excimer shows a characteristic band shifted to a longer wavelength compared to the corresponding monomer emission. This is illustrated in Fig. 11.2. The excimer band is broad and has no vibrational structure since the transition takes place to a nonbonded ground state. Excimer formation is often observed in aromatic hydrocarbons. The stability is particularly high in pyrene where the dissociation enthalpy D*H amounts to 42– 46 kJ mol–1 [16]. For compounds such as [2.2]-, [3.3]- and [3.2]naphthalenophanes, [2]naphthalene[2]paracyclophanes and [2.2](2,7)fluorenophane, the intramolecular p–p interaction between the arene rings in the excited singlet and triplet state was found to be highest if a maximal number of six-membered rings of the interacting arenes in a parallel orientation are completely eclipsed (see Section 11.6) [19]. Solutions of a p electron-excessive and a p electron-deficient compound, such as hexamethylbenzene and p-chloranil, usually show a long-wavelength absorption band that is not present in the solutions of the pure components. This absorption is assigned to the transfer of an electron from the donor D to the acceptor A and is called a charge transfer (CT) transition [17, 18]. Intramolecular CT interactions have been extensively studied in cyclophane systems as well-defined structures that allow the modelling of CT absorptions as a function of orientation and separation of A and D (see Section 11.10).

11.3

[n]Cyclophanes

UV absorption spectra of [n]cyclophanes are generally characterized by increasing bathochromic shifts and loss of fine structure as the lengths of the oligomethylene bridge decreases, and this effect is ascribed to the bending of the aromatic

277

278

11 UV/Vis Spectra of Cyclophanes Tab. 11.1 UV absorption data of [n]paracyclophanes (1) (in ethanol)

n

kmax (nm) (lg e)

12 10 9 8 7 6

221 223 226 230 245 253

(3.89) (3.91) (3.91) (3.83) (4) (4.0)

Ref. 255 261 264 268

(2.48) (2.42) (2.46) (2.43)

267 268 271 276

(2.62) (2.54) (2.57) (2.55)

275 275 279 282 283 296

(2.59) (2.50) (2.53) (2.48) (3) (2.8)

33 33 33 33 34 35

ring [20]. The absorption data of some [n]paracyclophanes (1) with short bridges are summarized in Tab. 11.1. The spectra of the larger members (n > 8) resemble that of 1,4-dipentylbenzene which was used as an open-chain reference model compound, but as the oligomethylene belt becomes shorter, the fine structure disappears and the bands move toward longer wavelength. Many other [n]arenophanes or [n]hetarenophanes are known with, for example, bridged naphthalene, pyridine, pyrazoline or quinoline rings. In general, their absorption spectra exhibit qualitatively similar behavior to that of the [n]paracyclophanes. Here we will only mention one more recent investigation. Bodwell et al. [21] have investigated a series of 1,n-dioxa[n](2,7)pyrenophanes (n = 7–12, 2–7) and have compared their absorption spectra with those of the parent [n](2,7)pyrenophanes (n = 7–9, 8–10). The structure of each member of the former series was determined crystallographically. Several spectroscopic properties were found to vary with the extent of the nonplanarity of the pyrene unit, and similar trends were observed for both series of compounds. The dioxapyrenophanes 2–7 were selected because the authors had previously observed that the parent [n](2,7)pyrenophanes are most difficult to crystallize in a form suitable for X-ray crystallography [22]. As for benzene, Clar has classified the absorption bands of pyrene [23]. A strong band at 242 nm (lg e = 4.95) is called the b' band and a second strong absorption at 273 nm (lg e = 4.73) is called the b band. A series of three bands at 306, 320 and 336 nm (lg e = 4.10, 4.51, 4.75) are termed the p bands. Finally, the a bands (352–372 nm) are difficult to observe because they have very low extinction coefficients. The UV/Vis spectra of the less strained pyrenophanes bear a strong resemblance to those of pyrene itself with all the bands having undergone a small red

Scheme 11.1

11.4 [m.n]Paracyclophanes

Scheme 11.2

shift that would be expected to occur upon substitution at the 2 and 7 positions. As the bridge becomes shorter and the pyrene unit becomes more distorted, a number of systematic changes were observed. The b' band is markedly red-shifted and the absorption becomes less intense. The b band experiences only a very slight red shift. The result of this is that the two bands move closer together such that the b band is observed only as a shoulder on the b' band in the spectrum of 3, and the two bands overlap completely in the spectrum of 2. The absorption maximum for this band (280 nm) is 23 nm red-shifted from the b' band of 7 (257 nm). The p bands do not appear to be significantly affected by the changes in the geometry of the pyrene system until the distortion becomes more pronounced, at which point there is a small blue shift. There is also a steady decrease in the intensity of the p bands with increasing distortion so that, in the spectrum of 2, they appear merely as poorly resolved humps. Similar trends are observed for the [n](2,7)pyrenophanes 8–10.

11.4

[m.n]Paracyclophanes 11.4.1

[1.1]Paracyclophane

Tsuji et al. [24] have generated the semistable [1.1]paracyclophane (11) and its bis(methoxycarbonyl) derivative (12) in an inert matrix at 77 K. Absorptions with kmax at 226, 237, 290 and 377 nm were observed for 11 and those with kmax at 256, 348 and 405 nm for 12. Quantum chemical geometrical optimization of 11, undertaken at the RHF–SCF, MP2 and B3LYP levels employing the 6-31G* basis set, indicated that the closest nonbonding interatomic distance between the aromatic rings is in a range of 236–240 pm, and the degree of bending of the benzene rings is comparable to that in [5]paracyclophane (1, n = 5), much less compared with that in [4]paracyclophane (1, n = 4). Calculations also support strong transannular electronic interactions between the p bonds of the aromatic moieties,

279

280

11 UV/Vis Spectra of Cyclophanes

Scheme 11.3

which lead to a significantly diminished HOMO–LUMO gap compared with that in p-xylene (13). 11.4.2

[2.2]Paracyclophane

Among the binuclear cyclophanes [2.2]paracyclophane (14) is by far the most thoroughly studied compound. In 1951 Cram and Steinberg [25] noted the abnormal electronic absorption spectrum of 14 (see Section 11.4.3, Fig. 11.4). 14 has absorption bands at 225 nm (lg e = 4.40), 244 (3.50, sh), 286 (2.51) and 302 (2.30). The band at 302 nm is well beyond the long-wavelength absorption of simple alkyl benzenes [9]. It has been termed the “cyclophane band” and is characteristic of the [2n]cyclophanes. The emission spectrum of 14 is also strikingly different from that of simple benzene derivatives. The unusual electronic structure of [2.2]paracyclophane (14) and the other smaller homologs (see Section 11.4.3) is reflected in their peculiar electronic spectra compared with those of open-chain compounds such as dibenzyl (15) or 1,4-bis(p-ethylphenyl)butane (16). These were explained by the distortions of the aromatic rings and by the rigidly enforced proximity of the two p systems [26, 27]. However, early models that were based solely on the p electron system and neglected the r core were unsatisfactory [28]. An explanation including r–p interaction (as a special case of hyperconjugation) was given by Gleiter in 1969 [29]. It yielded, even on a qualitative level, a satisfactory explanation of the electronic spectra, the ESR spectrum and the photochemical behavior of such compounds. The relevance of transannular electron-

Scheme 11.4

11.4 [m.n]Paracyclophanes

ic interactions is supported by the differences between the spectra of [4.2]paracyclophane (17, see Section 11.4.3) and the open-chain compounds 15 and 16. The latter cyclophane is assumed to be fairly strain-free, and therefore the observed spectral changes cannot be associated with aromatic ring deformations. The anomalous band shape has been explained by the assumption of large structural rearrangements on excitation or emission [30–32]. Comparable broadening and loss of fine structure is similar to those observed for strained [n]paracyclophanes containing a boat-shaped benzene ring [9, 33–36] (see Section 11.3). The presence of additional bands can be explained in principle by exciton and charge resonance interactions between the two chromophores. Similar transitions are observed in the emission spectrum from concentrated benzene solutions and have been assigned to excimer fluorescence [37–39]. However, the assignment of the individual absorption and emission bands for [2.2]paracyclophane (14) has proven a very difficult task, complicated not only by the complexity of the observed overlapping band structures, but also by the divergent results of different theoretical procedures [28]. A series of [2.2]paracyclophane derivatives such as 18–24 that hold chromophores of varying conjugation lengths has been investigated by Bazan and collaborators [40–42]. Comparison of the optical data from the paracyclophane dimers against data from monomeric units provided information on the effect of bringing together the two molecules into close proximity. For example, the optical properties of 21 and 22 (stilbene “dimers”) have been compared with those of 2,5-dimethylstilbene (25). For some of the investigated compounds the spectroscopic data are summarized in Tab. 11.2. The observed trends in absorption, fluorescence and radiative lifetime of these compounds were analyzed using collective electronic oscillators (CEO) representing the changes induced in the reduced single-electronic density matrix upon electronic excitation. Comparison of the CEO of the cyclophanes with the corresponding “mono-

Tab. 11.2 Summary of absorption and emission data of compounds 18–26 (in hexane at room temperature) [41]

Compound

kmax (nm) Absorption

kmax (nm) Emission a)

18 19 20 21 22 25 23 24 26

281 254 288 307 264, 325 294 369 355 349

374 394 386 412 401 338, 407, 407, 391,

355 430 430 415

a) For emission with vibrational structure the two highest energy peaks are reported.

281

282

11 UV/Vis Spectra of Cyclophanes

Scheme 11.5

Scheme 11.6

mers” provided an efficient method for tracing the origin of the various optical transitions by identifying the underlying changes in charge densities and bond orders. The absorptions of 21 and 22 show a red shift of &13 nm relative to the “monomeric” compound 25. More pronounced differences are evident in their respective emission spectra (21: 412, 22: 401, 25: 355 nm). Whereas 25 reveals vibrational fine structure, the bands of 21 and 22 are broad and featureless, reminis-

11.4 [m.n]Paracyclophanes

cent of excimer qualities. The same holds for compounds 19 and 20 and their “monomer” styrene. In the pseudo-ortho isomer 22 the proximity of the two styrene units appears to split the absorption band (264 and 325 nm) relative to 25. A red shift of 15–20 nm is observed in the absorption of the “dimers” 23 and 24 relative to 4-(2,5-dimethylstyryl)-4'-t-butylstilbene (26), but overall there is little difference in these spectra. Also the emissions of 23 and 24 are similar to the “monomer” 26 and display vibrational structure. A combination of spectroscopic and computational efforts produced the qualitative energy diagram shown in Fig. 11.3 to explain the photophysics of 21–24. It was found that in all cases the most significant absorption is attributed to the “monomer” chromophore, i.e. stilbene in the case of 21. There is a second “phane” excited state to consider which contains the paracyclophane core with through-space (p–p) delocalization. Emission from this state is broad and featureless and similar to that which characterizes excimers. Two situations could be distinguished with regard to photoluminescence. For smaller chromophores such as stilbene (21, 22), the energy of the localized excitation (S2) is higher than that of the state containing the paracyclophane core (Fig. 11.3 a). Energy migration by internal conversion after photon absorption transfers the excitation from the localized “monomer” (S2), and emission takes place from the “phane” state (S1). The latter carries a very weak oscillator strength to the ground state. Population of the “phane” state (S1) via energy transfer thus results in a relatively long-lived excited state. Thus, emission occurs from a state containing the through-space delocalized paracyclophane core. The second situation occurs when the energy of the “monomer” state (S1) is lower than that of the corresponding “phane” state (S2) (Fig. 11.3 b). This is the case for larger chromophores such as distyrylbenzene (23, 24). Under these circumstances there is no driving force for energy migration and the excitation re-

S2 S1

Fig. 11.3 Absorption and emission in compounds with (a) and without (b) internal con-

version

283

284

11 UV/Vis Spectra of Cyclophanes

mains localized at the initially generated state. Except for the spectral shift caused by the paracyclophane moiety, there is therefore negligible difference between the spectra of the parent compound 26 and the dimers 23 and 24. 11.4.3

[m.n]Paracyclophanes

Cram et al. [9] have studied the UV spectra of [m.n]paracyclophanes in which m and n were varied stepwise, one carbon at a time from m = n = 2 (14) to m = 5, n = 6 (27, 17, 28–35). Some spectra are shown in Fig. 11.4. This series of compounds permitted systematic variation of the distance between the two benzene rings, and as the most striking feature, characterizing the relationship between the spectra, a discontinuity was observed in the progression of the curves from normal (as compared to open-chain models such as 1,4-bis(p-ethylphenyl)butane, 16) to abnormal as the values of m and n become small. Thus 31 (m = n = 4) differs only slightly in its spectral properties from 34 (m = n = 5) but differs markedly from 29 (m = 3, n = 4). This discontinuity is apparent in both the region of high intensity (k&215 nm) and that of low intensity absorption (k&270 nm), the series of bands in both places moving toward longer wavelengths and lower intensities as the values of m and n become smaller than 4. A similar discontinuity was found in the spectra of the more unsymmetrical compounds in which n £ m–2, again the requirement for a normal spectrum being that both m and n be equal to or greater than 4. As the bands move toward longer wavelengths, they decrease in intensity, and the extinction coefficients e of even larger rings (m ³ 4, n ³ 4) are all about 10% below those of the open-chain models.

Scheme 11.7

11.5 Multibridged [2n]Cyclophanes and Related Compounds

lg e

Fig. 11.4 UV spectra of [m.n]paracyclophanes 14, 27–30, 32, 34, 35 and openchain model 16 in 95% ethanol. The curves, with the exception of 16 at the bottom, have been displaced upward on the ordinate axis by 0.5 lg e unit increments from the curve immediately below (reproduced with permission from ref. 9)

k/nm

11.5

Multibridged [2n]Cyclophanes and Related Compounds

The UV spectra of cyclophanes consisting of two benzene rings that are connected by several ethano bridges, such as 36–47, have been discussed by Boekelheide [43], and their electronic states have been analyzed by Spanget-Larsen [28]. The spectra of these compounds are quite different from those of the corresponding alkylbenzenes. Several theoretical and empirical attempts have been made to correlate and interpret the electronic spectra of [2n]cyclophanes. It has been mentioned above that the absorption at 302 nm of the prototype for the series, [22](1,4)cyclophane ([2.2]paracyclophane, 14), is termed the “cyclophane band”, and this band is characteristic of all [2n]cyclophanes. The two most important factors governing the position and intensity of the “cyclophane band” are the mean distance between the benzene rings, and the extent and nature of the deformation of the benzene rings. The long-wavelength absorption bands of the [2n]cyclophanes are summarized in Tab. 11.3.

285

286

11 UV/Vis Spectra of Cyclophanes

Scheme 11.8

Tab. 11.3 Long-wavelength absorptions of [2n]cyclophanes

Compound

kmax (nm) (lg e)

[22](1,4)Cyclophane (14) [22](1,3)Cyclophane (36) [23](1,3,5)Cyclophane (39) [23](1,2,3)Cyclophane (37) [23](1,2,4)Cyclophane (38) [24](1,2,3,4)Cyclophane (40) [24](1,2,3,5)Cyclophane (41) [24](1,2,4,5)Cyclophane (42) [25](1,2,3,4,5)Cyclophane (43) [26](1,2,3,4,5,6)Cyclophane (44) [22](1,3)(1,4)Cyclophane (45) [23](1,2,4)(1,2,5)Cyclophane (46) [23](1,2,4)(1,3,5)Cyclophane (47) 1,1,2,2,9,9,10,10-Octafluoro[2.2]paracyclophane (52) 1,1,2,2,9,9,10,10-Octamethyl-1,2,9,10-tetrasila[2.2]paracyclophane (54) anti-[22](1,3)Cyclophane-1,9-diene (50) [23](1,3,5)Cyclophane-1,9,17-triene (51) [1:2;9:10]Bismethano[2.2]paracyclophane (53)

286 (2.51) 272 (2.64) 258 (3.08) 272 (2.59) 290 (2.59) 283 (2.57) 287 (2.57) 294 (2.82) 294 (2.55) 296 (2.62) 284 (2.45) 285.5 (2.81) 235 (4.00) 282 (2.86)

302 (2.30) 277 (2.55) 312 (1.98) 275 (2.58) 309 (2.26) 297 (2.45) 308 (2.29) 303 (3.02) 313 (2.30) 311 (2.51) 292 (2.36) 293.5 (2.72) 300 (2.58) 308 (2.15)

223 (4.28)

263 (4.35)

48

252 (3.29) 260 (3.70)

280 (4.45) 325 (1.95) 304 (2.95)

134 124 46

Ref. 121 122, 123 124 125 126 127 128 129 130 131 132 133 133 47

Superphane (44), which has the shortest distance between decks (262.4 pm), but completely planar benzene rings, has its “cyclophane band” at 311 nm, whereas in [23](1,3,5)cyclophane (39) and [25](1,2,3,4,5)cyclophane (43), which have a greater distance between decks but distorted benzene rings, the “cyclophane band” is found at 312 and 313 nm, respectively. The effect of unsaturation in the bridges is not consistent. [22](1,4)Cyclophane1-ene (48) and [22](1,4)cyclophane-1,9-diene (49) have essentially the same UV ab-

11.5 Multibridged [2n]Cyclophanes and Related Compounds

Scheme 11.9

sorption spectrum as [22](1,4)cyclophane (14) itself [44]. However, anti-[22](1,3) cyclophane-1,9-diene (50), which exhibits only a small shift to longer wavelength, compared with anti-[22](1,3)cyclophane (36), shows a dramatic increase in intensity. On the other hand, [23](1,3,5)cyclophane-1,9,17-triene (51) shows a long-wavelength shift of 13 nm compared with 39, but virtually no change in intensity. The “cyclophane band” of 51 has the longest wavelength of any unsubstituted [2n]cyclophane. Modification of the ethano bridges of [2.2]paracyclophane (14) was also achieved by fluorination (52), replacing them by cyclopropane rings (53) or disilane groups (54). The photoelectron spectra of these compounds and that of 49 revealed variable r–p interactions between the bridges and the benzene decks [12, 45, 46]. The long-wavelength absorption band of 53 is little changed from that of 14 and the only evidence in the electronic spectrum of 53 to suggest an unusual r–p interaction is the strong increase in intensity of the “cyclophane band”. In the octafluoro derivative 52, the interaction of the p electron clouds of the aromatic rings is slightly weakened because of fluorination-induced steric hindrances [47]. And this is reflected in an increased intensity of the “cyclophane band” and a lowered extinction of the second band compared with compound 14. Because of the long Si– Si distances, the benzene rings in 54 are less distorted than in 14, however, the compound displays strong r–p mixing between the Si–Si bonds and the aromatic rings as evidenced by a large red shift in the UV spectrum [48]. The low energy singlet and triplet states for [2n]cyclophanes were discussed in terms of the results of a simple model calculation [28]. Experimental trends were explained under the assumption of significant r–p interaction involving the saturated bridges. This hyperconjugative interaction destabilizes low energy excimer states, in contrast to the usual red shift observed for alkylbenzenes. The observed near-constancy of the onset of the absorption spectra is explained by near-cancellation of through-bond and through-space contributions.

287

288

11 UV/Vis Spectra of Cyclophanes

11.6

[m.n]Arenophanes

In a similar way as the previous benzophanes, several other cyclophanes, which can be considered as arene “dimers”, have been investigated. Chromophores studied in this manner include the following of which only a few will be discussed here in some detail: naphthalenophanes [19, 49–53], anthracenophanes [51, 53, 54], fluorenophanes [55, 56], phenanthrenophanes [56, 57], and pyrenophanes [21, 22, 50, 51, 56, 58–60]. 11.6.1

[m.n]Naphthalenophanes

In a series of [2.2]-, [3.2]- and [3.3]naphthalenophanes, Haenel and Schweitzer [19] have studied the properties of the excited states in glasses such as methylcyclohexane at 1.3 K. Compared with the corresponding monomeric dimethylnaphthalenes, the electronic spectra of the naphthalenophanes exhibit bathochromic shifts together with the appearance of new bands and loss of vibrational structure [49, 52, 61–65]. Most compounds also show poorly structured fluorescence and phosphorescence spectra similar to the characteristic emission of intermolecular excimers. The authors concluded that the orientation of interacting aromatic p electron systems strongly influences the extent of the electronic interaction. In the excited singlet and triplet states of the cyclophanes, the electronic interactions, which stabilize the intramolecular excimer, increase if the chromophore planes are parallel and reach a maximum when their six-membered rings are completely eclipsed. As an example, in Fig. 11.5 the UV absorption spectra of two [2.2](2,6)naphthalenophanes (55, 56) are depicted together with that of 2,6-dimethylnaphthalene

Scheme 11.10

11.6 [m.n]Arenophanes Fig. 11.5 UV absorption spectra of (a) achiral [2.2](2,6)naphthalenophane (55), (b) chiral [2.2](2,6)naphthalenophane (56), and (c) 2,6-dimethylnaphthalene (57) in cyclohexane. The curves a and b have been displaced upward on the ordinate axis by 0.5 lg e unit increments from the curve immediately below (reproduced with permission from ref. 52.)

(57). The achiral isomer 55 exhibits a stronger bathochromic shift and hence a stronger p–p electronic interaction than the chiral isomer 56, which has crossed naphthalene units. In the case of the corresponding [3.3](2,6)naphthalenophanes (58, 59) this difference is not so clear, even though the spectra resemble those of 55 and 56. 11.6.2

[n.n]Pyrenophanes

In the absorption spectrum of [2.2](2,7)pyrenophane (60) [kmax = 328 nm (lg e = 4.46), 314 (4.12), 277 (4.19), 264 (4.28), 240 (4.95), in dioxane], compared with 2,7-dimethylpyrene, the absorption at longest wavelength appears strongly red-shifted as a broad, structureless band in the range of 380 to 430 nm [58]. This red shift and broadening is reduced in the absorption of the [3.3]pyrenophane 61 [kmax = 382 nm (sh, lg e &3.1), 368 (3.34), 328 (4.75), 314 (4.40), 271 (4.70), 259

Scheme 11.11

289

290

11 UV/Vis Spectra of Cyclophanes

(4.46), 237 (5.14), in dioxane] [59]. For the [4.4]pyrenophane 62 the long-wavelength start of the absorption shows a further hypsochromic shift to about 410 nm, and the vibrational structure is even more pronounced [kmax = 380 nm (lg e = 2.98), 365 (sh, &3.4), 342 (4.13), 327 (4.74), 314 (4.36), 302 (sh, &4.0), 273 (4.72), 261 (4.55), 239 (5.22), in dioxane] [59]. Thus, the absorption spectra reflect the decreasing transannular interactions in the series of [2.2]-, [3.3]- and [4.4]pyrenophanes.

11.7

Fluorinated Cyclophanes

Koga et al. [66] have investigated the electronic spectra of a series of fluoro[33] (1,3,5)cyclophanes (63–65) along with the parent 66 as a reference in chloroform. The p–p* absorption bands show blue shifts as the number of fluorine atoms is increased. Although in 63 the longest-wavelength band (kmax = 313 nm, lg e = 2.53) suffers a slight red shift compared with that of 66 (kmax = 308 nm, lg e = 1.94), the band shows a gradual blue shift as the number of fluorine atoms increases: 64 (kmax = 296 nm, lg e = 1.93), 65 (kmax = 291 nm, lg e = 2.27). In principle, the HOMO–LUMO gap estimated by ab initio MO calculations (HF/6-31G) (66: 11.03 eV, 63: 11.16 eV, 64: 11.19 eV, and 65: 11.52 eV) supports this phenomenon. The characteristic blue shift is explained in terms of significant lowering of the HOMO level and modest lowering of the LUMO level as the number of fluorine atoms increases. However, the authors could not explain the anomalous behavior of 63.

11.8

Heterocyclophanes

Heteroatoms can modify the physical and chemical properties of an arene without destroying its aromatic character. Since many heterocycles, in particular those with nitrogen atoms, play an important role in biochemistry, heterocyclophanes can be used – among other things – to model certain biosystems. In addition to

Scheme 11.12

11.8 Heterocyclophanes

p–p* transitions which are similar to those in the isoelectronic hydrocarbon, absorption spectra of nitrogen hetarenes are characterized by weak n–p* transitions at the long-wavelength end of the spectrum. A short overview on spectroscopic properties of heterophanes including UV analysis was given by Paudler and Bezoari [67]. Simple cyclophanes with a heterocyclic aromatic ring and a benzene ring, such as 4,7-diaza[2.2]paracyclophane (67) have been investigated by Staab and Appel [68]. Absorption and emission spectra demonstrate the p acceptor quality of pyrazine when linked in a [2.2]paracyclophane system vis-à-vis a strong electron donor. The stereoisomeric 12,15-dimethoxy-4,7-diaza[2.2]paracyclophanes 68 and 69 show CT absorptions at about 350–400 nm (in cyclohexane) which are markedly solvent dependent. Changing the solvent to methanol or acetic acid shifts the absorptions continuously to longer wavelengths. And this trend is increased in trifluoroacetic acid where the absorptions extend to above 600 nm (68: kmax = 440 nm, lg e = 2.61; 69: kmax = 460 nm, lg e = 2.01). The solvent dependence of the CT absorptions is ascribed to changes in the electron affinity of the pyrazine moiety by solvation and protonation. The different CT properties of the stereoisomers 68 and 69 are interpreted as orientation dependence of electron donor–acceptor interactions. Wisor and Czuchajowski have analyzed the UV spectra of [2]paracyclo[2](2,5)pyridinophane (70) and four isomeric [2.2](2,5)pyridinophanes (pseudogeminal 71, pseudopara 72, pseudoortho 73, pseudometa 74) [69]. Pyridine is isoelectronic with benzene and so are pyridinophanes with regard to benzophanes. However, while the changes in the UV spectrum of pyridine when compared to that of benzene concern only a small enhancement of the long-wavelength absorption which represents the p–p* transition (a band) and the new n–p* band covered by the former, a similar analogy between pyridinophanes and benzophanes cannot be expected, because of the interaction in pyridinophanes between the lone-pair electrons on the nitrogen atom and the p electron system in the ring, which appears in addition to the normal transannular electron interaction between the layered rings. The parapyridinophanes investigated show UV spectra of similar shape. Their spectroscopic features correspond to those of the isoelectronic [2.2]paracyclophane (14), and the four observed absorption bands correspond to p–p* transitions. The excimer transitions were characterized by localization numbers showing the contributing exciton-resonance and charge-resonance states. The transannular effect, which influences the third p–p* band most strongly, decreases in the order of pseudogeminal (71) and pseudopara (72, identical with

Scheme 11.13

291

292

11 UV/Vis Spectra of Cyclophanes

Scheme 11.14

[2.2]paracyclophane 14) > [2]paracyclo[2](2,5)pyridinophane (70) > pseudoortho (73) and pseudometa (74). Both n–p* transitions are totally localized in the pyridine ring(s), the localization number of the excitation being almost identical for all pyridinophanes. Heterocyclophanes in which two pyridine rings are connected by four ethano bridges were studied by Kang and Boekelheide [70]. The spectral properties of 5,12-diaza[24](1,2,4,5)cyclophane (75) and 5,15-diaza[24](1,2,4,5)cyclophane (76) provide evidence for a strong transannular interaction between the p electrons of the rings. The UV absorption spectra of 75 and 76 both show a long-wavelength band at 307 nm of high intensity (lg e = 3.79 and 3.72, respectively). This is exactly the region assigned to the “cyclophane band” of the [2n]cyclophanes [43] (see Section 11.5, Tab. 11.3). Zipplies and Staab have investigated p–p interactions in compounds such as [3.3](3,10)isoalloxazinophane (77) in which two flavin molecules are linked together by two trimethylene bridges [71]. Such interactions may be relevant for the mechanism of the enzymatic action of complex flavoproteins containing more than one flavin unit. Electron spectra including CT absorptions of semireduced states were reported for 77 and compared with the mono-bridged bis-flavin analog 78. The absorption spectra of 77 and 78 are very similar in showing, with approximately doubled extinction, the typical flavin bands [77: kmax = 430 nm (lg e = 4.26), 341 (4.23), 269 (4.79); 78: kmax = 434 nm (lg e = 4.27), 338 (4.20), 265 (4.83), in glycol]. On the other hand, in their semi-reduced states 77 and 78 give rise to very different CT absorptions. Solutions of 77 and 78 in glycol were fully reduced with sodium dithionite solution in water. No absorption > 530 nm was observed for the reduced states of 77 and 78 nor for 77 and 78 themselves. The reoxidation was achieved by admission of air in small doses until the new absorptions at longer wavelength reached a

Scheme 11.15

11.8 Heterocyclophanes

Scheme 11.16

maximum. In the case of 78 the semi-reduced state obtained in this way showed a strong broad absorption with kmax = 770 nm (lg e &2.83) reaching into the infrared beyond 1100 nm. This absorption was assigned to an intramolecular CT transition arising from flavin–flavin interaction in a conformation of 78 where the flavin units approach each other in a syn-like orientation, allowing favorable overlap between the oxidized and reduced isoalloxazine moieties. The importance of the intramolecular linking is demonstrated by the observation that monomeric 3,6,10trimethylisoalloxazine under the same conditions does not show a measurable absorption > 530 nm for the formally semi-reduced state. Becher, Stoddart and collaborators have performed binding studies between tetrathiafulvalene (TTF) derivatives and cyclobis(paraquat-p-phenylene) [72, 73]. The powerful p electron donor TTF (79) forms a strong green 1 : 1 complex with the p electron-accepting tetracationic cyclophane, cyclobis(paraquat-p-phenylene) (80), in both the solid and solution states. The complexation between a number of different p electron-donating TTF derivatives and 80 has been studied. The authors have carried out UV/Vis dilution measurements and correlated the maximum absorptions of the CT bands with the concentrations of the components. The results demonstrate that the strength of association between the donors (TTF derivatives)

Scheme 11.17

293

294

11 UV/Vis Spectra of Cyclophanes

Scheme 11.18

and the acceptor (80) is strongly dependent on the p electron-donating properties (measured by the first redox potential) of the TTF derivatives. In related studies the 4,4'-bipyridium dication has been replaced by the 2,7-diazapyrenium unit [74]. Capretta and Bell [75] have studied the influence of conformational distortion and deformation on the electronic excitation in [n](N 6,9)-6-aminopurinophanes (n = 8: 81, n = 9: 82, n = 10: 83), in which the nitrogen atom N-9 and the amino group on C-6 of the adenine core are linked by a polymethylene bridge. The UV spectra of these compounds were found to be markedly different from that of N 6,9-dinonyladenine [nonyl-(9-nonyl-9H-purin-6-yl)amine, 84] that was used as the undistorted standard. The absorptions are summarized in Tab. 11.4. The bathochromic shift of the p–p* transition, as the polymethylene chain length decreases, is the result of the bending of the aromatic plane, similar to the spectra of [n]paracyclophanes [20] (see Section 11.3). The smaller the bridging chain, the greater the deformation of the purine p system. This distortion reduces the energy difference between the p and p* states. The bathochromic shift of the n–p* transition is also a reflection of the molecular geometry, in particular with respect to the conformation of the amino group and its interaction with the aromatic p system. The maximum transition energy is required in 84 where the ni-

Tab. 11.4 UV absorption of N6,9-dinonyladenine (84) and [n](N6,9)-aminopurinophanes 81–83 (in methanol) [75]

Compound

p–p* kmax (nm) (lg e)

n–p* kmax (nm) (lg e)

84 81 82 83

209 218 (4.18) 216 (4.19) 213 (4.21)

269 278 (4.12) 276 (4.13) 273 (4.14)

11.8 Heterocyclophanes

trogen lone-pair is aligned with the purine p system. Altering the interaction by twisting the lone-pair out of conjugation, as in the aminopurinophanes (81–83), raises the energy of the n orbital, and the greater the degree of twisting, the less energy is required to induce a transition. The same structural features manifest themselves in a hyperchromic effect as the chain length increases. Sakata and coworkers [76–78] have investigated several purinophanes as models for stacking interactions of nucleic acid bases. Nucleic acid bases are stacked with interplanar distances of about 340 pm in the helical structure of DNA and this stacking brings about a significant decrease in the intensity of the longest-wavelength absorption band. In fact, the double helix absorbs roughly 40% less than does a mixture of the component monomers. This hypochromism has been widely used as evidence of stacked structures of various p systems, including nucleic acid bases in solution. Depending on the relative orientation of the transition moments, hypochromism (parallel stacking) or hyperchromism (linear array) is observed. Hypochromism is explained by the interaction between one particular electronic excited state of a given chromophore and the different electronic states of the neighboring chromophores. Twelve purinophanes 85–96, in which two purine rings are fixed with different modes of stacking by two or three polymethylene chains, have been prepared by the authors [76]. Five kinds of stacking geometries of the two component rings were determined by X-ray analysis and/or 1H NMR spectroscopy. The interplanar distances vary from 320 to 660 pm. All the purinophanes show high hypochromism (decrease in integrated absorption intensity compared with two molar monomeric references or one molar dimeric linear compound), and the maximum value was 47.6% for 92. This is the highest value so far observed for dimeric nucleic acid bases. The hypochromism values of the purinophanes 85–96

Scheme 11.19

295

296

11 UV/Vis Spectra of Cyclophanes

Scheme 11.20

Scheme 11.21

Scheme 11.22

are almost identical in four different media (ethanol, water, 0.1 N HCl, 0.1 N NaOH). As model systems for active site complexes in flavoenzymes, flavin and nicotinamide analogs were linked together in cyclophane skeletons of specific sterical structures, and UV/Vis spectroscopic results related to p–p interactions were reported [79].

11.9 Multi-layered Cyclophanes

11.9

Multi-layered Cyclophanes

A short overview on electronic spectra of multi-layered cyclophanes is included in a review by Misumi [80]. Electronic spectra, absorption, emission, and CT absorption of CT complexes, of nonsubstituted, multilayered [2.2]paracyclophanes with up to six stacked benzene rings have been measured by Otsubo et al. [81]. The absorption spectra exhibit increasingly strong bathochromic and hyperchromic shifts as the number of layers increases. In particular, these shifts are clearly expressed as the number of layers is increased from one to four, while the absorption bands become markedly structureless. Only minor changes in the spectra are observed when the number of layers is further increased. These spectral features are explained mainly by transannular p–p interaction or transannular delocalization, because the shifts of the absorption bands are little affected by the distortion of the benzene rings. Iwata et al. [82] have investigated the electronic spectra and the electronic structures of [2.2]paracyclophane (14), tetramethyl[2,2]paracyclophane (97), and the related triple- and quadrupole-layered cyclophanes (98 and 99). The absorption spectra were measured as n-heptane and 3-methylpentane solutions in the range of 170–400 nm. Polarized absorption spectra of a [2.2]paracyclophane single crystal were measured in the range of 200–333 nm for the ab plane and 267–333 nm for the ac plane. Strong bands corresponding to the 180 nm band of benzene were newly observed at 189, 201, 202 and 200 nm for 14, 97, 98 and 99, respectively. Otsubo et al. [83] have synthesized triple-layered [2.2][2.2]naphthalenophane (100) and investigated its structure and properties. In this compound, the outer naphthalene rings are bent into a boat form and the inner naphthalene ring is bent into a twist form. These naphthalenes are stacked in layers within van der Waals contact. Therefore, there is a strong transannular p electronic interaction between them. These interactions in 100 were evaluated in comparison with the double-layered homolog [2.2](2,6)naphthalenophane (56) [52] (see Section 11.6.1) and 2,6-dimethylnaphthalene (57) or 2,3,6,7-tetramethylnaphthalene (101) as a standard. 100 and 56 demonstrate the common spectral characteristics of layered cyclophanes,

Scheme 11.23

297

298

11 UV/Vis Spectra of Cyclophanes

Scheme 11.24

i.e. bathochromism, hyperchromism and broadening. These effects are more prominent with an increase of the layers. For example, the longest-wavelength absorption peaks of 57 and 101 appear at 326 and 325 nm, respectively. In contrast, 56 has a shoulder at 336 nm and 100 has its “cyclophane band” at 362 nm, indicating a strong transannular p electronic interaction in accord with the crystal structure. This is also confirmed for 100 by its CT absorption with tetracyanoethylene (TCNE) as an electron acceptor. 57 with TCNE exhibits two broad CT bands at kmax = 478 and 609 nm in dichloromethane. On the other hand, 56 displays the corresponding CT bands at 500 (sh) and 644 nm, and 100 at 514 and 683 nm. Evidently, the bathochromic shifts for these naphthalenophanes indicate the enhancement of their donor abilities caused by the transannular electronic interactions, which are stronger in 100 than in 56.

11.10

Donor–Acceptor Cyclophanes

Donor–acceptor cyclophanes have been extensively studied as models for intermolecular charge–transfer (CT) complexes. In contrast to the flexible intermolecular interactions of individual donor (D) and acceptor molecules (A), the cyclophane structures provide an intramolecular fixation of D and A units with well-defined and adjustable orientations and distances. Such systems have been systematically investigated mainly by Staab and coworkers. Their results have been published in more than 50 papers and briefly summarized some years ago [84]. A few examples will be presented here. 11.10.1

Donor–Acceptor Substituted [n.n]Paracyclophanes

The [2.2]paracyclophanes 102–105, the [3.3]paracyclophanes 106–108, and the [4.4]paracyclophanes 109 and 110 [85], all have the tetracyanobenzene (TCNB) unit as the common electron acceptor. They differ, however, in the strengths of the donor components and in the donor–acceptor distances. Thus, this series of

11.10 Donor–Acceptor Cyclophanes

Scheme 11.25

compounds enables the effect of these two parameters on CT absorptions in geometrically well-defined systems to be studied. The dependence of the CT absorption on donor strengths and donor–acceptor distances is easily recognized by the color of the compounds: in the [2.2]paracyclophane series the compound with the weakest donor part (102) is only yellow; increasing the donor strength by four methyl groups (104) changes the color to red; introducing two methoxy groups as strong donor substituents (103) results in a deep violet color; adding two more methoxy groups and, in this way, sterically hindering the mesomeric effect of all the methoxy substituents shifts the color back to orange-red. Keeping the dimethoxy-substituted donor constant and changing the donor–acceptor distance from [2.2]- to [3.3]- and [4.4]paracyclophanes leads to color changes from deep violet (103) to dark red (107) and orange (110). These qualitative observations are reflected in a quantitative way by the absorption data of the compounds (see below). In weak intermolecular electron donor–acceptor complexes there is only a relatively small perturbation of the electronic structures of the components. In these cases the short-wavelength part of the absorption spectra corresponds to the sum of the absorptions of the components, and only the CT absorption is added at longer wavelength. In the [2.2]- and [3.3]paracyclophane series there is, however, a stronger interaction between the p systems which leads already for the unsubstituted parent compounds (14, 28, see Section 11.4.3) to significant “phane-specific” changes in the absorption spectra: broadening of the absorption bands, loss of vi-

299

300

11 UV/Vis Spectra of Cyclophanes

brational structure, and appearance of new absorptions at longer and/or shorter wavelength. Of course, these phane-specific effects also operate in the substituted paracyclophanes. Only for the [4.4]paracyclophanes 109 and 110 the absorption at shorter wavelength can be rationalized by a correlation to the additive absorption of the components as is seen by comparison with the data of 1,2,4,5-tetracyano3,6-dimethylbenzene (111). The identification of CT absorptions for the donor–acceptor paracyclophanes is clearly possible either by their distinctive absorption bands at long wavelengths or by measurements of the solvent dependence of the fluorescence. Whereas for 102 the CT transition gives rise only to an absorption shoulder at 395 nm (lg e = 2.65), the strengthening of the donor part by the introduction of four methyl groups in 104 leads to a distinct CT absorption band at 440 nm (lg e = 2.73). 103, with the dimethoxybenzene unit, shows a CT absorption with a strong bathochromic shift to 520 nm (lg e = 2.38) (all data for chloroform solutions). By adding two more methoxy groups all four methoxy groups are rotated out of the coplanar arrangement with the aromatic ring as shown clearly by X-ray analysis [85]; consequently the electron-donating +M effect is inoperative leading for 105, compared with 103, to a strong hypsochromic shift of the CT absorption which appears as a shoulder only at 380 nm (lg e = 3.12). Going from [2.2]paracyclophanes to [3.3]paracyclophanes with the same substitution pattern, e.g. from 103 (kmax = 520 nm, lg e = 2.38) to 107 (kmax = 508 nm, lg e = 2.54) or from 104 (kmax = 440 nm, lg e = 2.73) to 108 (kmax = 434 nm, lg e = 2.76), only minor changes in the absorption wavelengths are observed. With the further increase in donor– acceptor distances in the corresponding [4.4]paracyclophanes, because the donor– acceptor overlap is reduced, hypsochromic shifts as well as reduced absorption intensities are observed. This is indicated by the data of the three homologous tetracyanodimethoxyparacyclophanes 103, 107 and 110 (kmax = 495 nm, lg e = 1.95). As one would expect for CT transitions leading to a highly polar first excited singlet state the fluorescence of these donor–acceptor cyclophanes shows a strong red shift with increasing polarity of the solvents. 11.10.2

[n.n]Paracyclophane Quinhydrones

Typical examples of the approach followed by the Staab group are pseudo-orthoand pseudogeminal [n.n]cyclophane quinhydrones and related D-A analogs [84, 86–90]. In a series of quinhydrones of [2.2]- to [6.6]paracyclophanes and their dimethyl ethers the dependence of CT absorption on distance, mutual orientation, and the degree of the fixation of the D and A components were studied. In these compounds, two para-disubstituted benzenes, one with two electron donor substituents and the other with two acceptor substituents, can interact in two different orientations, which are distinguished as pseudogeminal with eclipsed substituents and pseudoortho with noneclipsed substituents. A typical example is shown in Fig. 11.6 where the CT absorptions of the [3.3]paracyclophane quinhydrones 112 and 113 are compared [89]. The CT bands,

11.10 Donor–Acceptor Cyclophanes

Scheme 11.26

Fig. 11.6 Absorption spectra of [3.3]paracyclophane quinhydrones 112 and 113 in dioxane (reproduced with permission from ref. 89)

Scheme 11.27

301

302

11 UV/Vis Spectra of Cyclophanes

observed at long wavelengths, are similar to those present in the corresponding [2.2]paracyclophane quinhydrones (114/115) [86, 87]. However, the difference in the absorption caused by the different orientation of D and A found for the latter compounds, are even stronger expressed in 112 (kmax = 462 nm, lg e = 3.51) and 113 (kmax&500 nm, lg e = 2.02, and kmax = 356 nm, lg e = 3.28) [89]. The quinhydrone dimethyl ether 116 in the pseudogeminal structure has the most intense CT absorption band (kmax = 475 nm, lg e = 3.48) [89], because the D and A components obviously are optimally arranged for an intramolecular interaction. The CT absorption of the pseudoortho isomer 117 (kmax = 500 nm, lg e = 2.00, and kmax = 367 nm, lg e = 4.07) [89] is considerably less distinct. A similar strong dependence on donor–acceptor orientation for CT absorptions was measured for other [2.2]paracyclophanes such as 118/119 [91], 120/121 [92], 122/123 [93], as well as for [3.3]paracyclophanes such as 124/125 [94]. In each case, the pseudogeminal derivative gave a longer-wavelength CT band with a significantly higher extinction coefficient compared to the pseudoortho derivative. In all these cyclophanes, CT occurs predominantly in the excited state and does not contribute significantly to the electronic ground state. Even in the [3.3]paracyclophane 126 with N,N,N',N'-tetramethyl-p-phenylenediamine (TMPD) as a strong electron donor and 7,7,8,8-tetracyanoquinodimethane (TCNQ) as a strong acceptor, a ground state CT that would form a diradical-zwitterionic species was not observed [95]. In the next higher [n.n]paracyclophane quinhydrones (n = 4, 5, 6, 127/128) in which the donor–acceptor orientation is much less fixed than in lower members of this series (n = 2, 3), the pronounced difference between the intensities of the CT band of the pseudogeminal and pseudoortho stereoisomers has disappeared [90].

Scheme 11.28

11.10 Donor–Acceptor Cyclophanes

Scheme 11.29

11.10.3

[n.n]Metacyclophane Quinhydrones

Similar investigations of CT interactions as on donor–acceptor paracyclophanes have been performed on analogous [n.n]metacyclophanes (n = 2, 3). In contrast to paracyclophanes which have rather rigid structures, the metacyclophanes are conformationally flexible and can occupy syn and anti forms with quite different mutual orientations of the donor and the acceptor components. For the [3.3]metacyclophane quinhydrone dimethylethers 129 (syn) and 130 (anti) it was found that the syn isomer is the more stable [96]. 129 [kmax = 402 nm (lg e = 3.43), in chloroform; 388 nm (3.40), in dioxane] and 130 [kmax = 402 nm (lg e = 3.51), in chloroform; 487 nm (3.53), in dioxane] show solvent-dependent, broad, intensive CT absorptions in the range from 300 to 550 nm which for the pairs of stereoisomers, in spite of the very different donor–acceptor orientation, are surprisingly similar; the absorption intensity of the CT band, however, is considerably stronger for the anti than for their syn isomer. This and similar examples [97–102] demonstrate that no definite conclusion can be drawn from the wavelength and intensity of the CT band to the ground state stability of the donor–acceptor system. The closely related system 131/132, in which one oxygen atom of the p-benzoquinone unit is replaced by a dicyanomethylene group, has been studied in order to determine directly the relative importance of through-space and through-bond mechanisms for electron transfer reactions [103]. It was found that back-electrontransfer dynamics following photoexcitation of the ground state CT absorption band are the same in the two systems. The results provide direct evidence for the

O

Scheme 11.30

303

304

11 UV/Vis Spectra of Cyclophanes

O

Scheme 11.31

through-bond mechanism of electron transfer in bridged organic donor–acceptor systems. 11.10.4

Oligooxa[m.n]paracyclophane Quinhydrones and Related Compounds

Staab and collaborators have investigated the intramolecular oligooxa[3n.3]paracyclophane quinhydrones 133–136 (with n = 3 to 6) [104, 105] and the oligooxa[n.n]paracyclophane quinhydrones 137–141 (with n = 2 to 6) [106]. In particular, their interactions with alkali and alkaline earth metal and mercury(II) ions were studied by electron absorption spectroscopy. Some data are summarized in Tab. 11.5. While compounds 133 and 134 exhibit no or small changes in the intensity of the CT absorptions by complexation with metal ions, remarkable enhancements were observed for the pentaoxa[15.3]paracyclophane quinhydrone 135 and the hexaoxa[18.3]paracyclophane quinhydrone 136 when they were transferred to the corresponding crown-ether like complexes [105]. The macrocyclic oligooxa[n.n]paracyclophane quinhydrones 137–141 show only small absorption intensities in the wavelength range where the CT absorption of quinhydrones is to be expected (kmax&450 nm) [106]. For example, for 140 kmax = 447 nm with lg e = 1.99 (in chloroform) is observed. When 140, however, is complexed with sodium ions, this absorption band shows a fivefold increase in intensity to lg e = 2.68 and a bathochromic shift to kmax = 477 nm. This spectral change can be reversed by adding the stronger complex ligand 4,7,13,16,21-pen-

Tab. 11.5 CT absorptions of the oligooxa[3n.3]paracyclophane quinhydrones 133–136 (in chloro-

form) and their changes caused by cations [105] Added salt

none NaSCN KSCN Ba(SCN)2 Hg(SCN)2

kmax (nm) (lg e) 133

134

135

467 (2.73) 467 (2.73) 467 (2.73)

467 (2.69) 464 (2.66) 465 (2.68)

462 478 464 438 475

136 (2.51) (2.94) (2.57) (2.71) (2.94)

442 459 465 470

(2.33) (2.57) (2.55) (2.75)

11.10 Donor–Acceptor Cyclophanes

Scheme 11.32

Scheme 11.33

taoxa-1,10-diazabicyclo[8.8.5]tricosane (“Kryptofix 221”). The strong CT absorption of the sodium complex of 140 is explained by assuming a structure as shown in 142 in which a double complexation induces a stacking of the donor–acceptor rings at considerably shorter distances than in 140. Addition of NaSCN to chloroform solutions of the smaller ring systems 137–139 does not significantly change the intensity of the CT band; a small hypsochromic shift of this band is observed with 138 and 139. With the larger ring system 141, complexation with sodium ions does not result in a similar intensity increase of the CT absorption as with 140, whereas for 141 potassium ions show a much stronger complexation effect than sodium ions [141: k = 430 (sh, lg e = 1.71); 141 + KSCN: kmax = 455 mm (lg e = 2.43), in chloroform]. 11.10.5

Donor–Acceptor-Substituted [n.n]Cyclophanes with Extensive p-Electron Systems

CT interactions between D and A components possessing extensive p-electron systems have been investigated in the cyclophanes 143 and 144 consisting of pyrene

305

306

11 UV/Vis Spectra of Cyclophanes

Scheme 11.34

as D and 1,8;4,5-naphthalenetetracarboxdiimide as A [107]. The components D and A are linked by two tri- or tetramethylene chains from C2 and C7 of pyrene to the nitrogens in the corresponding positions of 1,8;4,5-naphthalenetetracarboxdiimide. In the absorption spectrum of 143, a very broad and structureless band in the range from about 420 to 660 nm, with kmax&529 nm and lg e = 3.24 (in chloroform), is observed, that cannot be attributed to the individual chromophores and thus must be assigned to a CT transition. On the other hand, in the short-wavelength region of the spectrum (250 to 400 nm) structured absorption bands are found, which are related to the individual absorptions of the components of 143, although they are broadened and shifted in comparison to the isolated donor and acceptors moieties. The corresponding [4.4]cyclophane 144 displays an absorption spectrum very similar to that of 143. There is again a broad structureless band extending from 430 to 660 nm, with kmax&520 nm and lg e = 3.00. Likewise, absorptions at short wavelengths are seen, corresponding to the components. The similarity between the absorption spectra of 143 and 144 is in marked contrast to the situation with normal donor–acceptor-[3.3]- and [4.4]paracyclophanes, where significant differences in the spectra are typically observed (see Sections 10.1 and 10.2) [85, 89, 90, 93–95, 108]. CT interactions and other spectroscopic properties have also been investigated in cyclophanes in which porphyrins form the donor part. As acceptors different groups such as quinones [109–116], 7,7,8,8-tetracyanoquinodimethane (TCNQ) [117], pyromellitic diimide [118], and others [119, 120] were utilized. The examples described in the preceding sections clearly demonstrate the versatility of UV/Vis spectroscopy in the study of cyclophanes. However, only few general principles with regard to the electronic properties of these compounds could be exemplified. To conclude it can be stated that many of the results are still awaiting their closer theoretical clarification.

11.11 References

11.11

References 1 2

3 4

5

6 7

8

9

10 11 12 13 14 15 16

17

18

19

D. J. Cram, J. M. Cram, Acc. Chem. Res. 1971, 4, 204–213. G. J. Bodwell, Angew. Chem. 1996, 108, 2221–2224; Angew. Chem. Int. Ed. Engl. 1996, 35, 2085–2088. R. Gleiter, D. Kratz, Acc. Chem. Res. 1993, 26, 311–318. F. Vögtle, Cyclophan-Chemie: Synthesen, Strukturen, Reaktionen; Einführung und Überblick, Teubner, Stuttgart, Germany, 1990. F. Vögtle, Cyclophane Chemistry: Synthesis, Structures and Reactions, Wiley, Chichester, 1993. F. Diederich, Cyclophanes, Royal Society of Chemistry, Cambridge, UK, 1991. D. J. Cram, J. M. Cram, Container Molecules and Their Guests, Royal Society of Chemistry, Cambridge, UK, 1994. H. Hopf, Classics in Hydrocarbon Chemistry, Ch. 12.3, Wiley-VCH, Weinheim, Germany 2000, pp. 337–368. D. J. Cram, N. L. Allinger, H. Steinberg, J. Am. Chem. Soc. 1954, 76, 6132– 6141. R. Hoffmann, Acc. Chem. Res. 1971, 4, 1–9. M. N. Paddon-Row, Acc. Chem. Res. 1982, 15, 245–251. R. Gleiter, W. Schäfer, Acc. Chem. Res. 1990, 23, 369–375. E. Clar, Ber. Dtsch. Chem. Ges. 1936, 69, 607–614. H. B. Klevens, J. R. Platt, J. Chem. Phys. 1949, 17, 470–484. T. Förster, K. Kasper, Z. Elektrochem. 1955, 59, 976–980. T. Förster, Angew. Chem. 1969, 81, 364– 374; Angew. Chem. Int. Ed. Engl. 1969, 8, 333. M. Klessinger, J. Michl, Lichtabsorption und Photochemie organischer Moleküle, VCH, Weinheim, Germany 1990. M. Klessinger, J. Michl, Excited States and Photochemistry of Organic Molecules, VCH Publ., New York, 1995. M. W. Haenel, D. Schweitzer, in Polynuclear Aromatic Compounds (Ed. L. B. Ebert), Vol. 217, Advances in Chemistry Series, American Chemical Society, Washington, D.C., 1988, pp. 333–355.

20

21

22

23 24

25 26

27

28 29 30 31

32

33 34 35 36

37

38

N. L. Allinger, J. T. Sprague, T. Liljefors, J. Am. Chem. Soc. 1974, 96, 5100– 5104. G. J. Bodwell, J. N. Bridson, M. K. Cyranski, J. W. J. Kennedy, T. M. Krygowski, M. R. Mannion, D. O. Miller, J. Org. Chem. 2003, 68, 2089–2098. G. J. Bodwell, J. J. Fleming, M. R. Mannion, D. O. Miller, J. Org. Chem. 2000, 65, 5360–5370. E. Clar, Spectrochim. Acta 1950, 4, 116– 121. T. Tsuji, M. Ohkita, T. Konno, S. Nishida, J. Am. Chem. Soc. 1997, 119, 8425–8431. D. J. Cram, H. Steinberg, J. Am. Chem. Soc. 1951, 73, 5691–5704. P. M. Keehn, in Cyclophanes (Eds. P. M. Keehn, S. M. Rosenfeld), Vol. 1, Academic Press, New York, 1983, pp. 69–238. D. J. Cram, R. H. Bauer, N. L. Allinger, R. A. Reeves, W. J. Wechter, E. Heilbronner, J. Am. Chem. Soc. 1959, 81, 5977–5983. J. Spanget-Larsen, Theoret. Chim. Acta (Berlin) 1983, 64, 187–203. R. Gleiter, Tetrahedron Lett. 1969, 51, 4453–4456. A. Ron, O. Schnepp, J. Chem. Phys. 1966, 44, 19–22. G. Melzer, D. Schweitzer, K. H. Hausser, J. P. Colpa, M. W. Haenel, Chem. Phys. 1979, 39, 229–235. W. Goldacker, D. Schweitzer, K. P. Dinse, K. H. Hausser, Chem. Phys. 1980, 48, 105–111. D. S. Cram, C. S. Montgomery, C. R. Knox, J. Am. Chem. Soc. 1966, 88, 515–525. A. D. Wolf, V. V. Kane, R. H. Levin, M. Jones, J. Am. Chem. Soc. 1973, 95, 1680. V. V. Kane, A. D. Wolf, M. Jones, J. Am. Chem. Soc. 1974, 96, 2643–2644. Y. Tobe, K. Kakiuchi, Y. Odaira, T. Hosaki, Y. Kai, N. Kassi, J. Am. Chem. Soc. 1983, 105, 1376–1377. T. V. Ivanova, G. Mokeeva, B. Sveshnikov, Opt. i Spektroskopiya 1962, 12, 586; Opt. Spectry. 1962, 12, 325–328. T. Azumi, S. P. McGlynn, J. Chem. Phys. 1964, 41, 3131–3138.

307

308

11 UV/Vis Spectra of Cyclophanes 39 40

41

42

43 44 45 46

47

48

49 50

51

52 53

54 55 56 57

T. Azumi, S. P. McGlynn, J. Chem. Phys. 1965, 42, 1675–1680. W. J. Oldham, Y. J. Miao, R. J. Lachicotte, G. C. Bazan, J. Am. Chem. Soc. 1998, 120, 419–420. a) G. C. Bazan, W. J. Oldham, R. J. Lachicotte, S. Tretiak, V. Chernyak, S. Mukamel, J. Am. Chem. Soc. 1998, 120, 9188–9204; b) G. C. Bazan, personal communication 2003. S. Wang, G. C. Bazan, S. Tretiak, S. Mukamel, J. Am. Chem. Soc. 2000, 122, 1289–1297. V. Boekelheide, Top. Curr. Chem. 1983, 113, 87–143. K. C. Dewhirst, D. J. Cram, J. Am. Chem. Soc. 1958, 80, 3115–3125. E. Heilbronner, J. P. Maier, Helv. Chim. Acta 1974, 57, 151–159. R. Gleiter, M. Eckert-Maksic, W. Schäfer, E. A. Truesdale, Chem. Ber. 1982, 115, 2009–2011. R. N. Nurmukhametov, L. V. Volkova, K. A. Mailyan, A. V. Pebalk, I. E. Kardash, A. Takahashi, Zh. Fiz. Khim. 1999, 73, 1861–1864; Russ. J. Phys. Chem. (Engl. Transl.) 1999, 73, 1676–1679. H. Sakurai, S. Hoshi, A. Kamiya, A. Hosomi, C. Kabuto, Chem. Lett. 1986, 1781–1784. M. W. Haenel, H. A. Staab, Chem. Ber. 1973, 106, 2203–2216. T. Otsubo, S. Mizogami, N. Osaka, Y. Sakata, S. Misumi, Bull. Chem. Soc. Jpn. 1977, 50, 1858–1862. S. Ishikawa, J. Nakamura, S. Iwata, M. Sumitami, S. Nagakura, Y. Sakata, S. Misumi, Bull. Chem. Soc. Jpn. 1979, 52, 1346–1350. N. E. Blank, M. W. Haenel, Chem. Ber. 1983, 116, 827–832. Y. Tobe, T. Takahashi, T. Ishikawa, M. Yoshimura, M. Suwa, K. Kobiro, K. Kakiuchi, R. Gleiter, J. Am. Chem. Soc. 1990, 112, 8889–8894. M. Sauer, H. A. Staab, Liebigs Ann. Chem. 1984, 615–617. M. W. Haenel, Tetrahedron Lett. 1976, 36, 3121–3124. J. P. Colpa, K. H. Hausser, D. Schweitzer, Chem. Phys. 1978, 29, 187–199. D. Schweitzer, K. H. Hausser, M. W. Haenel, Chem. Phys. 1978, 29, 181–185.

58 59

60 61 62 63 64 65 66 67

68 69 70 71 72

73

74

75 76

77

78

H. A. Staab, R. G. H. Kirrstetter, Liebigs Ann. Chem. 1979, 886–898. H. A. Staab, N. Riegler, F. Diederich, C. Krieger, D. Schweitzer, Chem. Ber. 1984, 117, 246–259. G. J. Bodwell, J. J. Fleming, D. O. Miller, Tetrahedron 2001, 57, 3577–3585. D. J. Cram, C. K. Dalton, G. R. Knox, J. Am. Chem. Soc. 1963, 85, 1088–1093. H. H. Wasserman, P. M. Keehn, J. Am. Chem. Soc. 1969, 91, 2374–2375. N. E. Blank, M. W. Haenel, Chem. Ber. 1981, 114, 1520–1530. N. E. Blank, M. W. Haenel, Chem. Ber. 1981, 114, 1531–1538. M. W. Haenel, Chem. Ber. 1978, 111, 1789–1797. T. Koga, M. Yasutake, T. Shinmyozu, Org. Lett. 2001, 3, 1419–1422. W. W. Paudler, M. D. Bezoari, in Cyclophanes (Eds. P. M. Keehn, S. M. Rosenfeld), Vol. 2, Academic Press, New York, 1983, pp. 359–441. H. A. Staab, W. K. Appel, Liebigs Ann. Chem. 1981, 1065–1072. A. K. Wisor, L. Czuchajowski, J. Phys. Chem. 1986, 90, 1541–1547. H. C. Kang, V. Boekelheide, J. Am. Chem. Soc. 1984, 106, 2672–2680. M. F. Zipplies, H. A. Staab, Tetrahedron Lett. 1984, 25, 1035–1038. D. Philp, A. M. Z. Slawin, N. Spencer, J. F. Stoddart, D. J. Williams, J. Chem. Soc. Chem. Commun. 1991, 1584–1586. M. B. Nielsen, J. O. Jeppesen, J. Lau, C. Lomholt, D. Damgaard, J. P. Jacobsen, J. Becher, J. F. Stoddart, J. Org. Chem. 2001, 66, 3559–3563. V. Balzani, A. Credi, S. J. Langford, A. prodi, J. F. Stoddart, M. Venturi, Supramol. Chem. 2001, 13, 303–311. A. Capretta, R. A. Bell, Can. J. Chem. 1995, 73, 2224–2232. F. Seyama, K. Akahori, Y. Sakata, S. Misumi, M. Aida, C. Nagata, J. Am. Chem. Soc. 1988, 110, 2192–2201. F. Hama, Y. Sakata, S. Misumi, M. Aida, C. Nagata, Tetrahedron Lett. 1982, 23, 3061–3064. K. Akahori, F. Hama, Y. Sakata, S. Misumi, Tetrahedron Lett. 1984, 25, 2379– 2384.

11.11 References 79

80

81

82

83

84

85 86 87 88 89

90 91

92 93 94 95

96 97 98

H. A. Staab, M. F. Zipplies, T. Müller, M. Storch, C. Krieger, Chem. Ber. 1994, 127, 1667–1680. S. Misumi, in Cyclophanes (Eds. P. M. Keehn, S. M. Rosenfeld), Vol. 2, Ch. 10, Academic Press, New York, 1983, pp. 573–628. T. Otsubo, S. Mizogami, I. Otsubo, Z. Tozuka, A. Sakagami, Y. Sakata, S. Misumi, Bull. Chem. Soc. Japan 1973, 46, 3519–3530. S. Iwata, K. Fuke, M. Sasaki, S. Nagakura, T. Otsubo, S. Misumi, J. Mol. Spectrosc. 1973, 46, 1–15. T. Otsubo, Y. Aso, F. Ogura, S. Misumi, A. Kawamoto, J. Tanaka, Bull. Chem. Soc. Jpn. 1989, 62, 164–170. H. A. Staab, in New Aspects of Organic Chemistry I (Eds. Z. Yoshida, T. Shiba, Y. Oshiro), Ch. 10, VCH, Weinheim, Germany 1989, pp. 227–236. H. A. Staab, C. Krieger, P. Wahl, K.-Y. Kay, Chem. Ber. 1987, 120, 551–558. H. A. Staab, W. Rebafka, Chem. Ber. 1977, 110, 3333–3350. H. A. Staab, C. P. Herz, H.-E. Henke, Chem. Ber. 1977, 110, 3351–3357. H. A. Staab, A. Döhling, C. Krieger, Liebigs Ann. Chem. 1981, 1052–1064. H. A. Staab, C. P. Herz, C. Krieger, M. Rentzea, Chem. Ber. 1983, 116, 3813– 3830. H. A. Staab, B. Starker, C. Krieger, Chem. Ber. 1983, 116, 3831–3845. H. A. Staab, R. Reimann-Haas, P. Ulrich, C. Krieger, Chem. Ber. 1983, 116, 2808–2826. H. A. Staab, H. Haffner, Chem. Ber. 1977, 110, 3358–3365. H. A. Staab, V. Taglieber, Chem. Ber. 1977, 110, 3366–3376. H. A. Staab, G. Gabel, C. Krieger, Chem. Ber. 1983, 116, 2827–2834. H. A. Staab, R. Hinz, G. H. Knaus, C. Krieger, Chem. Ber. 1983, 116, 2835– 2847. H. A. Staab, C. P. Herz, A. Döhling, C. Krieger, Chem. Ber. 1980, 113, 241–254. H. A. Staab, M. Joerns, C. Krieger, M. Rentzea, Chem. Ber. 1985, 118, 796–813. H. Vogler, L. Schanne, H. A. Staab, Chem. Ber. 1985, 118, 1254–1260.

99 100

101 102

103

104

105

106 107 108

109

110 111

112

113

114

115 116 117

118

H. A. Staab, W. R. K. Reibel, C. Krieger, Chem. Ber. 1985, 118, 1230–1253. H. A. Staab, L. Schanne, C. Krieger, V. Taglieber, Chem. Ber. 1985, 118, 1204– 1229. H. A. Staab, A. Döhling, C. Krieger, Tetrahedron Lett. 1991, 32, 2215–2218. T. Yamato, J. Matsumoto, N. Shinoda, S. Ide, M. Shigekuni, M. Tashiro, J. Chem. Res. S 1994, 178–179. S. H. Pullen, M. D. Edington, S. L. Studer-Martinez, J. D. Simon, H. A. Staab, J. Phys. Chem. A 1999, 103, 2740–2743. H. Bauer, J. Briaire, H. A. Staab, Angew. Chem. 1983, 95, 330–331; Angew. Chem. Int. Ed. Engl. 1983, 22, 334. H. Bauer, V. Matz, M. Lang, C. Krieger, H. A. Staab, Chem. Ber. 1994, 127, 1993–2008. H. Bauer, J. Briaire, H. A. Staab, Tetrahedron Lett. 1985, 26, 6175–6178. H. A. Staab, D.-Q. Zhang, C. Krieger, Liebigs Ann.-Recl. 1997, 1551–1556. H. A. Staab, G. H. Knaus, H.-E. Henke, C. Krieger, Chem. Ber. 1983, 116, 2785– 2807. H. A. Staab, T. Carell, Angew. Chem. 1994, 106, 1534–1536; Angew. Chem. Int. Ed. Engl. 1994, 33, 1466–1468. H. A. Staab, T. Carell, A. Döhling, Chem. Ber. 1994, 127, 223–229. H. A. Staab, A. Feurer, R. Hauck, Angew. Chem. 1994, 106, 2542–2545; Angew. Chem. Int. Ed. Engl. 1994, 33, 2428–2431. H. A. Staab, C. Krieger, C. Anders, A. Rückemann, Chem. Ber. 1994, 127, 231– 236. H. A. Staab, M. Tercel, R. Fischer, C. Krieger, Angew. Chem. 1994, 106, 1531– 1534; Angew. Chem. Int. Ed. Engl. 1994, 33, 1463–1466. H. A. Staab, A. Feurer, C. Krieger, A. S. Kumar, Liebigs Ann.-Recl. 1997, 2321– 2336. H. A. Staab, R. Hauck, B. Popp, Eur. J. Org. Chem. 1998, 631–642. H. A. Staab, B. Kratzer, S. Quazzotti, Eur. J. Org. Chem. 1998, 2149–2160. H. A. Staab, J. Weikard, A. Rückemann, A. Schwögler, Eur. J. Org. Chem. 1998, 2703–2712. H. A. Staab, S. Nikolic, C. Krieger, Eur. J. Org. Chem. 1999, 1459–1470.

309

310

11 UV/Vis Spectra of Cyclophanes 119 O. Korth, A. Wiehe, H. Kurreck, B. 120

121

122 123 124 125 126 127

Roder, Chem. Phys. 1999, 246, 363–372. D. R. Benson, R. Valentekovich, S.-W. Tam, F. Diederich, Helv. Chim. Acta 1993, 76, 2034–2060. D. J. Cram, N. L. Allinger, H. Steinberg, J. Am. Chem. Soc. 1954, 76, 6132– 6141. E. Heilbronner, Z.-Z. Yang, Top. Curr. Chem. 1983, 115, 1–55. K. Burri, W. Jenny, Helv. Chim. Acta 1967, 50, 1978–1993. V. Boekelheide, R. A. Hollins, J. Am. Chem. Soc. 1973, 95, 3201–3208. B. Neuschwander, V. Boekelheide, Israel J. Chem. 1980, 20, 288–290. A. F. Mourad, H. Hopf, Chem. Ber. 1980, 113, 2358–2371. J. Kleinschroth, H. Hopf, Angew. Chem. 1979, 91, 336–337; Angew. Chem. Int. Ed. Engl. 1979, 18, 329.

128 W. Gilb, K. Menke, H. Hopf, Angew.

129 130 131 132

133 134 135

Chem. 1977, 89, 177–178; Angew. Chem. Int. Ed. Engl. 1977, 16, 191. R. Gray, V. Boekelheide, J. Am. Chem. Soc. 1979, 101, 2128–2136. P. F. T. Schirch, V. Boekelheide, J. Am. Chem. Soc. 1981, 103, 6873–6878. Y. Sekine, V. Boekelheide, J. Am. Chem. Soc. 1981, 103, 1777–1785. D. J. Cram, R. C. Helgeson, D. Lock, L. A. Singer, J. Am. Chem. Soc. 1966, 88, 1324–1325. M. Nakazaki, K. Yamamoto, Y. Muira, J. Org. Chem. 1978, 43, 1041–1044. R. H. Mitchell, V. Boekelheide, J. Am. Chem. Soc. 1974, 96, 1547–1557. S. L. Mattes, S. Farid, Science 1984, 226, 917–921.

359

14

Molecular Electrochemistry of Cyclophanes Bernd Speiser

14.1

Introduction 14.1.1

Electron Transfer and Molecular Electrochemistry

As for many other classes of chemical compounds, the redox behavior of cyclophanes is of continuing interest. The transfer of electrons to (reduction) and from (oxidation) such molecules may generate stable products or reactive intermediates in a variety of oxidation states. Redox reactions can be induced by electrochemical means, where the electrons correspond to a current through an often inert electrode. Such investigation of the molecules’ redox properties is called molecular electrochemistry [1] and yields detailed information about the kinetics, thermodynamics, and mechanisms of the electron transfer steps and coupled chemical reactions [2]. These data can be further analyzed with respect to electronic interactions within molecules, stability and reactivity of various redox states, and more. Molecular electrochemistry thus provides insight into various aspects of behavior related to redox transitions. 14.1.2

Molecular Electrochemistry of Cyclophanes

Most molecules undergo redox processes, although at very different potentials. For example, hydrocarbons are often difficult to both oxidize and reduce. Thus, their formal redox potentials are rather positive or negative, respectively. On the other hand, if the molecular backbone carries substituents, which easily take up or release electrons, the redox potentials of the compound become much less extreme. Molecules with attached transition metal atoms are particularly susceptible to electron transfers, since these atoms are often stable in a variety of oxidation states.

Modern Cyclophane Chemistry. Edited by Rolf Gleiter, Henning Hopf Copyright © 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 3-527-30713-3

360

14 Molecular Electrochemistry of Cyclophanes

These general considerations are also true for cyclophanes and the present chapter will discuss the results of electrochemical studies on some hydrocarbon, functionalized and transition metal-complexed cyclophanes (for the latter, see also chapter 6). A general formulation for these classes of compounds is given as types A–E (R = substituent, X = hetero atom, L = general ligand, M = metal atom; the circles symbolize the cyclophane backbone). Although various general accounts of cyclophane chemistry are available, only a few (see e.g. [3, 4]) mention electrochemical redox reactions of such compounds in a systematic manner. 14.1.3

Methods of Molecular Electrochemistry

A variety of instrumental techniques is used in molecular electrochemistry, among which probably the most popular is cyclic voltammetry (CV). Space and scope of this chapter does not allow detailed discussion of this (or other) methods. Several general reviews have appeared [5–7]. For an understanding of the results discussed here, it may suffice to mention that a cyclic voltammogram is a current/potential curve recorded in the quiet (non-stirred), often non-aqueous solution of the compound in question. The electrode is subjected to a triangular variation of potential E (relative to a reference electrode with arbitrary potential 0 V), and the resulting current i is measured. In the cyclic voltammogram, i is plotted versus E generally in a clockwise manner. The onset of a redox process is usually indicated by a current peak, formed by the exponential increase of i with E, and the relaxation of the diffusion layer at the electrode with time t. Qualitative and quantitative analysis [2] of the curves leads to mechanistic information and numerical values for rate constants, formal redox potentials and so on. It is thus not surprising that CV is also the method of choice for the investigation of cyclophane molecular electrochemistry. An important qualitative piece of information gained from cyclic voltammograms is the “reversibility” of the redox process. Unfortunately, two meanings of this term in molecular electrochemistry are often not clearly distinguished in the literature. On the one hand, a redox reaction is termed reversible if it is more or less diffusion controlled, i.e. the electron transfer is fast with respect to transport processes in the solution. On the other hand, the same term is used to express the absence of followup reactions of the electron transfer product species. We will use the expanded term “chemical reversibility” for the second meaning. Such a chemically reversible process exhibits a cyclic voltammogram with peaks in opposite current directions for the oxidation and the reduction reaction, respectively. Absence of chemical reversibility is indicated by the absence of a reverse peak in the voltammogram. It is important to note that both types of reversibility depend on the time scale of the CV experiments, which is easily varied by adjusting the potential scan rate v.

14.2 Molecular Electrochemistry of Hydrocarbon Cyclophanes

From CV experiments of chemically reversible systems the formal redox potential E 0 of the redox process involved is simply estimated as the mean value of the potentials where the two peaks are located. Unfortunately, for highly irreversible (slow electron transfer) reactions, and in particular chemically irreversible processes, this is no longer possible, and E 0 determination is more complicated and less reliable. In such cases, computer simulation of the CV curve must often be performed [8]. One drawback in comparing results from different laboratories is the use of various reference electrodes, which are often difficult to compare in non-aqueous solvents (see e.g. [9] for a recent discussion, and the proposal to use ferrocene (fc) as a standard in non-aqueous solution [10]). Consequently, this review will mostly mention only relative potential variations reported within one or in closely related papers, rather than giving extensive tables of E 0 values. Also, experimental conditions are only mentioned if necessary for the understanding or in the context of depicted voltammograms. In all other cases, the reader is referred to the original literature.

14.2

Molecular Electrochemistry of Hydrocarbon Cyclophanes

Both electrochemical oxidation and reduction of cyclophane hydrocarbons (general type A) have been studied.

The oxidation of [2.2]metacyclophane 1 and some alkyl-substituted derivatives [11] is characterized by chemically irreversible processes, caused by follow-up reactions of the primarily formed radical cation. Products are tetrahydropyrene, dihydropyrene and pyrene in subsequent steps. Similar irreversibility was found for the oxidation of [2.2]paracyclophane 2 [12], however, no products were isolated. Chemical oxidation of the latter compound with various oxidants led to cleavage of the cyclophane moiety [13, 14] confirming the high reactivity of the cation radical. Despite the problems introduced by the chemical irreversibility of the redox process, Sato and coworkers [11, 12, 15] explained peak potential shifts by a combination of geometric strain and the well-known p–p interaction between the two aromatic systems in the cyclophane. Such interactions are particularly important in multilayer compounds (up to four aromatic decks, 3) which were studied by a combination of electrochemical and ESR-spectroscopic experiments [16].

361

362

14 Molecular Electrochemistry of Cyclophanes

Fig. 14.1 Cyclic voltammograms of 4 (left) and styrene (right) in CH3CN (potentials vs. fc, from

[17])

If the cyclophane system contains vinyl substituents, as in 4, the irreversible electrochemical oxidation occurs at potentials almost 1 V less positive than for styrene (Fig. 14.1) with a less extended p-system and even &100 mV less positive as compared to the unsubstituted parent compound 2. It results in oligo- and polymerization [17] via the radical cation along the side chain. Upon repeated cycling of the potential, the electrode surface becomes covered by an insulating polymer film, and the current decreases from cycle to cycle (Fig. 14.1, left). Various oligomers could be identified by liquid chromatography and electrospray MS. As expected, the reduction of 2 occurs at rather negative potentials (–3.0 V vs. a saturated calomel electrode [18]), leading to a radical anion, in accordance with chemical reduction by potassium metal [19]. Extension of the p-system by either substitution with styryl groups [20] or annelation with benzene rings [21] renders the molecules easier to reduce. In both cases, further reduction and uptake of up to four or six electrons, respectively, was observed.

Large paracyclophanes [22] with unsaturated bridges such as 5 or 6 show particularly interesting behavior: if their perimeter contains 4n p-electrons (e.g. 5) they can be reduced to stable dianions in a single, reversible two-electron step [22–25]. However, if 4n + 2 p-electrons are present in the perimeter of the neutral starting compound (e.g. 6), two separate, irreversible one-electron reduction peaks are found in the cyclic voltammogram [22]. In accordance with Hückel calculations [26], the dianions with 4n + 2 p-electrons are stabilized, at least if they attain a planar conformation. This stabilization makes the second reduction step from the monoanion to the dianion thermodynamically similar to or easier than the first one, a phenomenon now generally called “potential inversion” or “compression” [27].

14.3 Molecular Electrochemistry of Functionalized Cyclophanes

14.3

Molecular Electrochemistry of Functionalized Cyclophanes 14.3.1

Substituted Cyclophanes

The most simple functionalization of a cyclophane is substitution of the hydrocarbon backbone with substituents inducing redox activity (general type B). Again, the electronic interaction between the p-systems of the cyclophane parent of such molecules has been studied. Electron-withdrawing substituents at one of the aromatic rings in [2.2]paracyclophane render the molecule more difficult to oxidize and the oxidation potentials were correlated with Hammett r+p substituent constants [28]. This provides evidence of a nonbonding interaction of the substituted arene ring with the second one, from which the electron is removed during oxidation. The relation between the nature of the substituent and the oxidation potential was less straightforward for [2.2]metacyclophanes [11, 12, 15], but spectroscopic evidence from their charge-transfer complexes [29] suggested similar interactions. As a cautionary note, it should be mentioned here that all these conclusions were drawn from chemically irreversible half-wave or peak potential data. Fast reactions of the electrogenerated intermediates are also obvious in the oxidation and reduction of brominated [2.2]paracyclophanes [30]. Here reduction leads to the usual irreversible two-electronic C–Br bond cleavage [31], separately detected for each Br in most cases. Oxidation occurs in a chemically irreversible process, but simultaneous (i.e. at the same potential) for all halogen atoms. Despite the irreversibility of the reaction, frontier orbital considerations allowed to explain the relative position of the experimental potentials. Again, the p-interaction between the two arene moieties appears to be important.

Other electroactive substituents attached to [2.2]paracyclophane were based on thiophene units. Polythiophenes are an important class of conducting polymers with interesting, e.g. nonlinear optical or light-emitting properties [32]. [2.2]Paracyclophane-capped oligothiophenes 7 show enhanced stability of both the cation radical and dication oxidation products [33]. Delocalization of the positive charges into the cyclophane ring systems seems to be decisive. Similarly, a [2.2]paracyclophane was used to link two oligothienyl units, as in 8 [31, 34]. While 8 (R = H) is irreversibly oxidized to the radical cation with subsequent polymerization [34], 8 (R = CH3) undergoes a stepwise two-electron transfer to a bis-radical cation with a clear splitting of the two steps in differential pulse voltammograms

363

364

14 Molecular Electrochemistry of Cyclophanes

[35]. Thus, the positive charge induced by the first oxidation makes the removal of the second electron more difficult. In molecules with more than one redox-active moiety, this is usually taken as indication of electronic communication between the redox centers. Here, the interaction of the bis-thienyl units is channelled through the cyclophane p-system. The resulting species was regarded as a model for “stacked polarons”. No splitting of potentials was observed for the further oxidation to a bis-dication. 14.3.2

Cyclophanes with Non-metallic Heteroatoms as Ring Members

Enhanced electroactivity may also be introduced into cyclophanes by substituting ring carbon atoms by atoms bearing free electron pairs or having different electronegativity from carbon (general type C). Many reports about the electrochemistry of such cyclophanes have appeared, but only a selection with particular emphasis on the cyclophane character of the redox-active compounds is discussed here. Substitution by metal atoms will be summarized separately in Section 14.4.

A series of compounds (9–12) was synthesized by Wolff [36–38] based on the hexa-aminobenzene core. In all cases, nitrogen is the substituting atom in the linking chains of the cyclophane, and the N atoms are further substituted by methyl groups. These atoms have lone pairs, and the compounds are expected to be easily oxidized. Moreover, they possess basic character. If a single NH moiety is present (9), the oxidation of the molecules in CV appears chemically irreversible, owing to proton loss from the radical cation and protonation of the basic starting compound [39], in analogy to the oxidation of simple anilines [40].

Fig. 14.2 Cyclic voltammograms of 10 (left) and 12 (right) in CH3CN/CH2Cl2 (potentials vs. fc,

from [38])

14.3 Molecular Electrochemistry of Functionalized Cyclophanes

However, if the hexa-aminobenzene-derived cyclophanes are fully methylated as in 10 and 11 [38], the oxidation behavior (Fig. 14.2, left) becomes similar to that of hexakis(dimethylamino)benzene 13: Upon an initial two-electron oxidation, two one-electron steps occur at increasingly positive potentials. The resulting tetracation is quite persistent in a CH2Cl2/CH3CN electrolyte, and can be oxidized further at even higher potentials to a hexacation (not shown in Fig. 14.2). It is characteristic that the primary two-electron reaction shows broad peaks and a very large difference of peak potentials. This has been attributed to strong potential inversion of several 100 mV for both 10 and 13 [38, 41]. The transfer of the second electron is thermodynamically much easier than that of the first. The reason is the considerable geometric change induced when the planar aromatic ring in the neutral species attains a twisted bis-trimethincyanine structure 14 in the dication. Thus, the formation of an anti-aromatic 4p-system is avoided. As a further result of the structural rearrangement the electron transfers are unusually slow. The effect of the metacyclophane bridge in 10 as compared to 13 on the other hand is small.

This is still the case for flexible cyclophanes in which two hexa-aminobenzene units are linked by two alkyl chains (e.g. 11) [38]. Oxidation up to a dodecacation is observed. However, if a rigid structure with two closely stacked benzene rings is assembled (12), the voltammetric pattern completely changes (Fig. 14.2, right): the initial oxidation wave splits into two separate one-electron signals, clearly showing the electronic interaction between the two aromatic units held in close proximity (5.74 Å) [37]. A large group of redox-active cyclophanes is based on tetrathiafulvalene (TTF) units (tetrathiafulvalenophanes). Continuing interest [42] in such TTF derivatives relies on their electrical properties, while more recently the usefulness in the context of supramolecular chemistry has been emphasized. The parent system 15 is oxidized in two separated, reversible steps through a radical cation to a dication [43].

365

366

14 Molecular Electrochemistry of Cyclophanes

If the TTF unit is bent from planarity, an anodic shift of the potentials [44] and loss of chemical reversibility [44, 45] is observed. A particularly detailed cyclic voltammetric study [46] includes cis- and trans-isomers of [12]- and [14]tetrathiafulvalenophanes. The sterically more demanding, bent trans-isomers 16 are oxidized in an initial one-electron step (Fig. 14.3, left; peaks A1, C1). The resulting radical cations undergo a trans ? cis rearrangement, thus releasing steric strain. The cisradical cation formed is redox active at less positive potentials (&0.065 V difference in the formal potentials; peak A1'). The neutral cis-cyclophane, obtained upon re-reduction, cannot transform back into the starting trans-isomer. Scan rate variation additionally allowed the determination of rate constants for the isomerization processes (Fig. 14.3, right) which follows a variant of the “square scheme” mechanism [47]. Peaks A2 and C2 correspond to the second oxidation of the TTF unit. For tetrathiafulvalenophanes, which incorporate more than one TTF group [48], a structure-dependent behavior is related again to the interaction of the aromatic systems: if the two TTF units in a bis(tetrathiafulvalene) compound are apart far enough, and linked by a flexible chain (e.g. 17 (X = -(CH2)2-O-(CH2)2-)), both units are oxidized independently at the same potentials, resulting in two two-electron waves in CV [49]. On the other hand, if the linker structure distorts one TTF unit, as the right hand one in 17 (X = -(CH2)2-), the CV signals split and up to four peaks are observed [49]. Peak splitting is also apparent if the two TTF units do electronically interact (18) [50, 51]. The interaction is usually stronger in the oxidized form than

Fig. 14.3 Cyclic voltammogram (potentials vs. a standard calomel electrode, from Boubekeur et al. [46]; with permission) of 16 (n = 10) in CH2O2 (v = 1.0 V s–1) and electrode reaction mechanism (T = trans-, C = cis-isomers; electrons transferred are omitted) (redrawn after [46])

14.4 Molecular Electrochemistry of Organometallic Cyclophane Derivatives

in the neutral starting compound [52], in accordance with Hückel MO calculations [53]. The dominant interaction is Coulombic repulsion.

Even more complex molecules, like cage-type TTF-trimers [54], bis(tetrathiafulvalene) cuppedophanes [55] and to a lesser degree a pyrrolo-tetrathiafulvalene cage [56], fit into the same picture. Finally, more recently, “extended” tetrathiafulvalenes were incorporated into cyclophanes, and electrochemically oxidized [57, 58]. Again, the steric constraints appear highly important. In some cases, radical ionic intermediates of the oxidation process were electrocrystallized [59, 60] resulting in materials with interesting, e.g. ferromagnetic [60] behavior.

14.4

Molecular Electrochemistry of Organometallic Cyclophane Derivatives 14.4.1

A Possible Classification of Organometallic Cyclophane Derivatives

A large number of transition metal derivatives of cyclophanes have been investigated by electrochemical methods. In this section, we will first consider metallocenophanes, where the metal atom is part of a sandwich structure which forms itself the aromatic part of the cyclophane, as well as metallametallocenophanes, where such a compound additionally incorporates a metal atom in the aliphatic part of the ring (general type D). Then, cases are discussed, where a cyclophane forms a ligand to a metal atom, in particular in the form of a p-ligand (general type E). Finally, coordination compounds between a metal and a cyclophane will be mentioned. These form the basis of interesting applications, e.g. attempts to construct artificial molecular machines [61]. 14.4.2

Metallocenophanes and Metallametallocenophanes

Ferrocene (fc) itself is redox active and can be oxidized in a one-electron process. This reaction is often used as a standard for potentials in nonaqueous solvents [10] and reversibility, although it has neither an infinite electron transfer rate [62] nor is its oxidation product stable on a long time scale [63]. Nevertheless, in CV most often the reverse peak is observed, even in its derivatives.

367

368

14 Molecular Electrochemistry of Cyclophanes

If the two cyclopentadienyl (Cp) units in ferrocene are linked through one or more atoms, a ferrocenophane [64] is formed. The linking bridge may introduce electronic effects and geometrical changes, for example tilt the Cp rings with respect to each other, modulating the Fe ··· ring distance [65]. Consequently, the electron density at the Fe atom is changed. These effects influence the redox potential of the ferrocene unit. Keto groups in conjugation with the Cp unit increase E 0 for the oxidation [65], as expected.

It is particularly interesting to study the effects of the nature and the numbers of heteroatoms in the bridge of a ferrocenophane. The potential abilities of strained [1]ferrocenophanes 19 range from forming interesting materials [66] to being precursors for ring-opening polymerizations [67] or for redox-active host molecules [68]. The electron donor properties of Ge-[1]ferrocenophanes [66] were investigated by CV, in order to assess their ability to form charge transfer (CT) complexes. They show at least one reversible oxidation wave, however, at surprisingly high anodic potentials, making formation of such complexes difficult. Tin[1]ferrocenophanes [69, 70] reveal again reversible one-electron signals in CV with E 0 close to that of the parent ferrocene, while (dimethylsilylen)-[1]ferrocenophane appears to become chemically irreversible under comparable conditions [70]. Selena- and thia-[1]ferrocenophanes are also known [71] and show similar behavior as regards reversibility. These results indicate that larger atoms linking the two Cp rings introduce less distortion than the smaller ones. Further interesting effects were observed in the case of hetera[3]ferrocenophanes with S and Se as hetero atoms (20, X = S, Se, or CH2). Values of E 0 for various derivatives were compared to those for 1,1'-bis(methylthio)- and 1,1'bis(methylseleno)ferrocene [72]. If at least two chalcogen atoms are present in positions 1 and 3 of the bridge, the potentials moved in a characteristic way to more positive values by several hundred mV. In addition to the conventional mesomeric/inductive effects of the S and Se atoms, a through-space interaction between d-orbitals of the Fe and the chalcogen atoms is assumed. Aza[3]ferrocenophanes such as 21 [73–75] are easily oxidized at potentials close to the ferrocene oxidation itself. However, as a complication, (de)protonation reactions at the N atom play a role. It appeared that the Fe ··· N distance is inversely proportional to E 0 [73]. A second oxidation at higher potential is strongly dependent on the substituents at the N atom [74]. A change of the Fe oxidation state may be used to switch the binding capabilities of crown ethers linked to the fc subunit, e.g. in 22 [73]. Inversely, the effect of protonation at the N site on the fc redox potential is explained by a Coulombic model, and the maximum shift of E 0

14.4 Molecular Electrochemistry of Organometallic Cyclophane Derivatives

has been estimated as 390 mV (for a detailed discussion of the sensing application of such fc derivatives, see ref. [3]).

As proven by various voltammetric features, ansa-pyrazabole-bridged ferrocenes 23 [68] are reversibly oxidized in a one-electron process to the corresponding ferricinium cations without considerable structural changes. Effects of the substituents R at the boron atoms are, however, not purely inductive as indicated by their magnitude. Hyperconjugative p-interaction is assumed. Furthermore, the effect of the electron donating abilities of the linking heterocycles on E 0 is relatively small, but related to the pKb values of the N atoms. In metalla(cyclo)metallocenophanes additional geometric constraints and new reactivities are introduced by a transition metal atom replacing a methylene group in the phane-bridging chain [76]. Lindner synthesized and investigated a variety of osmium complexes of these types (osma- and ferracyclophanes, osmaferrocenophanes, and osmacycloferrocenophanes) with ferrocene units, as well as osmaruthenocenophanes including a ruthenocene unit.

The osmacyclophanes (e.g. 24) [77] do not seem to have been investigated electrochemically. For the heterobimetallic phanes in this series, however, the question of electronic interaction between the metal centers was probed by CV [78]. Model ferrocenes (e.g. 25) with two pendant Re(CO)5 groups linked by alkyl chains of 3–6 methylene groups showed a continuous increase of E 0 for the reversible, ferrocene based one-electron oxidation, levelling off with the chain length at a limiting value also found for bis(x-hydroxyalkyl)ferrocenes. This indicates a decreasing effect of both the Re(CO)5 or the OH groups. Such a limiting value is not obvious in the analogous series of osmaferrocenophanes (e.g. 26). Ring tilting effects on the ferrocene E 0 are definitely excluded on the basis of X-ray results. In the osmacycloferrocenophanes (e.g. 27) the Os core is so far removed from the ferrocene unit that hardly any effect is seen on E 0 [76]. However, considerable

369

370

14 Molecular Electrochemistry of Cyclophanes

Fig. 14.4 Cyclic voltammogram (left) and differential pulse voltammogram (right) of a Ga-linked tris(ferrocenophane) (potentials vs. fc, from Jutzi et al. [80] with permission)

shifts of E 0 with the distance between fc and the phenylene rings as well as the substitution pattern of the latter are noted. In ferrocenophanes incorporating several fc units, the interaction of the redoxactive moieties is very sensitive to the nature and length of the linking bridges. Bis(ferrocenes) in which the fc moieties are far removed do only show a single reversible oxidation [79]. On the other hand, ferrocenophanes with a single Ga atom linker [80, 81] are oxidized much easier than ferrocene itself with two separate signals for dimeric and three separate signals for trimeric species (Fig. 14.4) and the cationic intermediates are tentatively classified as Robin/Day class II-mixed valence compounds. 14.4.3

Cyclophanes as p-Complex Ligands

A particularly interesting class of cyclophane complexes is based on work by Boekelheide [82, 83] in an attempt to realize a polymeric transition metal atom stack held together by [2n]cyclophanes. Such stacks could possibly provide a new type of electrically conducting materials if the metal atoms attained different oxidation states (mixed valence) [84]. Due to electronic and stability reasons [85], research mainly concentrated on ruthenium as central atom. However, also Fe, Cr, as well as Os, and most recently, with different cyclophanes, Co, have attracted attention. The most simple examples of the Ru-cyclophane complexes are (g6-hexamethylbenzene)(g6-[2n]cyclophane)ruthenium(II) dications [86] e.g. 28. Their electrochemical reduction can best be explained by comparison to bis(g6-hexamethylbenzene)ruthenium(II) 29, which is easily reduced in a two-electron process to the neutral Ru(O) complex [87]. In order to keep the 18-electron count, one of the g6-ligands changes its hapticity into g4. Its aromatic system loses planarity. This occurs during the second electron transfer [88], whose rate is thus lower than that of the first one [86]. Simultaneously, potential compression (see Section 14.2) is apparent. The aromatic rings in [2.2]paracyclophane can easily attain the desired bent conformation [90] for g4-hapticity, which should be favorable for the reduction of complexes like 28. Indeed, the flexibility of the [2n]cyclophane in a series of complexes

14.4 Molecular Electrochemistry of Organometallic Cyclophane Derivatives 6

[68] governs the ease of reduction [90]. This is particularly true for bis(g [2.2]paracyclophane)ruthenium(II) (30) [90].

It was already observed early [86] that the solvent had a marked effect on the shape of the voltammograms: while in CH3CN several examples of fully overlapping two-electron waves were observed, in CH2Cl2 two separate one-electron signals occur. The CV for the reduction of 30 in propylene carbonate just shows the beginning of this splitting by a distortion of the CV curve shape, and becomes clearly visible in CH2Cl2 (Fig. 14.5) [91]. Computer simulations of the voltammograms indicate an increase of the difference of formal potentials DE 0 from 77 to 186 mV under these conditions [92]. This modulating effect on the E 0 (and the extent of potential compression) is related to the different stabilization of the three oxidation states by these solvents. In the case of 30, chemical reversibility of the redox process appears high in the voltammograms [91], indicating high stability of the Ru(0) form of the complex. Other Ru(0) complexes, however, are less stable [86]. Recently, mono- and di-vinyl- and methacrylate-substitution in the cyclophane ligands was investigated [93, 94]. The reduction of the corresponding Ru complexes proceeds similar as for the unsubstituted parent compounds, if one accounts for electronic and steric effects. However, in addition chemically irreversible oxidation leads to modification of the electrode surface with electroactive films [94]. This might provide an alternative access to the originally proposed polymers.

Fig. 14.5 Cyclic voltammograms (symbols; E vs. fc, v = 0.1 V s–1) of 30 in propylene carbonate

(left) and CH2Cl2 (right) compared with computer simulations (full lines)

371

372

14 Molecular Electrochemistry of Cyclophanes

Fig. 14.6 Cyclic voltammogram of tri-ruthenium complex 32 in propylene carbonate, potentials

vs. fc, v = 0.1 V s–1

In a second step towards polymetallocenes, Boekelheide studied di-ruthenium complexes in which the metal atoms were bound to the two interacting decks of a [2n]cyclophane, e.g. 31 [85, 95, 96], or a triple-layered analog [97]. It appears that the electrochemical reduction behavior strongly depends on the bridging central cyclophane. If this is [24](1,2,4,5)cyclophane, cyclic voltammetry shows two fully reversible two-electron signals, indicating reduction of the 4+ through a 2+ to a neutral complex, and the reverse oxidations [95]. The intermediate dicationic species was characterized as a net two-electron mixed-valence class II ion [95]. However, if the central bridging ligand is [2.2]paracyclophane, irreversibility on the dicationic stage leads to a follow-up product of the reduction, for which an unusually long C–C bond between the two decks of the cyclophane was apparent from structural analysis [96]. Other [2n]cyclophane analogs gave even more complex reduction behavior. Finally, tri-ruthenium complex 32 shows a complex voltammogram in propylene carbonate (Fig. 14.6), albeit isolation of reduction products proved impossible [90, 98]. The CV curve (supported by coulometric results) is consistent with three twoelectron reductions at increasingly negative potentials (peaks 1, 3, and 4), linked to first-order chemical steps. The product of these reactions is also reduced (peak 2; ECE-type mechanism). The less intense signals I and II at lower potentials are probably due to adsorption. The reductions in peaks 1 and 2 are clearly chemically reversible, in contrast to those in peaks 3 and 4. Multicycle experiments [98] indicate the onset of electrocrystallization on the electrode. Similar Fe complexes (however, with cyclopentadienyl ligands) were also studied electrochemically [99]. Here, in contrast to the Ru systems, a large potential separation between insertion of each single electron is characteristic. However, the bis(cyclopentadienyl) iron analog of 31 shows again a single two-electron reduction and intramolecular bond formation [100] similar to the Ru complexes. Further extension of this work includes [3n]cyclophane complexes [101]. A Ru(II) complex of hexabridged [26](1,2,3,4,5,6)cyclophane was already investigated voltammetrically in Boekelheide’s group [86] and found to comply with the considerations above. In recent years, other metal complexes of “superphanes” [102] (see also chapter 4) have been synthesized, among these maybe the most exotic being “superferrocenophane” [45](1,2,3,4,5)ferrocenophane [103, 104], which apparently has, however, not yet been characterized with respect to its electrochemical properties.

14.4 Molecular Electrochemistry of Organometallic Cyclophane Derivatives

Gleiter and coworkers have prepared and characterized electrochemically a series of CpCo-stabilized cyclobutadieno-superphanes of type 33 (see also chapter 4). While 33 (n = 5) shows a broad voltammetric signal, located at similar potentials as that of model compound 34 (see Fig. 14.7), this indicates two very closely spaced one-electron transfers. Consecutive shortening of the bridging alkane chains causes increasing splitting of the mono-electronic steps [105]. The potential of the first wave remains almost constant for the larger chains. However, in the compound with the shortest chain (33 (n = 1)), the initial oxidation shifts to potentials about 145 mV less positive than in 34.

These results indicate a considerable, distance-dependent interaction between the two Co atoms in the one-electron oxidation product of all examples. In 33 (n = 1) a split of the HOMOs in the neutral state is apparent due to the close proximity of the two halves of the molecule. The question of mixed valence behavior of such bi-nuclear complexes was answered in detail by optical and IR spectro-

Fig. 14.7 Redox potentials E 0 (vs. fc/fc+; dotted: irreversible process) for Co complexes 33 and 34 (courtesy R. Gleiter)

373

374

14 Molecular Electrochemistry of Cyclophanes

electrochemistry [106], revealing that 33 (n = 1) is at the Robin/Day class II/III interface (“almost delocalized” [107]). The importance of the experimental time scale of the experiments for this characterization as well as the difference between spin (localized) and charge (partially shared) distribution was emphasized. As for 33 (n = 1), the second electron transfer is chemically irreversible for various substituted derivatives [108]. However, the first step remains reversible. The potentials shift to less positive values with increasing methyl substitution of the Cp ligand due to the inductive effect of CH3-. If the second Cp ligand is then substituted by electron withdrawing CO2Me, the oxidation potential increases. This gives further evidence for charge delocalization between the two molecule fragments. CV data for a dimer of 35 linked across the Cp ligands reveal a strong interaction between the two central Co atoms [109]. Introduction of Si(CH3)3 groups into the bridge of these superphanes allows further fine-tuning of the distance between the cyclobutadiene units, and CV confirms the trends observed before for the carbocyclic systems [110]. It should be noted that similar splittings of formal potentials could also be observed in Co and Ni complexes of cyclophanes composed of cyclopentadienone units [111]. 14.4.4

Cyclophanes as Supramolecular Components

A wide field of research involving molecular electrochemistry of cyclophanes has been opened by Hünig’s [112] synthesis of bridged viologens, later used extensively by the groups of Balzani, Stoddart and others [113] [as an example, see cyclobis(paraquat-p-phenylene), 36]. These cyclophanes form one of the components in various catenanes and (pseudo) rotaxanes, which, in particular complexed to metal ions, are the basis of attempts to generate artificial molecular machines [4], or models for photosynthetic reaction centers. Research in this area is closely linked to electrochemical experiments. Owing to the extensive review literature on this subject (see e.g. [3, 4, 61, 114]), only the principle and some selected recent results will be discussed here.

Consider a rotaxane as symbolized by 37. Its axis contains two “docking stations” (rectangular symbols) for the cyclophane (symbolized by the elliptical unit; realized e.g. by 36) and two stoppers at the end (conical symbols; bulky substituents) which keep the cyclophane from sliding off the axis. A recent (metal-free)

14.4 Molecular Electrochemistry of Organometallic Cyclophane Derivatives

example [115] features a monopyrrolo-TTF and a 1,5-dioxynaphthalene unit as stations. The interpretation of cyclic and differential pulse voltammetric results includes an initial distribution of 36 between the two stations and, even if 36 encircles the dioxynaphthaline station, a folding of the supramolecular structure still to allow an “alongside” interaction of 36 with the TTF moiety. The oxidation signals of the TTF units (see also Section 14.3.2) in both isomers partially overlap and some results indicate the shuttling of 36 between the stations upon this oxidation. Sauvage et al. observed the rotation of cyclophanes as wheels around the axle of a [3]rotaxane induced by electrochemical redox transformations [116]. In this example, the different coordination preference of Cu(I) (tetra-coordination preferred) as compared to Cu(II) (stabilized by penta- or hexacoordination) is the reason for molecular movement upon a change of the metal oxidation state. As a perspective, such redox switching experiments not only open the way to molecular machines with input of electrical energy but also to molecular information processing [61]. Furthermore, electrochemical redox processes are able to trigger the decomplexation of a guest molecule from a redox-active host [117]: based on CV studies (shift of potentials; variation of peak heights, which are related to diffusion coefficients), it is concluded that the supramolecular complex 38 can be destroyed either by oxidation (of the fc guest; disruption of donor–acceptor interactions) or by reduction (one-electron process of the two bipyridinium units in the host). A second reduction of the bipyridinium moieties occurs after cleavage of the guest. All (de)complexation steps are chemically reversible and fast on the time scale of the experiments. In any of these cases, the electron transfer induces considerable structural rearrangements or molecular movements. Similar systems have been investigated as artificial photosynthetic reaction centers [118, 119]. For example, a 2,2'-bipyridine moiety in the cyclophane part in 39 acts as a bidental ligand for Ru2+. The resulting complex is incorporated into a catenane structure, and the crown ether introduces redox asymmetry. Consequently, the two viologen units are reduced at different potentials (difference &130 mV). The compound also shows light-induced transfer of electrons, and is thus regarded as a model for the two branches (“special pair”) in bacteriochlorophyll complexes.

375

376

14 Molecular Electrochemistry of Cyclophanes

14.5

Conclusions

Molecular electrochemistry proves to be widely applied in the chemistry of cyclophanes and provides a wealth of information. Many groups use electrochemical techniques to study the redox behavior of cyclophanes. In particular, cyclic voltammetry appears useful because E 0 is relatively easy to access. Moreover, but with more effort, mechanistic details can be revealed. Electrochemical data on cyclophanes have been crucial in a variety of contexts: They allow the general electron transfer chemistry of these compounds to be studied, yielding information about the redox behavior. This is not only important from a basic point of view, but also allows cyclophanes to be used as models for various types of electron transfer. The phenomenon of potential inversion or compression is apparent from several examples (see Sections 14.2, 14.3.2 and 14.4.3), indicating considerable structural rearrangements upon the redox process. The occurrence of mixed-valence oxidation states can be investigated, and unusual intramolecular two-electron processes (see Section 14.4.3) or translational and asymmetric redox processes (Section 14.4.4) can be detected. Another important aspect is the study of the interaction between various redox centers in a cyclophane. Cyclophanes provide the unique possibility to clearly define (and vary in a controlled way) the distance between redox centers. Various reports have been discussed in this chapter, where a relation between structure (i.e. in particular, distance) and redox activity is analyzed (see e.g. Sections 14.3.2 and 14.4.3). Finally, the redox properties are of prime importance with respect to characterization and improvement of materials properties of cyclophanes (e.g. Sections 14.3.1 and 14.3.2), and to pertinent applications, as in molecular switches, sensing, or artificial molecular machines (Section 14.4.4).

14.6

References J.-M. Savéant, Pure Appl. Chem. 1997, 69, 269–271. 2 B. Speiser, in A. J. Bard, M. Stratmann, H. Schäfer (eds.), Encyclopedia of Electrochemistry, Vol. 8 Organic Electrochemistry, Ch. 1, in press. 3 P. L. Boulas, M. Gómez-Kaifer, L. Echegoyen, Angew. Chem. 1998, 110, 226–258; Angew. Chem. Int. Ed. Engl. 1998, 37, 216– 247. 4 N. Armaroli, J.-C. Chambron, J.-P. Collin, C. Dietrich-Buchecker, L. Flamigni, J.-M. Kern, J.-P. Sauvage, in V. Balzani (ed.), Electron Transfer in Chemistry, Vol. 3, pp. 582–654, Wiley-VCH, Weinheim, 2001. 1

B. Speiser, Chem. in uns. Zeit 1981, 15, 62–67. 6 J. Heinze, Angew. Chem. 1984, 96, 823– 840; Angew. Chem. Int. Ed. Engl. 1984, 23, 831–847. 7 B. Speiser, in A. J. Bard, M. Stratmann, P. Unwin (eds.), Encyclopedia of Electrochemistry, Vol. 3 Instrumentation and Electroanalytical Chemistry, Ch. 2.1, pp. 81– 104, Wiley-VCH, 2003. 8 D. Britz, in A. J. Bard, M. Stratmann, P. Unwin (eds.), Encyclopedia of Electrochemistry, Vol. 3 Instrumentation and Electroanalytical Chemistry, Ch. 1.3, pp. 51– 71, Wiley-VCH, 2003. 5

14.6 References 9

10 11 12 13 14 15

16

17

18

19 20

21

22 23

24 25 26

27 28

29

M. Carano, B. Colonna, L. Echegoyen, F. Le Derf, E. Levillain, M. Salle, Supramol. Chem. 2003, 15, 83–85. G. Gritzner, J. Kuta, Pure Appl. Chem. 1984, 56, 461–466. T. Sato, M. Kamada, J. Chem. Soc. Perkin Trans. 2 1977, 384–388. T. Sato, K. Torizuka, J. Chem. Soc. Perkin Trans. 2 1978, 1199–1204. W. Adam, M. A. Miranda, F. Mojarrad, H. Sheikh, Chem. Ber. 1994, 127, 875–879. S. Sankararaman, H. Hopf, I. Dix, P. G. Jones, Eur. J. Org. Chem. 2000, 2711–2716. T. Sato, K. Torizuka, R. Komaki, H. Atobe, J. Chem. Soc. Perkin Trans. 2 1980, 561–568. H. Ohya-Nishiguchi, A. Terahara, N. Hirota, Y. Sakata, S. Misumi, Bull. Chem. Soc. Jpn. 1982, 55, 1782–1789. B. Gollas, I. Hesse, R. Lotz, H. Pasch, B. Speiser, I. Zagos, Liebigs Ann./Recueil 1997, 2255–2264. R. Jund, P. Lemoine, M. Gross, Angew. Chem. Suppl. 1982, 654–659; Angew. Chem. Int. Ed. Engl. 1982, 21, 305–306. F. Gerson, W. B. Martin Jr., J. Am. Chem. Soc. 1969, 91, 1883–1891. E. Shabtai, M. Rabinovitz, B. König, B. Knieriem, A. de Maijere, J. Chem. Soc., Perkin Trans. 2 1996, 2589–2595. O. Reiser, B. König, K. Meerholz, J. Heinze, T. Wellauer, F. Gerson, R. Frim, M. Rabinovitz, A. de Meijere, J. Am. Chem. Soc. 1993, 115, 3511–3518. B. Thulin, J. Chem. Soc. Perkin Trans. 1 1981, 664–667. K. Ankner, B. Lamm, B. Thulin, O. Wennerström, J. Chem. Soc. Perkin Trans. 2 1980, 1301–1305. U. Norinder, D. Tanner, O. Wennerström, Croat. Chem. Acta 1983, 56, 269–280. I. Raston, O. Wennerström, Acta Chem. Scand. B 1982, 36, 655–660. U. Norinder, D. Tanner, B. Thulin, O. Wennerström, Acta Chem. Scand. B 1981, 35, 403–406. D. H. Evans, K. Hu, J. Chem. Soc. Faraday Trans. 1996, 92, 3983–3990. T. Shono, A. Ikeda, J. Hayashi, S. Hakozaki, J. Am. Chem. Soc. 1975, 97, 4261–4264. T. Yamato, J.-I. Matsumoto, N. Shinoda, S. Ide, M. Shigekuni, M. Tashiro, J. Chem. Res. (S) 1994, 178–179.

30

*

31 32

33

34 35

36

37

38

39 40 41 42

43

44

45

46

K. P. Butin, A. A. Moiseeva, T. G. Magdesieva, E. V. Sergeeva, V. I. Rozenberg, V. G. Kharitonov, Russ. Chem. Bull. 1994, 43, 783–789; Izvest. Akad. Nauk Ser. Khim. 1994, 839–845. J.-M. Savéant, Adv. Electron Transfer Chem. 1994, 4, 53–116. T. A. Skotheim (ed.), Handbook of Conducting Polymers, Marcel Dekker, New York, 1986. L. Guyard, C. Dumas, F. Miomandre, R. Pansu, R. Renault-Méallet, P. Audebert, New J. Chem. 2003, 27, 1000–1006. L. Guyard, P. Audebert, Electrochem. Commun. 2001, 3, 164–167. F. Salhi, B. Lee, C. Metz, L. A. Bottomley, D. M. Collard, Org. Lett. 2002, 4, 3195–3198. J. J. Wolff, S. F. Nelsen, D. R. Powell, B. Nuber, Angew. Chem. 1992, 104, 899– 901; Angew. Chem. Int. Ed. Engl. 1992, 31, 882–884. J. J. Wolff, A. Zietsch, H. Irngartinger, T. Oeser, Angew. Chem. 1997, 109, 637–639; Angew. Chem. Int. Ed. Engl. 1997, 36, 621–623. J. J. Wolff, A. Zietsch, B. Nuber, F. Gredel, B. Speiser, M. Würde, J. Org. Chem. 2001, 66, 2769–2777. R. Mayer, B. Speiser, unpublished results. B. Speiser, A. Rieker, S. Pons, J. Electroanal. Chem. 1983, 147, 205–222. B. Speiser, M. Würde, C. MaichleMössmer, Chem. Eur. J. 1998, 4, 222–233. J. L. Segura, N. Martín, Angew. Chem. 2001, 113, 1416–1455; Angew. Chem. Int. Ed. Engl. 2001, 40, 1372–1409. H. Viertler, J. Gruber, V. L. Pardini, in H. Lund, O. Hammerich (eds.), Organic Electrochemistry, pp. 621–668, Marcel Dekker, New York, 4th edn, 2001. T. K. Hansen, T. Jørgensen, F. Jensen, P. H. Thygesen, K. Christiansen, M. B. Hursthouse, M. E. Harman, M. A. Malik, B. Girmay, A. E. Underhill, M. Bergtrup, J. D. Kilburn, K. Belmore, P. Roepstorff, J. Becher, J. Org. Chem. 1993, 58, 1359–1366. C. Wang, M. R. Bryce, A. S. Batsanov, J. A. K. Howard, Chem. Eur. J. 1997, 3, 1679–1690. K. Boubekeur, C. Lenoir, P. Batail, R. Carlier, A. Tallec, M.-P. Le Paillard, D. Lorcy, A. Robert, Angew. Chem. 1994,

377

378

14 Molecular Electrochemistry of Cyclophanes

47 48

49 50

51 52

53 54

55 56

57

58

59

60

61 62

63

64 65 66

106, 1448–1451; Angew. Chem. Int. Ed. Engl. 1994, 33, 1379–1381. D. H. Evans, Chem. Rev. 1990, 90, 739–751. J. Ippen, C. Tao-pen, B. Starker, D. Schweitzer, H. A. Staab, Angew. Chem. 1980, 92, 51–52; Angew. Chem. Int. Ed. Engl. 1980, 19, 67–69. D. E. John, A. S. Batsanov, M. R. Bryce, J. A. K. Howard, Synthesis 2000, 6, 824–830. J. Tanabe, G. Ono, A. Izuoka, T. Sugawara, T. Kudo, T. Saito, M. Okamoto, Y. Kawada, Mol. Cryst. Liq. Cryst. 1997, 296, 61–76. H. Fujiwara, E. Arai, H. Kobayashi, J. Mater. Chem. 1998, 8, 829–831. M. Adam, V. Enkelmann, H.-J. Räder, J. Röhrich, K. Müllen, Angew. Chem. 1992, 104, 331–333; Angew. Chem. Int. Ed. Engl. 1992, 31, 309–310. T. Otsubo, Y. Aso, K. Takimiya, Adv. Mater. 1996, 8, 203–211. P. Blanchard, N. Svenstrup, J. RaultBerthelot, A. Riou, J. Becher, Eur. J. Org. Chem. 1998, 1743–1757. J. O. Jeppesen, K. Takimiya, J. Becher, Org. Lett. 2000, 2, 2471–2473. K. A. Nielsen, J. O. Jeppesen, E. Levillain, N. Thorup, J. Becher, Org. Lett. 2002, 4, 4189–4192. D. Lorcy, D. Guérin, K. Boubekeur, R. Carlier, P. Hascoat, A. Tallec, J. Chem. Soc., Perkin Trans. 1 2000, 2719–2723. N. Godbert, A. S. Batsanov, M. R. Bryce, J. A. K. Howard, J. Org. Chem. 2001, 66, 713–719. A. S. Batsanov, D. E. John, M. R. Bryce, J. A. K. Howard, Adv. Mater. 1998, 10, 1360–1363. A. Izuoka, J. Tanabe, T. Sugawara, T. Kudo, T. Saito, Y. Kawada, Chem. Lett. 2002, 910–911. V. Balzani, A. Credi, Chem. Rec. 2001, 1, 422–435. W. R. Fawcett, M. Opallo, Angew. Chem. 1994, 106, 2239–2252; Angew. Chem. Int. Ed. Engl. 1994, 33, 2131–2143. J. Daschbach, D. Blackwood, J. W. Pons, S. Pons, J. Electroanal. Chem. 1987, 237, 269–273. A. Lüttringhaus, W. Kullick, Angew. Chem. 1958, 70, 438. Š. Toma, E. Solcˇániová, A. G. Nagy, J. Organomet. Chem. 1985, 288, 331–339. S. Zürcher, V. Gramlich, A. Togni, Inorg. Chim. Acta 1999, 291, 355–364.

67

68

69

70

71

72

73 74

75

76

77

78

79

80

81

82

83

S. C. Bourke, F. Jäkle, E. Vejzovic, K.-C. Lam, A. L. Rheingold, A. J. Lough, I. Manners, Chem. Eur. J. 2003, 9, 3042– 3054. E. Herdtweck, F. Jäkle, G. Opromolla, M. Spiegler, M. Wagner, P. Zanello, Organometallics 1996, 15, 5524–5535. F. Jäkle, R. Rulkens, G. Zech, D. A. Foucher, A. J. Lough, I. Manners, Chem. Eur. J. 1998, 4, 2117–2128. H. K. Sharma, F. Cervantes-Lee, J. S. Mahmoud, K. H. Pannell, Organometallics 1999, 18, 399–403. R. Rulkens, D. P. Gates, D. Balaishis, J. K. Pudelski, D. F. McIntosh, A. J. Lough, I. Manners, J. Am. Chem. Soc. 1997, 119, 10976–10986. H. Ushijima, T. Akiyama, M. Kajitani, K. Shimizu, M. Aoyama, S. Masuda, Y. Harada, A. Sugimori, Bull. Chem. Soc. Jpn. 1990, 63, 1015–1019. H. Plenio, J. Yang, R. Diodone, J. Heinze, Inorg. Chem. 1994, 33, 4098–4104. I. Yamaguchi, T. Sakano, H. Ishii, K. Osakada. T. Yamamoto, J. Organometallic Chem. 1999, 584, 213–216. T. Sakano, H. Ishii, I. Yamaguchi, K. Osakada, T. Yamamoto, Inorg. Chim. Acta 1999, 296, 176–182. E. Lindner, C. S. Ayasse, K. Eichele, M. Steinmann, Organometallics 2002, 21, 4217–4225. E. Lindner, M. W. Pitsch, R. Fawzi, M. Steinmann, Chem. Ber. 1996, 129, 639– 645. E. Lindner, I. Krebs, R. Fawzi, M. Steinmann, B. Speiser, Organometallics 1999, 18, 480–489. R. Altmann, O. Gausset, D. Horn, K. Jurkschat, M. Schürmann, M. Fontani, P. Zanello, Organometallics 2000, 19, 430–443. P. Jutzi, N. Lenze, B. Neumann, H.-G. Stammler, Angew. Chem. 2001, 113, 1470–1473; Angew. Chem. Int. Ed. Engl. 2001, 40, 1424–1427. A. Althoff, P. Jutzi, N. Lenze, B. Neumann, A. Stammler, H.-G. Stammler, Organometallics 2003, 22, 2766–2774. E. D. Laganis, R. G. Finke, V. Boekelheide, Tetrahedron Lett. 1980, 21, 4405– 4408. E. D. Laganis, R. H. Voegeli, R. T. Swann, R. G. Finke, H. Hopf, V. Boekelheide, Organometallics 1982, 1, 1415–1420.

14.6 References 84 J. K. Burdett, E. Canadell, Organometal85 86

87 88

89 90

91

92 93 94

95

96

97

98 99 100

101

102

lics 1985, 4, 805–815. V. Boekelheide, Pure Appl. Chem. 1986, 58, 1–6. R. G. Finke, R. H. Voegeli, E. D. Laganis, V. Boekelheide, Organometallics 1983, 2, 347–350. D. T. Pierce, W. E. Geiger, J. Am. Chem. Soc. 1992, 114, 6063–6073. J. Merkert, R. M. Nielson, M. J. Weaver, W. E. Geiger, J. Am. Chem. Soc. 1989, 111, 7084–7087. W. J. Bowyer, W. E. Geiger, J. Am. Chem. Soc. 1985, 107, 5657–5663. R. T. Swann, A. W. Hanson, V. Boekelheide, J. Am. Chem. Soc. 1986, 108, 3324–3334. B. Speiser, S. Dümmling, in J. Russow, G. Sandstede, R. Staab (eds.), Elektrochemische Reaktionstechnik und Synthese – Von den Grundlagen bis zur industriellen Anwendung, 1999, 33–39. B. Gollas, S. Dümmling, B. Speiser, unpublished results. B.Gollas,B.Speiser,J.Sieglen,J.Strähle, Organometallics 1996, 15, 260–271. B. Gollas, B. Speiser, I. Zagos, C. Maichle-Mössmer, J. Organomet. Chem. 2000, 602, 75–90. R. H. Voegeli, H. C. Kang, R. G. Finke, V. Boekelheide, J. Am. Chem. Soc. 1986, 108, 7010–7016. K.-D. Plitzko, B. Rapko, B. Gollas, G. Wehrle, T. Weakley, D. T. Pierce, W. E. Geiger Jr, R. C. Haddon, V. Boekelheide, J. Am. Chem. Soc. 1990, 112, 6545–6556. H. C. Kang, K.-D. Plitzko, V. Boekelheide, H. Higuchi, S. Misumi, Organomet. Chem. 1987, 321, 79–89. B. Krauss, B. Speiser, unpublished results. W. J. Bowyer, W. E. Geiger, V. Boekelheide, Organometallics 1984, 3, 1079–1086. H. Rabaâ, M. Lacoste, M.-H. DelvilleDesbois, J. Ruiz, B. Gloaguen, N. Ardoin, D. Astruc, A. Le Beuze, J.-Y. Saillard, J. Linares, F. Varret, J.-M. Dance, E. Marquestaut, Organometallics 1995, 14, 5078–5092. T. Satou, K. Takehara, M. Hirakida, Y. Sakamoto, H. Takemura, H. Miura, M. Tomonou, T. Shinmyozu, J. Organomet. Chem. 1999, 577, 58–68. R. Gleiter, D. Kratz, Acc. Chem. Res. 1993, 26, 311–318.

103 M. Hisatome, J. Watanabe, K. Yamaka-

104

105

106

107 108

109

110

111 112 113

114 115

116

117

118

119

wa, Y. Iitaka, J. Am. Chem. Soc. 1986, 108, 1333–1334. M. Hisatome, J. Watanabe, Y. Kawajiri, K. Yamakawa, Y. Iitaka, Organometallics 1990, 9, 497–503. R. Gleiter, H. Röckel, G. Pflästerer, B. Treptow, D. Kratz, Tetrahedron Lett. 1993, 34, 8075–8078. M. E.Stoll,S. R.Lovelace,W. E.Geiger,H. Schimanke, I. Hyla-Kryspin, R. Gleiter, J. Am. Chem. Soc. 1999, 121, 9343–9351. S. F. Nelsen, Chem. Eur. J. 2000, 6, 581– 588. R. Gleiter, R. Roers, J. Classen, A. Jacobi, G. Huttner, T. Oeser, Organometallics 2000, 19, 147–151. R. Roers, J. Hofmann, J. Classen, R. Gleiter, Tetrahedron Lett. 2001, 42, 5871– 5873. R. J. Schaller, G. Haberhauer, R. Gleiter, F. Rominger, Eur. J. Inorg. Chem. 2002, 2296–2304. P. Jutzi, U. Siemeling, A. Müller, H. Bögge, Organometallics 1989, 8, 1744–1750. W. Geuder, S. Hünig, A. Suchy, Tetrahedron 1986, 42, 1665–1677. P. L. Anelli, P. R. Ashton, R. Ballardini, V. Balzani, M. Delgado, M. T. Gandolfi, T. T. Goodnow, A. E. Kaifer, D. Philp, M. Pietraszkiewicz, L. Prodi, M. V. Reddington, A. M. Z. Slawin, N. Spencer, F. J. Stoddart, C. Vicent, D. J. Williams, J. Am. Chem. Soc. 1992, 114, 193–218. A. Vlcˇek Jr., Chemtracts – Inorg. Chem. 1999, 12, 322–329. J. O. Jeppesen, K. A. Nielsen, J. Perkins, S. A. Vignon, A. Di Fabio, R. Ballardini, M. T. Gandolfi, A. Venturi, V. Balzani, J. Becher, J. F. Stoddart, Chem. Eur. J. 2003, 9, 2982–3007. N. Weber, C. Hamann, J.-M. Kern, J.-P. Sauvage, Inorg. Chem. 2003, 42, 6780– 6792. V. Balzani, J. Becher, A. Credi, M. B. Nielsen, F. M. Raymo, J. F. Stoddart, A. M. Talarico, M. Venturi, J. Org. Chem. 2000, 65, 1947–1956. A. C. Benniston, P. R. Mackie, A. Harriman, Angew. Chem. 1998, 110, 376–378; Angew. Chem. Int. Ed. Engl. 1998, 37, 354–356. A. C. Benniston, P. R. Mackie, L. J. Farrugia, S. Parsons, W. Clegg, S. J. Teat, Platinum Metals Rev. 1998, 42, 100–105.

379

311

12

Electronic Circular Dichroism of Cyclophanes [1] Stefan Grimme and Arnold Bahlmann

12.1

Introduction

The optical properties of chiral molecules are of fundamental interest in chemistry and biochemistry. The main perspective of so-called “structure-chiroptic” relationships is their ability to determine the absolute configuration of chiral molecules which is, for example, of particular importance for compounds with pharmacological relevance. One of the most widely used methods in chiral spectroscopy is electronic circular dichroism (CD) where the difference of the absorption coefficients for left and right circular polarized light (De) is measured [2, 3, 4]. In electronic CD spectroscopy, one of the main quantities of interest is the rotatory strength, R, which determines the intensity of a CD band. In non-orientated media (gas phase, fluid solution), the rotatory strength for a transition between two electronic states, W0 and Wi (index 0 refers to the ground state), is given by the imaginary part of the dot-product between the electronic and magnetic dipole transition moment vectors ~ 0i † ^ i i ˆ j~ l0i j
j~ ljW i ihW 0 jmjW m0i j
cos …~ l0i m R0i ˆ Im hW 0 j^

…12:1†

ˆ are the electric and magnetic dipole moment operators (for where l^ and m further details the reader is referred to ref. [5]). Experimentally, one obtains R of a given CD absorption by integrating the area under the corresponding band. It can be shown that in cgs units Zk2 R ˆ 2:297
10

39 k1

De…k† dk k

…12:2†

where k is the radiation wavelength. Equations (1) and (2) are completely analoˆ is gous to UV spectroscopy where R is replaced by the oscillator strength f and m replaced by l^. Because CD signals can be positive or negative, an additional “dimension” is introduced into these spectra, compared with conventional UV/Vis spectroscopy, which makes CD relatively sensitive to details of the geometric and electronic structures of molecules. Modern Cyclophane Chemistry. Edited by Rolf Gleiter, Henning Hopf Copyright © 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 3-527-30713-3

312

12 Electronic Circular Dichroism of Cyclophanes

Although several semiempirical structure-chiroptic rules [6–8] have been suggested successfully in the past, it is clear that exploration of the full potential of the CD method in general requires an accurate theoretical treatment complementing experimental measurements. The first relatively extensive theoretical studies of the CD of larger molecules with quantum chemical methods based on modern density functional theory (DFT) appeared in the mid 1990s [9, 10] (for a recent review see also [11]). Since then, these [12] and other methods [10] have been developed further. The current status of theoretical CD spectroscopy employing modern quantum chemical methods of different degree of sophistication has recently been summarized in ref. [13] where a wide range of different chromophores have been investigated. Studies of the CD of chiral cyclophanes have a long history. First resolutions of the enantiomers of ring-substituted [2.2]paracyclophanes have been performed by Weigang and Nugent in 1969 [14]. Since then, a large number of chiral cyclophanes of different structure have been synthesized and at least, for the most prominent structural types, also experimental CD spectra are available ([n]paracyclophanes [15–18], orthometacyclophanes [19], [2.2]paracyclophanes [20– 22], [2.2]metacyclophanes [23–25]). Because their size is large and their electronic structure complicated, the first quantum chemical investigations on the CD spectra of (meta)cyclophanes appeared in 1993 [23] but were limited to the semiempirical level. Very early theoretical analysis of the lowest-lying CD transitions in [2.2]paracyclophanes can be found in ref. [26]. This work presents a comprehensive overview of the theoretical and experimental CD spectra of cyclophanes. Even if some of the systems considered have been investigated theoretically before, all structures and spectra have been recalculated using the most recent and most accurate quantum chemical methods available. After a brief description of the methods used (Section 12.2), the excited states of model systems (boat-shaped benzene, two interacting benzene moieties) are discussed in Section 12.3. This should provide the reader with the basic information about the electronic structures of cyclophanes which is necessary to understand the chiroptical properties of the more complex “real” molecules considered further. In Sections 12.4.1–3, theoretical results for substituted [n]cyclophanes, [2.2]paracyclophanes and [2.2]metacyclophanes are presented and discussed, together with the corresponding experimental data. Section 12.4.4 deals with cyclophanes containing aromatic fragments other than benzene which provides some insight how more severe symmetry breaking in cyclophanes influences their chiroptical properties.

12.2

Theoretical Methods

The accurate theoretical prediction of CD spectra constitutes a significantly more demanding task than the calculation of conventional UV spectra. Beside the excitation energies (DE), which have to be predicted with a relative accuracy of about 0.2–

12.2 Theoretical Methods

0.3 eV to avoid incorrect cancelation of neighboring bands of opposite sign, the elec~ 0i, and the angle between the tric and magnetic transition dipole moments ~ l0i and m two vectors must also be determined (cf. eq. 12.1). Naturally, CD spectroscopy deals with systems of relatively low symmetry (typically C2 point group at most) and in practice a large number of excited states are required to simulate CD spectra in the entire wavelength range, which is usually measured (200–700 nm). Even if we do not consider the inherent problems with an accurate quantum chemical description of excited states (as compared with ground states), in general it is clear that the theoretical simulation of CD spectra is a formidable task. Two very basic approximations are used throughout this work. First, all calculations refer to the gas phase, so that solvation effects are completely neglected. This is a very good approximation for all cyclophanes investigated, which are almost nonpolar, showing only small dipole moments. The solvation effects on the excitation energies of aromatic compounds in nonpolar solvents are usually of the order 0.1–0.2 eV (red shift compared with the gas-phase) and not very state-dependent. The second approximation concerns the geometries of the excited states. All excited state calculations have been performed at the ground state geometries, so the results correspond to vertical transitions and the excitation energies can be approximately identified as the band maxima in the experimental spectra. This also implies that any vibrational structure seen in the CD spectra and also vibrationally induced CD [4] cannot be predicted (although work along this line is in progress [27]). The quantum chemical method we mostly employ to obtain the excitation energies and transition moments is density functional theory (DFT), either time-dependent (TDDFT [28, 29]) or in combination with a multireference treatment (DFT/ MRCI [30]). The latter approach also includes higher excitations (doubles, triples, . . .) which are of particular importance for some aromatic p–p* states. In selected cases we also performed comparative calculations employing accurate ab initio MRCI and time-dependent coupled-cluster (CC2) [31] methods. In the TDDFT treatments, both the pure BP86 [32] and the hybrid B3LYP [33] functionals were used (DFT/MRCI is currently exclusively based on the BHLYP functional [34]). If not mentioned otherwise, all ground state geometries were completely optimized using second-order Møller-Plesset perturbation theory (MP2) [35]. This relatively costly approach is necessary because the geometries of cyclophanes are relatively sensitive to electron correlation effects [36]. Neither ring–ring (or ring–chain) distances nor some stereochemical details are accurately described by the (uncorrelated) Hartree-Fock (HF) or even DFT levels of theory (see [37] and Section 12.4.2). As one-particle basis sets we generally employ atom-centered Gaussian functions of at least valence triple-zeta quality (TZV [29]). For the geometry optimizations these were augmented by a set of (d,p)-polarization functions (TZVP). All excited state calculations were carried out with a specially designed TZV(2d) basis for carbon (and the TZVP basis for the other atoms). The outermost valence sp-functions were fixed at 2/3 of their value in the original TZV basis and all other valence exponents and contraction coefficients have been re-optimized at the HF level for the ground state of benzene. This basis set accounts for the often larger spatial extent of the excited states compared with the ground state. This basis set is denoted as TZV2P'. Rydberg

313

314

12 Electronic Circular Dichroism of Cyclophanes

states, which appear at about 6.3 eV excitation energy in benzene, are not important here because their CD intensities are negligible compared with those of the valence states appearing in the same energy region. The simulations of the CD spectra were performed by summing up rotatory strengths weighted Gaussian curves with a width at 1/e height of 0.2 eV. This is important not only to account for the (mainly vibrationally induced) width of the experimental bands but also to include the cancelation of close-lying transitions with opposite CD sign. Some spectra also include vertical lines indicating the position and rotatory strengths of the individual transitions.

12.3

Excited States of Model Compounds 12.3.1

Boat-type Deformation in Benzene

The question “How bent can a benzene be?” was always one of the main themes in cyclophane research, culminating in the synthesis and characterization of extremely strained [4]paracyclophane derivatives ([38], for early ab initio theoretical work see [39]). In this section, the effects of moderate boat-type deformation usually found in stable cyclophanes on their excited state properties will be investigated. These will serve as rough guidelines as to what band-shifts and intensity redistributions can be expected for the “real” systems considered later. Fig. 12.2 shows an orbital correlation diagram along the boat-type deformation angle a of a single benzene unit. For a definition of the most important geometrical parameters see Fig. 12.1. Starting from the parent compound with a degenerate HOMO (e1g) and LUMO (e2u) the deformation, which reduces the symmetry to C2v, splits the HOMO to b1 and b2 and the LUMO to a1 and a2 components, respectively. Because of the bonding interaction between the 1,4-positions in the a1 orbital, its energy decreases more strongly than the others. The LUMO+1 (a2) which has a nodal plane through C1–C4 is less influenced.

Fig. 12.1 Definition of important geometrical parameters in cyclophanes

Fig. 12.2 Orbital correlation diagram along the boat-type deformation for benzene at four different deformation angles a (DFT-B3LYP/TZV2P')

12.3 Excited States of Model Compounds 315

316

12 Electronic Circular Dichroism of Cyclophanes

The low-lying excited states of simple benzene derivatives can easily be understood in a four-state, four-orbital model. In benzene itself, the possible single excitations between HOMO and LUMO generate the lowest-lying B2u state (Lb in Platts’ [40] nomenclature). Next follow the B1u (La) state and the degenerate E1u states (Ba/b) which correspond to the only dipole-allowed transition from the ground state. Reducing the symmetry by deformation (or substituents) leads to a redistribution of the weights of the four single excitations in the excited states. The Lb and Bb states remain almost unchanged with almost equal contributions from HOMO-1 ? LUMO and HOMO ? LUMO+1 excitations while La and Ba gain more contribution from the HOMO ? LUMO excitation. This is clearly reflected by the calculated state energies shown in Fig. 12.3. The Lb (B1) state located initially slightly above 5 eV transition energy (exp. 4.9 eV (255 nm), see below) is effected much less than the La (B2) state. As expected, the splitting of the E1u components increases along the deformation co-

Fig. 12.3 State correlation diagram along the boat-type deformation for benzene at three different deformation angles a (TDDFT-B3LYP/TZV2P')

12.3 Excited States of Model Compounds

ordinate. The energy lowering is largest for these two states which can be explained by the strong anti-bonding character (which decreases as the p-orbital overlaps less). In summary, the boat-type deformation of the benzene rings in cyclophanes results in strong red shifts of all low-lying p–p* excited states. Most importantly, the components of E1u which carry most of the UV intensity are shifted above 200 nm where they can easily be detected experimentally. These theoretical predictions will be discussed together with the experimental data in the sections on the CD spectra. Before continuing, a closer look at the performance of various quantum chemical models to calculate excited state properties seems appropriate here. For this purpose the practising chemist has today a wide range of powerful techniques of varying cost and accuracy at his or her disposal. The small size of the benzene model allows a calibration of the “cheap” DFT methods in comparison to more sophisticated coupled-cluster (CC2) [31] or multi-reference CI (MRCI) [41] techniques. Such studies are of particular importance if one has to select a computational method which is also applicable to large chiral cyclophanes without any symmetry elements. In Tab. 12.1, excitation energies for the four lowest-lying benzene p–p* states at four different deformation angles a are presented. Beside the MRCI and CC2 results (which may serve as most accurate reference values), TDDFT data employing two different functionals (B3LYP and BP) and those from the DFT/MRCI method are discussed.

Tab. 12.1 Calculated excitation energies (in eV) for the four lowest-lying p–p* states of benzene at four different deformation angles a

State

a (deg)

TD-DFT B3-LYP

BP

CC2

DFT/MRCI (BH-LYP)

MRCI

B2u/1B1

0 10 20 30

5.40 5.34 5.16 4.83

5.26 5.20 5.03 4.72

5.27 5.21 5.02 4.69

5.05 4.97 4.77 4.44

4.88 4.84 4.57 4.32

B1u/1B2

0 10 20 30

6.07 6.01 5.80 5.39

6.00 5.94 5.74 5.34

6.56 6.49 6.24 5.77

6.25 6.18 5.96 5.55

6.52 6.57 6.29 5.79

E1u/2B1

0 10 20 30

7.01 6.87 6.57 6.20

6.91 6.79 6.49 6.11

7.28 7.14 6.84 6.47

7.11 6.95 6.65 6.28

7.30 7.10 6.75 6.41

E1u/2B2

0 10 20 30

7.01 6.89 6.63 6.38

6.91 6.80 6.57 6.31

7.28 7.16 6.93 6.72

7.11 7.00 6.76 6.53

7.30 7.19 7.18 6.66

317

318

12 Electronic Circular Dichroism of Cyclophanes

In general, we observe a relatively good agreement between the results of the five different methods. This holds especially for the red shifts when going from a = 0 to a = 30 8. The DE values of the E1u components are consistently predicted to decrease by about 0.8 and 0.6 eV, respectively, while La and Lb are red-shifted by about 0.5–0.6 eV [42]. The absolute excitation energies are slightly different with the various methods. The experimental DE of the Lb state for planar benzene (about 5 eV [43]) is matched almost exactly with the two MRCI approaches while CC2 and the two TDDFT methods are too high by 0.3–0.4 eV. The La and E1u states (exp. 6.2 and 7.0 eV) are a bit too high with the two ab initio methods (mainly due to basis set deficiencies). Of the two density functionals tested within the TDDFT method, there is no clear winner. The results from the BP functional are always 0.05–0.1 eV lower compared with those from B3LYP. The situation for the “real” systems (especially those with two aromatic rings) is different because non-local intra-annular excitations play an important role. These are better described by the hybrid functional B3LYP, which will be exclusively considered later. In Tab. 12.2, oscillator strengths and magnetic dipole transition moments for the deformed benzene model systems are shown. Beside the excitation energies discussed above, these values are the second important factor determining the sign and intensity of the CD bands when chiral perturbations occur. As found before, all methods provide more or less the same trends although distinct differ-

Tab. 12.2 Calculated oscillator strengths f and magnetic dipole transition moments m (atomic

units) for the four lowest-lying p–p* states of benzene at four different deformation angles a State

a (deg) TD-DFT

CC2

B3-LYP

DFT/MRCI (BH-LYP)

BP

f

m

f

m

f

m

f

m

B2u/1B1

0 10 20 30

0 0.0001 0.0017 0.0064

0 0.08 0.16 0.22

0 0.0001 0.0011 0.0043

0 0.07 0.13 0.18

0 0.0001 0.0010 0.0035

0 0.01 0.12 0.17

0 0.0001 0.0008 0.0031

0 0.08 0.15 0.20

B1u/1B2

0 10 20 30

0 0.001 0.012 0.034

0 0.16 0.37 0.46

0 0.001 0.010 0.031

0 0.13 0.28 0.40

0 0.002 0.029 0.059

0 0.14 0.28 0.36

0 0.002 0.018 0.043

0 0.15 0.32 0.44

E1u/2B1

0 10 20 30

1.180 0.472 0.384 0.316

0 0.28 0.31 0.25

1.101 0.467 0.374 0.301

0 0.21 0.25 0.21

0.618 0.521 0.417 0.339

0 0.21 0.25 0.21

0.580 0.494 0.399 0.325

0 0.26 0.31 0.26

E1u/2B2

0 10 20 30

1.180 0.475 0.370 0.280

0 0.28 0.27 0.14

1.101 0.472 0.370 0.278

0 0.22 0.22 0.12

0.618 0.523 0.389 0.280

0 0.20 0.18 0.08

0.580 0.478 0.366 0.268

0 0.27 0.26 0.13

12.3 Excited States of Model Compounds

ences between the DFT based and the CC2 methods are observed for the “forbidden” La (1B2) and Lb (1B1) transitions. In the point group D6h (a = 0), only the E1u transition is electrically allowed. Both moments decrease as a increases; the change when going from a = 0 to a = 10 is more emphasized for the TDDFT methods than with CC2 or DFT/MRCI. In contrast to the E1u components, the La and Lb states gain intensity as a increases. Especially the moments for the Lb transition increase dramatically, i.e. by one order of magnitude when going from a = 10 to a = 20 8. At intermediate deformation angles, both La and Lb can be considered as magnetically allowed transitions while the oscillator strengths are (especially for the Lb state) still relatively small. This resembles the situation of the n ? p* transition in carbonyl groups which exhibit a strong magnetic transition moment and often show a very small oscillator strength [4]. Therefore, the low-energy region of cyclophane CD spectra is expected to be relatively sensitive to small changes in geometry and/or substitution pattern as indeed found experimentally [14]. It should also be mentioned that in our achiral model electric and magnetic moments are always orthogonal to each other, lying in the x–y plane defined by the four carbon atoms 2, 3, 5 and 6 (mz components may occur in “real” systems where an admixture of states with A2 symmetry occurs, which are absent in the four-state model). In summary, our simple analysis for one boat-type deformed benzene unit predicts strongly red-shifted La and Lb transitions for [n]cyclophanes which should show weak intensity and probably varying CD sign depending on the actual structure. The components of the E1u state should show up above 200 nm with significant CD intensity due to strong electric transition moments. The small energy separation between these two states and the almost orthogonal alignment of the corresponding transition moments may give rise to exciton type (+/–)-CD-sign patterns. 12.3.2

Interacting Benzene Fragments

The excited states of [2.2]paracyclophanes (and to a lesser extent [2.2]metacyclophanes) are relatively complicated. Even if we neglect inter-ring (charge transfer) excitations, the number of excited states in a relatively small energy window is doubled compared with a simple [n]paracyclophane. Because of the close proximity of the aromatic rings, transannular effects are of particular importance. In Fig. 12.4, the frontier orbitals of two face-to-face oriented benzene rings at a distance of 3 Å (between the center of mass of the two rings) are shown. Linear combination of two e1g (HOMO) and e2u (LUMO) results in two degenerate sets with either inter-ring bonding (e1u HOMO-1, e2g LUMO) or antibonding (e1g HOMO, e2u LUMO+1) character. The missing double excitations between HOMO and LUMO in the Hartree-Fock or DFT approaches are responsible for the exaggerated inter-ring distances obtained by these methods. Compared with a single (planar) benzene unit, the HOMO is raised by about 1.3 eV while the LUMO is lowered by about 0.7 eV. Thus, even without any ring deformation, the excitation energies of [n.n]phanes should be lower than those of [n]paracyclophanes. As well

319

320

12 Electronic Circular Dichroism of Cyclophanes

as this through-space interaction, through-bond effects, caused by the necessary bridges in “real” systems, must be considered [ for more detailed discussion of this point see ref. [44] and Chapters 11 (Rademacher) on UV/Vis spectra and 10 (Muchall) on photoelectron spectra in this book]. In Fig. 12.5, the frontier orbitals of [2.2]paracyclophane (22PC, D2h symmetry, see Section 12.4.2) are displayed. Although these pictures closely resemble the schematic plots in Fig. 12.4, the admixture of bridge AOs to the HOMO-2 (b2u) and LUMO+1 (ag) is evident. Note also the strong bonding character of LUMO and LUMO+1. An orbital correlation diagram for various systems containing two benzene units is shown in Fig. 12.6. It also includes a “sandwich model” with a boat-type deformation of a = 10 8 (second column), twisted 22PC (see Section 12.4.2) and 4fluoro[2.2]paracyclophane (see Section 12.4.3). As expected, the boat-type deformation leads to an almost symmetrical splitting of the degenerate orbitals which is of about the same magnitude as for the one-ring model (see Section 12.3.1).

Fig. 12.4 Plots of the frontier orbitals of two interacting benzene molecules (D6h symmetry) at a distance of 3 Å (DFTB3LYP/TZV2P')

12.3 Excited States of Model Compounds

Fig. 12.5 Plots of the frontier orbitals of [2.2]paracyclophane (D2h symmetry, DFT-B3LYP/

TZV2P')

Fig. 12.6 clearly shows that the ethano bridges have by far the largest effect on the orbital energies, while both twisting of the benzene units (22PC with D2 symmetry, column 4) and including a fluorine substituent introduce little change. Compared with the deformed “sandwich” model, the energy of the lowest b2u MO of 22PC is strongly increased such that it becomes HOMO-2, the order of LUMO and LUMO+1 is changed and the energy of the highest b1u MO decreases by almost 1 eV. Hence, for [2.2]paracyclophane derivatives, a six orbital (9-state) model for the low- to mid-energy region of the CD spectra seems more appropriate.

321

322

12 Electronic Circular Dichroism of Cyclophanes

Fig. 12.6 Orbital correlation diagram for systems with two benzene units (DFT-B3LYP/TZV2P')

Tab. 12.3 shows the composition of the low-lying states of 22PC in D2h symmetry. Up to about 227 nm (2B2u), all states can be described by excitations out of the three occupied frontier orbitals. The five lowest states are dominated by a single one-electron excitation while the higher-lying states are mostly more complicated with several contributions of the same weight. The lowest B2g (1B2 in D2 symmetry) and the second state (B3g, 1B3) are of Lb character.

12.4 Theoretical and Experimental CD Spectra of Cyclophanes Tab. 12.3 Composition of the low-lying states of [2.2]paracyclophane in D2h symmetry, oscillator strengths f, magnetic transition dipole moments m (DFT-B3LYP/TZV2P') and comparison with experimental UV data

State

Excitation

Contributions (%)

f

m (au)

DEcalc (eV)

DEexp (eV)

1 B2g

b3g ? b1g b2g ? ag b2g ? b1g b3g ? ag b2u ? b1g b2g ? b1u b3g ? ag b2g ? b1g b2g ? ag b3g ? b1g b2u ? ag b3g ? b1u b2g ? b1u b2u ? b1g b3g ? au b2u ? b1u b3g ? b1u b2u ? ag b3u ? b1g b3g ? au b3u ? ag b2g ? b1u b2g ? au b3u ? b1g b2g ? ag

81.3 17.5 80.4 18.9 70.3 26.7 77.7 17.8 79.4 17.0 53.3 46.5 54.3 22.0 21.7 93.6 42.1 37.1 12.7 48.8 40.2 10.1 50.5 48.0 96.0

0

0.774

4.10

4.11

0

1.13

4.43

4.34

2·10–3

0

4.58

0

1.01

4.69

0

1.49

4.70

3·10–3

0

4.91

0.0330

0

5.15

5.09

0 0.144

2.65 0

5.40 5.45

5.51

0.0139

0

5.67

5·10–3

0

5.72

0

1.11

5.82

1 B3g 1 B3u 2 B3g 2 B2g 1 B2u 2 B3u

3 B3g 2 B2u

3 B3u

3 B2u 3 B2g

12.4

Theoretical and Experimental CD Spectra of Cyclophanes 12.4.1

[n]Cyclophanes 9,12-Dimethyl-4-oxa[7]paracyclophane The 9,12-dimethyl-4-oxa[7]paracyclophane (7PCDM) molecule was synthesized and separated into its enantiomers in 1997 [45], where a first interpretation of its CD spectrum and further stereochemical details were also reported. This compound has been chosen because the calculated deformation angle a of 14.38 is similar to those of [2.2]paracyclophanes considered later. A comparison of the experimental and theoretical CD spectra is shown in Fig. 12.7. 12.4.1.1

323

324

12 Electronic Circular Dichroism of Cyclophanes

Fig. 12.7 Comparison of experimental and theoretical (TDDFT-B3LYP/TZV2P') CD spec-

tra for (R)-9,12-dimethyl-4-oxa[7]paracyclophane

The experimental spectrum shows four resolved bands A–D which are very well reproduced by the calculation. As discussed in Section 12.3.1, the Lb states (band A) are in general predicted slightly too high in energy with TDDFT which explains the red shift of this band compared with experiment. The calculated CD intensities (which are usually higher than the experimental ones) are smaller here. Although bands C and D look like a typical exciton–CD couplet, it seems obvious that they are composed of several transitions which are not understandable in a simple four-state model. This finding is in agreement with the conclusions from ref. [13] that, although the overall description of CD bands with TDDFT is quite good, often too many states with little physical meaning are involved [46]. However, if one is mainly interested in an assignment of absolute configuration, these problems are of less importance.

[6]Paracyclophane-8-carboxylic Acid The [6]paracyclophane-8-carboxylic acid (C6PC) is the most strained cyclophane ever studied by CD spectroscopy [15, 16]. Compared with 7PCDM, the boat-type deformation angle a is larger by 5 8 resulting in a red shift of about 30 nm of the lowest CD band. Fig. 12.8 compares experimental and theoretical CD spectra where five bands A–E are clearly visible. The carboxy group in this compound introduces difficulties because the calculated CD spectra are rather sensitive to the dihedral angle of the carbonyl group with respect to the plane of the ring (u). This observation is understandable be12.4.1.2

12.4 Theoretical and Experimental CD Spectra of Cyclophanes

Fig. 12.8 Comparison of experimental and theoretical (TDDFT-B3LYP/TZV2P', B3LYP/TZVP

geometries) CD spectra for (R)-[6]paracyclophane-8-carboxylic acid

cause the ring and carbonyl p-systems interact rather strongly and the out-ofplane orientation of the substituent introduces helicity into the system. The calculations have been performed for two conformers (equal populations) with u values of about 0 and 1808, respectively. Except for the sign of the weak band A (Lb state) and the theoretically overestimated absolute intensities, the agreement between theory and experiment is very good. However, considering the flexibility of the COOH group, this seems to be a coincidence, and this system calls for a proper quantum chemical treatment of the nuclear degrees of freedom. 12.4.2

[2.2]Paracyclophanes [2.2]Paracyclophane Before the CD spectra of [2.2]paracyclophanes are discussed, a closer look at a particular stereochemical problem in these compounds is necessary. According to accurate MP2 calculations, the energy minimum for the parent compound 22PC occurs at a twisted structure with D2 symmetry (the twist angle c, measured as the rotation of one ring relative to the other around an axis perpendicular through the rings (see Fig. 12.1), is 9.4 8, the experimental estimate for this value is about 6 8). The D2h structure with perfectly parallel benzene rings corresponds to a transition state for the racemization of two chiral conformers. Because the barrier is very 12.4.2.1

325

326

12 Electronic Circular Dichroism of Cyclophanes

small (0.41 kcal mol–1, MP2/TZVP) corresponding to a permanent interconversion at ambient temperatures, 22PC can be considered as effectively achiral. This is supported by X-ray [47] data, which show that even at 93 K, the compound equilibrates between two structures with de-eclipsed methylene bridges. This structural feature is also present in substituted compounds [48] where it significantly affects the chiroptical properties. Mono-substituted [2.2]paracyclophanes thus exist in two diastereomeric forms as shown graphically in Fig. 12.10 for one enantiomer of 4fluoro[2.2]paracyclophane (F22PC). Because the energy difference between the two forms is very small (0.76 kcal mol–1 for F22PC at the MP2/TZV2P' level), the experimental measurements always correspond to an equilibrium mixture. Theoretically, this can be accounted for by two separate calculations of the CD spectra and subsequent Boltzmann-weighted averaging (for e.g. F22PC, the populations at room temperature are about 80% and 20%, respectively). This procedure is of particular importance because the CD spectra of the two conformers are sometimes nearly mirror images for substituted systems leading to strong cancelation effects of the CD bands. In Fig. 12.9, the calculated CD spectrum of the (P)-form of 22PC is shown and compared with a spectrum of the “sandwich” model which has the same twist but lacks the ethano bridges. Because the twisting of the two benzene moieties induces a significant helicity, the Cotton effects for 22PC are (even in the absence of planar-chirality inducing substituents) relatively large, reaching a De of about 25 l mol–1 cm–1 for the 4B3 band and even the lowest Lb transition shows relative-

Fig. 12.9 Calculated CD spectra of (P)-[2.2]paracyclophane and the twisted (c = 98) sandwich benzene model (a = 10 8, R = 3 Å, TDDFT-B3LYP/TZV2P')

12.4 Theoretical and Experimental CD Spectra of Cyclophanes

ly high intensity. This shows that the common classification of substituted [2.2]paracyclophanes as planar chiral is at least questionable. Furthermore, the sandwich model (see Section 12.3.2) shows much lower CD intensities and mostly different signs (for the same (P)-configuration) which confirms the conclusion in Section 12.3, that the orbitals of the bridges play a very important role for the chiroptical properties of these compounds.

4-Fluoro-[2.2]paracyclophane The CD spectra of mono-substituted [2.2]paracyclophanes were first measured by Nugent and Weigang [14] and Rosini et al. [49], from where the experimental spectrum of 4-fluoro-[2.2]paracyclophane (F22PC) has been taken. This compound has been selected because the substituent introduces no further conformational complications and interacts only moderately with the benzene p-system. Fig. 12.11 shows the calculated CD spectra of the two conformers whose structures are depicted in Fig. 12.10. The calculated CD spectrum for conformer A in Fig. 12.11 shows a strong resemblance to that of the unsubstituted phane (see Fig. 12.9) while that of conformer B is very different in the mid- to low-energy range (220–300 nm). An approximate mirror-image relation is only observed in the region below 220 nm where the transitions with high intensity are observed. One reason for the different behavior of the two conformers is that the steric interactions between the substituent and the bridge for conformer B result in a slight additional shift of the benzene rings relative to each other (see also Section 12.4.2.3). Fig. 12.12 shows a comparison of experimental and theoretical CD spectra. As outlined above, the theoretical data have been obtained by averaging the spectra of the two conformers. Experimentally, six bands A–F can be identified which can be also found in the simulation (although bands B and C are not separated enough). All in all, the agreement between theory and experiment can be considered as very satisfactory, bearing in mind that the averaged spectra are quite different and that there is also some error due to an inaccurate energy difference (populations). Note also that a thermodynamic averaging represents a relatively crude approximation for small barriers where a quantum mechanical treatment of the nuclear 12.4.2.2

Fig. 12.10 Calculated structures (MP2/TZVP) of the two conformers of (R)-4-fluoro[2.2]paracyclophane (A left, B right). The distance d between the fluorine substituent and one of the hydrogens in the ethano bridge is significantly shorter in the energetically higherlying structure A (2.43 vs. 2.78 Å)

327

328

12 Electronic Circular Dichroism of Cyclophanes

Calculated CD spectra (TDDFT-B3LYP/TZV2P') of the two conformers of (R)-4-fluoro-[2.2]paracyclophane

Fig. 12.11

Fig. 12.12 Comparison of experimental [20] and theoretical (TDDFT-B3LYP/TZV2P') CD spectra for (R)-4-fluoro[2.2]paracyclophane

12.4 Theoretical and Experimental CD Spectra of Cyclophanes

degrees of freedom along the racemization coordinate may be necessary (but is currently computationally not feasible).

4-Methyl[2.2]paracyclophane Although 4-methyl[2.2]paracyclophane (M22PC) seems to be similar to F22PC, the CD spectra of the two compounds differ substantially which can be understood only by detailed theoretical treatment. The CD spectrum of M22PC [14] (see Fig. 12.15) shows only three bands which have the same sign (for the same absolute configuration) and similar shape compared with those of F22PC (c.f. Figs. 12.12 and 12.15). Considering the quite different structures of the conformers involved, this finding seems to be accidental. The larger steric demand of the methyl group compared with a fluorine atom forces A into a different conformation. It has parallel-displaced [50] instead of twisted benzene rings (see Fig. 12.13) while conformer B is similar to that of F22PC (the twist angles are 8.5, 8.9 and 9.4 8 for M22PC, F22PC and 22PC, respectively; the corresponding value for conformer A of M22PC is < 18). As can be seen from Fig. 12.14, the CD spectra for the two conformers are relatively similar despite the structural differences mentioned above. Note that although the averaged CD spectrum is already in good agreement with experiment (see Fig. 12.15), the measured data compare even better with those calculated for the energetically higher-lying conformer A (negative band A, structure in band B, position of band C). Considering the limited accuracy of our MP2/TZVP treatment, it seems possible that the sign of the (very small) calculated energy difference of 0.34 kcal mol–1 is wrong. Nevertheless, this example clearly demonstrates the sensitivity of CD spectroscopy and the powerful interplay between theory and experiment to detect small structural changes in these complicated compounds. 12.4.2.3

Fig. 12.13 Optimized structures of the two conformers of (R)-4-methyl[2.2]paracyclophane (MP2/ TZVP)

329

330

12 Electronic Circular Dichroism of Cyclophanes

Calculated CD spectra (TDDFT-B3LYP/TZV2P') of the two conformers of (R)-4-methyl[2.2]paracyclophane

Fig. 12.14

Fig. 12.15 Comparison of experimental [14] and theoretical (TDDFT-B3LYP/TZV2P') CD spectra for (R)-4-methyl[2.2]paracyclophane

12.4 Theoretical and Experimental CD Spectra of Cyclophanes

12.4.3

[2.2]Metacyclophanes 1-Thia[2.2]metacyclophane Because the CD spectra of various metacyclophanes have already been extensively discussed in the literature, only two examples are presented here. The 1-thia [2.2]metacyclophane molecule was the first cyclophane CD spectrum ever studied by modern quantum chemical methods. Compared with [2.2]paracyclophanes, the benzene rings are here aligned almost parallel and the chirality is induced a) by a tilting of the rings against each other due to one longer bridging bond and b) by the sulfur atom in the bridge. The experimental spectrum displayed in Fig. 12.16 shows five CD bands A–E with very large calculated CD intensities up to De = 120 l mol–1 cm–1. As before, the agreement between theory and experiment is very satisfactory although a large, and as yet unexplained, discrepancy remains between the intensities. Although no conformational complications have to be considered here, the sulfur atom introduces energetically low-lying lone-pair–r* (band B) and p–r* (band C) states which have to be accounted for accurately. Especially the latter, which are responsible for the high-intensity band C, seem to be described very well by the TDDFT treatment. As in the [2.2]paracyclophanes, the lowest-lying band A is made out of the two locally excited Lb states. For a more detailed analysis of the importance of the intra-annular interactions in this compound, the reader is referred to ref. [23]. 12.4.3.1

Fig. 12.16 Comparison of experimental [23] and theoretical (TDDFT-B3LYP/TZVP, DFTBP86/TZVP geometry) CD spectra for (M)-1-thia[2.2]metacyclophane

331

332

12 Electronic Circular Dichroism of Cyclophanes

1-Thia-10-aza[2.2]metacyclophane This compound shows an interesting stereochemical detail. Replacing a CH2 group in the thiaphane discussed above, by a NH group results in two conformers A and B (see Fig. 12.17) which differ by an inversion of the hydrogen atom connected to the nitrogen center. Because the barrier has been estimated to be very small (about 5 kcal mol–1 [25]), a rapid interconversion is to be expected which is not accessible by dynamic NMR spectroscopy. Nor are the results from X-ray studies decisive, because the conformer B observed experimentally forms intermolecular hydrogen bonds in the solid which may be absent in the fluid phase. Fig. 12.17 shows a comparison of the experimental and the two simulated CD spectra. In the region below 210 nm, we observe distinct differences in the shape of the two theoretical spectra and also the double band between 220 and 240 nm is not present in the spectrum of B. Obviously, the spectrum for conformer A agrees much better with experiment than that of B indicating that A dominates the population in solution. Although this contradicts the calculated energy different between A and B (0.3 kcal mol–1 in favour of B, DFT-B3LYP/TZVP), we believe that the conclusion derived from the combined experimental and theoretical CD investigation is correct and that the neglected solvent affects the equilibrium significantly [51]. In general, the present results confirm the previous conclusion from ref. [25] although it is noted that the new TDDFT results agree much better with 12.4.3.2

Fig. 12.17 Comparison of experimental [23] and theoretical (TDDFT-B3LYP/TZVP, DFTBP86/TZVP geometry) CD spectra for (M)-thia-10-aza[2.2]metacyclophane

12.4 Theoretical and Experimental CD Spectra of Cyclophanes

experiment (especially for the intense double band) than those from the older semi-empirical treatments. 12.4.4

Cyclophanes with Two Different Aromatic Rings 14,17-Dimethyl[2](1,3)azuleno[2]paracyclophane The study of the intra-annular interactions in the cyclophanes hitherto considered is hampered by the fact that the interacting rings are similar in structure (mostly substituted benzenes) yielding a high density of states in a relatively small energy region. The investigation of a chiral meta-para[2.2]cyclophane (AZPC) with coupled benzene and azulene units seemed to remedy this problem because azulene itself has two low-lying states around 600 and 350 nm, respectively, which are well separated from any of the benzene excited states. The spectra shown in Fig. 12.18 show a very rich structure with as many as eight experimentally resolved CD bands A–H. Except for band D (the locally excited 3A1 state of the azulene moiety) and the very weak signal B, all bands are theoretically predicted with the correct sign and reliable absolute intensities. As outlined in detail in ref. [52], the CD intensities in the high-energy range (< 300 nm) are much larger for AZPC than for a dimethyl[7]paracyclophane model indicating an amplification of the chirality of the benzene unit by the azulene 12.4.4.1

Fig. 12.18 Comparison of experimental [52] and theoretical (TDDFT-B3LYP/TZV2P') CD spectra for (S)-14,17-dimethyl[2](1,3)azuleno[2]paracyclophane. Because of the large energy range, a linear energy scale (in eV) has been used

333

334

12 Electronic Circular Dichroism of Cyclophanes

chromophore. Furthermore, the sharp structure at about 3.2 eV could be assigned to a benzene(p)-azulene(p*) charge-transfer transition indicating strong intra-annular interactions. The AZPC system is furthermore a very prominent example how different the intensities in CD and UV spectra can be (see Fig. 12.3 of ref. [52]).

12.5

Conclusions

Cyclophanes exhibit several fascinating stereochemical aspects which can be analyzed successfully by a combination of theoretical and experimental electronic circular dichroism spectroscopy. Most surprisingly, the CD spectra of [2.2]paracyclophanes are mainly determined by the helical arrangement of the two benzene rings in the twisted conformers and through-bond interactions involving the ethano bridges. It could be shown that thermodynamical averaging of the spectra of two conformers provides almost quantitative agreement between theory and experiment. In this respect, time-dependent density functional theory (TDDFT) has proven as a robust and accurate tool to predict the excited state properties of various cyclophanes. It should be noted, however, that the performance of TDDFT to describe CD spectra is not uniform and that the method must be used with great care for other chiral (especially nonaromatic or mostly saturated) systems. In general, the CD spectra of cyclophanes are quite complicated and even if structural similarities exist between molecules, the shapes of the spectra may differ substantially. This may result in misleading interpretations if no theoretical support is present. One of the reasons for this is the relatively high density of electronic states in the energy region between 200 and 250 nm, giving rise to close-lying (canceling) bands with opposite sign. This inherent feature definitely excludes the use of simple structure-chiroptic relationships to analyze the experimental data. A common feature of most cyclophanes studied is the medium CD intensities which reach absolute De values typically between 20 and 60 l mol–1 cm–1. The main reason for this finding is the inherent magnitude of the electric and magnetic dipole transition moments of the basic benzene chromophore and the moderate chiral perturbations. For more complex cyclophanes including polarizable hetero atoms or built out of different aromatic units, as e.g. studied in thia[2.2] metacyclophanes and an azulenophane, the situation is different because of the distinct properties of the chromophores. In the latter case, “true” inter-ring charge-transfer excitations and a strong amplification of CD intensities by the other chromophore could be detected for the first time. It should finally be mentioned that only the very basic sensitivity of CD spectroscopy on the structural and electronic effects (in contrast to conventional UV spectroscopy) allows a detailed insight into such complicated molecules as the cyclophanes.

12.7 References

12.6

Acknowledgement

We thank Dr. C. Mück-Lichtenfeld and C. Diedrich for helpful discussions and technical assistance. This work was financially supported by the Deutsche Forschungsgemeinschaft in the framework of the SFB 424 “Molekulare Orientierung als Funktionskriterium in chemischen Systemen”.

12.7

References 1 2

3

4

5 6

7 8

9 10

11

12 13 14

Part of the Diploma Thesis of A. Bahlmann, Universität Münster, 2003. A. Rauk, in: Encyclopedia of Computational Chemistry, John Wiley & Sons, New York, 1998. K. Nakanishi, N. Berova, R. W. Woody, eds. Circular Dichroism, VCH, Weinheim, 1994. D. A. Lightner, J. E. Gurst, Organic Conformational Analysis and Stereochemistry from Circular Dichroism Spectroscopy, Wiley-VCH, New York, 2000. A. E. Hansen, T. D. Bouman, Adv. Chem. Phys., 1980, 44, 545. W. Moffitt, R. B. Woodward, A. Moscowitz, W. Klyne, C. J. Djerassi, J. Am. Chem. Soc., 1961, 83, 4013. A. I. Scott, A. D. Wrixon, Tetrahedron, 1970, 26, 3695. A. Moscowitz, E. Charney, U. Weiss, H. J. Ziffer, J. Am. Chem. Soc., 1961, 83, 4661. S. Grimme, Chem. Phys. Lett., 1996, 259, 128. F. Furche, R. Ahlrichs, C. Wachsmann, E. Weber, A. Sobanski, F. Vögtle, S. Grimme, J. Am. Chem. Soc., 2000, 122, 1717. F. Vögtle, S. Grimme, J. Hormes, K.-H. Dötz, N. Krause, in Final Report of the Sonderforschungsbereich “Wechselwirkungen in Molekülen”, S. D. Peyerimhoff, ed., Wiley-VCH, New York, 2002. F. Furche, J. Chem. Phys., 2001, 114, 5982. C. Diedrich, S. Grimme, J. Phys. Chem. A, 2003, 107, 2524. M. J. Nugent, O. E. Weigang, Jr., J. Am. Chem. Soc., 1969, 91, 4556.

15 16

17 18 19

20 21

22

23

24

25

26 27

28

W. Tochtermann, U. Vagt, G. Snatzke, Chem. Ber., 1985, 118, 1996. W. Tochtermann, G. Olsson, A. Mannschreck, G. Stühler, G. Snatzke, Chem. Ber., 1990, 123, 1437. I. Pischel, M. Nieger, A. Archut, F. Vögtle, Tetrahedron, 1996, 30, 10034. S. Grimme, I. Pischel, S. Laufenberg, F. Vögtle, Chirality, 1998, 10, 147. C. Niederalt, S. Grimme, S. D. Peyerimhoff, A. Sobanski, F. Vögtle, M. Lutz, A. L. Spek, M. J. van Eis, W. H. de Wolf, F. Bickelhaupt, Tetrahedron Asymmetry, 1999, 10, 2153. H. Falk, P. Reich-Rohrwig, K. Schlögl, Tetrahedron, 1970, 26, 511. J. Issberner, M. Böhme, S. Grimme, M. Nieger, W. Paulus, F. Vögtle, Tetrahedron Asymmetry, 1996, 8, 2223. U. Wörsdörfer, F. Vögtle, M. Nieger, M. Waletzke, S. Grimme, F. Glorius, A. Pfaltz, Synthesis, 1999, 4, 597. S. Grimme, S. D. Peyerimhoff, S. Bartram, F. Vögtle, A. Breest, J. Hormes, Chem. Phys. Lett., 1993, 213, 32. D. Wortmann-Saleh, S. Grimme, B. Engels, D. Müller, F. Vögtle, J. Chem. Soc., Perkin Trans., 1995, 2, 1185. S. Grimme, I. Pischel, F. Vögtle, M. Nieger, J. Am. Chem. Soc., 1995, 117, 157. O. E. Weigang, M. J. Nugent, J. Am. Chem. Soc., 1969, 91, 4555. M. Dierksen, Berechnung der vibronischen Struktur von Elektronenspektren großer Moleküle, Diploma Thesis, Universität Münster, 2003. M. E. Casida in: Recent Advances in Density Functional Methods, ed. D. P. Chong, World Scientific, Singapore, 1995, 155.

335

336

12 Electronic Circular Dichroism of Cyclophanes 29 30 31 32

33

34 35 36

37 38

39 40 41

42

R. Bauernschmidt, R. Ahlrichs, Chem. Phys. Lett., 1995, 256, 454. S. Grimme, M. Waletzke, J. Chem. Phys., 1999, 111, 5645. O. Christiansen, H. Koch, P. Jørgensen, Chem. Phys. Lett., 1995, 155. J. P. Perdew, Phys. Rev. B, 1988, 38, 3098; A. D. Becke, J. Chem. Phys., 1993, 98, 5648. P. J. Stephens, F. J. Devlin, C. F. Chabalowski, M. J. Frisch, J. Phys. Chem., 1994, 98, 11623. A. D. Becke, J. Chem. Phys., 1995, 98, 1272. C. Møller, M. S. Plesset, Phys. Rev., 1934, 46, 618. See also Chapter 13 (Herges) in this book. Our MP2/TZVP geometries are in general in very good agreement with experimental X-ray data. For example, in [2.2]paracyclophane, the calculated distances (see Fig. 14.1) are (exp. in parentheses): c: 1.56 Å (1.58 Å), f: 2.76 Å (2.78 Å), e: 3.08 Å (3.09 Å). R. A. Pascal, J. Phys. Chem., 2001, 105, 9040. T. Tsuji, M. Okuyama, M. Ohkita, H. Kawai, T. Suzuki, J. Am. Chem. Soc., 2003, 125, 951. S. Grimme, J. Am. Chem. Soc., 1992, 114, 10542. J. R. Platt, J. Chem. Phys., 1949, 17, 484. I. Shavitt, in Modern Theoretical Chemistry Vol. 3: Methods of Electronic Structure Theory, ed. H. F. Schaefer III (Plenum, New York, 1977). For the triplet La state, the experimental red shift when going from 0 to 158 is about 0.4 eV which compares very well with our result of 0.2–0.3 eV. See H. Hopf, M. Haase, J. Hunger, W. Tochtermann, M. Zander, Chem. Phys. Lett., 1986, 127, 145; N. L. Allinger, T. J. Walter, J. Am. Chem. Soc., 1972, 94, 9267; N. L. Allinger, J. T. Sprague, T. Liljefors, J. Am. Chem. Soc., 1974, 96, 5100.

43 44 45 46

47 48

49

50

51

52

J. B. Birks, Photophysics of Aromatic Molecules, Wiley, New York, 1970. S. Grimme, I. Pischel, S. Laufenberg, F. Vögtle, Chirality, 1998, 10, 147. J. Spanget-Larsen, Theoret. Chim. Acta, 1983, 64, 187–203. The reason for this is the wrong asymptotic behavior of the DF potentials. For details and further references see S. Grimme, M. Parac, Chem. Phys. Chem., 2003, 4, 292. For details see Chapter 9 (Irmgartinger) in this book. A search in the Cambridge Crystallographic Database revealed that out of 30 substituted [2.2]paracyclophanes (only structures with sterically non-demanding substituents have been considered), seven show significant twist and six exhibit parallel-displaced benzene rings as found in our geometry optimization of methyl[2.2]paracyclophane. C. Rosini, R. Ruzziconi, S. Superchi, F. Fringuelli, O. Piermatti, Tetrahedron Asymmetry, 1998, 9, 55. Note that a structure with parallel-displaced rings is energetically lowest for the benzene dimer and that a minimum for a twisted structure similar to [2.2]paracyclophane has not been found at all, see M. O. Sinnokrot, W. F. Valeev, C. D. Sherill, J. Am. Chem. Soc., 2002, 124, 10887. The calculated dipole moment for conformer A (2.6 D) is 0.9 D larger than that of B. A DFT-B3LYP/TZVP calculation for the ground state including the COSMO continuum solvation model for a solvent with DK = 30 gave an inverted order, i.e. A is now 0.2 kcal mol–1 more stable than B which agrees with the conclusion from CD spectroscopy. S. Grimme, W. Mennicke, F. Vögtle, M. Nieger, J. Chem. Soc., Perkin Trans., 1999, 2, 521.

337

13

Fully Conjugated Beltenes (Belt-Like and Tubular Aromatics) R. Herges

13.1

Introduction

In “classical” aromatic systems the p orbitals are perpendicular with respect to the ring plane. It is one of the main goals of cyclophane chemistry to bridge these aromatic systems and to investigate the properties of the bent aromatics. Belt-like and tubular aromatic systems can be viewed as an extreme kind of cyclophanes, where both ends are extended and bent in such a way that the two ends meet each other. The p orbitals are now perpendicular with respect to the surface of a cylinder and the inner lobes of the p orbitals point towards the axis of the system (Fig. 13.1). The bent pyrenophane 2 was synthesized by Bodwell et al. [1] and the belt-like system 3 was proposed as an “interesting target” by Vögtle et al. [2, 3]. As in real life, molecular belts are short sections of tubes. The discovery of carbon nanotubes by Iijima initiated a very extensive and concentrated research on tubular structures. Based on this fund of knowledge, it is straightforward to look at the conjugate belts from the point of view of carbon nanotube chemistry. In principle three different types of conjugated beltenes are conceivable: armchair, chiral and zig-zag beltenes. The shortest belt-like cut of a zig-zag nanotube is a cyclacene (4). In the chiral belt 5 four tetracene units are angularly annulated in a cyclic fashion. 6 is a reduced version of the Vögtle belt 3. The “waist” of fullerenes (complying with the 5-ring rule) is also defined by a belt. C60 and C70 contain (5,5) armchair belts and the D2 symmetric C76, the smallest chiral fullerene, includes a chiral (9,1) belt. None of these fully-conjugated beltenes, exclusively formed by annulated benzene rings, has been synthesized so far by rational synthesis. Dresselhaus et al. [4, 5] proposed a vector notation for nanotubes which can be applied to conjugated belts as well. Formally, the tubes or belts are generated by rolling up a graphite sheet (Fig. 13.4). The tubes are characterized by two integer numbers n and m. Rolling along the horizontal “zig-zag” line and superimposing the points (0,0) and (n,0) leads to a (n,0) zig-zag nanotube or a [n]cyclazene. If the zig-zag line and rolling vector a form Modern Cyclophane Chemistry. Edited by Rolf Gleiter, Henning Hopf Copyright © 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 3-527-30713-3

338

13 Fully Conjugated Beltenes (Belt-Like and Tubular Aromatics)

Fig. 13.1 Normal aromatics, cyclophanes and belt-like conjugated systems (Vögtle belt)

Fig. 13.2 Examples for zig-zag (4), chiral (5) and armchair (6) conjugated belts. Ac-

cording to the vector notation of carbon nanotubes (see Fig. 13.4 below), 4 is a (12,0), 5 a (13,3) and 6 a (7,7) belt

13.1 Introduction

Fig. 13.3 Belts in fullerenes

Fig. 13.4 Vector notation of belts and tubes. The structures are characterized by two integer numbers n and m. Rolling up the “chick-

en wire” from (0.0) to (n,m) and superimposing these two points leads to a (n,m) nanotube

an angle of 30 8, armchair nanotubes are generated. For example, superimposing (0,0) and (10,10) gives an (10,10) armchair nanotube which is one of the most abundant types in commercially-available samples of carbon nanotubes. Thus, the “Vögtle belt” (3) would be a short piece of a (6,6) armchair nanotube, and compound 6 (Fig. 13.2) is a cut from a (7,7) armchair nanotube. Rolling angles a between 0 8 and 30 8 lead to chiral tubes with n=m and n, m=0. Structure 5 is a substructure of a (13,3) chiral nanotube. Because of the six-fold symmetry of the hexagonal

339

340

13 Fully Conjugated Beltenes (Belt-Like and Tubular Aromatics)

Fig. 13.5 Two types of non-annulated conjugated belts derived from armchair nanotubes: all-Z-[0n]paracyclophanes (7) and (n)benzo[4n]annulenes (8)

Fig. 13.6 Cyclo[36]carbon 9, [16,16]paracyclophyne 10, cyclic [8,8]paraphenylacetylene 11 and a cyclic paraphenylacetylene containing 2,6-naphthylene bridges 12

13.2 Approaches Towards Fully-conjugated Beltenes

grid, we only have to consider angles between 0 and 30 8. At a = 60 8 again a zig-zag nanotube is formed. However, consider the case when a (13,3) belt (a = 8.99 8) and a (3,13) belt (a = 60 8–8.99 8 = 51.01 8) (black dots in Fig. 13.4) are enantiomers. The (3,13) belt defines a left hand helix and the (13,3) system a right hand helix. Fullerenes like C60 and C70 contain (5,5) armchair type belts (Fig. 3). [0n]Paracyclophanes (7) and all-Z-(n)benzo[4n]annulenes (8) are non-annulated substructures of armchair nanotubes. Belt-like structures can also be derived from another allotropic form of carbon, the cyclo[n]carbons [6]. The paraphenylacetylenes (11 and 12) were first synthesized by Kawase and Oda et al. [7–10] and the paracyclophane 10 was generated at low temperatures by Tobe et al. [11]. These cyclophynes are described in detail in Chapter 1.

13.2

Approaches Towards Fully-conjugated Beltenes

There are a number of theoretical investigations of the properties of cyclacenes [12–16]; however, synthetic verification has yet to come. Cyclacenes have been the target of extensive synthesis projects, mainly from the group of Stoddard et al. [17–19] (Scheme 13.1) and Corey et al. [20, 21] (Scheme 13.2). Both groups used the Diels-Alder reaction as the key step in the formation of the belt-shape molecular framework. However, neither group has been able to convert the cyclohexadiene rings into benzene units to generate a fully-conjugated belt-like system.

Scheme 13.1

Stoddard’s strategy to synthesize cyclacenes

341

342

13 Fully Conjugated Beltenes (Belt-Like and Tubular Aromatics)

Scheme 13.2

Corey’s strategy to synthesize cyclacenes

When they attempted to synthesize belt-shaped polymers, Schlüter et al. obtained a cyclic but not fully-conjugated beltene derivative 13 [22].

13.3

Belt-like Benzoannulenes (8)

All-Z-tribenzo[12]annulene 15 [23] was synthesized by Iyoda et al. [24] and Vollhard et al. [25] by [2 + 2 + 2] cycloreversion of tris(benzocyclobutadieno)benzene 14 (Scheme 13.3). Like other p spherands 15 forms 1 : 1 complexes with Ag+ and Cu+ [26]. Higher homologs, the tetrabenzo[16]- (16) and pentabenzo[20]annulene (17) were prepared by VCl3/Zn-mediated cyclization of the corresponding dialdehydes, stereochemical conversion of the diol and Corey/Winter olefination [27].

13.3 Belt-like Benzoannulenes (8)

Scheme 13.3

Scheme 13.4

Scheme 13.5

The structure of 15 is rather rigid and probably has C3v symmetry. 16 at room temperature exhibits a fast equilibrium between two C2v symmetric conformations. Coalescence in the 1H NMR spectrum was found at about 08C. The larger pentabenzo[20]annulene 17 is even more flexible and shows no coalescence up to –908C.

343

344

13 Fully Conjugated Beltenes (Belt-Like and Tubular Aromatics)

Scheme 13.6

13.4

[0n]Paracyclophanes 7

Two mesomeric forms of [0n]paracyclophanes [28] are conceivable: polyphenylene (7 a) and quinoid (7 b) structures. The smaller members of the series are quinoid. Larger systems tend to twist the benzene rings out of conjugation and adopt a cyclic oligophenylene structure. In [0,0]paracyclophane 18 and its tetrabenzo analog 19, the benzoid mesomeric structure 18 b does not contribute significantly to the overall wavefunction. Both 18 and 19 are substructures of the smallest conceivable armchair nanotube.

Scheme 13.7

Scheme 13.8

13.4 [0n]Paracyclophanes 7 2,5

A derivative of [0,0]paracyclophane (18), (3,4)-dicyano tricyclo[4.2.2.2 ]dodeca1,3,5,7,9,11-hexaene 20, was recently generated and trapped in situ by Tsuji et al. [29]. Tetradehydrodianthracene (TDDA) 19 was synthesized by Greene et al. [30, 31]. To avoid a nucleophilic addition of the base needed for elimination, the extremely reactive double bonds are trapped in situ by addition of azide, and the double bonds are regenerated under non-nucleophilic conditions. The distance between the two double bonds in TDDA 19 is 2.4 Å, which is considerably smaller than the sum of the van der Waals’ radii of the corresponding sp2 carbon atoms. The pyramidalization angle of the olefinic C atoms is 35 8, which is unusually large. Through-bond and through-space interaction and the pyramidalization lead to a strongly enhanced reactivity towards nucleophiles and an increased reactivity towards electrophiles (Fig. 13.7). Hence, in Diels–Alder reactions 19 is more reactive as an electron-deficient dienophile than maleic anhydride [32] and it readily reacts with tetrazenes according to a [4 + 2] cycloaddition with inverse electron demand [33]. Electrophiles add either in a transannular (anti) or via ring opening (syn). The ratio depends on the polarity of the solvent [34]. The reaction with nucleophiles is also unusual. In contrast to non-activated olefins, 19 reacts even at low temperatures, e.g. with MeLi at –78 8C, in a transannular or 1,2 fashion. The epoxide reacts in a similar way [35].

Scheme 13.9

345

346

13 Fully Conjugated Beltenes (Belt-Like and Tubular Aromatics)

Scheme 13.10 Greene’s synthesis of tetradehydrodianthracene

Fig. 13.7 The electronic structure of TDDA 19

13.4 [0n]Paracyclophanes 7

Scheme 13.11 Reaction of 19 with electrophiles

Scheme 13.12 Reaction of 19 with nucleophiles

Scheme 13.13 Reaction of 19 with alkynes

347

348

13 Fully Conjugated Beltenes (Belt-Like and Tubular Aromatics)

The strained double bonds of 19 are also susceptible to photochemically-induced [2 + 2] cycloadditions. Reaction with acetylene gives the corresponding cyclobutene and thermal reaction with benzyne leads to a benzocyclobutene. The latter compound exhibits the longest C-C bond, 1.713 Å, that has ever been found in a pure and uncharged hydrocarbon [36]. Photochemical [2 + 2] cycloaddition with ethylene gives a cyclobutane, which upon heating undergoes cycloreversion. In this overall metathesis reaction an anthraquinodimethane is formed. Because of the steric hindrance of the inner hydrogens, these compounds cannot be planar. Of the three possibilities to avoid the steric problem: twisting the rings and cis- and trans-pyramidalization, the latter conformation is preferred [37]. Upon reaction with cyclic olefins the cyclobutane intermediate cannot be isolated and cyclophane-like bridged anthraquinodimethanes are formed, which are fixed in a cis-pyramidalized conformation. Reactions with conjugated cyclic polyenes furnish fully-conjugated belt-like systems. The smallest annulene, cyclobutadiene, can be generated in situ and undergoes cycloaddition reactions. However, Diels-Alder reaction with a-pyrone and pyridazine as synthetic equivalents of cyclobutadiene offer a more convenient route to the target product [38]. The quinoid double bonds in 20 are twisted by more than 20 8 and are quantitatively epoxidized at room temperature upon standing under air. Even though benzene is not known to readily undergo cycloaddition or even metathesis reactions it reacts with 19 to form the belt-like conjugated metathesis product 21 in 28% isolated yield [37].

Scheme 13.14 Photochemical reaction of 19 with ethylene

13.4 [0n]Paracyclophanes 7

Scheme 13.15 Reaction of 19 with cyclic olefins

Scheme 13.16 Synthesis of fully-conjugated belts by metathesis of 19 with annulenes

Scheme 13.17 Reaction of 19 with a-pyrone and pyridazine

In contrast to “normal” aromatics, the olefinic C–H bonds point up and down alternately with respect to the ring plane, and the inner lobes of the p orbitals point towards the axis of the tube.

349

350

13 Fully Conjugated Beltenes (Belt-Like and Tubular Aromatics)

Scheme 13.18 Metathesis reaction of 19 with benzene forming the fully-conjugated belt 21

Scheme 13.19 Formation of Kammermeierphane 22

13.4 [0n]Paracyclophanes 7

The double bonds in cyclic conjugated bianthraquinodimethane 20 again react with TDDA 19 in a metathesis reaction to give the fully-conjugated belt-like structure 22 which was named Kammermeierphane [38]. The inner lobes of the p orbitals in 22 are perpendicular with respect to a deformed cyclinder. To avoid a 20-electron anti-aromaticity, the quinoid double bonds and the two single bonds connecting the two bianthraquinodimethane halves are localized. TDDA 19 also reacts with itself in a ring enlargement dimerization metathesis reaction and forms a tube-like structure in which four anthracenylidene groups are connected by four double bonds to a [04]paracyclophane [39]. 23 was named “picotube” because it is a substructure of a carbon nanotube. According to X-ray analysis, 1H (two signals) and 13C NMR spectra (four signals) the symmetry is D4h. However, theoretical calculations and low temperature IR spec-

Scheme 13.20 Ring enlargement dimerization of 19 to form the “picotube” 23

Scheme 13.21 “Picotube” 23 is a substructure of a [4,4] armchair carbon nanotube

351

352

13 Fully Conjugated Beltenes (Belt-Like and Tubular Aromatics)

Scheme 13.22 Time-averaged D4h symmetry

of picotube 23

Scheme 13.23 Quinoid and benzoid structure of 23

tra revealed that there is only a time-averaged D4h structure which is a fast equilibrium of two D2d structures [40]. Obviously, steric reasons prevent the structure from completely distorting towards a benzoid system. The twist of the central double bonds (17.7 8) notwithstanding, the bond lengths of these bonds (1.367 Å) and the neighboring single bonds (1.489 Å) indicate a quinoid structure 23 a.

Scheme 13.24 Pyrolysis of picotube 23

13.4 [0n]Paracyclophanes 7

Scheme 13.25 Reaction products of the Friedel-Crafts alkylation of 23 with t-BuCl/AlCl3

Attempts to form a closed tube by cyclodehydration failed. Upon flash pyrolysis the quinoid bonds and the neighboring single bonds are cleaved [41]. The benzene rings of picotube 23 readily react under Friedel-Crafts conditions; usually yielding a mixture of regioisomers. With an excess of t-BuCl/AlCl3, however, two main products are formed which bear eight t-Bu groups in the sterically most favorable position [42]. 24 has D4 symmetry and is therefore chiral. Separation of the enantiomers was achieved by HPLC on a chiral stationary phase. Picotube 23 is relatively stable towards oxidation (no reaction with MCPBA at 20 8C or heating under air up to 350 8C) but can be reduced with Li metal to the tetra-anion, which is extraordinarily stable (even in boiling THF!). Two of the Li atoms are inside the tube and two complex with two of the double bonds from outside the tube [43]. The parent systems of [0n]paracyclophanes have not yet been synthesized, but calculations predict that the [05]paracyclophane has a belt-like quinoid D5h sym-

Scheme 13.26 Structure of tetra-anion of 23. The Li counterions are indicated as spheres

353

354

13 Fully Conjugated Beltenes (Belt-Like and Tubular Aromatics)

Scheme 13.27 B3LYP/6-31G*-calculated structures of [05]- and [06]paracyclophane

metric structure and the larger systems such as the [06]paracyclophane twist the benzene rings towards Dnd symmetry [44].

13.5

Möbius Belts

Since the prediction of Heilbronner in 1964 [45], that annulenes with a Möbius twist should be aromatic with 4n and antiaromatic with 4n + 2 electrons, a number of belt-like Möbius systems have been proposed, such as the Möbius cyclacenes [46, 47] and Möbius coronene [48]. The first stable Möbius-annulene was synthesized by Herges et al. by combining the belt-like conjugated tetradehydrodianthracene and a “normal” p system using ring enlargement metathesis [49]. The twist in the p system is stabilized by the rigid belt-like conjugated bianthraquinodimethane unit. Tricyclooctadiene was used as a synthetic equivalent of cyclooctatetraene. Upon irradiation a ladderane compound is formed which opens in sequence of [2 + 2] cycloreversion and two electrocyclic reactions to form the Möbius annulene.

Scheme 13.28 Hypothetical Möbius-twisted [15]cyclacene and Möbius coronene

13.6 Conjugated Belts from Fullerenes

Scheme 13.29 Strategy for the synthesis of a Möbius

annulene

Scheme 13.30 Synthesis of a Möbius belt

13.6

Conjugated Belts from Fullerenes

Another approach to synthesize conjugated belts starts from fullerenes [50]. Nakamura et al. used an organocopper reagent to convert C60 into a 50 p electron cyclopentadiene compound. The acidic cyclopentadiene hydrogen had to be protected by a cyano group before performing a second copper-mediated phenylation which leaves a 40 p electron-conjugated belt (substructure of a (5,5) armchair nanotube). Since conjugated belts are straightforward interim points on the route to a rational synthesis of carbon nanotubes we expect a rapid increase in synthetic activities aimed at these targets.

355

356

13 Fully Conjugated Beltenes (Belt-Like and Tubular Aromatics)

Scheme 13.31 Synthesis of a [10]cyclophenacene from C60. Fully-conjugated rings are grey

shaded

13.7

References 1

2 3

4 5

6 7

G. J. Bodwell, T. J. Houghton, J. W. Jason, M. R. Mannion, Angew. Chem. 1996, 108, 1418–1420; Angew. Chem. Int. Ed. Engl. 1996, 35, 2121–2123. F. Vögtle, Top. Curr. Chem. 1983, 115, 157. Review: a) A. Schröder, H.-B. Mekelburger, F. Vögtle, Top. Curr. Chem. 1994, 172, 180–201; b) L. T. Scott, Angew. Chem. 2003, 115, 4265–4267; Angew. Chem. Int. Ed. Engl. 2003, 42, 4133–4135. M. S. Dresselhaus, G. Dresselhaus, R. Saito, Carbon 1995, 33, 883. M. S. Dresselhaus, G. Dresselhaus, P. C. Eklund, Science of Fullerenes and Carbon Nanotubes, Academic Press, San Diego, 1996. Review: N. S. Goroff, Acc. Chem. Res. 1996, 29, 77. T. Kawase, H. R. Darabi, M. Oda, Angew. Chem. 2003, 108, 2803–2805; Int. Ed. Engl. 1996, 35, 2664–2666.

8

9

10

11

12 13

T. Kawase, Y. Seirai, H. R. Darabi, M. Oda, Y. Sarakai, K. Tashiro, Angew. Chem. 2003, 115, 1659–1662; Int. Ed. Engl. 2003, 42, 1624–1628. T. Kawase, K. Tanaka, N. Fujiwara, H. R. Darabi, M. Oda, Angew. Chem. 2003, 115, 1662–1666; Int. Ed. Engl. 2003, 42, 1624–1628. T. Kawase, K. Tanaka, Y. Seirai, N. Shiono, M. Oda, Angew. Chem. 2003, 115, 5755–5758; Int. Ed. Engl. 2003, 42, 5597–5600. Y. Tobe, R. Furukawa, M. Sonoda, T. Wakabayashi, Angew. Chem. 2001, 113, 4196–4198; Int. Ed. Engl. 2001, 40, 4072– 4074. E. Heilbronner, Helv. Chim. Acta 1954, 37, 921–935. J. Aihara, J. Chem. Soc. Perkin Trans. 2 1994, 971–974.

13.7 References 14

15

16 17

18

19

20

21 22

23

24

25 26

27 28 29

30

31

I. Gutman, P. U. Biedermann, V. IvanovPetrovic, I. Agranat, Polycycl. Arom. Compounds 1996, 8, 189–202. H. S. Choi, K. S. Kim, Angew. Chem. 1999, 111, 2400–2402; Int. Ed. Engl. 1999, 38, 2256–2258. K. N. Houk, P. S. Lee, M. Nendel, J. Org. Chem. 2001, 66, 5517–5521. F. H. Kohnke, A. M. Z. Slawin, J. F. Stoddart, D. J. Williams, Angew. Chem. 1987, 99, 941; Int. Ed. Engl. 1987, 26, 892. P. R. Ashton, G. R. Brown, N. S. Isaacs, D. Giuffrida, F. H. Kohnke, J. P. Mathias, A. M. Z. Slawin, D. R. Smith, J. F. Stoddart, D. J. Williams, J. Am. Chem. Soc. 1992, 114, 6330–6353. P. R. Ashton, U. Girreser, D. Giuffrida, F. H. Kohnke, J. P. Mathias, F. M. Raymo, A. M. Z. Slawin, J. F. Stoddart, W. J. Williams, J. Am. Chem. Soc. 1993, 115, 5422–5429. R. M. Cory, C. L. McPhail, A. Dikmans, J. J. Vittal, Tetrahedron Lett. 1996, 1983– 1986. R. M. Cory, C. L. McPhail, Tetrahedron Lett. 1996, 1987–1990. A. Godt, V. Enkelmann, A. D. Schlüter, Angew. Chem. 1989, 101, 1704; Angew. Chem. Int. Ed. Engl. 1989, 28, 1680. Y. Kuwatani, T. Yoshida, A. Kusaka, M. Oda, K. Hara, M. Yoshida, H. Matsuyama, M. Iyoda, Tetrahedron 2001, 57, 3567–3576. M. Iyoda, Y. Kuwatani, T. Yamauchi, M. Oda, J. Chem. Soc., Chem. Commun. 1988, 65. D. L. Mohler, K. P. C. Vollhardt, S. Wolff, Angew. Chem. 1990, 102, 1200. T. Yoshida, Y. Kuwatani, K. Hara, M. Yoshida, H. Matsuyama, M. Iyoda, S. Nagase, Tetrahedron Lett. 2001, 42, 53–56. Y. Kuwatani, T. Yoshida, A. Kusaka, M. Iyoda, Tetrahedron Lett. 2000, 359–363. M. N. Jagadeesh, A. Makur, J. Chandrasekhar, J. Mol. Model. 2000, 6, 226–233. T. Tsuji, M. Okuyama, M. Ohkita, T. Imai, T. Suzuki, Chem. Comm. 1997, 2151–2152. R. L. Viavatenne, F. D. Greene, L. D. Cheung, R. Majeste, L. M. Trefonas, J. Am. Chem. Soc. 1974, 96, 4342–4343. The procedure given in ref. 30 is not reproducible (Herges et al. unpublished and

32

33

34 35

36

37

38

39

40

41 42

43

44

45 46 47

numerous private communications) even though the X-ray structure of Greene et al. was reproduced by a modification of the procedure (Herges, unpublished). S. Kammermeier, H. Neumann, F. Hampel, R. Herges, Liebigs Ann. Chem. 1996, 1795–1800. J. Sauer, J. Breu, U. Holland, R. Herges, H. Neumann, S. Kammermeier, Liebigs Ann./Receuil 1997, 1473–1479. R. Herges, H. Neumann, Liebigs Ann. Chem. 1995, 1283. R. Herges, H. Neumann, F. Hampel, Angew. Chem. 1994, 106, 1024–1026; Angew. Chem. Int. Ed. Engl. 1994, 33, 993. S. Kammermeier, P. G. Jones, R. Herges, Angew. Chem. 1997, 109, 1825– 1828 b; Angew. Chem. Int Ed. Engl. 1997, 36, 1757–1760. S. Kammermeier, R. Herges, Angew. Chem. 1996, 108, 470–472; Angew. Chem. Int. Ed. Engl. 1996, 35, 417–419. S. Kammermeier, P. G. Jones, R. Herges, Angew. Chem. 1997, 109, 2317– 2319; Int. Ed. Engl. 1997, 36, 2200–2202. S. Kammermeier, P. G. Jones, R. Herges, Angew. Chem. 1996, 108, 2834– 2836; Angew. Chem. Int. Ed. Engl. 1996, 35, 2669–2671. R. Herges, M. Deichmann, J. Grunenberg, G. Bucher, Chem. Phys. Lett. 2000, 327, 149–152. M. Deichmann, C. Näther, R. Herges, Org. Lett. 2003, 5, 1269–1271. R. Herges, M. Deichmann, T. Wakita, Y. Okamoto, Angew. Chem. 2003, 115, 1202–1204; Angew. Chem. Int. Ed. Engl. 2003, 42, 1170–1172. N. Treidel, M. Deichmann, T. Sternfeld, T. Sheradsky, R. Herges, M. Rabinowitz, Angew. Chem. 2003, 115, 1204– 1208; Angew. Chem. Int. Ed. Engl. 2003, 42, 1172–1176. M. N. Jagadeesh, A. Makur, J. Chandrasekhar, J. Mol. Modelling 2000, 6, 226–233. E. Heilbronner, Tetrahedron Lett. 1964, 1923–1928. S. Martin-Santamaria, H. S. Rzepa, J. Chem. Soc., Perkin 2 2000, 2378–2381. J.-M. Andre, B. Champagne, E. A. Perpete, M. Guillaume, Int. J. Quant. Chem. 2001, 84, 608–616.

357

358

13 Fully Conjugated Beltenes (Belt-Like and Tubular Aromatics) J. Cz. Dobrowolski, J. Chem. Inf. Comput. Sci. 2002, 42, 490–499. 49 D. Ajami, O. Oeckler, A. Simon, R. Herges, Nature 2003, 426, 819–821. 48

50

E. Nakamura, K. Tahara, Y. Matsuo, M. Sawamura, J. Am. Chem. Soc. 2003, 125, 2834–2835.

381

15

NMR Spectra of Cyclophanes Ludger Ernst and Kerstin Ibrom

15.1

Introduction and Scope

NMR spectroscopy of cyclophanes has been treated extensively in the literature and a number of reviews exist that have adequately covered the field until the middle of 1999. An early review was given by Smith (1964) in his book “Bridged aromatic compounds” [1]. An often-cited two-volume book edited by Keehn and Rosenfeld [2] includes a chapter by Mitchell on the NMR properties and conformational behavior of cyclophanes [3] which treats the literature up to the end of 1981. That book also contains chapters on [n]cyclophanes [4], heterophanes [5], nonbenzenoid cyclophanes [6], and multilayered cyclophanes [7], in which the authors discuss the NMR aspects of their respective classes of compounds. Some time ago, one of the authors of the present chapter summarized the literature on the NMR spectroscopy of cyclophanes that appeared between 1982 and mid-1999 [8]. Hence, there is now only need to cover approximately the last four years. Still, in view of the limited space available, a number of restrictions had to be made with respect to the literature references included. As in our previous review [8], the focus is on the NMR properties of the cyclophanes themselves, not on the changes that occur when the cyclophanes interact with other molecules. Thus host–guest or supramolecular interactions upon NMR spectra are not covered. Accordingly, crown ether derivatives of cyclophanes and their thio analogs, polyazacyclophanes, calixarenes and analogous compounds, the cavitands, the carcerands, the spherands or the rotaxanes etc. are not treated as the main interest in these compounds lies in their suitability for such interactions. Also, cyclophanes possessing only bridges spanning ortho-positions are not considered to be proper cyclophanes because most of them do not show the characteristic NMR spectroscopic properties usually associated with cyclophanes, such as shielding of protons positioned above/below the planes of aromatic rings. Orthocyclophanes are therefore also omitted. Chemical Abstracts online searches were carried out for papers containing the combination of the terms “NMR” (or related ones) and “phane” or “cyclophane” (or related terms). The latest such search was performed in the “CA File” updated on 6th March 2003. Thus, the literature reviewed here spans the period from the Modern Cyclophane Chemistry. Edited by Rolf Gleiter, Henning Hopf Copyright © 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 3-527-30713-3

382

15 NMR Spectra of Cyclophanes

middle of 1999 to approximately the end of 2002. Some references from 2003 are also included. The type of search carried out implies that the respective author must have considered the NMR-related content of his/her paper important enough to mention it in the title, abstract or keywords. The papers found by this method were then inspected and judged whether they contained interesting enough information to be included in this chapter. Admittedly, certain personal preferences may have influenced the selection process. The material presented in this article is arranged according to classes of compounds. This makes it easy for the reader to find information related to particular molecules. Cross references to other sections are given when papers deal with compounds belonging to different classes.

15.2

[n]Phanes

There are two main reasons why phanes are interesting to NMR spectroscopists. The first consists in unusual 1H chemical shifts which are caused by the magnetic anisotropy of the aromatic system(s) in these molecules. The second reason lies in the mobility of the bridges which is often restricted because of their shortness and therefore brings the rate of existing conformational processes into the range observable by NMR spectroscopy. The main interest in the [n]phanes concerns metacyclophanes and paracyclophanes with short bridges by which the aromatic ring is forced out of planarity and the chemical and spectroscopic behavior of the molecule is altered relative to cyclophanes possessing longer bridges. 15.2.1

[n]Metacyclophanes

The shortest-bridged [n]metacyclophanes that are isolatable and stable compounds are those with n = 5. The parent hydrocarbon [9] and some ring-substituted halogen derivatives [10] have been described earlier. The 8,11-dichloro- (1 a) and the 11-chloro-N-tosyl-3-aza[5]metacyclophane (1 b) are present exclusively in the exoconformation, in which the bridge is pointing away from the benzene ring [11]. This was deduced from the vicinal H,H-coupling constants in the bridge and from NOEs. The 1H-NMR spectra were unchanged between –60 and +120 8C. At higher temperatures decomposition occurred. By way of contrast, the 3-carba analogs contained 11–15% of the exo-conformer [10]. A Karplus analysis of the bridge proton NMR signals at room temperature showed ketone 2 also to exist as an exo/ endo-equilibrium mixture [12]. Variable temperature spectra and line shape analysis revealed that, at –67 8C, the ratio is 80 : 20 (88 : 12 for the compound with CH2 instead of C=O). The activation parameters for the conformational exchange are DHz = 58.2 kJ mol–1 and DSz = 15.9 J mol–1 K–1 (DHz = 48.5 kJ mol–1 and DSz = – 23.0 J mol–1 K–1, respectively for the oxygen-free compound).

15.2 [n]Phanes

The endo-conformation of [5]metacyclophane, endo-3, is one of the examples with protons over benzene rings for which Martin et al. [13] performed GIAO-SCF ab initio calculations to estimate 1H-NMR shieldings and to derive a relatively simple equation for predicting shieldings in such situations. Other examples are mentioned in Sections 15.2.2 and 15.3. Rüdiger and Schneider [14] computed the chemical shifts of all protons in exo- and endo-3 using their program SHIFT, which takes into account classical equations describing magnetic anisotropy, electric effects, and sterically induced charge separation. The average deviations between predicted and experimental shifts were 0.30 and 0.57 ppm for the exo- and the endo-conformer, respectively. For further examples, see Section 15.4. As far as [6]metacyclophanes are concerned, Bickelhaupt and coworkers [15] performed a careful 1H- and 13C-NMR study at –60 8C of [6]metacyclophane (4 a) and its chloro derivatives 4 b and 4 c. While, at room temperature, the hexamethylene chain cannot flip from one side of the aromatic ring to the other, it does undergo a pseudorotation movement on the same side of the ring, for which DHz = 45.6 kJ mol–1 and DSz = –20.9 J mol–1 K–1 were determined by variable temperature (VT) 13C-NMR spectroscopy for 4 c. At –60 8C the molecule is frozen in an unsymmetrical conformation (loss of the symmetry plane through C-9, C-12, and the midpoint of the C-3/C-4 bond) and all protons and carbon nuclei are nonequivalent. Complete assignment of all 1H and 13C shifts was achieved by 2D shift correlation and 1D NOE techniques and most of the J(H,H) values in the bridge were determined. The latter define the conformation of the six-membered bridge. A VT 1H- and 13C-NMR study of a [6]heterometacyclophane, the 1,3-dimethyluracil-annulated [6](2,4)pyridinophane 5 a [16], indicated a barrier DGz to bridge

383

384

15 NMR Spectra of Cyclophanes

flipping of ca. 84 kJ mol–1 (150 8C) and also the existence of a lower-energy process with DGz = 42 kJ mol–1 (–30 8C), which was interpreted as the pseudorotation of the bridge as in the cases of the preceding compound and the analogous 6-phenyl[6](2,4)pyridinophane [17]. The longer bridges in 5 b and 5 c are rapidly flipping at room temperature as seen from the chemical equivalence of the benzylic protons, yet introduction of the intra-annular methyl group as in 5 d causes so much hindrance that even the 11-membered bridge does not flip rapidly on the NMR time scale at 150 8C. These authors also reported work much along the same lines for [7]metacyclophanes 6 a–c [18]. For these compounds the barriers to bridge flipping are DGz (Tc) = 47.3 (–10 8C), 49.0 (0 8C), and 51.0 kJ mol–1 (–5 8C), respectively. Some [n](3,5)isoxazolophanes 7 were also studied in a similar manner [19]. The barriers DGz to bridge flipping are 77.8 kJ mol–1 at 100 8C for 7 a with the 8-membered bridge and 48.1 kJ mol–1 at –10 8C for 7 b with the 9-membered bridge while the barrier in the methyl compound 7 c was too high to be determined (Tc > 150 8C). The pseudorotation processes of the oligomethylene chains required DGz values of 46.7 kJ mol–1 at –10 8C, 38.1 kJ mol–1 at –70 8C, and 36.0 kJ mol–1 at –80 8C in the order mentioned.

Compounds 8 [20] and 9 [21] are both dibenzoannulated [n]metacyclophanes (n = 8 and 12, respectively) containing a butadiyne fragment in the bridge. Both possess

15.2 [n]Phanes

an intra-annular proton that is distinctly deshielded by the magnetic anisotropy of the opposite diyne unit, cf. the d values in the formulae. In the precursor of 8, in which the central C–C bond does not yet exist, the proton in question has a chemical shift of 7.89 ppm.

Two other metacyclophanes, 10 a and 10 b, and the N-protonated form, 10 c, of the latter, all possessing rather complex rigid 13-membered bridges, have been studied by VT 1H NMR in order to determine the barrier to aromatic ring passage past the bridge [22]. Surprisingly, the barrier DGz is smaller for the benzene ring in the first compound (28.9 kJ mol–1 at –116 8C) than for the pyridine ring in the second (36.4 kJ mol–1 at –78 8C), so the different steric hindrance of C–H vs. N cannot be invoked as the explanation. A substantial increase of the barrier is caused by protonation of the pyridine nitrogen: DGz for 10 c is 55.2 kJ mol–1 at 21 8C. Two corresponding [13]paracyclophanes are mentioned in Section 15.2.2.

15.2.2

[n]Paracyclophanes

The diester 11 of [6]paracyclophane is another example of compounds with protons over a benzene ring for which the chemical shifts have been computed by Martin et al. [13] (see the previous section). Abraham and coworkers extended their earlier model for calculating the 1H chemical shifts of aliphatic compounds to aromatic ones by including ring current effects and deshielding steric effects for crowded protons [23]. Two cyclophane examples were included. [10]Paracy-

385

386

15 NMR Spectra of Cyclophanes

clophane (12) showed good agreement between predicted and experimental proton shifts and for the in-phane 13 the authors demonstrated that aromatic carbons exert very different steric effects than aliphatic ones. (We note that the structure of 13 given in ref. [23] is incorrect.)

The [13]paracyclophanes 14 and 15 have much higher barriers to aromatic ring flipping than the meta-analogs mentioned above [22]. The DGz value for the naphthalene derivative was determined from coalescence of the benzylic and endo-proton signals and from the 19F signals of the CF3 groups to be 68.6–70.3 kJ mol–1 at 49–90 8C while that for the nitro-p-phenylene group is 76.1–77.4 kJ mol–1 at 97–108 8C (from benzylic and endo-proton signals).

NMR spectra of some tri- and tetra-aza[n](1,4)naphthalenophanes (n = 9, 11, 12, 14), 16 a–d, have been reported [24]. Apart from 16 d, bridge flipping is slow at room temperature as indicated by the chemical nonequivalence of the benzylic protons (Dd = 0.89–1.21 ppm). Only at –67 8C is the flipping of the 14-membered bridge in 16 d slow enough on the NMR time scale for decoalescence to occur, DGz = 38.9 kJ mol–1). A high-pressure 1H-NMR study of some [n]paracyclophanes, 17 a–c, [n] = 12,13, was carried out by Yamada et al. [25]. Increasing the hydrostatic pressure P from 3 to 392 MPa at 133 8C increased the rate constant of internal rotation of the pphenylene ring from 35 to 75 s–1, indicating a negative activation volume DVz, i.e. the transition state of the rotation (bridge near the plane of the aromatic ring) occupies less space than the ground state (bridge perpendicular to the ring). The increase in k was diminished with increasing P. There were some differences in the pressure-dependent behavior of 17 a, 17 b, and 17 c which could not be explained satisfactorily.

15.2 [n]Phanes

15.2.3

Other [n]Phanes

The [n](1,6)heptalenophanes (n = 577), 18 a–c, possess rigid bridge conformations [26]. The bridge proton chemical shifts are unspectacular and show no effect of a potential paramagnetic ring current of the anti-aromatic system (12 p-electrons). Reduction of 18 a with lithium furnished the 14 p-electron dianion 18 a2– which behaves as an aromatic system although the perimeter of the molecule cannot be planar. The central methylene group of the bridge is shielded by Dd = 1.28 ppm by formation of the anion. Perimeter protons H-3,8 are also shielded (by 1.42 ppm), but here it is difficult to separate the effects of charge density and of aromatization.

A nice series of [n](2,7)pyrenophanes (n = 7–9), 19 a–c, was studied by Bodwell and coworkers [27]. As the length of the tether decreases and the bend in the pyrene unit increases, the first three pairs of methylene protons (counting from the aromatic system) are increasingly shielded: d(H–a) = 2.84, 2.59, 2.30; d(H–b) = 1.10, 0.88, 0.45; d(H–c) = 0.05, –0.69, –1.38 for 19 c, 19 b, and 19 a, respectively. Model considerations suggest that along this series the protons mentioned move further into the shielding zone of the pyrene system as the tether length decreases. The highest shielding is observed for H–e (d = –2.08), which is only present in 19 c. The aromatic protons are moderately shielded in the same order. This is, however, difficult to explain, because in spite of the strong bend of the pyrene unit it is still fully aromatic as evidenced by the ring current effect upon the central methylene

387

388

15 NMR Spectra of Cyclophanes

protons. Two-electron reduction of 19 a with lithium metal afforded dianion 20 instead of the expected anti-aromatic 16 p-pyrenophane dianion [28]. The structure of 20 was inferred from its NMR data, among other things from the shift of the C-1 signal into the aliphatic region (d = 33.9) and its 1J(C,H) coupling constant of 163.2 Hz. Both speak for a cyclopropane-type carbon atom. Moreover, the 3J(H,H) coupling of 9.1 Hz between the equivalent protons H-1 and H-8 show that C-1 and C-8 must be connected by a r-bond. The spectra of the dianion were fully assigned by 2D techniques.

Bodwell’s group also extended their earlier investigation of 1,n-dioxa[n](2,7)pyrenophanes, 21, with n = 7 and 8 to compounds with n = 9–12 [29]. The chemical shifts of the bridge protons, the aromatic protons and the aromatic carbons were discussed in terms of the degree of bending of the pyrene unit. The first example of a bridged cis-10 b,10 c-dihydropyrene is 22, which represents a [14]annulene with a 12-membered bridge [30]. The comparison of its proton chemical shifts (see the values given in the formula) with those of cis-dimethyldihydropyrene showed that the bridge does not significantly affect the geometry of the ring system. Also, the bridge proton shifts are very similar to those of pyrenophane 21 (n = 12).

In the dithia[9]corannulenophane 23 the aliphatic bridge is held in a rigid conformation above the aromatic system [31]. The diastereotopicity of the protons in all methylene groups demonstrates that there is no jump-rope motion of the chain around the corannulene nucleus. The protons pointing into the inside of the corannulene bowl are strongly shielded (see the dH values in the formula). The extreme shift of –3.64 ppm corresponds to a 5 ppm shielding relative to the shift of the corresponding hydrogen in 1,5-pentanedithiol (see Section 15.4 for a related [3.3]phane).

15.3 [2.2]Phanes

15.3

[2.2]Phanes

Publications on NMR spectroscopy of [2.2]phanes that appeared within the review period predominantly deal with [2.2]paracyclophanes although [2.2]phanes with other aromatic units than simple benzene rings attracted a considerable amount of interest. Some work on [2.2]metaphanes was also reported. For example Martin et al. [13] performed GIAO-SCF ab initio calculations to estimate 1H-NMR shieldings in the anti-conformation of [2.2]metacyclophane 24 which is one more case of molecules with protons over benzene rings. Other examples are mentioned in Sections 15.2.1 and 15.2.2.

As for [2.2]paracyclophanes, Battiste et al. [32] characterized a number of Diels–Alder adducts of 4,5-dehydro- and 4,5,15,16-bis(dehydro)-1,1,2,2,9,9,10,10-octafluoro[2.2]paracyclophanes (25 a–d, 26 a–b, 27, 28 a–c, 29) by 1H-, 13C-, and 19FNMR spectroscopy. The stereochemistry was proven by the NOEs that were observed between the aromatic protons of the [2.2]paracyclophane system and protons of the substituent. Furthermore the paracyclophane protons in the unsubstituted ring that point towards the aromatic ring of the naphthalene and anthracene units in 25 b (d = 5.78) and 25 c (d = 5.84) are highly shielded in comparison with 25 a (d = 6.91) because they are located in the shielding region of the opposite aromatic ring. Due to a through-space contribution 25 a–c exhibit remarkably large spin–spin coupling constants J(F,C) between one of the bridge fluorine atoms at C-2 and the bridge-head carbon atom of the substituent, for example 10.5 Hz in 25 a.

389

390

15 NMR Spectra of Cyclophanes

The same group also reported through-space 19F,19F spin–spin couplings that were found for 30 in which the CF3 group is coupled with two of the proximate bridge fluorines, J(F,F) = 26.8 and 19.5 Hz [33]. Further work on hetero-annularly disubstituted octafluoro[2.2]paracyclophanes by Dolbier and Roche that was quoted as a private communication in ref. [8] has been published in the meantime [34].

15.3 [2.2]Phanes

The parent compound, [2.2]paracyclophane (31), does not adopt a fixed eclipsed D2h conformation but exists as a rapidly equilibrating mixture of two degenerate twisted D2 conformers. This degeneracy is lifted in an ar-monosubstituted [2.2]paracyclophane 32 and the equilibrium is shifted to one side such that the syn-hydrogen H-2s at the bridge carbon ortho to the substituent evades the close contact with the latter that is present in 32 a [35]. The extent of the equilibrium shift depends on the size of the substituent R and can be recognized from the change of the two trans vicinal H,H coupling constants in the near ethano bridge: 3 J(1 a,2s) decreases and 3J(1s,2 a) increases relative to the value of 4.1 Hz in the parent compound. The extreme values observed so far are 1.5 Hz and 7.2 Hz, respectively, when the substituent is a nitro group. When a second substituent is introduced in the position pseudo-geminal to the first one, the two conformers of the disubstituted [2.2]paracyclophane 33 are degenerate again and the trans 3 J(H,H) coupling has practically the same value as in 31 independent of the nature of the substituent (4.1 Hz for the dichloro and the dibromo compound, 4.2 Hz for the difluoro and the diiodo compound) [36].

The condensation of 4-formyl[2.2]paracyclophane with cyclopentadiene yields 4-(6fulvenyl)[2.2]paracyclophane as two different isomers 34 a (major) and 34 b (minor) [37]. These arise from restricted rotation about the C–C bond connecting the fulvenyl substituent with the phane which is caused by unfavourable steric interactions between the bulky substituent and the phane bridges. An activation barrier of 109 kJ mol–1 (70 8C) was derived by means of equilibration 1H-NMR experiments.

391

392

15 NMR Spectra of Cyclophanes

More complex structures are formed in the reaction of dehydrobenzene with the four isomeric ar,ar'-diethenyl[2.2]paracyclophanes which furnished different [2.2]phenanthrenophanes 35–37, as well as mono-annulated products 38 a–c [38]. Educts and products were characterized by 1D and 2D 1H- and 13C-NMR methods. Careful consideration of substituent and anisotropic shielding effects allowed the identification of the particular stereoisomers.

Extensive NMR investigations were also done by Minuti et al. who prepared a series of helicenophanes 39–43 whose structures were deduced from fully assigned 1D and 2D 1H- and 13C-NMR spectra [39]. Additionally, NOE measurements proved the cis-relationship of H-5 a, H-8 a and H-8 b in 40 and between H-8 b and the unsubstituted benzene ring. NOEs also gave information about the regiochemistry of the carbonyl group in 42 and 43. The unusually large chemical shift of one of the C-12 protons in 41 is attributed to the spatial proximity of the magnetically anisotropic carbonyl group. Similar investigations were made with [2.2]paracyclophanes 44–46 bearing even larger polycyclic aromatic units [40]. Furthermore the helicenophanes 47–50 containing five-membered rings and 51– 52 were studied in detail using NOE experiments for the determination of the exact regio- and stereochemistry [41, 42].

15.3 [2.2]Phanes

393

394

15 NMR Spectra of Cyclophanes

[2]Paracyclo[2](2,7)pyrenophane (53), a [2.2]phane consisting of a benzene and a pyrene unit, was studied by Bodwell et al. [43]. In this compound the aromatic protons of the benzene deck are strongly shielded by the pyrene moiety that is much wider than the benzene ring. They absorb at d = 5.54, roughly 1.5 ppm upfield from p-xylene (d = 7.07). The corresponding [2]metacyclo[2](2,7)pyrenophane (54) was submitted to a two-electron reduction of which Aprahamian et al. [28] obtained the surprising dianion 55. The arguments for the structure are the same as for the analogous product 20 with an aliphatic chain instead of the meta-substituted ring (see Section 15.2.3).

[2.2]Phanes lacking simple benzene units are the 1,8-fluorenophanes 56–58. They possess a rigid conformation [44]. This is obvious from their proton NMR spectra. Two different chemical shifts are observed for the 9-protons, one of which is strongly shielded (56: d = –0.69, 57: d = –1.59, 58: d = –0.80) because of the magnetic anisotropy of the opposite aromatic ring. For the corresponding dithiaphanes, see Section 15.4.

15.3 [2.2]Phanes

The dibenzochrysenophanes 59 and 60 have even more complex structures which Yano et al. [45] characterized by 1H-NMR spectroscopy. They show four distinctly separated pairs of biphenyl aromatic protons due to restricted rotation. The remarkable upfield shifts of the biphenyl protons (d = 5.37, 6.31, 6.87, 7.09) and of the dibenzochrysene protons H-a and H-g (d = 7.43 and 8.54, respectively) in 59 compared to the corresponding ones in 4,4'-dimethylbiphenyl (d = 7.23, 7.47) and 3,11-dimethyldibenzo[c,l]chrysene (H-a, H-g : d = 8.85 and 9.12, respectively) clearly indicate an oblique stacking structure for 59. Similar observations were made for 60.

The carbazolophanes 61 and 62 were also studied by 1H-NMR spectroscopy [46]. From the remarkable high-field shifts of H-4 (d = 6.74 in 61 and 6.22 in 62) and

395

396

15 NMR Spectra of Cyclophanes

H-a (d = 4.75 in 61 and 5.20 in 62) relative to p-xylene, 1,4-dimethylnaphthalene and 9-ethyl-3,6-dimethylcarbazole the authors inferred that these protons lie directly above the opposite aromatic ring.

15.4

[3.3]Phanes

[3.3]Phanes are interesting due to their conformational flexibility which is often observable by NMR methods. Bridge flipping processes may be analyzed. In the case of [3.3]metacyclophanes ring flipping is possible which leads to syn/anti-isomers. Therefore many publications deal with metacyclophanes. Mitchell, for example, re-examined the temperature dependence of the 1H- and 13C-NMR spectra of 2,11-dithia[3.3]metacyclophane (63) [47]. One process observed is the interconversion of the syn-chair,chair and syn'-chair,chair isomers (DGz = 40 kJ mol–1 at –75 8C) that exchanges the benzylic protons. Calculations show that this process proceeds through the anti-isomer in which bridge wobbling occurs that only requires little activation energy. Hence, the experimentally determined barrier is essentially that for syn/anti interconversion. The second process observed is the transformation of the syn-chair,chair into the syn-boat,chair isomer which becomes slow on the NMR time scale at about the same temperature. At –90 8C two sets of 1H- and 13C-NMR signals are observed in the intensity ratio of 2 : 1.

15.4 [3.3]Phanes

Satou and Shinmyozu were also interested in the bridge flipping processes of [3.3]metacyclophanes. They reported a NMR study of 2,11-diaza[3.3](3,5)pyridinophane (64) with an aza- instead of a thia-bridge [48]. Like 63 this compound adopts the syn-conformation. In combination with deuterium exchange, VT 1H-NMR experiments revealed that the syn-chair,boat- and the syn-boat,boat-isomer are present in approximately a 1 : 1 ratio at –90 8C. The assignment of the bridge proton signals to the individual isomers is done on the basis of their geminal coupling constants. The free energy barrier DGz for the bridge inversion amounts to 43.5 kJ mol–1 at –55 8C. Another interesting example of a [3.3]heterophane is [3](3,6)pyridazino[3](1,3)indolophane (65) synthesized by Bodwell and Li [49]. The authors conclude from the chemical shift of the proton at the indole deck (d = 5.78) and from the comparison of the shielding effects of the pyridazine protons (d = 6.13 and 6.33) with that of 3,6-dimethylpyridazine (d = 7.20) and indolophane 66 that the ring conformation is predominantly exo in solution.

The 1H-NMR chemical shifts of the syn- and anti-isomers of diketometacyclophane 67 and of metaparacyclophane 68 were computed on the basis of classical shielding calculations by use of the computer program SHIFT [14] already mentioned in Section 15.2.1.

Laali et al. performed low temperature protonation studies of dithia[3.3]metacyclophanes 69–70 with and without intra-annular methyl groups [50]. They assigned

397

398

15 NMR Spectra of Cyclophanes

all 13C- and 1H-NMR signals of educts and products by 1D and 2D methods. NOE measurements, in particular, served for the determination of the stereochemistry. The products retain the syn-conformation. Protonation of the sulfur bridges causes pronounced deshielding of the aromatic protons whereas the Dd(13C) imply a limited charge delocalization into the aromatic rings. Monothia[3.2]metacyclophanes were also investigated, see Section 15.5.

Internally substituted dithia[3.3]metacyclophanes were also examined by Moriguchi, Tsuge and their coworkers. They treated the conformational preferences of 9nitro-2,11-dithia[3.3]metacyclophane (71) and its reduction product 72 [51, 52]. The authors concluded from the chemical shift of H-18, d = 7.46, that 71 prefers the syn-conformation. The remarkable shielding of H-18 in 72, d = 4.93, however, indicates that this proton is affected by the ring current of the opposite ring and that 72 “is not the syn-form” [51].

The same group prepared the three dithiafluorenophanes 73–75 which are 1,8bridged by different polynuclear aromatic units [44]. The macrocyclic rings in 73 and 74 are large enough to permit fast flipping of the aromatic components on the NMR time scale at room temperature. This is consistent with a singlet for the 9-protons and two singlets for the methylene bridges. VT 1H-NMR measurements gave a DGz of 43 kJ mol–1 at –30 8C for the ring flipping in 74. The same DGz was determined for 73, but at a lower coalescence temperature (–50 8C). In contrast, no ring inversion is observed for 75 even at 150 8C because the t-butylated pyrene unit is too large. Consequently, two different chemical shifts are observed for the 9-protons, one of which is strongly shielded (d = –0.48) because of the magnetic anisotropy of the opposite pyrene unit.

15.4 [3.3]Phanes

[3.3]Metacyclophanes that carry C=O instead of S bridges are the field research of Yamato et al. [53]. They determined the syn/anti-conformer ratio in a series of 6-tbutyl-9-methoxy-[3.3]metacyclophane-2,11-diones 76 by 1H-NMR spectroscopy. The different isomers could be distinguished mainly by the chemical shifts of the 9methoxy protons which are significantly shielded in the anti-isomers. For example, d(OMe) = 3.15 for anti-76 b and 3.43 for syn-76 b. Furthermore, the shieldings of the aryl hydrogens and the t-butyl-protons by the adjacent, face-to-face benzene ring give strong evidence for the syn-compounds. Similar arguments are used by Shinmyozu et al. [54]. They deduced the anticonformation of [3.3]metacyclophane-2,11-dione (77) from the strong shielding of the intra-annular proton (d = 5.78). Variable temperature 1H-NMR experiments (down to –90 8C) produced no change in the spectrum and indicated fast bridge wobbling and ring flipping processes. See Section 15.5 for conformational studies of several [3.2]metacyclophanes.

399

400

15 NMR Spectra of Cyclophanes

A more exotic example of a dithia[3.3]phane is the dithiacorannulenophane 78. The endo-protons of the p-phenylene ring of 78 experience a large upfield shift of 5 ppm because they are only 2.3 Å above the plane of the rim carbons [31]. See Section 15.2.3 for a related [n]phane.

15.5

[m.n]Phanes (m > 2, n ³ 2)

This section gives a survey over NMR investigations of [m.n]phanes other than [2.2]- and [3.3]phanes. Like the [3.3]phanes, these compounds are generally more flexible than the [2.2]phanes. Hence, the majority of NMR spectroscopic investigations concentrate on their conformational behavior. For example, Laali et al. performed low temperature protonation studies of the monothia[3.2]cyclophanes 79 which exist in the anti-conformation [50]. Similar results were obtained as for 69 and 70 (see Section 15.4). However, upon protonation of 79 b the 1H chemical shift of the internal methyl group changes from d = 0.83 to 2.49, suggesting that, in contrast to 79 a, the conformation of 79 b is altered from anti to syn.

As in 79, the anti-conformation is also preferred by [3.3]metacyclophane-2,11dione (77) that was mentioned in Section 15.4. In contrast to 77, the ring inversion processes in the [3.2]metacyclophanes 80 and 81 were amenable to VT 1HNMR studies [54]. The energy barriers DGz for the anti-anti'-interconversion were estimated to be 64.9 (62 8C, 80 a), 70.7 (74 8C, 81 a), and ca. 59 kJ mol–1 (34 8C,

15.5 [m.n]Phanes (m > 2, n ³ 2) z

81 b). For 80 b, DG was difficult to determine but was expected to be slightly lower than for 80 a. The lower barriers of the pyridinophanes were explained with the lower steric bulkiness of the nitrogen lone pair in comparison with a hydrogen atom.

In contrast to 80 and 81, the 1,8-naphthylene-bridged [3.2]paracyclophane 82 with a cyclobutane calliper is a rather rigid [3.2]phane. This compound and its derivatives 83–84 were characterized by 1D and 2D 1H- and 13C-NMR techniques [55]. NOE difference spectra assisted to prove the assignment of the aromatic protons of the cyclophane unit and the stereochemistry of the alkyl bridge. They also clearly reflect the parallel nature of the cofacial rings in the more flexible [4.3]phanes 83 a and 83 b. A notable feature in the 1H-NMR of these two compounds is that the shielding of the cofacial protons is diminished due their flexibility (82: d = 6.54, 6.29, 83 a: d = 6.75, 6.62, 83 b: d = 6.75, 6.64). Co4(CO)9 complexation of 82 occurs at the exterior of one of the cofacial rings and leads to a significant upfield shift of the p-complexed ring and also to a shielding of the opposite ring in 84.

A cyclobutane bridge is also present in phenanthrenophane 85 which was obtained as a mixture of the isomers syn- and anti-85 [56]. They could be readily distinguished by their proton NMR spectra. The higher symmetry of syn-85 results in only eight different chemical shifts in the aromatic region. NOE interactions between the methylene protons and H-1 of the phenanthrene unit reveal the stereochemistry of the cyclobutane ring as depicted.

401

402

15 NMR Spectra of Cyclophanes

Iwamoto et al. carried out the conformational analysis of a metacyclophane with even longer bridges, namely 1,4,11,14-tetraoxa[4.4]metacyclophane (86) [57]. 1HNMR spectra taken at –90 8C showed two singlets in a ratio of 1 : 1 for the internal protons. Magnetization transfer experiments disclosed that these signals belonged to two different conformers. Three plausible conformers were generated by molecular modelling and their chemical shifts estimated using 1,3-dimethoxybenzene as a model compound. Optimum agreement between calculated and observed chemical shifts was achieved for the isomers with C2v and Ci symmetry. Hence, these structures play an important role in the conformational equilibrium of 86.

The spontaneous formation of a [4.4]paracyclophane was reported by Shull et al. [58]. The simple 1H-NMR spectrum of L-p-boronophenylalanine 87 in D2O changes to a rather complex one in DMSO-d6 because the compound tends to aggregate in this solvent. At high concentrations (> 90 mM) oligomeric structures are formed whereas at low concentrations (< 50 mM) a paracyclophane dimer 88 dominates in which one molecule 87 chelates head-to-tail with a second molecule. The complexity of the 1H-NMR spectra indicates that there is a second stereocenter present which can only be the boron atom. The proton signals are significantly shifted to higher field compared with the monomer and four different chemical shifts are observed for the aromatic ring. These findings are in agreement with the proposed cyclophane structure. The signals coalesce pairwise at 141 8C corresponding to a DGz of 86.2 kJ mol–1. The authors suppose that the temperature-dependent changes of the spectra are not induced by phenyl ring rotation but by a more complex dynamic process.

15.5 [m.n]Phanes (m > 2, n ³ 2)

Two dynamic conformational processes have to be considered for acceptor-porphyrin cyclophanes such as 89, a complete rotation of the acceptor unit and a pendular motion of the whole acceptor-containing bridge from C-5 to C-15 leading to two equivalent unsymmetrical conformations [59]. The latter is observed in 89 a and is monitored by the broadening of the 1H-NMR signal of the methine protons of the porphyrin (180–210 K). This dynamic behavior is not shown by 89 b due to its more pronounced rigidity that is caused by the shorter bridges. A comparison of the 1H-NMR spectra also reflects the different conformational behavior of 89 a and 89 b. The aromatic proton of the pyromellitic part absorbs at d = 5.01 in 89 b and d = 6.85 in 89 a, however they resonate at d = 8.09 and 8.22 in the open chain precursors 90 a and 90 b, respectively. So in both cases they are shielded considerably by the aromatic ring current of the porphyrin. The proton in 89 a is less affected than in 89 b, presumably not only because the bridges and the distance to the porphyrin are shorter but because of different conformations of 89 a and 89 b.

403

404

15 NMR Spectra of Cyclophanes

Gonzalez et al. observed restricted rotation about the biaryl ether bonds in 91 a and 91 c (four different chemical shifts for the para-substituted ring), but ring rotation is fast in 91 b (two chemical shifts for the para-substituted ring) [60]. According to these authors, the larger bond angle of a sp2-carbon atom compared to a sp3-carbon, induces less strain in the macrocycle of 91 b, hence allowing rotation of the p-phenylene ring. The hindered rotation in 91 c, despite the presence of the carbonyl group, is explained by the bulkiness of the glycosyl substituent. Atropisomerism is also found for 92 in which the conformers 92 a and 92 b differ from each other by the relative arrangement of the nitro group [61]. The exact stereochemistry of these compounds was determined by NOE studies (NOESY cross peaks between H-12 and H-18 for 92 a and between H-15 and H-18 for 92 b).

Stereochemical questions were also of interest in a series of papers by Yamato et al. who investigated numerous [n.2]metacyclophane-enes and –ynes, 93, with varying lengths of the aliphatic bridge and different substitution patterns of the aromatic rings and the olefinic bridge [62–65]. Some of the enes were subjected to addition reactions. The variety of compounds is too large to be presented here in detail. Some [n.1]metacyclophanes, 94, were also synthesized. The authors studied the conformation of the compounds by 1H-NMR spectroscopy. Syn- and anti-isomers were readily distinguished by chemical shift arguments. In the anti-conformers the internal proton or an internal methoxy group exhibits strong shielding because it is situated in the shielding region of the second aromatic ring. In the syn-conformer, however, the other aromatic protons are slightly shielded in comparison with the anticonformers due to the ring current of the cofacial ring. The same holds true for a t-butyl substituent in the 5-position. VT proton NMR investigations served to probe the conformational stabilities and the ring flipping processes.

15.5 [m.n]Phanes (m > 2, n ³ 2)

Molecules with longer bridges were described by Bauer et al. They investigated a series of oligooxa[8.8]-, -[11.11]-, -[14.14]-, -[17.17]-, and -[20.20]phenothiazine bipyridinium cyclophanes 95 as well as the corresponding permethylene[3.3]- and -[4.4]phanes 96 [66]. Because the phenothiazine system is unsymmetrical, the oligooxacyclophane bipyridinium protons exhibit four chemical shifts, two for the outer and two for the inner protons of the bipyridinium unit. Upon shortening of the bridges from 95 e to 95 b, the bipyridinium hydrogens are increasingly shielded compared with dimethylbipyridinium bis(tetrafluoroborate) and transannular NOEs were observed for 95 c–e. The authors concluded that these compounds have a growing tendency to adopt a conformation in which the phenothiazine and the bipyridinium units are arranged perpendicularly. 95 a is not quite in line with the other compounds because of a different but unknown conformation. NOE experiments also show that in the [3.3]- and [4.4]cyclophanes 96 with even shorter bridges the bipyridinium and the phenothiazine decks are not arranged with their longitudinal axes parallel as might be expected, but are slightly distorted. Ring rotation is possible at room temperature in 96 b, but not in 96 a.

405

406

15 NMR Spectra of Cyclophanes

15.6

[mn]Phanes

This section deals with cyclophanes of the type (-arylene-alkylene-)n, i. e. [m.m.m. . . .]phanes, abbreviated [mn]phanes, not to be confused with the multiply bridged [mn]phanes (see Section 15.7). The most common group of [mn]phanes are the [1.1.1.1]metacyclophanes or [14]phanes or [4]calixarenes and their homologs. As stated in the introduction these are not covered here. The [34]allenophane 97 described by Krause and coworkers [67] can exist as four diastereomers of different topologies, RRRR, RRRS, RRSS or RSRS, because each allene unit can have the R- or the S-configuration. Although the synthetic precursors did show this phenomenon and, correspondingly, showed very complex NMR spectra, product 97 had only a single set of 1H- and 13C-NMR signals. The addition of Eu(fod)3/Ag(fod) shift reagent did not induce any chemical shift differences either.

The 1H-NMR spectrum of the macrocyclic bis(bipyridino)phane 98 showed an interesting temperature behavior [68]. While there was little effect on the chemical shifts of the o-phenylene protons, the protons of the 2,2'-bipyridine units, most

15.6 [mn]Phanes

notably H-3, suffered increasing deshielding as the temperature was raised from 0 to 120 8C. Molecular modelling calculations indicated that in the minimum energy conformation of 98 the molecule is highly twisted with both bipyridine units in very close proximity. The magnetic anisotropy of one bipyridine unit therefore causes larger shieldings than normal of H-3 and the other protons of the opposite bipyridine unit. With increasing temperature the molecule is untwisted and the resulting larger average distance between the bipyridine moieties induces deshielding of their protons. Logically then, such temperature effects were not observed in the precursor molecule of 98, in which the two butadiyne C–C bonds do not yet exist.

Rabinovitz and coworkers investigated the 1H and 13C chemical shift changes that occur when cyclophanes 99–102 are reduced to their anions [69]. 99 is reduced to an aromatic dianion with 26 p-electrons in the periphery, displaying a diamagnetic ring current (d of the inner/outer protons: ca. –7/+9.5 ppm) and to an antiaromatic tetra-anion (28 p) with a paramagnetic ring current (d inside/outside: ca. +16/+3 and +1 ppm). The corresponding 1H shifts for 1002– (34 p) are: d inside/ outside = ca. –4 and –6/+10 ppm and for 1004– (36 p): d inside/outside = ca. 18/0 and +2 ppm. The even larger ring system 101 could be reduced to a hexa-anion that had similar shifts to the dianion for the outside protons (ca. 9 ppm) but more highly shielded inner protons (–6 ppm) than the dianion (–2 ppm). The general finding of this work is that the limiting number of p-electrons for aromatic/antiaromatic behavior of annulenes is far higher than the accepted number of 26.

407

408

15 NMR Spectra of Cyclophanes

Compound 103 is a photochemically produced dimer of the precursor [23](2,4)pyrenophanetriene [70]. It is itself a phane in the plane of the paper and in the direction perpendicular to it. It possesses some interesting shifts of the protons Ha (inner aromatic proton, d = 10.56), Hb (inner cyclobutane proton, d = 7.55(!)) and Hc (outer cyclobutane proton, d = 5.88) which may be attributed to deshielding ring current and steric effects.

15.7 Other Phanes

15.7

Other Phanes

This section gives a survey of NMR spectroscopic studies of cyclophanes that do not fit into the aforementioned sections. For example, Yasutake and collaborators analyzed the conformational properties of hexadeuterio-5,7,9-trimethyl[3.3.3](1,3,5)cyclophane (104 a) [71]. At –90 8C only one isomer is frozen out. It possesses C3-symmetry with all CD2 groups pointing into the same direction and gives rise to two pairs of doublets for the two different kinds of benzylic methylene groups. The energy barrier DGz(–48 8C) for the bridge flipping process amounts to 41.4 ± 0.8 or 43.5 ± 0.8 kJ mol–1 (slightly different values from the two different coalescence processes). In contrast, the related trione 104 b shows no temperature-dependence of its spectra down to –90 8C, suggesting lower energy barriers for the CH2C(O)CH2 than for the CH2CH2CH2 bridge flipping process.

The dynamic process not of the bridges but the rotational motion of the benzene rings in the diazabicyclophanes 105–107 was examined by Kunze et al. [72]. The spectra of 105 are invariant to cooling or heating. The temperature-dependent 1H-NMR spectra of 106, however, reveal that both benzene rings prefer a “face conformation” which gives rise to two different signals for the aromatic protons (d = 7.04, 6.87). The coalescence of these signals (d = 6.96) at 48 8C is attributed to fast internal motion of the aromatic rings on the NMR time scale [DGz(48 8C) = 65.3 kJ mol–1]. In the case of 107, two different dynamic processes are detected. Below –82 8C the singlet of the aromatic protons resolves into two separate lines of equal intensity. This observation is rationalized by a predominant or frozen “lateral” conformation in which the one set of aromatic protons is shielded by the ring current of the neighboring rings. A value of 38.9 kJ mol–1 is given for DGz(–82 8C) of the ring rotation. Furthermore, the individual protons of the methylene groups become chemically nonequivalent below –20 8C which could be explained by a frozen torsional motion of the entire bridge.

409

410

15 NMR Spectra of Cyclophanes

As for other phanes, a large number of papers deal with metallocenophanes, in particular with ferrocenophanes (see also Chapter 6). A large variety of elements were incorporated in the bridge such as silicon, phosphorus, tin, nitrogen and, of course, carbon. For example, the 1H-NMR spectra of the phenyl-substituted ferrocenophanes 108 and 109 exhibit remarkably small chemical shifts for the cyclopentadienyl protons cofacial to the junction between the 4-nitrophenyl group and the other cyclopentadienyl ring: d = 3.40 (108) and 3.19 (109) [73]. For a reason, the authors assume that these protons reside in the shielding region of the nitrophenyl ring.

Several articles deal with shorter bridged ferrocenophanes in which phosphorus is used as a bridging atom. Manners’ group undertook a multinuclear MR study of the borane adducts 110 b and 110 c of the phosphorus-bridged [1]ferrocenophane 110 a [74]. The appearance of the 1H- and 13C-NMR spectra is similar, the most striking exception being the 1:1:1:1-quartet of the borane hydrogens in the proton spectrum of 110 b. In all 1H-NMR spectra the tetravalent phosphorus gives rise to four resonances for the Cp ligands. The most remarkable feature is the increasing shielding of the ipso-phenyl and ipso-Cp carbons as the phosphorus substituent becomes more electron-withdrawing (110 a: d = 138 and 18.5, 110 b: d = 129 and 16.3, 110 c: d = 121 and &10, respectively). A similar tendency was found for the 31P chemical shifts (110 b: d = 42, 110 c: d = 19). For 110 b and 110 c boron–phosphorus spin–spin coupling could be observed in the 11B- as well as in the 31P-NMR spectra [1J(P,H)&39 Hz in 110 b and &154–157 Hz in 110 c] although they are poorly resolved for 110 b.

The groups of Herberhold and Wrackmeyer were also interested in phospha[1]ferrocenophanes, particularly in 111 and 112 a and its derivatives 112 b–e [75]. These compounds were studied by 1D and 2D 1H-, 13C-, 15N-, 31P- and 195Pt-NMR techniques. Isotope induced chemical shifts, 1D12/13C(31P) and 1D14/15N(31P) (the latter was obtained with the Hahn echo-extended experiment) were determined and were proposed as a diagnostic tool for extreme bonding situations as in [1]ferrocenophanes. 1J(195Pt,31P) coupling constants and 195Pt chemical shifts were

15.7 Other Phanes

crucial for the identification of the complexes 112 d and 112 e which contain twoand three-coordinate Pt(0).

Compound 111 and a series of structurally related silicon-bridged [1]ferrocenophanes 113 and 114 a–e were also investigated by 57Fe-NMR spectroscopy because the 57Fe-NMR data should, in principle, serve as valuable diagnostic tools for ferrocenophanes [76, 77]. Generally, the 57Fe nuclei in 114 a–114 e are shielded compared with the corresponding open-chain ferrocenes. Substitution of methyl groups at silicon by chloro ligands is accompanied by increased shielding which was attributed to a distortion of the sandwich structure. However, the increase is more pronounced in the ferrocenophanes (for example, d(57Fe) = 1519.8, 1462.2 and 1394.0 in 114 a, 114 c and 114 e, respectively) than in open-chain compounds (Dd < 20 ppm). Methyl and phenyl groups have comparable influences on the 57Fe chemicals shifts. The comparison of the 57Fe chemical shifts of the disila[2]- and sila[1]ferrocenophanes 113 and 114 e reveals a smaller shielding for the less strained [2]ferrocenophane. The iron nucleus is even less shielded than in ferrocene itself (d(57Fe) = 1884.7, 1519.8 and 1541.7 in 113, 114 a and ferrocene, respectively). The two-bond spin–spin coupling constants between iron and silicon and between iron and phosporus were determined for the first time: 2 57 J( Fe,29Si) = 2.7 Hz in 114 c and 2J(57Fe,31P) = 3.4 Hz in 111.

Herberhold’s and Wrackmeyer’s groups also studied [3]ferrocenophanes 115 a–d, 116, and 117 a–c by multinuclear magnetic resonance (1H, 13C, 31P, 119Sn) [78]. The data indicate flexible structures for 115 a–c but not for 115 d and 116 in which dynamic processes are slow at ambient temperature and at –10 8C respectively (DGz(–10 8C) = 67.7 kJ mol–1). The low activation barriers speak against inversion of the pyramidal phosphorus as the reason for signal averaging. Rather, rotation

411

412

15 NMR Spectra of Cyclophanes

of the Cp ligands in their plane relative to the other Cp ring is assumed. In 117 a–c dynamic processes are slow at room temperature. In order to explore the influence of the ferrocene structure on the NMR parameters of the heavy nuclei 31P and 119Sn the authors compared the 31P and 119Sn chemical shifts as well as the coupling constants in which these nuclei are involved with appropriate open-chain compounds. They therefore also carried out sign determinations of these coupling constants, for example of 1J(119Sn,31P), by 2D heteronuclear shift correlations. The NMR data show that there is no particular strain in the aforementioned structures and that the bond angles at phosphorus are comparable to those of non-cyclic compounds.

15.8

Conclusion

As shown in the preceding sections, cyclophanes continue to be an exciting and often challenging object of NMR investigations. Today, the combination of higher magnetic field strengths and the multitude of available 1D and 2D NMR experiments allow many detailed assignments and analyses of spectra, yielding in-depth information that was unthought of 25 years ago. Unfortunately, however, many authors shy away from the effort of extracting the full information content of their spectra. This is highly deplorable since NMR spectra of cyclophanes do reveal many structural, in particular stereochemical, details as indicated at the beginning of this chapter. Numerous low-quality papers have therefore been omitted from this review. Nevertheless, the variability of cyclophane structures will provide a never-dwindling supply of interesting molecules for NMR studies in the years to come. 15.9

References B. H. Smith, Bridged Aromatic Compounds, Academic Press, New York, 1964. 2 P. M. Keehn, S. M. Rosenfeld (eds.), Cyclophanes, Vols. 1 and 2, Academic Press, New York, 1983. 3 R. H. Mitchell, in Cyclophanes, P. M. Keehn, S. M. Rosenfeld (eds.), Academic Press, New York, 1983, 1, 239– 310. 1

S. M. Rosenfeld, K. A. Choe, in Cyclophanes, P. M. Keehn, S. M. Rosenfeld (eds.), Academic Press, New York, 1983, 1, 311–357. 5 W. W. Paudler, M. D. Bezoari, in Cyclophanes, P. M. Keehn, S. M. Rosenfeld (eds.), Academic Press, New York, 1983, 2, 359–441. 4

15.9 References 6

7

8 9

10

11

12

13

14 15

16 17 18

19 20 21 22

23

S. Itô, Y. Fujise, Y. Fukazawa, in Cyclophanes, P. M. Keehn, S. M. Rosenfeld (eds.), Academic Press, New York, 1983, 2, 485–520. S. Misumi, in Cyclophanes, P. M. Keehn, S. M. Rosenfeld (eds.), Academic Press, New York, 1983, 2, 573–628. L. Ernst, Prog. Nucl. Magn. Reson. Spectrosc. 2000, 37, 47–190. L. A. M. Turkenburg, W. H. de Wolf, F. Bickelhaupt, W. P. Cofino, K. Lammertsma, Tetrahedron Lett. 1983, 24, 1821–1824. L. W. Jenneskens, F. J. J. de Kanter, L. A. M. Turkenburg, H. J. R. de Boer, W. H. de Wolf, F. Bickelhaupt, Tetrahedron, 1984, 40, 4401–4413. D. S. van Es, M. J. van Eis, S. N’Krumah, N. Gret, A. Egberts, M. de Rijke, F. J. J. de Kanter, W. H. de Wolf, F. Bickelhaupt, A. L. Spek, Heterocycles, 2001, 54, 799–824. D. S. van Es, N. Gret, M. de Rijke, M. J. van Eis, F. J. J. de Kanter, W. H. de Wolf, F. Bickelhaupt, S. Menzer, A. L. Spek, Tetrahedron 2001, 57, 3557–3565. N. H. Martin, N. W. Allen III, J. C. Moore, J. Mol. Graphics Mod. 2000, 18, 242–246. V. Rüdiger, H.-J. Schneider, Magn. Reson. Chem. 2000, 38, 85–89. G. W. Wijsman, F. J. J. de Kanter, W. H. de Wolf, F. Bickelhaupt Eur. J. Org. Chem. 2001, 2743–2748. H. Yamamoto, H. Takeda, M. Nitta, Heterocycles 2000, 53, 1891–1903. N. Kanomata, M. Nitta, J. Chem. Soc., Perkin Trans. 1 1990, 1119–1126. H. Miyabara, T. Takayasu, M. Nitta, J. Chem. Soc., Perkin Trans. 1 1999, 3199– 3205. H. Yamamoto, M. Nitta, Heterocycles 2000, 53, 27–35. S. K. Collins, G. P. A. Yap, A. G. Fallis, Org. Lett. 2002, 4, 11–14. Y. Tobe, J. Kishi, I. Ohki, M. Sonoda, J. Org. Chem. 2003, 68, 3330–3332. D. N. Butler, M. Shang, D. Mann, M. R. Johnston, ARKIVOC 2001, 2(12), 27–34. R. J. Abraham, M. Canton, M. Reid, L. Griffiths J. Chem. Soc., Perkin Trans. 2 2000, 803–812.

24

25

26 27

28

29

30

31 32

33

34 35 36

37

38 39

40

41

M. I. Burguete, B. Escuder, E. GarcíaEspaña, S. V. Luis, J. F. Miravet, Tetrahedron 2002, 58, 2839–2846. H. Yamada, S. Yonehara, S. Tanaka, F. Muro, A. Watanabe, K. Nishikawa, J. Am. Chem. Soc. 2001, 123, 279–284. F. W. Grimm, K. Hafner, Tetrahedron 2000, 56, 4239–4247. G. J. Bodwell, J. J. Fleming, M. R. Mannion, D. O. Miller, J. Org. Chem. 2000, 65, 5360–5370. I. Aprahamian, G. J. Bodwell, J. J. Fleming, G. P. Manning, M. R. Mannion, T. Sheradsky, R. J. Vermeij, M. Rabinovitz, J. Am. Chem. Soc. 2003, 125, 1720– 1721. G. J. Bodwell, J. N. Bridson, M. K. Cyran´ski, J. W. J. Kennedy, T. M. Krygowski, M. R. Mannion, D. O. Miller, J. Org. Chem. 2003, 68, 2089–2098. G. J. Bodwell, J. N. Bridson, S.-L. Chen, R. A. Poirier, J. Am. Chem. Soc. 2001, 123, 4704–4708. T. J. Seiders, K. K. Baldridge, J. S. Siegel, Tetrahedron 2001, 57, 3737–3742. M. A. Battiste, J.-X. Duan, Y.-A. Zhai, I. Ghiviriga, K. A. Abboud, W. R. Dolbier Jr., J. Org. Chem. 2003, 68, 3078–3083. A. J. Roche, J.-X. Duan, W. R. Dolbier Jr., K. A. Abboud, J. Org. Chem. 2001, 66, 7055–7058. A. J. Roche, W. R. Dolbier Jr., J. Org. Chem. 2000, 65, 5282–5290. L. Ernst, Liebigs Ann. 1995, 13–17. Correction: Liebigs Ann. 1996, 153. K. El Shaieb, V. Narayanan, H. Hopf, I. Dix, A. Fischer, P. G. Jones, L. Ernst, K. Ibrom, Eur. J. Org. Chem. 2003, 567– 577. S. Sankararaman, H. Hopf, I. Dix, P. G. Jones, Eur. J. Org. Chem. 2000, 2703–2709. A. A. Aly, H. Hopf, L. Ernst, Eur. J. Org. Chem. 2000, 3021–3029. L. Minuti, A. Taticchi, A. Marrocchi, L. Costantini, E. Gacs-Baitz, Tetrahedron: Asymmetry 2001, 12, 1179–1183. A. Taticchi, L. Minuti, A. Marrocchi, D. Lanari, E. Gacs-Baitz, Tetrahedron: Asymmetry 2002, 13, 1331–1335. L. Minuti, A. Taticchi, A. Marrocchi, D. Lanari, A. Broggi, E. Gracs-Baitz, Tetrahedron: Asymmetry 2003, 14, 481–487.

413

414

15 NMR Spectra of Cyclophanes 42

43 44

45

46

47 48 49 50

51 52

53

54

55

56

57 58

59

A. Marrocchi, L. Minuti, A. Taticchi, I. Dix, H. Hopf, E. Gracs-Baitz, P. G. Jones, Eur. J. Org. Chem. 2001, 4259– 4268. G. J. Bodwell, D. O. Miller, R. J. Vermeij, Org. Lett. 2001, 3, 2093–2096. A. Tsuge, R. Nada, T. Moriguchi, K. Sakata, J. Org. Chem. 2001, 66, 9023– 9025. K. Yano, M. Osatani, K. Tani, T. Adachi, K. Yamamoto, H. Matsubara, Bull. Chem. Soc. Jpn. 2000, 73, 185–189. K. Tani, K. Matsumura, K. Hori, Y. Tohda, H. Takemura, H. Ohkita, S. Ito, M. Yamamoto, Chem. Lett. 2002, 934–935. R. H. Mitchell, J. Am. Chem. Soc. 2002, 124, 2352–2357. T. Satou, T. Shinmyozu, J. Chem. Soc., Perkin Trans. 2 2002, 393–397. G. J. Bodwell, J. Li, Org. Lett. 2002, 4, 127–130. K. K. Laali, T. Okazaki, R. H. Mitchell, J. Chem. Soc., Perkin Trans. 2 2001, 745– 748. T. Moriguchi, K. Sakata, A. Tsuge, Chem. Lett. 1999, 167–168. T. Moriguchi, K. Sakata, A. Tsuge, J. Chem. Soc., Perkin Trans. 2 2001, 934– 938. T. Yamato, K. Tsuchihashi, N. Nakamura, M. Hirahara, K. Tanaka, Can. J. Chem. 2002, 80, 510–516. H. Isaji, M. Yasutake, H. Takemura, K. Sako, H. Tatemitsu, T. Inazu, T. Shinmyozu, Eur. J. Org. Chem. 2001, 2487– 2499. K. K. Laali, G. I. Borodkin, T. Okazaki, Y. Hayashida, Y. Nakamura, J. Nishimura, J. Chem. Soc., Perkin Trans. 2 2000, 2347–2350. Y. Nakamura, T. Fujii, S. Sugita, J. Nishimura, Chem. Lett. 1999, 1039– 1040. H. Iwamoto, Y. Yang, S. Usui, Y. Fukazawa, Tetrahedron Lett. 2001, 42, 49–51. B. K. Shull, D. E. Spielvogel, R. Gopalaswamy, S. Sankar, P. D. Boyle, G. Head, K. Devito, J. Chem. Soc., Perkin Trans. 2 2000, 557–561. H. A. Staab, S. Nikolic, C. Krieger, Eur. J. Org. Chem. 1999, 1459–1470.

60 61 62

63

64 65

66

67

68 69

70 71

72 73

74

75

76

77

78

G. I. Gonzalez, J. Zhu, J. Org. Chem. 1999, 64, 914–924. A. Bigot, M. E. T. H. Dau, J. Zhu, J. Org. Chem. 1999, 64, 6283–6296. T. Yamato, K. Fujita, K. Okuyama, H. Tsuzuki, New J. Chem. 2000, 24, 221– 228. T. Yamato, K. Fujita, K. Futatsuki, H. Tsuzuki, Can. J. Chem. 2000, 78, 1089– 1099. T. Yamato, K. Fujita, T. Abe, H. Tsuzuki, New J. Chem. 2001, 25, 728–736. T. Yamato, K. Fujita, H. Tsuzuki, J. Chem. Soc., Perkin Trans. 1 2001, 2089– 2097. H. Bauer, F. Stier, C. Petry, A. Knorr, C. Stadler, H. A. Staab, Eur. J. Org. Chem. 2001, 3255–3278. S. Thorand, F. Vögtle, N. Krause, Angew. Chem. 1999, 111, 3929–3931; Angew. Chem. Int. Ed. 1999, 38, 3721–3723. P. N. W. Baxter, J. Org. Chem. 2001, 66, 4170–4179. E. Shabtai, O. Segev, R. Beust, M. Rabinovitz, J. Chem. Soc., Perkin Trans. 2 2000, 1233–1241. C. Schnorpfeil, H. Meier, M. Irie, Helv. Chim. Acta 2001, 84, 2467–2475. M. Yasutake, K. Araki, M. Zhou, R. Nogita, T. Shinmyozu, Eur. J. Org. Chem. 2003, 1343–1351. A. Kunze, S. Bethke, R. Gleiter, F. Rominger, Org. Lett. 2000, 2, 609–612. O. Kadkin, C. Näther, W. Friedrichsen, J. Organomet. Chem. 2002, 649, 161– 172. C. E. B. Evans, A. J. Lough, H. Grondey, I. Manners, New J. Chem. 2000, 24, 447– 453. M. Herberhold, F. Hertel, W. Milius, B. Wrackmeyer, J. Organomet. Chem. 1999, 582, 352–357. B. Wrackmeyer, O. L. Tok, A. Ayazi, F. Hertel, M. Herberhold, Z. Naturforsch. 2002, 57 b, 305–308. B. Wrackmeyer, A. Ayazi, H. E. Maisel, M. Herberhold, J. Organomet. Chem. 2001, 630, 263–265. M. Herberhold, U. Steffl, B. Wrackmeyer, Z. Naturforsch. 1999, 54 b, 57–62.

415

16

Strained Heteroatom-Bridged Metallocenophanes Ian Manners and Ulf Vogel

16.1

Introduction

Strained organic molecules have fascinated chemists for over a century. Much less is known about strained molecules that contain elements other than carbon, especially transition metals. The creation of strained rings (e.g. 1–4) based on a metallocene nucleus – species that can be considered under the broad umbrella of metallocyclophanes – was first achieved in the 1960s when [2]ferrocenophanes containing a two-atom hydrocarbon bridge were successfully prepared by Rinehart et al. and later by Lentzner and Watts [1, 2]. Despite the fact that previous workers had regarded the possibility of connecting the two cyclopentadienyl ligands of ferrocene by a single atom bridge as being impossible due to the strain that would be generated, Osborne successfully isolated the first [1]ferrocenophane 2 with a silicon bridge in 1975 [3]. Analogous species with a single germanium (3) and a phosphorus bridge (4) were reported by Osborne and by Seyferth in 1980 [4, 5].

Strained species such as 1 and 2 possess high energy, ring-tilted structures, unlike ferrocene itself in which the cyclopentadienyl ligands are approximately parallel. The discovery, in our laboratories in the early 1990s, that [1]- and [2]ferrocenophanes such as 2 and 1 undergo ring-opening polymerization (ROP) reactions to yield high molecular weight, processable polymers with interesting properties provided additional impetus for the broad development of this area [6, 7]. Strained [1]ferrocenophanes with a variety of bridges including sulfur, selenium, tin, and boron have been reported and the range of [2]ferrocenophanes has been similarly expanded. Novel [1]ferrocenophanes with zirconium in the bridge have also been prepared, and attempts to generalize the range of metallocenophanes to ruthenocenophanes, cobaltocenophanes and heteroatom-bridged bis(benzene)chromium complexes have been described (see also chapter 6 on Endohedral Metalcomplexes of Cyclophanes). Metallocenophanes based on titanocene and zirconocene are also Modern Cyclophane Chemistry. Edited by Rolf Gleiter, Henning Hopf Copyright © 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 3-527-30713-3

416

16 Strained Heteroatom-Bridged Metallocenophanes

well known but these possess an inherently tilted structure due to the tetrahedral metal center and attempts to induce strain have been unsuccessful, an observation that has been rationalized theoretically. Ferrocenophanes and ruthenocenophanes with more than three atoms in the bridge also have negligible strain. In the following sections of this chapter the synthesis, properties, and reactivity of strained metallocenophanes is surveyed. Some of the salient features of the ring-opened polymers are also outlined.

16.2

Synthesis and Structure 16.2.1

[1]Ferrocenophanes

For the synthesis of strained [1]ferrocenophanes 5 two different routes have been used. The first involves the reaction of a heteroelementdichloride with dilithioferrocene, and in the second, the ligand system is synthesized and then metalated (to give 6) and reacted with FeCl2. Since dilithioferrocene can be obtained as a N,N,N',N'-tetramethylethylenediamine (TMEDA) adduct on a large scale, the first route is the more convenient and versatile and is therefore more widely used.

The most striking structural feature of [1]ferrocenophanes, induced by the presence of the heteroatom bridge, is the tilting of the two Cp-rings out of the parallel arrangement found in ferrocene and unstrained ferrocenophane derivatives. The structural distortion can be characterized by the angles a, b, d and h and the distance of the heteroatom to the iron atom (Fig. 16.1). The most important value is usually the tilt-angle a which describes the angle between the two planes of the Cp-rings (Tab. 16.1). The tilt-angle a is influenced not only by the size of the bridging atom, but also by electronic effects. Substitution of electron donating groups on the Cp-rings lowers the tilt-angle significantly in the case of silicon-bridged [1]ferrocenophanes. For all known structurally characterized [1]ferrocenophanes to date, the distance between the heteroatom and the iron atom is larger than the sum of the respective Van der Waals radii, but in some cases a small dative bond contribution from the heteroatom to the iron atom may exist.

16.2 Synthesis and Structure

Fig. 16.1 Important angles used for comparing strained ferrocenophanes

Tab. 16.1 Comparison of tilt-angles in some heteroatom-bridged [1]ferrocenophanes

Bridging group

a (8)

b (8)

Ref.

BN(SiMe3)2 S PPh Se SiMe2 SiPh2 GeMe2 GePh2 Zr(tBuC5H4)2

32.4(4) 31.05(10) 26.9 26.4(2) 20.8(5) 19.2 19.0(9) 16.6 6.0

33.7(2)/34.0(2) 29.1(1)/28.8(1) 32.3 30.0(2) 37.0(6) 40.0 37.7(5)/35.9(5) 38.0 40.1

8 9 10 9 11 12 13 14 15

Group 13 Bridged [1]Ferrocenophanes The only examples of group 13 bridged [1]ferrocenophanes that have been synthesized to date possess a boron atom in the bridge. These species are also the only [1]ferrocenophanes currently known with a first row element as a bridging atom. All the three known compounds [Fe(g5-C5H4)2BR] (R = N(SiMe3)2, NiPr2, N(SiMe3)tBu) feature an amido-ligand on the boron atom [8, 16]. They have been synthesized using fcLi2 · TMEDA and Cl2BR. The tilt-angle a between the two planes of the Cp-rings in [Fe(g5-C5H4)2BN(SiMe3)2] (32.4(4)8) is the highest found for any [1]ferrocenophane (Fig. 16.2). 16.2.1.1

Group 14 Bridged [1]Ferrocenophanes After being synthesized first by Osborne [3] the vast majority of research carried out on strained [1]ferrocenophanes has been directed at silicon-bridged derivatives [Fe(g5-C5H4)2SiR2]. These compounds are usually synthesized by reaction of the corresponding diorganylsilicondichloride R2SiCl2 with fcLi2 · TMEDA. A large number of substituents on the silicon atom can be accessed by this direct meth16.2.1.2

417

418

16 Strained Heteroatom-Bridged Metallocenophanes

Fig. 16.2 Structure of [Fe(g5-C5H4)2BN(SiMe3)2] in the solid state

od. Those that cannot be synthesized by this method can often be introduced by the reaction of the appropriate nucleophilic reagent and [Fe(g5-C5H4)2SiCl2] or [Fe(g5-C5H4)2SiMeCl] (7), respectively [17]. An example is provided by the preparation of 8.

These silicon-bridged [1]ferrocenophanes exhibit structures with typical ring tiltangles around 20 degrees. A systematic study of the structure of silicon-bridged [1]ferrocenophanes with increasing methylation of the Cp-rings shows that the tilt-angle gets smaller with increasing methylation. This effect was attributed to the strengthening of the iron–Cp bond with methylation. With stronger Fe-ring bonding, deviation from a parallel arrangement leads to a greater energy penalty. Therefore increasingly methylated ferrocenophanes generally adopt less tilted structures [18]. Comprehensive theoretical and experimental studies on the change of the electronic structure with increasing tilt-angle have also been carried out [19]. Higher congeners of silicon can also act as the bridging atoms. For example the germanium-bridged [1]ferrocenophanes [Fe(g5-C5H4)2GeR2] (R = Me, Et, Ph) have been synthesized [4, 13]. Tin-bridged ferrocenophanes require more sterically demanding substituents on the tin atom to enable their isolation [20, 21]. They could be isolated with mesityl, t-butyl or even larger substituents, but on the at-

16.2 Synthesis and Structure 5

tempted synthesis of [Fe(g -C5H4)2SnR2] (R = Me, Et, Ph) only oligomeric material was obtained. Unlike other [1]ferrocenophanes the tin-bridged species show the tendency to polymerize in solution even at room temperature. This behavior has been attributed to trace impurities of TMEDA, which are always present at the synthesis of these compounds [22, 23].

Group 15 Bridged [1]Ferrocenophanes Phosphorus-bridged [1]ferrocenophanes were first reported by Seyferth and Osborne and are readily available via the reaction of the respective organylelementdichloride and fcLi2 · TMEDA [4, 5]. A large number of different substituents at the bridging phosphorus atom have been introduced, even chiral substituents such as (–)-menthyl [24]. The phosphorus atoms in the bridge can be oxidized using elemental sulfur to yield the corresponding phosphine sulfide (10). The lone pair on the phosphorus atom can further be used for coordination to a transition metal fragment like W(CO)5 (to give 9). 16.2.1.3

Arsenic-bridged [1]ferrocenophanes like [Fe(g5-C5H4)2AsPh] are also known and can be synthesized in a similar manner to their phosphorus analogs [5, 25].

Group 16 Bridged [1]Ferrocenophanes For group 16 bridged [1]ferrocenophanes only sulfur and selenium are known as the bridging atoms. They can be synthesized by the reaction of FcLi2 · TMEDA with S(O2SPh)2 and Se(S2CNEt2)2, respectively [9, 26]. Both compounds exhibit a highly strained structure with tilt-angles of 31.05(10) and 26.4(2) degrees, the structure of the sulfur-bridged compound is shown in Fig. 16.3. 16.2.1.4

16.2.2

Other [1]Metallocenophanes

Although most of the work concerning strained [1]metallocenophanes has been focused on ferrocenophanes, a few other systems have also been investigated. The bisbenzenechromium and -vanadium moieties can also be used to synthesize strained [1]metallocenophanes. Examples include the silicon-bridged [1]metalloarenophanes [M(g6-C6H5)2SiR2] (11) (M = Cr, V; R = Ph, Me; R2 = (CH2)2CH2) [27–29].

419

420

16 Strained Heteroatom-Bridged Metallocenophanes

Fig. 16.3 Solid state structure of a sulfur-bridged [1]ferrocenophane

The structure of these compounds is also characteristic of a strained ring system where the two benzene rings are tilted out of a parallel arrangement. The tilt-angles are smaller than the corresponding ferrocene systems, for example [Cr(g6C6H5)2SiMe2] shows a tilt-angle of 16.6(3) degrees [28]. 16.2.3

[2]Ferrocenophanes

Several methods for the synthesis of [2]ferrocenophanes have been developed. Like the [1]ferrocenophanes they are accessible by the reaction of fcLi2 · TMEDA and a corresponding dichloride. Another method involves the synthesis of the dilithiated ligand systems and subsequent reaction with FeCl2. For the carbon-heteroatom-bridged [2]ferrocenophanes another general route was established: pentamethylferrocene [Fe(g5-C5H5)(g5-C5Me5)] (12) can be dilithiated to give 13 which can then be reacted with an appropriate heteroelementdichloride to give [Fe(g5C5H4)(g5-C5Me4)CH2ERn] (14). This route has been applied for ERn = SiMe2, PPh, PMes and S [30].

16.2 Synthesis and Structure

These compounds show tilt-angles between 11.8(1)8 (ERn = SiMe2) and 18.5(1)8 (ERn = S). The solid state structure of [Fe(g5-C5H4)(g5-C5Me4)CH2S] is shown in Fig. 16.4. [2]Ferrocenophanes can also be synthesized by insertion of a Pt(0)-fragment into the Si–C bond of a strained [1]ferrocenophane: The complexes [Fe(g5C5H4)2SiMe2PtLn] (L = COD, n = 1; L = PEt3, n = 2) have been synthesized by reaction of [Fe(g5-C5H4)2SiMe2] with Pt(PEt3)3 or Pt(COD)2, respectively [31–33].

The complex [Fe(g5-C5H4)2SiMe2Pt(COD)] (16) can act as a precatalyst for the polymerization of [Fe(g5-C5H4)2SiMe2] (15). A similar insertion of a platinum(0)-species was shown to occur for phosphorus-bridged [1]ferrocenophanes [34].

Fig. 16.4 Solid state structure of [Fe(g5-C5H4)(g5-C5Me4)CH2S]

421

422

16 Strained Heteroatom-Bridged Metallocenophanes

16.2.4

Other [2]Metallocenophanes

By moving from iron to its higher congener ruthenium the strain in the [2]metallocenophanes can be significantly increased. To date only strained [2]ruthenocenophanes with bridging carbon atoms are known; they were synthesized by the reaction of the dilithiated ligand system with cis-[RuCl2(dmso)4] [35]. Comparison of the tilt-angle in the [2]ruthenocenophane [Ru(g5-C5H4)2(CH2)2] (29.6(5)8) with the corresponding iron compound (21.68) shows an increase of a by ca. 8 degrees.

16.3

Ring-Opening Polymerization of Strained Ferrocenophanes 16.3.1

Stoichiometric Insertion and Ring-Opening Reactions

Ferrocenophanes undergo a variety of stoichiometric ring-opening reactions. Perhaps the most important reaction is the anionic ring-opening using an organyllithium compound. Phosphorus-bridged [1]ferrocenophanes 4 are attacked by the alkyllithium reagent at the heteroatom yielding a ring-opened product with a negatively charged Cp-ring (17). This intermediate can be reacted with an electrophile in an onepot reaction to give an asymmetrically substituted 1,1'-ferrocene such as 18 [5, 25].

For silicon-bridged ferrocenophanes the reaction with alkyllithium leads to a similar reaction. But in some cases the metalation step is not necessary and the ferrocenophane can be directly reacted with a suitable electrophile. Reaction of 15 with R2BCl leads directly to [Fe(g5-C5H4BR2)(g5-C5H4SiMe2Cl)] (19) [36].

Another synthetically important reaction is the ring-opening reaction of siliconbridged [1]ferrocenophanes with HCl leading to ferrocenylchlorodiorganylsilanes [37]. With suitable reagents it is not only possible to observe ring-opening reactions, but also insertion reactions into the heteroatom–carbon bond. As noted above, the Pt(0)complex [Pt(COD)2] (COD = cycloocta-1,5-diene) reacts with [Fe(g5-C5H4)2SiMe2] (15) with insertion into the carbon–silicon bond to give 16 [33].

16.3 Ring-Opening Polymerization of Strained Ferrocenophanes 5

The boron-bridged complex [Fe(g -C5H4)2BNiPr2] (20) reacts with metal(0) carbonyl complexes at 25 8C with cleavage of the Fe–Cp bond [8]. Reaction of this species with Fe2(CO)9 for example yields 21, a boron-bridged derivative of the well-known dimer [Cp(CO)2Fe]2.

The tin-bridged [1]ferrocenophane [Fe(g5-C5H4)2SntBu2] (22), which possesses a much weaker Cp–heteroelement bond, reacts with Fe2(CO)9 to give insertion into the Cp–Sn bond to form the [2]ferrocenophane 23. With Co2(CO)8 a CO insertion into the carbon–tin bond is observed to yield 24 [38].

16.3.2

Ring-Opening Polymerizations (ROP) Thermal ROP As noted above, strained [1]silaferrocenophane monomers 25 are readily available on a substantial (> 100 g) scale from the reaction of fcLi2 · TMEDA with dichloroorganosilanes RR'SiCl2. In 1992 the thermal ring-opening polymerization (ROP) of these monomers was demonstrated and this yielded high molecular weight polyferrocenylsilanes 26 for the first time [6]. 16.3.2.1

Since the initial ROP discovery a wide range of silicon-bridged [1]ferrocenophanes 25, with either symmetrically or unsymmetrically alkyl- or aryl-substituted silicon atoms, have been prepared and polymerized to afford high molecular weight poly-

423

424

16 Strained Heteroatom-Bridged Metallocenophanes

ferrocenylsilanes 26. A large number of other [1]- and [2]ferrocenophanes can also be polymerized thermally. Examples include germanium-, tin-, phosphorus- and sulfur-bridged [1]ferrocenophanes as well as the [2]metallocenophanes [Fe(g5C5H4)2(CH2)2] and [Ru(g5-C5H4)2(CH2)2]. Thermal copolymerization can be used to prepare random copolymers from mixtures of different silicon-bridged [1]ferrocenophanes and also copolymers of ferrocenophanes and other monomers such as cyclic silanes and silicon-bridged bis(benzene)chromium complexes [28, 39].

However, thermal ROP of metallocenophanes at elevated temperatures provides virtually no control over molecular weight, and the molecular weight distribution is broad with polydispersity indices (PDI = Mw/Mn) of around 1.5–2.5.

Living Anionic ROP of Silicon-Bridged [1]Ferrocenophanes In 1994 it was shown that highly purified silicon-bridged [1]ferrocenophanes undergo living anionic ROP at 25 8C using initiators such as nBuLi in THF [40]. This has permitted the synthesis of polyferrocenylsilanes 29 (E = H) with controlled molecular weights and narrow molecular weight distributions (polydispersities < 1.3) and has also allowed the preparation of end-functionalized polymers 29 (E = H) and the first block copolymers such as 30 containing skeletal transition metal atoms (Scheme 16.1) [41]. 16.3.2.2

Scheme 16.1 Controlled architectures by living anionic polymerization of 15

16.3 Ring-Opening Polymerization of Strained Ferrocenophanes

The mechanism is believed to proceed through an initial attack of the organyllithium compound at the bridging atom which leads to a ring-opening reaction and the formation of a ferrocene with a negatively charged Cp-ring (28). This anion can attack another monomer and hence propagate polymerization. Other [1]ferrocenophanes which can be polymerized by anionic initiation include phosphorus- and sulfur-bridged compounds [9, 42].

Transition Metal-Catalyzed ROP of [1]Ferrocenophanes Transition metal-catalyzed ROP of silicon-bridged [1]ferrocenophanes, which occurs in solution at room temperature, has been reported using Pt(II), Pt(0), Rh(I), and Pd(II) precatalysts [43, 44]. Significant understanding of the mechanism of these transition metal-catalyzed reactions has also been forthcoming and colloidal metal appears to be the true catalyst [45]. Transition metal-catalyzed ROP is a mild method which does not require monomer with an exceptional degree of purity and the method has now been developed to the stage where considerable control over polymetallocene architectures is possible [46]. For example, [1]silaferrocenophanes with different cyclopentadienyl rings (31) undergo transition metal-catalyzed ROP to yield the regioregular, crystalline polyferrocenylsilane 33 whereas thermal ROP affords a regio-irregular amorphous material 32 (Scheme 16.2). 16.3.2.3

Scheme 16.2 Synthesis of a regioregular polyferrocenylsilane by transition metal

catalysis

Addition of sources of Si–H bonds such as Et3SiH to the polymerization mixture allows molecular weight control. The process is believed to occur via competition of the Si–H bond and the strained Si–C bond of the [1]ferrocenophane for oxidative-addition at the metal center. In the case of 15, addition of the Si–H bond followed by reductive elimination affords the Et3Si–fc- and Si–H-terminated polymeric product (34).

425

426

16 Strained Heteroatom-Bridged Metallocenophanes

If polysiloxanes with Si–H groups in the main chain or as termini are used in place of Et3SiH then graft or block copolymers can be prepared. Moreover, the use of cyclic tetrasiloxanes as sources of the Si–H functionalities allows the preparation of novel polyferrocenes with star architectures with a cyclosiloxane core. The remarkable convenience by which complex organometallic polymer architectures can be constructed compensates for the broader polydispersities (Mw/Mn = 1.5–1.8) compared with materials prepared via anionic polymerization (Mw/Mn < 1.3).

Other Initiation Methods for ROP In addition to the thermal, anionic and transition metal-catalyzed polymerizations mentioned above, a cationic polymerization mechanism has also been found to work for some systems. For example, the tin-bridged ferrocenophanes [Fe(g5C5H4)2SnR2] (R = Mes, tBu) can be induced to polymerize by the addition of a catalytic amount of electrophilic species like H+ or Bu3Sn+ [22]. The [2]ferrocenophane [Fe(g5-C5H4)(g5-C5Me4)CH2S] can also be polymerized by a cationic mechanism using MeOTf or BF3 · Et2O as initiators. The mechanism for this polymerization is likely to proceed via the methylation of the sulfur atom and generation of a cationic intermediate, which can then be ring-opened by reaction with another monomer molecule [30]. Miyoshi discovered that the transition metal complexes of phosphorus-bridged ferrocenophanes [Fe(g5-C5H4)2PPh(MLn)] (MLn = W(CO)5, MnCp(CO)2, Mn(g5-C5H4Me) (CO)2) can be polymerized photolytically by irradiation with UV light. The mechanism for this reaction is believed to proceed via Fe–Cp bond cleavage [47, 48]. 16.3.2.4

16.4

Properties and Applications of Ring-Opened Polyferrocenes 16.4.1

Polyferrocenylsilanes

The prototypical polyferrocenylsilane, the dimethylderivative 26 (R = R' = Me), is an amber, film-forming thermoplastic (Fig. 16.5) which possesses melt transitions (Tm) in the range of 122–143 8C [49]. This material possesses a Tg at 33 8C, which is remarkably low for a polymer with a bulky unit such as ferrocene in the main chain. The ability of the iron atom in ferrocene to act as a freely rotating “molecular

16.4 Properties and Applications of Ring-Opened Polyferrocenes

Fig. 16.5 Polyferrocenyldimethylsilane as a powder and processed into a shape

ball-bearing” probably plays a key role in generating the observed conformational flexibility. The manner in which these polymer chains pack in the solid state for this polymer has attracted significant interest and has been examined by detailed conformational calculations and X-ray structural studies of well-defined oligomers and polymers by O’Hare, Manners, Foster, and Pannell, respectively [50–53]. Polyferrocenyldimethylsilane can be melt-processed into shapes above 150 8C (Fig. 16.5) and can be used to prepare crystalline, nanoscale fibers (diameter ca. 150 nm) by electrospinning in which an electric potential is used to produce an ejected jet from a solution of the polymer in THF which subsequently stretches, splays, and dries. The nanofibers of different thickness show different colors due to interference effects similar to those in soap bubbles [52].

Like the dimethyl material, other symmetrically substituted polyferrocenylsilanes with short (C2–C5) alkyl chains at silicon will also crystallize and similar melting behavior is observed. In contrast to the polyferrocenylsilanes with shorter C2–C5 alkyl chains, the n-hexyl analog 26 (R = R' = n-hexyl) is an amber, gummy amorphous material with a Tg of –26 8C. Alkoxy-substituted polyferrocenylsilanes such as 35 (R = R' = n-hexyloxy) are also amorphous and possess Tg values down to –51 8C. Polyferrocenylsilanes with longer alkoxy side chains (e.g. 35, R = R' = n-octadecyloxy) have been found to crystallize and form lamellar structures with interdigitated side groups [54].

427

428

16 Strained Heteroatom-Bridged Metallocenophanes

In addition to their morphological behavior, considerable effort has also been directed towards investigating and understanding the properties of the resulting polyferrocenylsilane materials the vast majority of which are soluble in organic solvents despite their very high molecular weights. Cyclic voltammetric studies of polyferrocenylsilanes generally show the presence of two reversible oxidation waves spaced by ca. 0.25 V in a 1 : 1 ratio providing clear evidence for the existence of interactions between the iron atoms [55]. The proposal that initial oxidation occurs at alternating iron sites along the main chain, followed by oxidation of those remaining at a higher potential, has been supported by recent detailed work on individual members of a series of model oligomers, with between 2 and 9 ferrocene units [51]. Comparisons of the behavior of polyferrocenylsilanes with analogous polymers suggest that the metal–metal interactions are mediated by the silicon spacer [55]. Polyferrocenylsilanes also possess interesting hole transport properties and partial oxidation leads to a 1010 increase in electrical conductivity up to semiconductor values (ca. 10–3–10–4 S cm–1) [56]. Thin films of the polymers are attracting attention as, for example, chemomechanical sensors, electrochromic materials, electrode mediators, and materials of variable refractive index [56, 57].

Controlled crosslinking of polyferrocenylsilanes has been achieved to yield solvent swellable, redox-active gels. The crosslinking can be controlled by using specific amounts of the spirocyclic [1]ferrocenophane 36 in the polymerization mixtures [58, 59]. The swelling of the gels can be controlled by the degree of oxidation of the ferrocene centers. This has allowed the preparation of redox-tunable photonic crystals where the position of the Bragg diffraction peak can be controlled. Such structures are of interest for display applications, and have been termed photonic ink (P-ink) [60]. Functionalization of polyferrocenylsilanes has been achieved by several methods. For example, ROP of [1]silaferrocenophane 25 (R = Me, R' = H) with a Si–H functionality affords 37 which undergoes metal-catalyzed hydrosilylation reactions. This methodology has been recently used to prepare thermotropic liquid crystalline polyferrocenylsilanes such as 38 [61].

R = (CH2)xO

16.4 Properties and Applications of Ring-Opened Polyferrocenes

To illustrate another approach, substitution of Si–Cl bonds present in polyferrocenylsilane 39 permits access to water soluble polyferrocenylsilanes such as 41 [62].

Polyferrocenylsilanes function as preceramic polymers and have been shown to yield interesting magnetic ceramic composites at 500 – 1000 8C which contain small Fe particles in a SiC/C matrix [63, 64]. The use of such involatile but processable polymeric precursors to ceramics is potentially an attractive way of circumventing the difficulty of processing ceramic materials into desired shapes. Such structures can be patterned on the micron scale using soft lithography approaches [65, 66]. Nanostructured magnetic ceramic materials can be obtained by pyrolysis of polyferrocenylsilanes within the channels of the mesoporous silica MCM-41 [67]. Polyferrocene block copolymers in which the blocks are immiscible, self-assemble to form phase-separated domains in the solid state or to form micellar aggregates in selective solvents for one of the blocks [68]. These materials are attracting attention for nanopatterning applications where, for instance, nanoscopic ceramic structures can be created on surfaces [69, 70]. For example, cylindrical micelles of polyferrocenyldimethylsilane-b-polydimethylsiloxane (PFS-b-PDMS) are formed in hexane, a selective solvent for PDMS, and consist of a wire-like core of semiconducting PFS surrounded by a sheath or corona of insulating PDMS. Such structures are of interest as semiconducting nanowires. Pyrolysis or plasma etching of such structures offers the possibility of generating magnetic wire-like structures with a silica coating (Fig. 16.6) [71, 72].

Fig. 16.6 Nanowires fabricated using a PFS-b-PDMS block copolymer

429

430

16 Strained Heteroatom-Bridged Metallocenophanes

16.4.2

Other Polymetallocenes via ROP

The ROP route has also been extended to the synthesis of other polymers from [1]ferrocenophane precursors. Polyferrocenylgermanes 42 were first reported in 1993 and possess similar properties to their silicon analogs [13]. Tin-bridged [1]ferrocenophanes also undergo thermal ROP to yield high molecular weight polyferrocenylstannanes 43 [20]. Interestingly, this ROP reaction also occurs in solution at room temperature in the presence of amines [22].

Polyferrocenylphosphines 44 are also accessible via the thermal ROP of phosphorus-bridged [1]ferrocenophanes such as 4 [73]. In addition, the latter species has also been shown to undergo living anionic ROP to yield high molecular weight materials, including block copolymers (e.g. 45) [42, 74]. These results contrast with those of the previous studies in the early 1980s from which it was concluded that ROP of phosphorus-bridged [1]ferrocenophane 4 does not occur and that only the formation of oligomers is possible using ring-opening methods [75].

ROP of chalcogen-bridged [1]ferrocenophanes have also been developed. The sulfur-bridged [1]ferrocenophane 46 (R = H) is a remarkably strained species with a tilt-angle of ca. 31 8C (Fig. 16.3). This compound and the dimethylated analog undergo ROP to yield polyferrocenylsulfides, 47 [9]. The iron–iron interactions in the resulting polymers are significantly greater than in the analogous polyferrocenylsilanes as shown by cyclic voltammetry.

Although disilane-bridged [2]ferrocenophanes (e.g. 48) possess very small ring-tiltangles (ca. 4–58) and have been found to be insufficiently strained to undergo

16.5 References

thermal ROP, hydrocarbon-bridged [2]ferrocenophanes 1 possess strained ringtilted structures (tilt-angles = ca. 218) as a result of the smaller atomic radius of carbon compared with silicon [76]. These species have been found to yield polyferrocenylethylenes 49 via thermal ROP at 250–300 8C. As a consequence of the presence of a more insulating bridge, these polymers show much smaller interactions between the iron atoms (redox coupling ca. 0.09 V) compared with polyferrocenylsilanes. Nevertheless, significant antiferromagnetic interactions are present in TCNE-oxidized samples of these materials [77].

The inclusion of the larger ruthenium atom into a metallocenophane structure gives rise to even more tilted and strained molecules and hydrocarbon-bridged [2]ruthenocenophanes 50 (tilt angles ca. 29–308) undergo thermal ROP to yield polyruthenocenylethylenes 51. These materials exhibit significantly different electrochemistry from their iron analogs [35]. The [2]thiacarboferrocenophane 52, which possesses a strained structure with a tilt-angle of ca. 188, also undergoes cationic ROP with initiators such as MeOSO2CF3 or BF3 · OEt2. The resulting ring-opened polymers 53 are soluble if Mn < 6000 [30].

16.5

References 1

2 3 4

5 6 7 8

K. L. Rinehart, Jr.; A. K. Frerichs, P. A. Kittle, L. F. Westman, D. H. Gustafson, R. L. Pruett, J. Am. Chem. Soc. 1960, 82, 4111. H. L. Lentzner, W. E. Watts, Tetrahedron 1971, 27, 4343. A. G. Osborne, R. H. Whiteley, J. Organomet. Chem. 1975, 101, C27. A. G. Osborne, R. H. Whiteley, R. E. Meads, J. Organomet. Chem. 1980, 193, 345. D. Seyferth, H. P. Withers, Jr.; J. Organomet. Chem. 1980, 185, C1. D. A. Foucher, B. Z. Tang, I. Manners, J. Am. Chem. Soc. 1992, 114, 6246. I. Manners, Can. J. Chem. 1998, 76, 371. A. Berenbaum, H. Braunschweig, R. Dirk, U. Englert, J. C. Green, F. Jäkle,

9

10

11

12

A. J. Lough, I. Manners, J. Am. Chem. Soc. 2000, 122, 5765. R. Rulkens, D. P. Gates, D. Balaishis, J. K. Pudelski, D. F. McIntosh, A. J. Lough, I. Manners, J. Am. Chem. Soc. 1997, 119, 10976. I. R. Butler, W. R. Cullen, F. W. B. Einstein, S. J. Rettig, A. J. Willis, Organometallics 1983, 2, 128. W. Finckh, B. Z. Tang, D. A. Foucher, D. B. Zamble, R. Ziembinski, A. Lough, I. Manners, Organometallics 1993, 12, 823. H. Stoeckli-Evans, A. G. Osborne, R. H. Whiteley, Helv. Chim. Acta 1976, 59, 2402.

431

432

16 Strained Heteroatom-Bridged Metallocenophanes 13

14

15

16

17 18

19

20

21

22

23

24 25 26

27

28

29

D. A. Foucher, M. Edwards, R. A. Burrow, A. J. Lough, I. Manners, Organometallics 1994, 13, 4959. H. Stoeckli-Evans, A. G. Osborne, R. H. Whiteley, J. Organomet. Chem. 1980, 194, 91. R. Broussier, A. Da Rold, B. Gautheron, Y. Dromzee, Y. Jeannin, Inorg. Chem. 1990, 29, 1817. H. Braunschweig, R. Dirk, M. Müller, P. Nguyen, R. Resendes, D. P. Gates, I. Manners, Angew. Chem., Int. Ed. Engl. 1997, 36, 2338. A. Berenbaum, A. J. Lough, I. Manners, Organometallics 2002, 21, 4415. J. K. Pudelski, D. A. Foucher, C. H. Honeyman, A. J. Lough, I. Manners, S. Barlow, D. O’Hare, Organometallics 1995, 14, 2470. S. Barlow, M. J. Drewitt, T. Dijkstra, J. C. Green, D. O’Hare, C. Whittingham, H. H. Wynn, D. P. Gates, I. Manners, J. M. Nelson, J. K. Pudelski, Organometallics 1998, 17, 2113. F. Jäkle, R. Rulkens, G. Zech, D. A. Foucher, A. J. Lough, I. Manners, Chem. Eur. J. 1998, 4, 2117. H. K. Sharma, F. Cervantes-Lee, J. S. Mahmoud, K. H. Pannell, Organometallics 1999, 18, 399. T. Baumgartner, F. Jäkle, R. Rulkens, G. Zech, A. J. Lough, I. Manners, J. Am. Chem. Soc. 2002, 124, 10062. F. Jäkle, R. Rulkens, G. Zech, J. A. Massey, I. Manners, J. Am. Chem. Soc. 2000, 122, 4231. H. Brunner, J. Klankermayer, M. Zabel, J. Organomet. Chem. 2000, 601, 211. D. Seyferth, H. P. Withers, Jr.; Organometallics 1982, 1, 1275. J. K. Pudelski, D. P. Gates, R. Rulkens, A. J. Lough, I. Manners, Angew. Chem., Int. Ed. Engl. 1995, 34, 1506. C. Elschenbroich, A. BretschneiderHurley, J. Hurley, A. Behrendt, W. Massa, S. Wocadlo, E. Reijerse, Inorg. Chem. 1995, 34, 743. K. C. Hultzsch, J. M. Nelson, A. J. Lough, I. Manners, Organometallics 1995, 14, 5496. C. Elschenbroich, J. Hurley, B. Metz, W. Massa, G. Baum, Organometallics 1990, 9, 889.

30

31

32 33

34

35

36 37

38

39

40 41 42

43

44 45

46

47 48 49

R. Resendes, J. M. Nelson, A. Fischer, F. Jäkle, A. Bartole, A. J. Lough, I. Manners, J. Am. Chem. Soc. 2001, 123, 2116. K. Temple, A. J. Lough, J. B. Sheridan, I. Manners, J. Chem. Soc., Dalton Trans. 1998, 2799. J. B. Sheridan, A. J. Lough, I. Manners, Organometallics 1996, 15, 2195. J. B. Sheridan, K. Temple, A. J. Lough, I. Manners, J. Chem. Soc., Dalton Trans. 1997, 711. T. Mizuta, M. Onishi, T. Nakazono, H, Nakazawa, K. Miyoshi, Organometallics 2002, 21, 717. J. M. Nelson, A. J. Lough, I. Manners, Angew. Chem., Int. Ed. Engl. 1994, 33, 989. F. Jäkle, A. Berenbaum, A. J. Lough, I. Manners, Chem. Eur. J. 2000, 6, 2762. M. J. MacLachlan, M. Ginzburg, J. Zheng, O. Knöll, A. J. Lough, I. Manners, New J. Chem. 1998, 22, 1409. A. Berenbaum, F. Jäkle, A. J. Lough, I. Manners, Organometallics 2002, 21, 2359. E. Fossum, K. Matyjaszewski, R. Rulkens, I. Manners, Macromolecules 1995, 28, 401. R. Rulkens, Y. Ni, I. Manners, J. Am. Chem. Soc. 1994, 116, 12121. Y. Ni, R. Rulkens, I. Manners, J. Am. Chem. Soc. 1996, 118, 4102. T. J. Peckham, J. A. Massey, C. H. Honeyman, I. Manners, Macromolecules 1999, 32, 2830. Y. Ni, R. Rulkens, J. K. Pudelski, I. Manners, Macromol. Rapid Commun. 1995, 16, 637. N. P. Reddy, H. Yamashita, M. Tanaka, Chem. Commun. 1995, 2263. K. Temple, F. Jäkle, J. B. Sheridan, I. Manners, J. Am. Chem. Soc. 2001, 123, 1355. P. Gómez-Elipe, R. Resendes, P. M. Macdonald, I. Manners, J. Am. Chem. Soc. 1998, 120, 8348. T. Mizuta, M. Onishi, K. Miyoshi, Organometallics 2000, 19, 5005. T. Mizuta, Y. Imamura, K. Miyoshi, J. Am. Chem. Soc. 2003, 125, 2068. I. Manners, Chem. Commun. 1999, 857.

16.5 References 50

51

52

53

54

55

56 57

58

59 60

61 62 63

64

S. Barlow, A. L. Rohl, S. Shi, C. M. Freeman, D. O’Hare, J. Am. Chem. Soc. 1996, 118, 7578. R. Rulkens, A. J. Lough, I. Manners, S. R. Lovelace, C. Grant, W. E. Geiger, J. Am. Chem. Soc. 1996, 118, 12683. Z. Chen, M. D. Foster, W. Zhou, H. Fong, D. H. Reneker, R. Resendes, I. Manners, Macromolecules 2001, 34, 6156. V. S. Papkov, M. V. Gerasimov, I. I. Dubovik, S. Sharma, V. V. Dementiev, K. H. Pannell, Macromolecules 2000, 33, 7107. P. Nguyen, G. Stojcevic, K. Kulbaba, M. J. MacLachlan, X.-H. Liu, A. J. Lough, I. Manners, Macromolecules 1998, 31, 5977. D. A. Foucher, C. H. Honeyman, J. M. Nelson, B. Z. Tang, I. Manners, Angew. Chem., Int. Ed. Engl. 1993, 32, 1709. K. Kulbaba, I. Manners, Macromol. Rapid Commun. 2001, 22, 711. L. I. Espada, M. Shadaram, J. Robillard, K. H. Pannell, J. Inorg. Organomet. Polym. 2000, 10, 169. K. Kulbaba, M. J. MacLachlan, C. E. B. Evans, I. Manners, Macromol. Chem. Phys. 2001, 202, 1768. M. J. MacLachlan, A. J. Lough, I. Manners, Macromolecules 1996, 29, 8562. A. C. Arsenault, H. Míguez, V. Kitaev, G. A. Ozin, I. Manners, Adv. Mater. 2003, 15, 503. X.-H. Liu, D. W. Bruce, I. Manners, J. Organomet. Chem. 1997, 548, 49. K. N. Power-Billard, I. Manners, Macromolecules 2000, 33, 26. R. Petersen, D. A. Foucher, B.-Z. Tang, A. Lough, N. P. Raju, J. E. Greedan, I. Manners, Chem. Mater. 1995, 7, 2045. K. Kulbaba, A. Cheng, A. Bartole, S. Greenberg, R. Resendes, N. Coombs, A. Safa-Sefat, J. E. Greedan, H. D. H. Stöver, G. A. Ozin, I. Manners, J. Am. Chem. Soc. 2002, 124, 12522.

65

66

67

68 69

70

71

72

73

74 75

76 77

M. Ginzburg, M. J. MacLachlan, S. M. Yang, N. Coombs, T. W. Coyle, N. P. Raju, J. E. Greedan, R. H. Herber, G. A. Ozin, I. Manners, J. Am. Chem. Soc. 2002, 124, 2625. J. Y. Cheng, C. A. Ross, V. Z.-H. Chan, E. L. Thomas, R. G. H. Lammertink, G. J. Vancso, Adv. Mater. 2001, 13, 1174. M. J. MacLachlan, P. Aroca, N. Coombs, I. Manners, G. A. Ozin, Adv. Mater. 1998, 10, 144. J. A. Massey, K. N. Power, M. A. Winnik, I. Manners, Adv. Mater. 1998, 10, 1559. K. Temple, K. Kulbaba, K. N. PowerBillard, I. Manners, K. A. Leach, T. Xu, T. P. Russell, C. J. Hawker, Adv. Mater. 2003, 15, 297. J. Y. Cheng, C. A. Ross, E. L. Thomas, H. I. Smith, G. J. Vancso, Adv. Mater. 2003, 15, 1599. J. A. Massey, M. A. Winnik, I. Manners, V. Z.-H. Chan, J. M. Ostermann, R. Enchelmaier, J. P. Spatz, M. Möller, J. Am. Chem. Soc. 2001, 123, 3147. X.-S. Wang, A. Arsenault, G. A. Ozin, M. A. Winnik, I. Manners, J. Am. Chem. Soc. 2003, 125, 12686. C. H. Honeyman, D. A. Foucher, F. Y. Dahmen, R. Rulkens, A. J. Lough, I. Manners, Organometallics 1995, 14, 5503. L. Cao, I. Manners, M. A. Winnik, Macromolecules 2001, 34, 3353. H. P. Withers, Jr., D. Seyferth, J. D. Fellmann, P. E. Garrou, S. Martin, Organometallics 1982, 1, 1283. J. M. Nelson, H. Rengel, I. Manners, J. Am. Chem. Soc. 1993, 115, 7035. J. M. Nelson, P. Nguyen, R. Petersen, H. Rengel, P. M. Macdonald, A. J. Lough, I. Manners, N. P. Raju, J. E. Greedan, S. Barlow, D. O’Hare, Chem. Eur. J. 1997, 3, 573.

433

435

17

Cyclophanes as Templates in Stereoselective Synthesis Valeria Rozenberg, Elena Sergeeva, and Henning Hopf

17.1

Introduction

The planar chirality of [2.2]paracyclophane derivatives (A, Scheme 17.1) was described by Cram as early as 1955 [1]. However, for a long time stereochemical studies involving these compounds were restricted to the synthesis of various monosubstituted compounds, the determination of their absolute configuration and the study of their chiroptical properties, and their use as chiral markers in reaction mechanisms investigations [2]. No application in stereoselective synthesis was reported up to the early 1990s despite the wide application of planar chiral ferrocene derivatives [3] and arene transition metal complexes [4] to which chiral cyclophanes are related. In the last few years, however, the full potential of substituted [2.2]paracyclophanes for stereosynthesis has been realized, since these compounds possess not only configurational stability up to 200 8C, are stable towards the action of acids and bases and require no special handling techniques, but are also often easily prepared, and the diversity of possible chiral structures, which may be generated by changes in type and position of the substituents, is huge. Not surprisingly, therefore, since ca. 1990 planar chiral paracyclophanes have been employed with rapidly growing success. The present state of knowledge and experience in this area are summarized in this chapter. The review covers not only the application of chiral paracyclophanes as ligands in asymmetric catalysis [5] but also their employment as stoichiometric chiral auxiliaries and their use in stereoselective reactions at the side chain/bridge of the molecule. Routes to novel chiral template structures are also included.

17.2

Chiral [2.2]Paracyclophanes: Nomenclature and Stereochemical Assignment

Any monosubstituted [2.2]paracyclophane is chiral: if a substituent is introduced into one of the benzene rings the derivative possesses planar chirality [6] (structure A in Scheme 17.1); if the substituent is incorporated into one of the ethano bridges, the derivative is centrally chiral (structure B). Increasing the number of Modern Cyclophane Chemistry. Edited by Rolf Gleiter, Henning Hopf Copyright © 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 3-527-30713-3

436

17 Cyclophanes as Templates in Stereoselective Synthesis

Scheme 17.1 Chiral structures attractive as ligands for stereoselective synthesis

substituents rapidly generates a plethora of novel chiral structures [7]. If one just takes those chiral [2.2]paracyclophanes into account that bear two functional groups located on the same side of the plane passing through the four bridge carbon atoms and being positioned as close as possible – to allow e.g. coordination with a metal atom or effect each other sterically (see below) – inter alia the four patterns C–F arise (Scheme 17.1). Situation C we will call syn-latero, whereas D, E, and F carry the long-used prefixes ortho, pseudo-ortho, and pseudo-gem. The list of structures shown in Scheme 17.1 is by no means complete; a possibly interesting ligand for stereoselective synthesis could be one with two (equal or different) vicinal substituents pointing in the same direction (cis-orientation). For the description of planar chiral compounds a set of rules within the CahnIngold-Prelog (CIP) system has been proposed [6 b], which has subsequently been extended and revised [6 c]. Still, following the development of cyclophane chemistry to increasingly complex systems, more subtle nomenclature problems have evolved, and these have been addressed recently [8]. In this review we will adhere to the original CIP-system [6 b–d] which begins by selecting a pilot atom and then chooses a chiral plane containing a maximum number of atoms. Thus in cyclophane derivative A atoms 2, 3, and 4 lie in the chiral plane, while atom 1 is the pilot atom. Since following the sequence 2, 3, 4 generates a counter-clockwise movement, compound A has the S-configuration, the p designating the fact that one is dealing with planar chirality (Scheme 17.2). The other examples in Scheme 17.2 are self-explanatory. In cases where the description of the chirality may change because of a different precedence of groups X and Y during successive transformations both chiral planes will be marked by adding the numbers of the in-plane carbon atoms to which the substituents are

17.4 Monosubstituted [2.2]Paracyclophanes as Chiral Inductors

Scheme 17.2 Stereochemical description of several important chiral cyclophane derivatives

attached, thus structure F in Scheme 17.2 is (4Sp,13Rp)-configured. The same is done for the specification of the chiral centers in laterally substituted derivatives, e.g. such a compound might possess the (Sp,2R)-configuration. That the second chiral element is a center is not particularly pointed out. Analogously we will mark chiral centers in any substituents attached to the cyclophane nucleus.

17.3

The Resolution of Representative Mono- and Disubstituted [2.2]Paracyclophanes

For a long time chiral paracyclophane derivatives were represented by a number of monosubstituted compounds derived from resolved 4-carboxy[2.2]paracyclophane 2 [1, 2 f, 9]. With the development of the massive use of chiral cyclophanes in stereoselective synthesis this situation has changed dramatically, and a large number of multiply-substituted, enantio pure [2.2]paracyclophanes have been prepared. These are either used as such or employed for the creation of more elaborate chiral ligands. Applying various techniques of resolution, the key enantiomerically pure disubstituted paracyclophane derivatives assembled in Scheme 17.3 have been prepared. The table also contains the respective resolving method. In particular cases the chiral compounds were obtained in enantiomerically or diastereomerically pure form by HPLC resolution or by preparative chromatography; these cases will be mentioned at the appropriate place.

17.4

Monosubstituted [2.2]Paracyclophanes as Chiral Inductors

Enantiomers of monosubstituted [2.2]paracyclophanes are usually prepared by resolution of the key compounds 2–6 (Scheme 17.3) which are then chemically modified as needed. The synthesis of the racemic compounds and their resolution have been summarized in several original papers [1, 2 f, 9–16] and a recent review [5]. The analytical resolution of a wide range of monosubstituted [2.2]paracyclophanes by HPLC and enantioselective gas chromatography has been described [25].

437

438

17 Cyclophanes as Templates in Stereoselective Synthesis

Scheme 17.3

Key resolved mono- and disubstituted [2.2]paracyclophanes

Scheme 17.4

Enantioselective alkylation of [2.2]paracyclophane amides

The application of monosubstituted paracyclophanes as chiral auxiliaries will be described by a selection of representative examples. Racemic and homochiral N-substituted amines 14 give rise to the corresponding amides 15, which were subsequently employed as model compounds in stereoselective reactions (Scheme 17.4) [26]. Enolization of 15 followed by reaction with an electrophile EX produced the substituted amides 16 with stereoselective formation of an additional chiral center. Alkylation (82–92% ee) and phenylthiolation (72–97% ee) were more efficient than

17.4 Monosubstituted [2.2]Paracyclophanes as Chiral Inductors

chlorination (58–98% ee) and bromination (18–78% ee). The stereoselectivity of the two latter reactions was in several cases significantly reduced due to equilibration at the a-CHHal position. From the nonhalogenated amides 16 the chiral amines 14 were recovered by hydrolysis in quantitative yield, i.e. they had functioned as efficient chiral auxiliaries. Unfortunately, the yields and enantiomeric purity of the alkylated acids were not reported. Preliminary results on the application of the diallylborinates 17 (in turn obtained by stereoselective allylboration of rac- or (Sp)-4-acetyl[2.2]paracyclophane 6) (see Section 17.7) as chiral auxiliaries have been reported [27]. Racemic (Rp*,R*)17 provided the allylboration product of 4-formyl[2.2]paracyclophane 3, the carbinol 18, in 50% de, whereas optically pure (Sp,S)-17 lead to 19 with 18% ee (Scheme 17.5). The chiral auxiliary was recovered as the diastereomerically pure carbinol 20 and repeatedly used in the same reaction. Other examples involve the application of monosubstituted [2.2]paracyclophane derivatives in catalytic processes. A comprehensive study was, for example, undertaken to evaluate the applicability of the planar chiral hydroxamic acids (Sp)-21 to catalyze the asymmetric epoxidation of allylic alcohols (Scheme 17.6) [28]. The influence of several factors, such as the substituent R at the nitrogen atom, the oxygen source (including a chiral one), the structure of the substrate, the ratio of the reaction components, temperature, etc. was studied and this allowed one to optimize the reaction conditions, finally achieving an enantio-purity of the epoxide 22 of up to 72%.

Scheme 17.5

Stoichiometric allylboration of aldehydes

Scheme 17.6

Catalytic epoxidation of allylic alcohols

439

440

17 Cyclophanes as Templates in Stereoselective Synthesis

Scheme 17.7

Asymmetric autocatalytic formation of a pyrimidyl alcohol

According to a multinuclear NMR study of this system [29], the moderate level of enantioselectivity can be explained by the formation of two reactive intermediate diastereomeric alkylperoxo vanadium(V) complexes with comparable reactivity which are produced in 3 : 1 ratio. Furthermore these NMR studies showed that the ligand-to-metal ratio should be 1.5 to achieve optimal results in this asymmetric catalysis (lower than the ratio proposed by Sharpless [30 a] and by Yamamoto [30 b], who claimed up to five-fold excess for oxidations of this type). One of the most efficient applications of monosubstituted chiral paracyclophanes is provided by the initiation of the enantioselective addition of diisopropylzinc to 2-alkynyl-pyrimidine-5-carbaldehyde 23 (Scheme 17.7) [31]. The initial formation of the corresponding pyrimidyl alkanol 24, induced by the acid (Sp)-2, its methyl ester (Sp)-25, and the methylketone (Sp)-6, and further autocatalytic amplification of the enantiomeric excess occur with a comparable high level of selectivity not only in the presence of 4.8 mol% but also in significantly smaller amounts (0.24–0.095 mol%) of the enantio pure or highly enriched (88–96% ee) inductors [31 a]. Moreover, even moderate (27–29%) or low ee (2.5%) of the initiator was enough to reach quite a high enantio-purity of the resulting alcohol. It is interesting to note that the chiral hydrocarbons 26 and 27 can also be used in these reactions, and that highly enantiomerically enriched 5-pyrimidyl alkanols with a reversed sense of enantioselectivity are obtained as compared with the additions in which [2.2]paracyclophane ligands containing heteroatoms were employed [31 b, c].

17.5

Disubstituted [2.2]Paracyclophanes as Chiral Inductors 17.5.1

Ortho- and syn-latero-Substituted Derivatives

The general route to chiral ortho-disubstituted [2.2]paracyclophane derivatives consists in further substitution of monosubstituted compounds, and these are normally derivatives of 4-hydroxy- (4) or 4-carboxy[2.2]paracyclophane (2).

17.5 Disubstituted [2.2]Paracyclophanes as Chiral Inductors

The dominant methods for the substitution of the phenol 4 are its highly orthoregioselective formylation [17] and acylation [19] and two alternative ortho-lithiations of the corresponding carbamoyloxy (4 a) [32] or MOM-derivatives (4 b) [15]. Enantiomerically pure compounds could successfully be obtained both by resolution [17–19, 33] and by synthesis from the resolved phenol 4 [13 b, 14, 15, 19]. The syntheses of 5-formyl-4-hydroxy[2.2]paracyclophane [14, 15, 17, 18, 32] (7, FHPC, a chiral analog of salicylaldehyde) and the related ortho-hydroxyketones [19], 5-acetyl- (8, AHPC) and 5-benzoyl-4-hydroxy[2.2]paracyclophane (28, BHPC), respectively, determined the development of studies involving auxiliaries and ligands of the Schiff base type. The Cu-complex 29 of the imine of FHPC with glycine was involved in an addition reaction with aldehydes (Scheme 17.8) [17, 18]. As a result, the highly diastereoselective (77–98% de) formation of b-hydroxy-a-amino acids 30 with two asymmetric centers was observed. Only the syn-isomers were produced, and the configuration of the amino acid center was determined by the chirality of the paracyclophanyl fragment. The asymmetric synthesis of a-alkyl-a-amino acids 32 with ees up to 82% was also carried out by alkylation of the Schiff bases 31 of FHPC and its derivatives with amino acid esters (Scheme 17.9) [18]. For the model synthesis of a-methylphenylalanine it was shown that for (Rp)-FHPC-derivatives the alkylation occurs from the re-face of the intermediate Li-enolate. Substitution of the phenolic hydrogen atom by alkyl groups reverses the stereochemical outcome of the reaction. The size of the existing and incoming substituents (R1 and R2) also had an influ-

Scheme 17.8 Diastereoselective synthesis of b-hydroxy-a-amino acids

Scheme 17.9

Stereoselective synthesis of a-alkyl-a-amino acids

441

442

17 Cyclophanes as Templates in Stereoselective Synthesis

ence on the enantiomeric excess and defined the absolute configuration of the produced amino acid. In both cases the enantiomers of FHPC were recovered in high chemical yield with unchanged enantiomeric purity and could hence be reused. The possibility of combining planar chiral aldehyde or ketone components with both achiral and chiral amines, amino alcohols or diamines provided different types of bi-, three- and tetradentate ligands. Bidentate imines of AHPC (33) and BHPC (34), easily obtained as diastereomerically pure compounds during the resolution of the parent ketones [19], were successfully applied as catalysts for stereoselective addition of organozinc derivatives to aldehydes (Scheme 17.10) and imines (Scheme 17.11) [34]. In the diethylzinc addition to aromatic aldehydes 35 good selectivities ((R)-36, 81–88% ee) were obtained with the (Sp,S)-34 ligand whereas its diastereomer (Rp,S)-34 was less effective ((S)-36, 0–77% ee, Scheme 17.10) [34 a]. Significant efficiency of all these ligands in the catalytic addition to aliphatic and branched aldehydes (86–98% ee for cyclohexane- and pivaloylaldehydes) should be emphasized. The alkenylzinc addition to aldehydes was carried out with the optimal ligand of this series, (Rp,S)-33 (Scheme 17.10) [34 b]. For aromatic and aliphatic aldehydes better results (38, 86–98% ee) were obtained with 1-octyne as the alkenylzinc precursor 37, whereas with the bulky tert-butylacetylene and 3-hexyne, respectively, only moderate selectivities (64–75%) were observed. The improvement of

Scheme 17.10 Asymmetric organozinc addition to aldehydes

17.5 Disubstituted [2.2]Paracyclophanes as Chiral Inductors

the results to the level of 88–89% ee was obtained by using dimethylzinc as a transmetalation reagent or by selection of the catalyst. In another application this set of ligands was employed in the asymmetric addition of diethylzinc to imines 39 (Scheme 17.11, R2 = Et) [34 c]. All ligands except (Sp,S)-33 were effective in the model reaction (40, R2 = Et, 92–95% ee). A number of different electron-rich, electron-poor and hindered imino substrates (R1) were then screened in the reaction catalyzed by (Rp,S)-33; it showed a high level of selectivity (89–95% ee). The protocol elaborated was then applied to the enantioselective synthesis of diarylmethyl amines by the addition of the mixed Et2Zn/Ph2Zn reagent to the same imino substrates (Scheme 17.11, R2 = Ph) [34 d]. In this series (Rp,S)-34 was the ligand of choice and the corresponding arylamines 40 were obtained with ees in the range 81–97%. The other type of imino ligand is represented by the tridentate amino alcohols 41–46 (Scheme 17.12). These chiral ligands were obtained either from racemic FHPC and a chiral amino alcohol followed by separation of the diastereomeric mixture by HPLC [35] or from enantiomerically pure FHPC, AHPC and BHPC with both achiral and chiral amino alcohols [35, 36]. In the V-catalyzed sulfoxidation of thioesters 47 with hydrogen peroxide a matched-mismatched effect was observed: in the case of (Sp,S)-41 almost racemic sulfoxide 48 was formed, although a “matched” ligand (Rp,S)-41 was also not as effective ((S)-48, 46–48% ee) as were other leucinol derivatives with additional asymmetric centers or a chiral axis (70– 78% ee) [35]. A set of such tridentate ligands was also tested in the diethylzinc addition to aldehydes 35 [36]. Aldimines, both with or without additional asymmetric centers (42 and 43), catalyzed formation of the alcohol, though with very low selectivity. At the same time ketimine type ligands (44–46) impressively influenced the stereochemical result of the addition. Again a matched-mismatched effect was observed for diastereomers 45 and 46, and better selectivities (36, 92–93% ee) were achieved with the (Sp,S)-ligands both for benzaldehyde and cyclohexanaldehyde.

Scheme 17.11 Asymmetric organozinc addition to imines

443

444

17 Cyclophanes as Templates in Stereoselective Synthesis

Scheme 17.12 Tridentate aminophenols and their application

A modular approach consisting in combining different carbonyl and diamine components was applied to the synthesis of a variety of salen ligands which were classified by the structural and configurational features into four types: I. Structurally and configurationally symmetric; II. Structurally unsymmetric, configurationally symmetric; III. Structurally and configurationally unsymmetric; IV. Structurally symmetric, configurationally unsymmetric (Scheme 17.13) [37]. Salens of type I are C2 symmetrical and could therefore be synthesized by straightforward methods from FHPC, AHPC or BHPC and achiral or chiral diamines [33, 37–38]. For the synthesis of salens of the three other types a stepwise condensation methodology which includes prior synthesis of a hemisalen unit 49 from the diamine and the first carbonyl component, followed by the reaction of the remaining free amino group with a second carbonyl component (structurally and/or configurationally different from the first) was applied [37, 39]. Note that salen ligands of type IV are chiral only in the case of a chiral diamine part. Salens of type I, derived from the enantiomers of FHPC and ethylene- and cyclohexadiamines (EDA and (1R,2R)-CHDA), respectively, (Sp)- or (Rp)-50 and diastereomeric (Sp)- or (Rp)-51 (the description of the diamine moiety is omitted for clarity), possess Lewis acid-activity in the Ti(IV)-catalyzed asymmetric trimethylsilylcyanation of benzaldehyde (Scheme 17.14) [38]. A stereoselectivity of up to 84% ee was observed for the salen ligands 50 with achiral diamine, whereas the diastereomeric ligands 51 were less selective and a matched-mismatched situation was observed in this case (23 and 49% ee for the cyanohydrins of opposite configuration). The catalytic complexes were recovered from the reaction mixture and reused causing a very similar level of asymmetric induction.

17.5 Disubstituted [2.2]Paracyclophanes as Chiral Inductors

Scheme 17.13 Four types of novel chiral salen ligands

Scheme 17.14 Asymmetric trimethylsilylcyanation of benzaldehyde in the presence of various salen ligands

A collection of 15 salen ligands of all four types as well as the hemisalen intermediates 49 was investigated in diethylzinc addition to aldehydes (Scheme 17.15) [39]. Among the salens containing achiral diamine components those derived from FHPC and BHPC (symmetric as well unsymmetric) demonstrated low (50 and 53) to moderate (54 and diastereomeric 56) enantioselectivity in contrast to (Sp)-52 which afforded the best result with 72% ee. For salens with four chiral elements some basic features were recognized. In this series the salens with a combination of (Rp)-AHPC and (R,R)-CHDA (ligands (Rp)- and (Rp,Sp)-55, (Rp,Rp)- and (Rp,Sp)-57) show a matched effect and lead to the formation of (R)-1-phenylpropanol (36, R = Ph) with 58–72% ee whereas other combinations reveal a mismatched effect (0–53% ee of 36). The salens (Sp)-52 and

445

446

17 Cyclophanes as Templates in Stereoselective Synthesis

Scheme 17.15 The efficiency of the salen ligands 50–57 in diethylzinc addition to benzaldehyde

(Rp,Sp)-57 in the addition of Et2Zn to cyclohexancarbaldehyde provide the product with 70 and 94% ee. Hemisalens of a type 49, such as [(Sp)-AHPC](R,R)-CHDA and [(Rp)BHPC](R,R)-CHDA, catalyzed the addition of Et2Zn to form (S)-36 and (R)-36 with 60 and 65% ee, respectively. It should be noted that presently no stable hemisalens could be obtained and applied to any type of asymmetric reaction. Traditional applications of salen type ligands include asymmetric cyclopropanation, epoxidation etc., reactions which demand the participation of transition metals. Salen complexes of Cu(II) with the diastereomeric ligands (Sp)-51 and (Rp)51 were prepared and chiral stationary phases based on these complexes demonstrated the ability to discriminate between different aromatic substrates [40 a]. The complexes of Cu(II) with racemic (Rp*)-52 and achiral (Rp,Sp)-52 as well as Mn(III) with (Rp)-52 were also obtained [40 b] and all complexes characterized by X-ray structural analysis. The application of such types of complexes in asymmetric catalysis is envisaged. A number of enantio- and diastereomerically pure aminophenols 61–63 were synthesized (Scheme 17.16) [41]. In these derivatives the chiral environment can be organized by the planar chiral fragment and modified by the presence of one or two additional chiral centers in the alkylamino group. The developed approaches to various primary, secondary, and tertiary aminophenols include the reduction of the Schiff bases of FHPC, AHPC or BHPC, addition of MeLi to FHPC aldimines, reductive amination of AHPC and transformations of the ortho-lithiated phenol 4. Aside from the above imino- and aminophenol type auxiliaries and ligands the other important ortho-substituted planar chiral compound, (Rp)-[2.2]paracyclophan-[4,5-d]-1,3-oxazol-2-(3H)-one (64), was obtained from the resolved phenol 4 and used as a chiral auxiliary (Scheme 17.17) [42]. Easily available by N-acylation of the oxazol moiety, chiral precursors 65–67 were employed in aldol, Michael or Diels-Alder reactions. The aldol reaction occurs selectively at the si-face, and the aldol addition products (propionic acids) 68 were obtained with 71% ee (R1 = H) and

17.5 Disubstituted [2.2]Paracyclophanes as Chiral Inductors

Scheme 17.16 Diastereomerically pure aminophenols as novel N,O-ligands

90% ee, the syn/anti-ratio amounting to 80 : 20 (R1 = Me). The Michael addition adduct was obtained with > 99% re-selectivity, and from it the enantio pure butanoic acid 69 was released. Diels-Alder addition of cyclopentadiene to 67 was found to occur with a high degree of endo-selectivity. However, the Et2AlCl2-catalyzed reaction at low temperatures generated the endo-adduct 70 A (95.6%), whereas the uncatalyzed reaction yielded the face-reversed endo-adduct 70 B (85.6%) as the main product. In all reactions the chiral auxiliary 64 was recovered quantitatively and reused. The selectivity observed was attributed to the spatial relationship between a prochiral center and the C9–C10-ethano bridge of the paracyclophanyl moiety.

Scheme 17.17 Stereoselective aldol, Michael and Diels-Alder reactions

447

448

17 Cyclophanes as Templates in Stereoselective Synthesis

An alternate way to ortho-substituted [2.2]paracyclophanes consists in the ortho-metalation of oxazolines 71 (R1 = H) derived from racemic 4-carboxy[2.2]paracyclophane 2 and optically pure amino alcohols (Scheme 17.18). Both lithiation [43] and palladation [44] of (Rp,S)-71 (R = iso-Pr or tert-Bu) were regioselective and afforded the corresponding metalated (Sp,S)-72 (the stereochemical descriptors change in these cases due to the higher precedence of the incoming groups). In turn, oxazolines (Sp,S)-71 could be metalated both in ortho- and benzylic position of the ethano bridge (so-called lateral substitution). For the palladation the conditions of the regioselective introduction of the metal in both positions were elaborated. Further transformations of the intermediate metalated oxazolines produced the N,S-, N,Se- or N,Pligands 73–75. Ligands 73 and 74 were examined in the Pd-catalyzed allylic alkylation of 76 [43]. Surprisingly, superior results were obtained with “benzylic” (or syn-latero) ligands (Sp,2R,19S)-73 and 74 ((R)-77, 93–94% ee), whereas the diastereomeric ortho-substituted ligands (Sp,S)- and (Rp,S)-73, 74, respectively, led to the products of opposite configuration with moderate selectivity ((R)-77, 54–57% ee, (S)-77, 63– 73% ee). Other laterally substituted [2.2]paracyclophanes became available by development of two further highly regio- and syn-diastereoselective approaches [45]. The respective protocols include the anionic Fries rearrangement of ortho-protected O([2.2]paracyclophanyl) diisopropylcarbamate or direct lithiation of N-tert-butyl-4[2.2]paracyclophancarboxamide followed by an electrophilic quench.

Scheme 17.18 Stereoselective alkylation of allylic acetates

17.5 Disubstituted [2.2]Paracyclophanes as Chiral Inductors

17.5.2

Pseudo-ortho-Derivatives

The synthesis of pseudo-ortho-disubstituted planar chiral paracyclophanes is easily performed by lithiation of the corresponding dibromide 11 [21, 46–48] followed by trapping of the generated organolithium intermediate with an appropriate electrophile. The way to C2-symmetrical ligands of this type was opened up by the preparation of the enantiomers of the bisphosphine [2.2]PHANEPHOS, 78 (Scheme 17.19), which was obtained via the intermediate phosphinoxide derivative 10, its resolution with d- or l-tartaric acid and further reduction to the corresponding ligands (Rp)- or (Sp)-78 [21]. An alternative synthesis of racemic PHANEPHOS was carried out by direct reaction of the mentioned lithiated intermediate with Ph2PCl [47]. The application of the PHANEPHOS enantiomers as chiral ligands in various stereoselective reactions presently defines the main stream of activity in this direction. Hydrogenation of the dehydro amino acid methyl esters 80, catalyzed by the Rh-complex (Rp)-79, was found to occur under very mild conditions [21]. Various substrates with different substitution patterns were reduced with enantioselectivities of 91–99.6%. This catalyst also demonstrated superior activity over DuPHOS (86% ee in contrast to 56% ee) in the asymmetric synthesis of an intermediate for the HIV protease inhibitor Crixivan by hydrogenation of the corresponding tetrahydropyrazin. Several mechanistic aspects of the catalytic process were studied by NMR spectroscopy [49]. As one of the steps to the bromopyrrol alkaloid Manzacidin the hydrogenation of the C=C double bond was carried out; in the presence of catalytic amounts of (Rp)-79 it proceeded in 50% stereoselectivity [50].

Scheme 17.19 Stereoselective hydrogenation of dehydro amino acid methyl esters

449

450

17 Cyclophanes as Templates in Stereoselective Synthesis

Scheme 17.20 Asymmetric reduction of various b-ketoesters

For the asymmetric hydrogenation of b-ketoesters 83 a highly effective catalytic system (Sp)-82 derived from [2.2]PHANEPHOS-Ru(II)bis(trifluoroacetate) with tetrabutylammonium iodide as a halide source was introduced (Scheme 17.20) [51]. Reduction of a number of substrates occurred with high enantioselectivities ((R)84, 95–96% ee) under milder conditions (lower pressure and temperatures) than with catalysts containing other phosphin ligands. Based upon (Sp)-PHANEPHOS (Ar = Ph), and relative to it, (Sp)-xylyl-PHANEPHOS (Ar = 3,5-(CH3)2C6H3) and chiral diamines, the efficient catalytic system 85 has been formed and employed for asymmetric hydrogenation of various carbonyl compounds (Scheme 17.21) [52]. A pronounced matching effect was observed for the combination of (Rp)-diphosphine and (S)- or (S,S)-diamine ligands in 85 and vice versa. A wide range of substrates, aromatic, heteroaromatic and a,b-unsaturated ketones 86, was reduced to the corresponding carbinols 87 with high enantioselectivity (91–99% ee) using [((Sp)-xylyl-PHANEPHOS)Ru((R,R)-DPEN)Cl2]. The effectiveness of this system is comparable with the Noyori’s best BINAP-Ru-DAIPEN complex; however, it allows to use the less expensive diamine ligands DPEN and DACH. An X-ray investigation of the complex rac-[Pd(4,12-bis(diphenylphosphino)[2.2]paracyclophane)Cl2) and analysis of literature data revealed that the P–Pd–P bite angle (103.69(6)8) is ideal for reactions such as cross-coupling and carbonylation in which reductive elimination constitutes the rate-determining step [47]. Buchwald/Hartwig amination of the racemic dibromide 11 using (Sp)-PHANEPHOS/Pd2dba3 as a chiral catalyst can kinetically resolve the substrate due to different reaction rates of (Sp)- and (Rp)-11 [22]. The chiral catalyst facilitates the amination of the enantiomer of its own chirality, viz. (Sp). The unreacted (Rp)-11 could be isolated from the complex reaction mixture in 42% chemical yield and 93% ee, thus making the enantiomers of bromide 11 available to be used as chiral starting materials. In an application the enantiomerically pure dibromide (Sp)-11 was converted to the phosphonites (Sp)-88 having a paracyclophane backbone and two additional P/O-heterocycles derived from both achiral and chiral diols (Scheme 17.22) [48].

17.5 Disubstituted [2.2]Paracyclophanes as Chiral Inductors

Scheme 17.21 Asymmetric hydrogenation of aromatic, heteroaromatic and a,b-unsaturated ketones

Scheme 17.22 Asymmetric hydrogenation of dehydro amino acids and methyl esters

451

452

17 Cyclophanes as Templates in Stereoselective Synthesis

Scheme 17.23 Potential chiral ligands with pseudo-ortho substitution pattern

The rhodium complexes of the biphenoxy-derivative 89 and the (Sp,R)-diastereomer of the binaphthoxy-derivative 90 efficiently catalyze the asymmetric hydrogenation of various dehydro amino acids and their esters 92 with high stereoselectivity ((S)-93, 93–99% ee), whereas similar complexes of (Sp,S)-90 and (Sp,R)-91 were both less reactive and less selective in this reaction ((S)-93, 74 and 46% ee, respectively). (Sp,S)-91 did not form a Rh-complex at all. Another chiral ligand with C2-symmetry, the corresponding diphenol (Rp)- and (Sp)-PHANOL 12 (R = OH), was introduced recently (Scheme 17.23) [23]. Di-hydroxymethyl-[2.2]paracyclophane 13 (R = CH2OH) was resolved into enantiomers with (1S)-camphanic acid [24]. Pseudo-ortho-disubstituted carboxylic acid 94 (R = COOH) has been described in racemic form so far only [46]. Interesting chiral derivatives 95 with a combination of ortho-, pseudo-ortho- and pseudo-gem-substitution pattern were obtained in enantiomerically pure form [53]. All these ligands as yet have not found application as chiral promoters. 17.5.3

Pseudo-gem-Derivatives

[2.2]Paracyclophane derivatives with pseudo-gem-arrangement of the functional groups (like their ortho-disubstituted counterparts) are chiral only when these substituents are different. The regioselective introduction of substituents into pseudogem position relative to certain functional groups was discovered in the late 1960s [54]. Since then this regioselectivity, studied mostly in electrophilic substitution reactions of 4-carboxy[2.2]paracyclophane derivatives (esters, amides, oxazolines) [55–59] or 4-tosyl[2.2]paracyclophane [60], has increasingly been used for the synthesis of chiral compounds. However up to now only very few examples of the application of such types of compounds as chiral inductors have been reported in the chemical literature. The asymmetric synthesis of linalool 98, reported in 1991 by Reich and Yelm, using the optically pure diselenide 96 as a recoverable chiral auxiliary, was the first example of the use of planar chiral paracyclophane derivatives in stereoselective synthesis [60]. In this reaction the intermediate 97, obtained from 96 and geranylchloride, was subjected to oxidation with m-CPBA followed by a [2,3]-sigmatrop-

17.5 Disubstituted [2.2]Paracyclophanes as Chiral Inductors

Scheme 17.24 Asymmetric synthesis of linalool (98)

ic rearrangement of the resulting selenoxide and isolation of the product (S)-98 and recovery of 96 (Scheme 17.24). The optical yield depended on whether the selenoxide intermediates were equilibrated or not. Product (S)-98 was obtained in 64% ee when the reaction was carried out under thermodynamic control. For the synthesis of bromo-substituted oxazolins 99 two alternative approaches from 4-carboxy[2.2]paracyclophane, namely pseudo-geminal bromination of the intermediate amide followed by cyclization or previous synthesis of the oxazoline derivative such as 71 and its regioselective bromination were elaborated. From racemic substrates several ligands 100 a–d were obtained (Scheme 17.25) [56]. The oxazolin-mediated approach was successfully applied to the synthesis of various N,O- and N,P-ligands (Scheme 17.25) [57, 58]. Diastereomeric bromo-oxazolins 101 having an additional chiral center were easily separated by chromatography. Lithiation followed by reaction with an appropriate electrophile resulted in

Scheme 17.25 Pseudo-gem-disubstituted ligands obtained by a oxazolinmediated route

453

454

17 Cyclophanes as Templates in Stereoselective Synthesis

Scheme 17.26 Allylic alkylation with ligands 103 and 104

formation of compounds 102 and 103. It is interesting to note that these ligands with a similar arrangement of substituents have a different stereochemical descriptor for the planar chiral paracyclophanyl moiety owing to a different precedence in the two pairs oxazolin/Ph2C(OH) and oxazolin/Ph2P, respectively. The N,O-ligands 102 were tested in diethylzinc additions to benzaldehyde [57]. For diastereomeric ligands strongly pronounced matched-mismatched effects were observed. Ligands (4Sp,13Rp,19S)-102 a,c and (4Rp,13Sp,19R)-102 d showed higher activity and selectivity in formation of 1-phenylpropanol (87–93% ee) than diastereomers (4Rp,13Sp,19S)-102 a,c and (4Sp,13Rp,19R)-102 d (5–7% ee); however, (4Sp,13Rp,19S)-102 b was not very effective either. With the matched ligand (4Sp,13Rp,19S)-102 c a number of aromatic aldehydes was converted into the corresponding carbinols 36 in high chemical yields and 81–95% ee. The efficiency of the N,P-ligands 103 was tested in palladium-catalyzed allylic alkylation reactions (Scheme 17.26) [58]. Although all ligands catalyzed the reaction with high reactivity, only a moderate level of enantioselectivity was observed. The best result (62% ee) was obtained with (4Sp,13Rp,19S)-103 a under optimized conditions. Modification of the arylphosphine moieties by introduction of p-Me or pOMe groups (ligands 104 b, c) allowed an increase of stereoselectivity to a level of 69 and 90%, respectively.

17.6

Chiral Templates from Substituted [2.2]Paracyclophanes as Building Blocks

In the preceding paragraphs it was shown how chiral [2.2]paracyclophanes, mostly of the mono-, ortho-, pseudo-ortho- or pseudo-gem-type, could be used as ligands and auxiliaries in asymmetric synthesis. In an extension of these studies, these chiral units are now embedded in other frameworks or are used as building block for the construction of more complicated systems. Thus starting from achiral polyamines and (Rp)-FHPC a set of dendrimers of type 105 bearing 4, 8 or 16 planar chiral units (Scheme 17.27) was obtained with the aim of applying these complex structures in homogeneous catalysis [33]. Racemic [61] and optically pure 4-formyl[2.2]paracyclophane [12] were involved in the construction of the porphyrin core of 106. Mn(III)-complexes of a mixture of four optically pure isomeric porphyrins were investigated in the epoxidation of prochiral alkenes providing the corresponding epoxides with moderate enantioselectivity (22–31% ee) [12].

17.6 Chiral Templates from Substituted [2.2]Paracyclophanes as Building Blocks

Scheme 17.27 Chiral [2.2]paracyclophanyl units as parts of more complex molecular fra-

meworks

The quinolinophane 107 possessing planar chirality (Scheme 17.28) was tested as a chiral ligand in asymmetric diethylzinc addition to aldehydes [62]. In the presence of 10 mol% of (Sp)-107 the corresponding (R)-1-aryl-1-propanols 36 were obtained with ees from 30 to 75%. N-Salicyliden-derivatives of 4-amino[2.2]paracyclophane 108 were evaluated for the copper-catalyzed cyclopropanation of styrene derivatives with diazoesters [63]. The best result (trans/cis-ratios up to 67.8% with ee of the trans-isomer 67.8%) was achieved for the reaction of unsubstituted styrene with the tert-Bu-substituted ligand 108. In preliminary experiments the Ti(IV)-complex of the diastereomerically pure bridged bisphenol 109 showed some activity in enantioselective diethylzinc additions to benzaldehyde (up to 36% ee of 1-phenylpropanol) [64]. Diastereomerically pure b-diketone 110, constructed from two planar chiral paracyclophanyl moieties, was prepared to be used as a chiral ligand [65]. It should be noted that techniques developed for symmetrical compounds such as 109 and 110 can also be used for the synthesis of their unsymmetrical analogs bearing certain aryl substituents instead of one paracyclophanyl fragment. The intramolecular combination of the planar chiral [2.2]paracyclophane system and 2,2’-bipyridine or 2-(2pyridyl)-quinaline resulted in the formation of new types of ligands, 111 and 112 [66]. These ligands catalyzed the cyclopropanation of styrene with good chemical yields and low to moderate enantioselectivity (111: trans/cis 1.9 : 1, ee of trans-isomer: 10%; ee of cis-isomer: 23%; 112: trans/cis 2 : 1, ee of trans- and cis-isomers: both 26%). Other examples of an intramolecular alliance of two types of chirality are represented by compounds 113–115 [67]. In these bidentate ligands a conformationally flexible biphenyl unit (in principle axially chiral) is generated by the contribution of one aromatic ring of a [2.2]paracyclophane unit, which possesses planar chirality, and ortho-substituted benzene ring. The second functional group could be placed in any aromatic ring of the paracyclophane subunit, e.g. in ortho-, pseudo-ortho- or pseudo-gem-position, thus allowing a fine-tuning of the chiral environment. Stereose-

455

456

17 Cyclophanes as Templates in Stereoselective Synthesis

Scheme 17.28 Bifunctional ligands with a planar chiral [2.2]paracyclophanyl unit as building

block

lective pinacol coupling of enantiomerically pure carbonyl derivatives of [2.2]paracyclophane or their imines makes chiral diamines 116 and diols 117 easily available in one step [68]. b-Diketone 118, which exists predominantly in the enol form, was obtained and resolved into enantiomers to be used as a chiral precursor for new chiral diene [69], however, it can also be regarded itself as a potential chiral ligand. Emphasis could, furthermore, be laid on compounds of type 119, which are the first examples of paracyclophane ligands with a cyclohexadiene substructure [70]. These compounds are easily available from optically pure [2.2]paracyclophane-4,7quinone by stereoselective addition of functionally substituted arylmagnesium halides (see Section 17.7). Ortho-substituents in aromatic rings together with geminal hydroxy-groups could participate in the interaction with metals during catalysis. This was demonstrated with the model reaction of diethylzinc addition to benzaldehyde in which 71% ee was achieved with ligand 119 (X = NMe2).

17.7 Stereoselective Reactions in the Side Chain of the Paracyclophanyl Moiety

17.7

Stereoselective Reactions in the Side Chain of the Paracyclophanyl Moiety

The potential of the planar chiral [2.2]paracyclophanyl moiety as a chiral inductor has been repeatedly demonstrated by a number of processes allowing one to generate a novel chiral center in the side chain of the molecule in a stereoselective manner. The advantage of the paracyclophanyl system consists in the possibility to investigate the stereoselectivity of these processes on racemic models initially and then (if necessary) apply the discovered features to the optically pure compounds. Thus the reduction of C=O bond of the rigid annelated compound 120 (Scheme 17.29) with LiAlH4 produced endo- and exo-carbinols in 54 and 27% isolated yields, respectively [2 f ]. The reduction of the cyclic a-ketoester 121 was found to occur stereoselectively and produce the diol of the relative (Rp*,R*)-configuration with de corresponding to 95% (depending on the reaction conditions) [14]. The reduction of the C=C double bond of the polyheterocyclic compound 122 with NaBD4 was completely stereoselective [71]. The analysis of the products obtained in all cases reveal the shielding effect of the aromatic protons of the [2.2]paracyclophane ring that directs the incoming H (or D) towards the unshielded face of the double bond. As mentioned above the diastereomeric Schiff bases 33 and 34 were reduced to provide novel N,O-ligands of the aminophenol type [41]. In all cases conditions for stereoselective reduction were optimized. Both the stereoselectivity of the reactions and the configurations at the newly produced chiral centers were governed by the location of the imino groups in syn-position to the corresponding hydroxy groups, and the attack of LiAlH4 occured from the side of the C=N double bond which was not shielded by the protons of the unsubstituted [2.2]paracyclophane ring. Several examples of stereoselective nucleophilic addition to various derivatives of [2.2]paracyclophane have also been described (Scheme 17.30). Thus addition of organomagnesium- and organolithium reagents to (Rp)-FHPC 7 [72] takes place completely stereoselectively and leads to carbinols of (Rp,S)-configuration. At the same time the introduction of bulky groups into ortho-position to the carbonyl group (compounds 123 a–c) reverses the stereochemistry of addition, and in the case of 123 c furnishes a product with completed inverted stereochemistry at the novel center of chirality. Diastereomeric diols of both configurations can be obtained after deprotection of the hydroxy group. The addition of MeLi to aldimine

Scheme 17.29 Model systems for reduction and hydrogenation studies

457

458

17 Cyclophanes as Templates in Stereoselective Synthesis

Scheme 17.30 Model systems for nucleophilic addition studies

(Sp,S)-124 is also stereoselective and produces the corresponding aminophenol (Sp,R,S)-63 (R = Me) [41 a]. This method complements the reduction of the ketimine (Sp,S)-33 which should give rise to the formation of (Sp,S,S)-63 (R = Me). Allylboration of FHPC (7) is as stereoselective as addition of organometallic compounds to this substrate (see Scheme 17.5) [18]. It is noteworthy that allylboration of the methylketone 6 is also stereospecific [18], whereas nucleophilic addition to 6 leads to a racemic mixture of the corresponding carbinols [73]. The ratio of the diastereomeric carbinols obtained from the aldehyde 3 depends on the allylboration conditions, the highest ratio obtained was 87 : 13. [2.2]Paracyclophan-4,7-quinone (125, Scheme 17.30) represents another type of carbonyl compound, here two carbonyl groups are incorporated into the [2.2]paracyclophane skeleton. The conformational rigidity and the shielding effect of the aromatic ring cause exclusive regio- and stereoselective formation of cisdiols of type 119 with endo-orientation of the hydroxy groups under nucleophilic addition to quinone 125 or its allylboration [74]. Optically pure 125 [75] could be regarded as a chiral auxiliary itself (for example, in asymmetric redox reactions) as well as a starting material for the design of novel chiral ligands [70]. A series of publications describes Diels-Alder addition to (S)-4-vinyl [2.2]paracyclophane [76]. A number of enantio pure condensed polycyclic [2.2]paracyclophane derivatives were obtained with anti-endo-diastereoselectivity with a wide variety of dienes (1,4-benzoquinone and heterocyclic quinones, substituted naphthoquinones, substituted 2-inden-1-ones etc.). The selectivity was again explained by the shielding effect of the unsubstituted [2.2]paracyclophane ring.

17.8

Concluding Remarks

Although the chirality of [2.2]paracyclophanes was already noted and used in the pioneering studies by Cram and coworkers it took nearly 40 years before these configurationally stable compounds were used on a broad scale in stereoselective synthesis. One reason for this long delay doubtlessly has to do with the development of efficient synthetic methods to prepare derivatives of all types. After four

17.9 References

decades of an intense synthetic effort one can regard the problem of preparing all types and kinds of functionalized [2.2]paracyclophanes as largely solved. On the other hand chemists for a very long time were just too fixed on the chiral center and have neglected other elements of chirality, be it the axis or plane of chirality or helical structures. In fact, [2.2]paracyclophanes incorporating all these chiral elements are known in the meantime. If one looks at the [2.2]paracyclophane nucleus as a kind of superatom – like the adamantane core, or cubane, or even C60 – to which substituents of practically any functionality may be connected, then one begins to realize that the use of cyclophanes in stereoselective synthesis has just only been tapped. Compared with the other superatoms mentioned above, the cyclophane core has the advantage of being both aliphatic and aromatic (saturated and unsaturated) allowing to employ the reactions characteristic of both classes of reactions. We can therefore safely predict that many more applications of these interesting layered compounds will be forthcoming.

17.9

References D. J. Cram, N. L. Allinger, J. Am. Chem. Soc., 1955, 77, 6289–6294. 2 (a) D. J. Cram, F. L. Harris, Jr., J. Am. Chem. Soc., 1967, 89, 4642–4651; (b) H. Falk, K. Schlögl, Angew. Chem., 1968, 80, 405–406; Angew. Chem. Int. Ed. Engl., 1968, 7, 383–384; (c) H. J. Reich, D. J. Cram, J. Am. Chem. Soc., 1969, 91, 3517– 3526; (d) O. E. Weigang, Jr., M. J. Nugent, J. Am. Chem. Soc., 1969, 91, 4555– 4556; (e) M. J. Nugent, O. E. Weigang, Jr., J. Am. Chem. Soc., 1969, 91, 4556– 4558; (f) H. Falk, P. Reich-Rohrwig, K. Schlögl, Tetrahedron, 1970, 26, 511– 527; (g) M. H. Delton, D. J. Cram, J. Am. Chem. Soc., 1972, 94, 2741–2478; (h) R. E. Gilman, M. H. Delton, D. J. Cram, J. Am. Chem. Soc., 1972, 94, 2748–2482; (i) A. Guest, P. H. Hoffman, M. J. Nugent, J. Am. Chem. Soc., 1972, 94, 4241– 4246; (j) J. Tribout, R. H. Martin, M. Doyle, H. Wynberg, Tetrahedron Lett., 1972, 2839–2842; (k) K. Schlögl, Top. Curr. Chem., 1984, 125, 27–62. 3 (a) O. Delacroix, J. A. Gladysz, Chem. Commun., 2003, 665–675; (b) C. Bolm, J. P. Hildebrand, K. Mun˘iz, N. Hermanns, Angew. Chem., 2001, 113, 3382– 3407; Angew. Chem. Int. Ed. Engl. 2001, 40, 3284–3308; (c) K. Mun˘iz, C. Bolm, Chemistry: A European Journal, 2000, 6, 1

2309–2316; (d) G. C. Fu, Acc. Chem. Res., 2000, 33, 412–420; (e) C. J. Richards, A. J. Locke, Tetrahedron: Asymmetry, 1998, 9, 2377–2407; (f) A. Togni, Chimia, 1996, 50, 86–93; (g) G. Wagner, R. Herrmann, in: Ferrocenes: Homogeneous Catalysis. Organic Synthesis, Materials Science, A. Togni, T. Hayashi, eds., VCH, Weinheim, 1995, pp 173–218; (h) T. Hayashi, in: Ferrocenes: Homogeneous Catalysis. Organic Synthesis, Materials Science, A. Togni, T. Hayashi, eds., VCH, Weinheim, 1995, pp 105–142; (i) Y. Butsugan, S. Araki, M. Watanabe, in: Ferrocenes: Homogeneous Catalysis. Organic Synthesis, Materials Science, A. Togni, T. Hayashi, eds., VCH, Weinheim, 1995, pp 143–169. 4 (a) S. E. Gibson, H. Ibrahim, Chem. Commun., 2002, 2465–2473; (b) C. Bolm, K. Mun˘iz, Chem. Soc. Rev., 1999, 28, 51– 59. 5 S. E. Gibson, J. D. Knight, Org. Biomol. Chem., 2003, 1, 1256–1269. 6 (a) R. S. Cahn, C. K. Ingold, V. Prelog, Experientia, 1956, 12, 81; (b) R. S. Cahn, C. K. Ingold, V. Prelog, Angew. Chem., 1966, 78, 413–443; Angew. Chem. Int. Ed. Engl., 1966, 5, 385–410; (c) V. Prelog, G. Helmchen, Angew. Chem., 1982, 94, 614–630; Angew. Chem. Int. Ed. Engl.,

459

460

17 Cyclophanes as Templates in Stereoselective Synthesis

7

8 9

10

11

12

13

14

15

16

17

1982, 21, 567–583; (d) G. Helmchen in Methods in Organic Synthesis (HoubenWeyl), Stereoselective Synthesis, G. Helmchen, R. W. Hoffmann, J. Mulzer, E. Schaumann, eds., Thieme, Stuttgart, 1996, Chapter 1. D. J. Cram, R. B. Hornby, E. A. Truesdale, H. J. Reich, M. H. Delton, J. M. Cram, Tetrahedron, 1974, 30, 1757–1768. P. J. Pye, K. Rossen, Tetrahedron: Asymmetry, 1998, 9, 539–541. (a) J. L. Marshall, L. Hall, Tetrahedron, 1981, 37, 1271–1275; (b) P. G. Jones, T. Laue, H. Hopf, Acta Cryst., 1993, C49, 1195–1197. V. Rozenberg, N. Dubrovina, E. Sergeeva, D. Antonov, Yu. Belokon’, Tetrahedron: Asymmetry, 1998, 9, 653–656. (a) T. Izumi, T. Hinata, J. Chem. Tech. Biotechnol., 1992, 55, 227–231; (b) D. Pamperin, H. Hopf, C. Syldatk, M. Pietzsch, Tetrahedron: Asymmetry, 1997, 8, 319–325. S. Banfi, A. Manfredi, F. Montanari, G. Pozzi, S. Quici, J. Mol. Catal. A: Chem., 1996, 113, 77–86. (a) A. Cipiciani, F. Fringuelli, V. Mancini, O. Piermatti, A. M. Scappini, R. Ruzziconi, Tetrahedron, 1997, 53, 11853– 11858; (b) D. Pamperin, C. Schulz, H. Hopf, C. Syldatk, M. Pietzsch, Eur. J. Org. Chem., 1998, 1441–1445; (c) D. Pamperin, B. Ohse, H. Hopf, M. Pietzsch, J. Mol. Catal. B: Enzym., 1998, 5, 317– 319; (d) A. Cipiciani, F. Bellezza, F. Fringuelli, M. G. Silvestrini, Tetrahedron: Asymmetry, 2001, 12, 2277–2281. V. Rozenberg, T. Danilova, E. Sergeeva, E. Vorontsov, Z. Starikova, A. Korlyukov, H. Hopf, J. Org. Chem., 2002, 468–477. A. Cipiciani, F. Fringuelli, V. Mancini, O. Piermatti, F. Pizzo, R. Ruzziconi, J. Org. Chem., 1997, 62, 3744–3747. (a) L. Minuti, A. Taticchi, A. Marrocchi, Tetrahedron: Asymmetry, 2000, 11, 4221–4225; (b) P. Dorizon, C. Martin, J.C. Daran, J.-C. Fiaud, H. B. Kagan, Tetrahedron: Asymmetry, 2001, 12, 2625–2630. V. Rozenberg, V. Kharitonov, D. Antonov, E. Sergeeva, A. Aleshkin, N. Ikonnikov, S. Orlova, Yu. Belokon’,

18

19

20

21

22

23

24

25

26

27

28

29 30

Angew. Chem. 1994, 106, 106–107; Angew. Chem. Int. Ed. Engl., 1994, 33, 91–92. D. Yu. Antonov, Yu. N. Belokon’, N. S. Ikonnikov, S. A. Orlova, A. P. Pisarevsky, N. I. Raevsky, V. I. Rozenberg, E. V. Sergeeva, Yu. T. Struchkov, V. I. Tararov, E. V. Vorontsov, J. Chem. Soc., Perkin Trans. I, 1995, 1873–1879. V. Rozenberg, T. Danilova, E. Sergeeva, E. Vorontsov, Z. Starikova, K. Lysenko, Yu. Belokon’, Eur. J. Org. Chem., 2000, 3295–3303. V. I. Rozenberg, D. Yu. Antonov, E. V. Sergeeva, E. V. Vorontsov, Z. A. Starikova, I. V. Fedyanin, C. Schulz, H. Hopf, Eur. J. Org. Chem., 2003, 2056– 2061. P. J. Pye, K. Rossen, R. A. Reamer, N. N. Tsou, R. P. Volante, P. J. Reider, J. Am. Chem. Soc., 1997, 119, 6207–6208. K. Rossen, P. J. Pye, A. Maliakal, R. P. Volante, J. Org. Chem., 1997, 62, 6462– 6463. D. C. Braddock, I. D. MacGilp, B. G. Perry, J. Org. Chem., 2002, 67, 8679– 8681. P. G. Jones, J. Hillmer, H. Hopf, Acta Crystallographica, E, 2003, 59, O24–O25; J. Hillmer, Ph. D. Dissertation, Braunschweig, 1991. (a) H. Hopf, W. Grahn, D. G. Barrett, A. Gerdes, J. Hilmer, J. Hucker, Y. Okamoto, Y. Kaida, Chem. Ber., 1990, 841–845; (b) W. A. König, B. Gehrcke, D. H. Hochmuth, C. Mlynek, H. Hopf, Tetrahedron: Asymmetry, 1994, 5, 347–350. A. Pelter, H. Kidwell, R. A. N. C. Crump, J. Chem. Soc. Perkin Trans. I, 1997, 3137–3139. N. V. Vorontsova, V. I. Rozenberg, E. V. Vorontsov, O. L. Tok, Yu. N. Bubnov, Russ. Chem. Bull., 2000, 49, 912–919 (Izv. Akad. Nauk, Ser. Khim., 2000, 914– 921). (a) C. Bolm, T. Kühn, Synlett, 2000, 899–901; (b) C. Bolm, O. Beckmann, T. Kühn, C. Palazzi, W. Adam, P. B. Rao, C. R. Saha-Möller, et al., Tetrahedron: Asymmetry, 2001, 12, 2441–2446. K. P. Bryliakov, E. P. Talsi, T. Kühn, C. Bolm, New J. Chem., 2003, 27, 609–614. (a) R. C. Michaelson, R. E. Palermo, K. B. Sharpless, J. Am. Chem. Soc. 1977,

17.9 References

31

32 33

34

35 36

37

38

39

40

99, 1990–1992; (b) Y. Hoshino, H. Yamamoto, J. Am. Chem. Soc. 2000, 122, 10452–10453. (a) S. Tanji, A. Ohno, I. Sato, K. Soai, Org. Lett., 2001, 3, 287–289; (b) K. Soai, I. Sato, Chirality, 2002, 14, 548–554; (c) I. Sato, A. Ohno, Y. Aoyama, T. Kasahara, K. Soai, Org. Biomol. Chem., 2003, 1, 244–246. H. Hopf, D. C. Barrett, Liebigs Ann., 1995, 449–451. J. Issberner, M. Böhme, S. Grimme, M. Nieger, W. Paulus, F. Vögtle, Tetrahedron: Asymmetry, 1996, 7, 2223–2232. (a) S. Dahmen, S. Bräse, J. Chem. Soc. Chem. Commun, 2002, 26–27; (b) S. Dahmen, S. Bräse, Org. Letters, 2001, 3, 4119–4122; (c) S. Dahmen, S. Bräse, J. Am. Chem. Soc., 2002, 124, 5940–5941; (d) N. Hermanns, S. Dahmen, C. Bolm, S. Bräse, Angew. Chem. 2002, 114, 3844– 3846; Angew. Chem. Int. Ed. Engl., 2002, 41, 3692–3694. A. H. Vetter, A. Berkessel, Tetrahedron Lett., 1998, 39, 1741–1744. T. I. Danilova, V. I. Rozenberg, E. V. Sergeeva, Z. A. Starikova, S. Bräse, Tetrahedron: Asymmetry, 2003, 14, 2013– 2019. T. I. Danilova, V. I. Rozenberg, E. V. Vorontsov, Z. A. Starikova, H. Hopf, Tetrahedron: Asymmetry, 2003, 14, 1375– 1383. Yu. N. Belokon’, M. Moscalenko, N. Ikonninkov, L. Yashkina, D. Antonov, E. Vorontsov, V. Rozenberg, Tetrahedron: Asymmetry, 1997, 8, 3245–3250. T. I. Danilova, V. I. Rozenberg, E. V. Sergeeva, E. V. Vorontsov, Z. A. Starikova, S. Bräse, Modern Trends in Organometallic Catalytic Chemistry, Mark Vol’pin (1923–1996), Memorial International Symposium, Moscow, Russia, 2003, May 16– 23, p 122. T. I. Darilova, V. I. Rozenberg, Z. A. Starikova, S. Bräse, Tetrahedron: Asymmetry, 2004, 15, 223–229. (a) S. A. Orlova, N. S. Ikonnikov, T. D. Churkina, V. I. Rozenberg, A. A. Aleshkin, D. Yu. Antonov, Yu, N. Belokon’, Russ. Chem. Bull., 1997, 46, 1120–1124 (Izv. Acad. Nauk, Ser. Khim., 1997, 1169–1173); (b) Z. A. Starikova, M. Yu. Antipin, T. I. Danilova, E. A.

41

42

43

44 45

46

47

48 49

50 51

52

53

54 55

Permyakov, V. I. Rozenberg, Russ. J. Inorg. Chem., 2002, 47, 1486–1494 (Zhurn. Neorg. Khim., 2002, 47, 1623–1631). (a) S. Dahmen, S. Bräse, Tetrahedron: Asymmetry, 2001, 12, 2845–2850; (b) V. I. Rozenberg, T. I. Danilova, E. V. Sergeeva, I. A. Shouklov, Z. A. Starikova, H. Hopf, K. Kühlein, Eur. J. Org. Chem., 2003, 432–440. (a) F. Fringuelli, O. Piermatti, F. Pizzo, R. Ruzziconi, Chem. Lett., 2000, 38– 39; (b) A. Cipiciani, F. Fringuelli, O. Piermatti, F. Pizzo, R. Ruzziconi, J. Org. Chem., 2002, 67, 2665–2670. X.-L. Hou, X.-W. Wu, L.-X. Dai, B.-X. Cao, J. Sun, J. Chem. Soc. Chem. Commun., 2000, 1195–1196. C. Bolm, K. Wenz, G. Raabe, J. Organomet. Chem., 2002, 662, 23–33. T. Focken, H. Hopf, V. Snieckus, I. Dix, P. G. Jones, Eur. J. Org. Chem., 2001, 2221–2228. D. Yu. Antonov, E. V. Sergeeva, E. V. Vorontsov, V. I. Rozenberg, Russ. Chem. Bull., 1997, 46, 1897–1900 (Isv. Akad. Nauk, Ser. Khim. 1997, 2001–2004). P. W. Dyer, P. J. Dyson, S. L. James, C. M. Martin, P. Suman, Organometallics, 1998, 17, 4344–4346. A. Zanotti-Gerosa, C. Malan, D. Herzberg, Org. Lett., 2001, 3, 3687–3690. (a) R. Giernoth, H. Heinrich, N. J. Adams, R. J. Deeth, J. Bargon, J. M. Brown, J. Am. Chem. Soc., 2000, 122, 12381–12382; (b) H. Heinrich, R. Giernoth, J. Bargon, J. M. Brown, Chem. Comm., 2001, 1296–1297. P. M. Wehn, J. Du Bois, J. Am. Chem. Soc., 2002, 124, 12950–12951. P. J. Pye, K. Rossen, R. A. Reamer, R. P. Volante, P. J. Reider, Tetrahedron Lett., 1998, 39, 4441–4444. M. J. Burk, W. Hems, D. Herzberg, C. Malan, A. Zanotti-Gerosa, Org. Lett., 2000, 2, 4173–4176. A. Pelter, B. Mootoo, A. Maxwell, A. Reid, Tetrahedron Lett., 2001, 42, 8391– 8394. H. J. Reich, D. J. Cram, J. Am. Chem. Soc., 1969, 91, 3505–3516. (a) A. Pelter, R. A. N. C. Crump, H. Kidwell, Tetrahedron Lett., 1996, 37, 1273– 1276; (b) A. Pelter, R. A. N. C. Crump,

461

462

17 Cyclophanes as Templates in Stereoselective Synthesis

56

57

58

59 60 61

62 63

64

65

66

67

68

H. Kidwell, Tetrahedron: Asymmetry, 1997, 8, 3873–3880. A. Marchand, A. Maxwell, B. Mootoo, A. Pelter, A. Reid, Tetrahedron, 2000, 56, 7331–7338. X.-W. Wu, X.-L. Hou, L.-X. Dai, J. Tao, B.-X. Cao, J. Sun, Tetrahedron: Asymmetry, 2001, 12, 529–532. X.-W. Wu, K. Yuan, W. Sun, M.-J. Zhang, X.-L. Hou, Tetrahedron: Asymmetry, 2003, 14, 107–112. H. Zitt, I. Dix, H. Hopf, P. G. Jones, Eur. J. Org. Chem., 2002, 2298–2307. H. J. Reich, K. E. Yelm, J. Org. Chem., 1991, 56, 5672–5679. (a) L. Czuchajowski, M. Lozynski, J. Heterocycl. Chem., 1988, 25, 349–350; (b) L. Czuchajowski, S. Goszczynski, D. E. Weeler, A. K. Wisor, T. Malinski, J. Heterocycl. Chem., 1988, 25, 1825–1930. R. Ruzziconi, O. Piermatti, G. Ricci, D. Vinci, Synlett., 2002, 747–750. (a) D. S. Masterson, T. L. Hobbs, D. T. Glatzhofer, J. Mol. Catalysis A: Chem., 1999, 145, 75–81; (b) D. S. Masterson, D. T. Glatzhofer, J. Mol. Catalysis A: Chem., 2000, 161, 65–68. V. I. Rozenberg, D. Y. Antonov, R. P. Zhuravsky, E. V. Vorontsov, V. N. Khrustalev, N. S. Ikonnikov, Yu. N. Belokon’, Tetrahedron: Asymmetry, 2000, 11, 2683–2693. V. I. Rozenberg, N. V. Dubrovina, E. V. Vorontsov, E. V. Sergeeva, Yu. N. Belokon’, Tetrahedron: Asymmetry, 1999, 10, 511–517. (a) U. Wörsdörfer, F. Vögtle, M. Nieger, M. Waletzke, S. Grimme, F. Glorius, A. Pfaltz, Synthesis, 1999, 597–602; (b) U. Wörsdörfer, F. Vögtle, F. Glorius, A. Pfaltz, J. Prakt. Chem., 1999, 341, 445–448. V. I. Rozenberg, D. Yu. Antonov, R. P. Zhuravsky, E. V. Vorontsov, Z. A. Starikova, Tetrahedron Lett., 2003, 44, 3801– 3804. (a) E. V. Sergeeva, V. I. Rozenberg, D. Yu. Antonov, E. V. Vorontsov, Z. A. Starikova, H. Hopf, Tetrahedron: Asym-

69

70

71

72

73 74

75

76

metry, 2002, 13, 1121–1123; (b) E. V. Sergeeva, V. I. Rozenberg, D. Yu. Antonov, E. V. Vorontsov, Z. A. Starkikova, H. Hopf, Modern Trends in Organometallic Catalytic Chemistry, Mark Vol’pin (1923– 1996), Memorial International Symposium, Moscow, Russia, 2003, May 16–23, p 142. L. Minuti, A. Taticchi, C. Rosini, D. Lanari, A. Marrocchi, S. Superchi, Tetrahedron: Asymmetry, 2002, 13, 1257– 1263. N. V. Vorontsova, D. Yu. Antonov, V. I. Rozenberg, E. V. Vorontsov, Z. A. Starikova, Yu. N. Bubnov, Modern Trends in Organometallic Catalytic Chemistry, Mark Vol’pin (1923–1996), Memorial International Symposium, Moscow, Russia, 2003, May 16–23, p 95. R. Yanada, M. Higashikava, Y. Miwa, T. Taga, F. Yoneda, Tetrahedron: Asymmetry, 1992, 3, 1387–1390. E. V. Sergeeva, V. I. Rozenberg, E. V. Vorontsov, T. I. Danilova, Z. A. Starikova, A. I. Yanovsky, Yu. N. Belokon’, H. Hopf, Tetrahedron: Asymmetry, 1996, 7, 3445–3454. D. J. Cram, H. P. Fischer, J. Org. Chem. 1965, 30, 1815–1819. N. V. Vorontsova, V. I. Rozenberg, E. V. Vorontsov, D. Yu. Antonov, Z. A. Starikova, Yu. N. Bubnov, Izv. Akad. Nauk. Ser. Khim. 2002, 8, 1369–1375 (Russ. Chem. Bull. 2002, 51, 1483–1490). N. Vorontsova, V. Rozenberg, E. Vorontsov, D. Antonov, Z. Starikova, Eur. J. Org. Chem., 2003, 761–770. (a) L. Minuti, A. Taticchi, A. Marrocchi, L. Costantini, E. Gacs-Baitz, Tetrahedron: Asymmetry, 2001, 12, 1179–1183; (b) A. Taticchi, L. Minuti, D. Lanari, E. Gacs-Baitz, Tetrahedron: Asymmetry, 2002, 13, 1331–1335; (c) L. Minuti, A. Taticchi, A. Marrocchi, D. Lanari, A. Broggi, E. Gacs-Baitz, Tetrahedron: Asymmetry, 2003, 14, 481–487; (d) L. Minuti, A. Taticchi, D. Laneri, A. Marrochi, E. Gacs-Baitz, Tetrahedron: Asymmetry, 2003, 14, 2387–2392.

While this Chapter was edited the following publications dealing with the use of chiral cyclophane ligands in asymmetric synthesis have appeared: D. C. Braddock, I. D. MacGilp, B. G. Perry, Synlett, 2003, 1121–1124, C. Bolm, T. Focken, G. Raabe, Tetrahedron: Asymmetry, 2003, 14, 1733–1746.

485

19

From Cyclophanes to Molecular Machines Amar H. Flood, Yi Liu, and J. Fraser Stoddart

19.1

Introduction

Molecular machines represent a novel area of chemical research that began, at least in our laboratory, with the preparation [1] of a tetracationic cyclophane (Fig. 19.1), specifically, the p-electron-deficient cyclobis(paraquat-p-phenylene) (CBPQT4+) 14+. There are two key properties of 14+ that have allowed its chemistry to flourish in our hands and also in those of others [2]: 1) it can interact with guests by p–p stacking and charge–transfer interactions [3] and, in the case of appropriate guests, by using C–H · · · O interactions [4] as well; 2) the presence of a rigid cavity helps to trap the guests, giving inclusion complexes. In other words, both energetic and steric effects have made this particular cyclophane very attractive for applications. In this chapter, we will outline the key achievements, from the emergence of this tetracationic cyclophane in 1988, through its infancy in the 1990s as a facile vehicle for supramolecular exploration, forward to its adolescence, as catenanes and rotaxanes started taking a strong hold of this tetracationic cyclophane, up to the present years of its adult life, earning its keep as the central movable part in numerous molecular machines, harnessed to date in molecular memories, but ready to do real mechanical work in the near future. 19.1.1

Control over the Location and Motion of Moving Parts in Molecular Machines

A molecular machine [5] is a multicomponent system in which the reversible movement of the components can be controlled by an external stimulus (S). In particular,

Fig. 19.1 Structural formula of the tetracationic cyclophane 14+ and its graphical representation Modern Cyclophane Chemistry. Edited by Rolf Gleiter, Henning Hopf Copyright © 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 3-527-30713-3

486

19 From Cyclophanes to Molecular Machines

Fig. 19.2 Graphical representations of the three primary design motifs of molecular

machines

when a machine is composed of 1) moving parts, and when it is provided with 2) an energy supply, and following 3) a signal to start, can 4) perform work, it is of paramount importance to be able to control the relative locations and motions of the moving parts. It is through the evolving chemistry of the tetracationic cyclophane 14+ that the element of control has been established. First, we describe our development of host–guest complexation and the ensuing preparative chemistry leading to pseudorotaxanes, catenanes and rotaxanes. Then, we discuss how we have introduced control (Fig. 19.2) over the motions of components in nondegenerate rotaxanes (Type I), and nondegenerate catenanes (Type II), as well as over the dethreading/rethreading of pseudorotaxanes (Type III). Practically, the movement results in a change in properties which produce a signal that allows the operation of the machine to be monitored. The outside stimuli can be photons, electrons, or chemical species, to generate photochemically-, electrochemically- and chemically-driven molecular machines, respectively. The story that will unfold in this chapter covers two decades of research developments in the design, customization and optimization of molecules interlocked with the tetracationic cyclophane. These developments are a testament to the fact that the tetracationic cyclophane has been, and will continue to be an active key component in building nanoscale molecular machines with controllable movements.

19.2

The Creation of the Tetracationic Cyclophane

Our first synthesis [1] of the tetracationic cyclophane 14+ followed directly from the early research of Hünig [6]. In the Würzburg laboratories, three cyclophanes that may be considered as constitutionally isomeric forms of 14+, were synthe-

19.2 The Creation of the Tetracationic Cyclophane

Scheme 19.1

Synthesis of the tetracationic cyclophane 1·4PF6 and the formation of the complex

[14] · 4PF6

sized. These compounds’ acceptor properties were found [6 b] to resemble those of the paraquat component. The cyclophanes were modified by us [1] to produce (Scheme 19.1) yet another isomeric form, namely cyclobis(paraquat-p-phenylene). The compound was obtained as its tetrakis(hexafluorophosphate) salt in 12% yield, after refluxing a solution of the bis(hexafluorophosphate) salt 2 · 2PF6 and 1,4-bis(bromomethyl)benzene (3) in MeCN, followed by counterion exchange using aqueous NH4PF6 solution. Whereas 1 · 4PF6 is soluble in MeCN and MeNO2, exchanging the counterion to afford the tetrachloride 1 · 4Cl, confers aqueous solubility upon 14+ but renders it insoluble in MeCN and MeNO2. This counterion-dependent solubility provides the possibility of investigating the receptor properties of 14+ in both aqueous and non-aqueous media, as well as in the solid state. The X-ray crystal structure of 1 · 4PF6 reveals (Fig. 19.3) that 14+ adopts [1] a rigid rectangular box-like conformation of dimensions 10.3 ´ 6.8 Å in the solid state. Although it was noted [7] that each cyclophane ring is stacked one on top of each other to form continuous open channels, it was the inner dimensions of these channels that was exploited in the host–guest chemistry we investigated in the early days.

Fig. 19.3 a) Structure of 14+ in the crystal; b) Space-filling representation of the structure of 14+ in the solid state

487

488

19 From Cyclophanes to Molecular Machines

19.3

Host–Guest Chemistry with the Tetracationic Cyclophane 19.3.1

Location Control: Host–Guest Chemistry

Host–guest complexation of an aromatic donor included inside the tetracationic cyclophane 14+ is a strict self-assembly process that is thermodynamically driven by the formation of noncovalent bonds, aided by the macrocyclic effect – that is, the location of one molecule inside another can be controlled by noncovalent bonding between each component molecule. A high stability constant Ka for one component relative to the other reflects the integrity of the 1 : 1 complex. The host–guest chemistry of the tetracationic cyclophane 14+, which involves complexation between it, the receptor (host) and the substrate (guest), is the traditional starting point to explore its molecular recognition properties. 14+ turns out to be a multipurpose host which can bind effectively with a wide range of substrates [7]. It is interesting to note that, prior to pursuing these complexation studies, the investigation of its molecular recognition led to a useful synthetic outcome, specifically, a new and improved template-directed synthesis of 14+. Moreover, the early work on the host–guest chemistry of 14+ was crucial to the ultimate development of artificial molecular machines. The first assessment of the molecular recognition properties of 14+ involved [7] three isomeric p-electron-rich donors, namely, the dimethoxybenzenes, of which 1,4-dimethoxybenzene (4) is the most striking. This p-electron-rich aromatic compound and the p-electron-deficient tetracationic cyclophane 14+ are representative p-donors and p-acceptors, respectively. As a consequence of the donor-acceptor nature of the 1 : 1 complex, charge-transfer interactions contribute to the noncovalent bonding that drives the formation (Scheme 19.1) of the inclusion complex [14]4+. 1 : 1 Stoichiometry was established in MeCN solution for [14]4+ with a Ka value of 17 M–1. The presence of a red color in the solution indicated the charge-transfer interaction between the p-electron-deficient host and the p-electron-rich guest. This charge transfer interaction was retained in the solid state, as indicated by the red color of single crystals. The crystal structure (Fig. 19.4) of the 1 : 1 complex [14]4+ revealed that the neutral hydroquinoid guest is inserted centrosymmetrically through the middle of the tetracationic cyclophane’s cavity. The principal intermolecular bonding interactions in this complex are 1) p–p stacking and charge-transfer interactions [3] between the p-electron-rich substrate and the p-electron-deficient bipyridinium units in the receptor and 2) “T-type”, edge-to-face C–H · · · p interactions [8], involving the orthogonally-aligned dimethoxybenzene molecule and the para-phenylene units that bridge the paraquat components in the tetracationic cyclophane 14+. The discovery of its inclusion complexation led to an exploration to find out which guests are recognized by the tetracationic host 14+. It was found that 14+ is an excellent receptor for a wide range of guests containing p-electron-rich aromatic rings, such as dioxynaphthalene-based compounds [9], biphenyl [10], benzidine [10] and

19.3 Host–Guest Chemistry with the Tetracationic Cyclophane

Fig. 19.4 The X-ray crystal structure of the 1 : 1 complex [14]4+

indole [11] and their derivatives [12] in both organic and aqueous solutions (Tab. 19.1). The tetracationic cyclophane 14+ was also found to recognize numerous small bioactive molecules by forming stable inclusion charge-transfer complexes. These include amino acids possessing electron-rich aromatic subunits [13], neurotransmitters [14] and phenyl D-glycopyranosides [15]. Tetrathiafulvalene (TTF), a well-known p-electron donor [16], was found to form a 1 : 1 inclusion complex with 14+ in both solution and in the solid state [17]. TTF undergoes two consecutive oneelectron oxidation processes, a key property that anticipates its use in constructing molecular machines with electrochemically controllable internal movements of com-

Tab. 19.1 Comparison of association constants (Ka) between different p systems and the tetra-

cationic cyclophane 14+ in MeCN determined by 1H NMR or UV-visible spectroscopy at 298 K.

Substrates

Ka/M–1 18

768

140

1 044

150

10 000

489

490

19 From Cyclophanes to Molecular Machines

ponents. It is interesting to note that TTF and its derivatives are among the very few non-aromatic compounds that complex strongly with 14+. The variety of donors that form inclusion complexes with the tetracationic cyclophane 14+ testifies to the importance of a planar electron-rich p-system to achieve association. However, the binding strength between the guest and 14+ is not solely determined by the strength of the p-donor. Subtle structural differences, conferred by substituents attached to the donor’s core, were also found to play a surprisingly dramatic role in determining the binding strength. A systematic investigation [18] of the effect of long chains, which were attached to the aromatic core of several donor molecules, on the binding affinity of guests with 14+ was conducted. It was found that the presence of oxygen atoms on the donor’s side arms was responsible for a majority (Tab. 19.2) of the observed binding as a result of C–H · · · O interactions [4] between the a-hydrogen atoms of the pyridium rings in 14+ and the oxygen atoms of the polyether chains in the guest. Repetitive insertion of ethylene glycol units into the side chains indicates that the binding affinity of the guest toward the tetracationic cyclophane increases with elongation by one ethylene glycol unit, before reaching a plateau after the attachment of two ethylene glycol units: elongation of the side chains with further ethylene glycol units has little effect on Tab. 19.2 a) Comparison of association constants (Ka) between hydroquinone derivatives with different side arms and the tetracationic cyclophane 14+ in MeCN determined by UV-visible spectroscopy at 298 K.

Ka/M–1

17 340 3400 290 3200 28 320 180 54 3200 22 1200 a) Reprinted with permission from J. Org. Chem. 1996, 61, 7298–7303. © 1996 American Chemical Society

19.3 Host–Guest Chemistry with the Tetracationic Cyclophane

the strength of the complexation. This side-chain effect suggests that the C– H · · · O interactions were satisfied, both sterically and electronically, in the case when the side chains are diethylene glycol units because they are sufficient to support the strongest possible binding. There is a further important point. Neither the aromatic core nor the side arms of the guest bind well by themselves, but together they exert a substantial cooperative effect which directs the formation of a stable inclusion charge transfer complex. In a separate study [19], the influence of the donor’s electronic properties on the strength of complexation between a number of different TTF derivatives and the tetracationic cyclophane 14+ has been investigated. The results demonstrate that the strength of association between the donors (TTF derivatives) and the acceptor (14+) is strongly dependent on the strength of the donor, which is determined by 1) the first redox potential, and 2) the size of the donor’s p surface. The formation of strong inclusion complexes between 14+ and p-electron-rich substrates was recognized [20] as the signal to use appropriate donors as templates to direct the formation of the host molecule. Template-directed synthesis [21] has proved to be a very efficient strategy in optimizing the yield of 14+. In the presence of a template, such as the hydroquinone-based diol 5 (Scheme 19.2), the yield of the cyclophane was raised from 12% to 45% when the reaction was carried out in DMF. The yield was further improved to 62% when the reaction was performed under ultra-high pressure (10 kbar). Under the same reaction conditions, but using a different template, that is, the 1,5-dioxynaphthalene-based poly-

Scheme 19.2

Template-directed synthesis of the tetracationic cyclophane 1 · 4PF6

491

492

19 From Cyclophanes to Molecular Machines

ether 6, the efficiency of the template-directed synthesis was raised to a remarkable yield of 81%. The extensive investigation of the molecular recognition properties of 14+ not only contributed to advances in its synthesis using template-directed protocols, but also provided the background for its inclusion in a range of mechanically interlocked molecules.

19.4

Catenane Chemistry with the Tetracationic Cyclophane

The ability of the tetracationic cyclophane 14+ to form inclusion complexes provides us with the unique opportunity to construct large, ordered molecular assemblies such as catenanes and rotaxanes, using the templating actions inherent in the interlocked compounds themselves as they are formed. The first catenane incorporating the tetracationic cyclophane was synthesized [22] in the remarkably high yield of 70%, simply by stirring a mixture of the bis(hexafluorophosphate) salt 2 · 2PF6, bisparaphenylene[34]crown-10 (7) and 1,4bis(bromomethyl)benzene (3) in MeCN for two days. A proposed mechanism for the formation of the [2]catenane is shown in Scheme 19.3. Alkylation of 2 · 2PF6 with the dibromide 3 affords a tricationic intermediate which is bound by the macrocyclic polyether 7 with a pseudorotaxane-like geometry. The subsequent macrocyclization to give the tetracationic cyclophane component is template-directed by the p-electronrich component 7 of the pseudorotaxane-like intermediate to afford the [2]catenane 8 · 4PF6 after counterion exchange. The X-ray crystal structure (Fig. 19.5 a) of 8 · 4PF6 shows quite beautifully the mutually interlocked nature of the two component rings. A p-donor/p-acceptor/p-donor/p-acceptor stack (Fig. 19.5 b) is formed all the way along one of the crystallographic directions by the complementary aromatic units, which are separated by the interplanar p–p distances of about 3.5 Å. In solution, the tetracationic cyclophane and macrocyclic polyether ring components of the [2]catenane 84+ are free to circumrotate through each others’ cavities. Exchange of the “alongside” and “inside” hydroquinone rings (Process I in Scheme 19.4) is achieved by circumrotation of the macrocyclic polyether component through the cavity of the tetracationic cyclophane. Exchange of the “alongside” and “inside” bipyridinium units (Process II in Scheme 19.4) is achieved by circumrotation of the tetracationic cyclophane through the cavity of the macrocyclic polyether component. The free energies of activation associated with Processes I and II have been obtained by variable temperature 1H NMR spectroscopic investigations. While these investigations reveal a certain degree of freedom for the circumrotation processes at higher temperatures, their motions are found to be frozen out at lower temperatures. Provided it is possible to control the circumrotational motions of the rings, then it should be possible to adapt degenerate catenanes to operate like machines at the molecular level. Macrocyclic polyethers incorporating a variety of p-electron-rich aromatic units, such as 1,5-dioxynaphthalene (DNP) [23] and resorcinol [24] ring systems, have

19.4 Catenane Chemistry with the Tetracationic Cyclophane

Scheme 19.3

The template-directed synthesis of the [2]catenane 8 · 4PF6

Fig. 19.5 a) The crystal structure of the [2]catenane 84+ and b) the p-donor/p-acceptor/ p-donor/p-acceptor stack formed in the solid state

also been shown to act successfully as templates for the formation of the tetracationic cyclophane, thus generating [2]catenanes. In addition, introduction of two different p-electron-rich aromatic residues within the same macrocyclic polyether component of a [2]catenane leads to a wide range of molecular structures [25]. In fact, there is a large family of [2]catenanes based on the tetracationic cyclophane 14+ which incorporate different p-donors in the macrocyclic polyether’s constitution. Not only can two different p-donors be incorporated, but the positions of their substitution by the polyether linkages can also be varied and even further modifications can be brought about by changing the lengths of each polyether

493

494

19 From Cyclophanes to Molecular Machines

Scheme 19.4

Dynamic circumrotational processes associated with the [2]catenane 84+

in solution

chain [26]. The myriad of [2]catenanes reflects the common theme that characterizes the synthesis of interlocked molecular compounds: it is modular. In those [2]catenanes where there are two donors with differing affinities for binding inside the cavity of the tetracationic cyclophane, two translational isomers are observed. Depending on the relative affinities of each donor for the cyclophane, two different kinds of translational isomerization can be observed. When one of the donors displays a markedly higher affinity for the tetracationic cyclophane, it is localized in the cavity almost exclusively and an “all-or-nothing” situation ensues. For example, a [2]catenane with TTF and DNP units in the macrocyclic polyether [27] exists as only one translational isomer with the TTF inside the cavity of the cyclophane as determined in solution by 1H NMR spectroscopic studies and in the solid state by X-ray crystallography. In the second case, where the two donors compete for binding inside the cyclophane, an unequal population of the two translational isomers is observed. For example, when DNP and 1,4-hydroquinone (HQ) are in the macrocyclic polyether [25 a], there is a 65 : 35 distribution in CD3COCD3 solution where the translational isomer with the HQ ring located inside the cavity of the cyclophane predominates over the translational isomer of the DNP unit inside. The characterization of the translational isomer distribution provides insight for the design and construction of molecular switches wherein control over the location of the two rings needs to be enforced.

19.4 Catenane Chemistry with the Tetracationic Cyclophane

19.4.1

Going for Gold – The Story of Olympiadane

The remarkably efficient template-directed syntheses of the [2]catenanes provided the foundation to extend this approach to the self-assembly of [n]catenanes incorporating more than two interlocking ring components. Increasing the size of either the p-electron-rich or the p-electron-deficient macrocyclic components of the [2]catenane 84+ made it possible to construct [28] a large number of different higher catenanes, such as a [5]catenane (known as Olympiadane) and a branched [7]catenane [29]. Two enlarged macrocycles based on p-electron-rich and p-electron-deficient components, respectively, were designed for use in the self-assembly of a [3]catenane. An additional 1,5-dioxynaphthalene unit was incorporated into the macrocyclic polyether 9, which was used as the template for directing the synthesis of the enlarged tetracationic cyclophane, cyclobis(paraquat-4,4'-biphenylene) [29]. Reaction (Scheme 19.5) of the bis(hexafluorophosphate) salt 10 · 2PF6, in which the two bipyridinium units are now separated by the longer bitolyl spacer unit, with the dibromide 11 in the presence of 9 at room temperature and ambient pressure, gave a [3]catenane 134+ in a yield of 10%, together with a small amount (3%) of the [2]catenane 124+. In the [3]catenane 134+, two of the three p-electron-rich aromatic units of each macrocyclic polyether are not occupied by a tetracationic cyclophane. As a result, additional tetracationic cyclophanes can subsequently have their synthesis directed by the “free” aromatic templates remaining on each of the two pelectron-rich macrocyclic polyether components, giving rise to higher [n]catenanes [30] (n = 4–7). Further reaction of the [3]catenane 134+, with the bis(hexafluoro-

Scheme 19.5 Self-assembly of the [3]catenane 13 · 4PF6 incorporating an enlarged tetracationic cyclophane

495

496

19 From Cyclophanes to Molecular Machines

phosphate) salt 2 · 2PF6 and the dibromide 3 at ambient temperature and pressure, gave the [4]catenane 148+ and Olympiadane – the [5]catenane 1512+ – in yields of 31% and 5%, respectively. Remarkably, on using ultra-high pressure (Scheme 19.6), a [7]catenane 1820+ was also isolated in 26% yield, together with Olympiadane 1512+ (30%), another [5]catenane 1612+, which is isomeric with Olympiadane, in only trace amounts, and a [6]catenane 1716+ (28%). No [4]catenane 148+ was isolated from the reaction mixture. The X-ray structural analysis of Olympiadane 1512+ reveals (Fig. 19.6 a) the nature of five macrocycles that are interlocked linearly with each other, thus facilitating maximized potential for p–p stacking between the components. The p–p stacking interactions within the Olympiadane are augmented by a series of C–H · · · p and C–H · · · O interactions (not shown in Fig. 19.6 a). Despite the absence of any intermolecular p–p stacking interactions involving the smaller tetracationic cyclophanes, the [5]catenane molecules aggregate to form sheets that are essentially coplanar with the mean planes of these smaller cyclophanes. Furthermore, the X-ray structural analysis of the [7]catenane 1820+ reveals (Fig. 19.6 b) a molecular structure that offers every one of the anticipated recognition sites that were designed into the interlocked molecule

Scheme 19.6

The self-assembly of [n]catenanes starting from the [3]catenane 13 · 4PF6

19.5 Rotaxane Chemistry with the Tetracationic Cyclophane

Fig. 19.6 The space-filling representation of the X-ray crystal structure of a) Olympiadane 1512+ and b) a [7]catenane 1820+

for p–p stacking, that is, all of the six p-electron-rich 1,5-dioxynaphthalene units are utilized in a mutually compatible manner.

19.5

Rotaxane Chemistry with the Tetracationic Cyclophane

Rotaxanes can be prepared using the same strategies as those employed for the self-assembly of catenanes. The mutual recognition of p-electron-deficient and pelectron-rich aromatic units can be utilized for the interlocking of the tetracationic cyclophane around a linear dumbbell-shaped component under template control.

Scheme 19.7 Two strategies, a) clipping and b) threading, for making a [2]rotaxane incorporating the tetracationic cyclophane

497

498

19 From Cyclophanes to Molecular Machines

Simple [2]rotaxanes, incorporating p-electron-rich aromatic cores, can be self-assembled using two different procedures (Scheme 19.7). Clipping is the specific name given to the formation of a tetracationic cyclophane around the dumbbell component by utilizing the same template-directed protocol as that used successfully in the [2]catenane synthesis. An alternative approach relies on threading an unstoppered dumbbell through the cavity of the tetracationic cyclophane to form, in the first instance, a pseudorotaxane. A rotaxane is formed subsequently by covalently bonding bulky stoppers onto the ends of the rod. In both cases, the template effect of the p-electron-rich guest dictates the outcome of the reactions to give the desired [2]rotaxanes. Our investigations on the host–guest chemistry of the tetracationic cyclophane 14+ had already established a range of different p-electron-rich guests. From the list of guests, it was possible to synthesize [2]rotaxanes incorporating different aromatic units, such as HQ [18, 31], DNP [18], benzidine and 4,4'-biphenol [10]. While both approaches to rotaxane formation have been employed, clipping has become the preferred method of choice because of its relative simplicity. A two-station [2]rotaxane can be prepared by adding a second p-electron-rich donor into the dumbbell component. The presence of a second station enables “shuttling” of the tetracationic cyclophane to occur along the rod section of the dumbbell. The first molecular shuttle 194+, which relies upon this design, was prepared [32] in 32% yield by clipping (Scheme 19.8) the two components of the cyclophane around the two-station dumbbell 20. At room temperature in CD3COCD3, the tetracationic cyclophane moves back and forth like a shuttle (Scheme 19.9) between the two identical HQ rings on the rod 500 times per second, a speed which corresponds to an energy barrier of 13.0 kcal mol–1. This shuttle has served as the prototype for the construction of intricate molecular machines.

Scheme 19.8

The template-directed synthesis of the molecular shuttle 19 · 4PF6

19.6 Switchable Rotaxanes, Catenanes and Pseudorotaxanes

Scheme 19.9 The shuttling process and energy barrier for the movement of the tetracationic cyclophane between the two degenerate recognition sites, A and B, in the [2]rotaxane 194+

19.5.1

Location Control – Catenanes and Rotaxanes

Mechanically interlocked molecules demonstrate the high degree of control over the mutual location of the tetracationic cyclophane with respect to different donors that was first observed in the host–guest chemistry of 14+. The higher level of complexity represented by the two interlocked rings in catenanes and the interlocked ring and dumbbell components in rotaxanes allows for circumrotations and shuttling, respectively. These movements give rise to different translational isomers that are in dynamic equilibrium. Introducing two donors with differing affinities for the cyclophane offers the opportunity to isolate only one of the two possible translational isomers. Furthermore, by considering the options to turn off the noncovalent bonding, the cyclophane can be induced to move to the other donor, thus providing the opportunity to control its movements.

19.6

Switchable Rotaxanes, Catenanes and Pseudorotaxanes

The evolving chemistry of 14+ has provided examples of location control and a demonstration of dynamically moving molecular parts. In order to realize a molecular machine, control over the movements of the tetracationic cyclophane had to be addressed through the development of switchable rotaxanes, catenanes and pseudorotaxanes. In particular, the movements have to be attained by switching off and on the recognition elements binding the donor and acceptor together, using chemical, electrochemical and photochemical means to do so.

499

500

19 From Cyclophanes to Molecular Machines

19.6.1

Controllable Molecular Shuttles

Removing the symmetry in the two-station molecular shuttle by inserting nonidentical “stations” within the polyether thread portion of the dumbbell component provides the opportunity to address the two stations selectively. The stronger station’s binding with the tetracationic cyclophane can be removed by chemical or electrochemical means, allowing the cyclophane to move to the other weaker station. The first switchable molecular shuttle 214+ we reported [33] incorporated (Scheme 19.10) benzidine and 4,4'-biphenol units as the two p-electron-rich donors. The rotaxane 214+ exists as two translational isomers in a ratio of 84 : 16 in CD3CN solution at –44 8C, with the tetracationic cyclophane located preferentially on the more p-electron-rich benzidine ring system. When an excess of deuterated trifluoroacetic acid is added to the solution, only one [21 · 2D]6+ of a number of possible translational isomers, in which the 4,4'-biphenol residue is encircled by the cyclophane, is observed. This dramatic change in structure is brought about by protonation of the basic nitrogen atoms associated with the benzidine ring system. Addition of deuterated pyridine returns the solution to neutrality, and reinstates the previous distribution of translational isomers in 214+. Furthermore, this switchable molecular shuttle can also undergo reversible electrochemical switching. Monoelectronic oxidation of the more p-electron-rich benzidine residue switches the structure to one in which the radical form 21·5+ displays the encircle-

Scheme 19.10 The chemically and electrochemically controllable switching of the [2]rotax-

ane 214+

19.6 Switchable Rotaxanes, Catenanes and Pseudorotaxanes

ment of the 4,4'-biphenol unit selectively by the tetracationic cyclophane. Both the chemical and electrochemical switching processes are completely reversible. This [2]rotaxane demonstrates the reversible switching off and on of one of the molecular recognition units, concomitant with the movement of the cyclophane ring from one station to the other. An amphiphilic molecular shuttle 22 · 4PF6 incorporating a monopyrrolo-TTF unit and a DNP ring unit within its rod section was designed in order to utilize the reversible oxidation of the monopyrrolo-TTF unit to turn off its binding with the tetracationic cyclophane. The rotaxane was synthesized [34] using a clipping strategy (Scheme 19.11). The dumbbell-shaped compound 23 was used as the template for the formation of the encircling tetracationic cyclophane from its precursor 2 · 2PF6 and 1,4-bis(bromothyl)benzene (3). The [2]rotaxane 22 · 4PF6 was isolated as an analytically pure brown solid after column chromatography. 1H NMR and UV/Vis absorption spectroscopic investigations indicated the coexistence of the two translational isomers: one with the cyclophane encircling the monopyrrolo-TTF unit (22 · 4PF6-Green), and the other with the cyclophane encircling the DNP ring system (22 · 4PF6-Red). The translational isomer 22 · 4PF6-Red can be isolated by preparative thin layer chromatography and has been shown subsequently to isomerize slowly in solution to 22 · 4PF6-Green, until such times as an

Scheme 19.11 The synthesis of the slow-shuttling [2]rotaxane 22 · 4PF6

501

502

19 From Cyclophanes to Molecular Machines

approximately 1 : 1 mixture exists at equilibrium. The first order kinetics associated with this shuttling process has been measured and corresponds to an activation free energy barrier (DGˆj ) of 24 kcal mol–1. The shuttling process of 22 · 4PF6 provides a working efficiency of no more than 50% to a molecular switch. In order to achieve higher efficiencies, we needed to build a two-station shuttle 1) that favors thermodynamically one translational isomer so that the cyclophane resides preferentially on one site, 2) in which the shuttling process between the two inequivalent states is extremely slow, reflecting a reasonably high activation-free energy barrier between the two states, and 3) where there exists a clear means to switch between the two states when an appropriate stimulus is applied. The three requirements for an efficient molecular machine have been fulfilled by using TTF and DNP as the two donor units in a [2]catenane as well as in a [2]rotaxane. 19.6.2

Switchable Catenanes

A [2]catenane 24 · 4PF6 comprising (Scheme 19.12) a tetracationic cyclophane and a macrocyclic polyether ring with two p-electron-rich recognition sites, namely a DNP ring system and a TTF unit, has been constructed [27]. The X-ray crystal structure of 24 · 4PF6 reveals that the TTF unit resides preferentially inside the cavity of the tetracationic cyclophane, an observation which is consistent with the solution-state behavior indicated by the 1H NMR and UV/Vis spectra. The circum-

Scheme 19.12 A [2]catenane 244+ that can be switched by both chemical and electrochemical means

19.6 Switchable Rotaxanes, Catenanes and Pseudorotaxanes

rotation (Scheme 19.12) of the macrocyclic polyether component through the cavity of the tetracationic cyclophane can be induced reversibly by oxidizing and then reducing the TTF unit. When the TTF unit is oxidized, a charge-charge repulsive force between its single (or double) positive charge and the four positive charges on the cyclophane drives the circumrotation of the macrocyclic polyether to position the DNP ring system inside the cavity of the tetracationic cyclophane. Upon reduction of the oxidized TTF unit back to the neutral state, the macrocyclic polyether once again circumrotates through the cavity of the tetracationic cyclophane, returning the [2]catenane to its original state. The switching process can be controlled by both chemical and electrochemical means. 19.6.3

Switchable Rotaxanes

Aside from the circumrotary motion of the cyclophane observed in catenanes, the shuttling motion of the cyclophane in rotaxanes can also be controlled by the judicious design of the recognition sites located within the polyether rod component. A [2]rotaxane 25 · 4PF6 was designed [35] and synthesized (Scheme 19.13) with a tetracationic cyclophane encircling its dumbbell component incorporating TTF and DNP recognition sites, such that the cyclophane is located exclusively around the TTF unit in the ground state. The redox-switching process in solution was probed by UV/Vis and 1H NMR spectroscopies. When the TTF unit is oxidized, the cyclophane moves a remarkable 3.7 nm along the rigid p-terphenyl spacer from the TTF to the DNP recognition site. The reduction of the TTF+·/TTF2+ back to its neutral state, switches the TTF unit back on as a strong donor-based recognition site, thus allowing the concomitant movement of the cyclophane back to its starting position, the ground state of the [2]rotaxane. 19.6.4

Photochemically Switchable Pseudorotaxanes

In addition to the controllable movements of the tetracationic cyclophane in rotaxanes and catenanes, the dethreading/rethreading of p-electron-rich guests through the cavity of 14+ in a pseudorotaxane also constitutes mechanical motion. The dethreading/rethreading mechanism of the [2]pseudorotaxane [126]4+ can be photochemically stimulated [36, 37] and so can be likened to a photo-driven molecular switch. The [2]pseudorotaxane [126]4+ is formed (Scheme 19.14) by adding the thread-like 1,5-dioxynaphthalene derivative 26 to an aqueous solution of the chloride salt of the tetracationic cyclophane 14+. In the presence of a “sacrificial” reductant (Red, triethanolamine), the photosensitizer (P, 9-anthracenecarboxylic acid) can be excited (Process 1) by light and transfer an electron to one of the bipyridium units in the cyclophane (Process 2 in Scheme 19.14). The oxidized species P+ can be rapidly scavenged by the sacrificial reductant to prevent the back electron-transfer reaction and hence the pseudorotaxane remains reduced. As a consequence, the chargetransfer interaction between the thread and the ring is permanently lowered and the

503

504

19 From Cyclophanes to Molecular Machines

Scheme 19.13 Switching of the [2]rotaxane 254+

dethreading process takes place spontaneously. When oxygen is allowed to enter the irradiated solution, the reduced cyclophane is promptly re-oxidized and 26 again threads through its cavity. This controllable pseudorotaxane represents a prototype at the supramolecular level for a simple molecular machine in which the changes in the relative positions of the components can be followed readily by monitoring differences in absorption and luminescence spectra. 19.6.5

Control of Motion

It has been by investigating non-degenerate two-station rotaxanes and catenanes that we have learnt how to control the movements of the tetracationic cyclophane with respect to encircled dumbbells and threaded rings. Control over the cyclophane’s location is established by giving markedly different affinities to the two donors and by increasing the kinetic barrier to shuttling. This “all-or-nothing” situ-

19.7 Electronic Devices Containing Molecular Switches

Scheme 19.14 The photochemically controllable dethreading/rethreading processes of the pseudorotaxane [126]4+. Process 1 indicates the photo-excitation of the photosensitizer P and Process 2 indicates the subsequent electron transfer from P* to one of the bipyridium units in the cyclophane

ation can then be capitalized upon by turning off one of the recognition sites, thus allowing the cyclophane to relocate around the unchanged donor unit. By utilizing an electrochemically active donor, this movement can be electrochemically powered in a repeating manner by turning on or off an electrical switch. These characteristics constitute all of the requirements for an artificial molecular machine to be put to work in a functioning system. The first demonstration was in the fabrication of molecular memory.

19.7

Electronic Devices Containing Molecular Switches

Catenanes and rotaxanes incorporating the tetracationic cyclophane have been used [38] to fabricate molecular switches that form the key active component in electronic devices. With the ability to control the locations and the relative movements of the cyclophane now well established, the molecular switches have been incorporated [39] into devices with crossbar architectures. A variety of switchable catenanes and rotaxanes have been deployed [38, 39] in a range of devices in order to investigate 1) the influence of molecular structure on device function and 2) to explore the opportunities for miniaturizing the overall device. The idea of introducing molecules into electronic devices is not new [40]. It is a logic step that

505

506

19 From Cyclophanes to Molecular Machines

follows from the miniaturization of electronic components predicted by Moore’s law. Researchers, in academia and industry alike, are now considering alternatives, such as the development of nanoelectronic devices based on molecular-scale switches [41–43] and nanowire-based systems [44]. While these latter systems might be more readily integrable with current microelectronics and represent a natural evolution in CMOS technology, they lack the flexibility and performance, which is represented by the internal structure, inherent in a switchable molecule. The simplest device that can be imagined consists of a single crossbar of two electrodes, between which are sandwiched molecular switches [44]. 19.7.1

A [2]Catenane-Based Electronic Device

The first fabrication of a crossbar device [38], utilizing a redox-switchable interlocked molecule, was based (Fig. 19.7) on the switchable [2]catenane 244+. The general method for constructing such devices relies upon depositing a LangmuirBlodgett (LB) monolayer of closely-packed molecular switches onto a polysilicon electrode. A top electrode of Ti, followed by Al, is subsequently vapor-deposited on top of the monolayer. Amphiphilicity is required to direct the formation of the LB monolayers. In [2]catenanes, hydrophilicity is conferred from the tetracationic cyclophane with its high ionic character, and hydrophobicity is obtained (Fig. 19.7 a) by employing a co-surfactant in the form of dimyristoylphosphatidyl anion (DMPA–) as the counterion [45]. High quality, closely-packed films produced from switchable [2]catenanes are essential for preventing the Ti, which is deposited on top, from penetrating through the monolayer and causing a short circuit. In this example, the fabrication of a working device relies on customizing the molecular switches based on the recommendation from device manufacturers. Such a crossdisciplinary interaction also operated in reverse and thus the devices were modified in order to accommodate the peculiarities of the molecules. The molecular devices display switching between high and low conductance states. Each device is interrogated and characterized by applying a ‘write’ voltage, V and recording the ‘read’ current, I. Of the two I–V measurements, the remnant molecular signature is used to characterize 1) the threshold voltages that are required for switching and 2) the magnitudes of the high and low current values. The binary switching behavior allows for the device’s performance in an electronics context to be measured. The remnant molecular signature (Fig. 19.7 b) tracks the read current (y axis) as the writing voltage (x axis) is cycled around a loop in 40 mV steps, from 0.0 V to +2.0 V, down to –2.0 V, and back to 0.0 V. In order to record the read current a read voltage of +100 mV is applied in between each of the 40 mV pulses. For the [2]catenane-based crossbar, a low current is recorded, corresponding to the OFF state of the device, until the threshold voltage of +2 V is reached, whereupon a higher current is read, effectively switching the device into an ON state. The ON state is maintained until the threshold voltage for switching the device back into a low current or OFF state at –1 V is reached. The binary behavior is recorded (Fig. 19.7 c) in a separate experiment by reading the current after the

19.7 Electronic Devices Containing Molecular Switches

Fig. 19.7 a) An electronic device based on the bistable [2]catenane 244+; b) rem-

nant molecular signature; c) binary switching behavior of device and d) the proposed nanoelectromechanical switching mechanism

device is alternately cycled between ON and OFF states by writing at the threshold voltages. These data reveal that, in addition to the reversible voltage-gated switching of the device ON and OFF more than 10 times, the ON state is metastable, displaying a temperature-dependent resetting of the device back to the OFF state. A molecule-based nanoelectromechanical switching mechanism, which is consistent with all of the data, has been proposed (Fig. 19.7 d) to account for the device’s observed experimental behavior. The OFF state corresponds to the co-conformer with the TTF unit inside the cyclophane. In the device, application of a +2 V bias to the Si electrode generates an oxidized form (TTF+· or TTF2+) of the TTF unit

507

508

19 From Cyclophanes to Molecular Machines

in the [2]catenane. Just as in solution, the resulting charge-charge repulsive force drives a circumrotational process, thus positioning the DNP ring system inside the tetracationic cyclophane. When the bias is lowered to +100 mV for the purposes of reading the device, the charge on the molecule is neutralized and yet the DNP ring system remains trapped inside the cavity of the cyclophane on account of the mutually attractive noncovalent interactions. This new co-conformer is responsible for the high-conductance ON state. The device is observed to decay to the OFF state in a manner that is qualitatively similar to the movements of the tetracationic cyclophane in bistable [2]rotaxanes, self-assembled [46] onto a gold electrode. Therefore, the decay rate is concomitant with the thermally activated circumrotation of the TTF unit back into the central cavity of the cyclophane. Alternatively, applying a reverse bias of –2 V can allow a circumrotation to take place that also leads back to the thermodynamically favored co-conformer. The net reducing voltage causes the reduction of the tetracationic cyclophane, removing its ability to form strong noncovalent bonds with the crown ether component, thus allowing the facile formation, by circumrotation, of the most stable co-conformer. 19.7.2

Bistable [2]Rotaxane Electronic Devices

Crossbar devices built around (Fig. 19.8) the amphiphilic bistable rotaxane 274+ have been investigated [39] for their ability to switch. The switching displayed by

Fig. 19.8 Structural formula and graphical representation of the amphiphilic bistable rotaxane 274+ used in electronic devices

19.7 Electronic Devices Containing Molecular Switches

this [2]rotaxane in solution is almost identical to that displayed by the switchable [2]catenane 244+. Similarly, the devices display an almost one-to-one correlation in terms of remnant molecular signatures and the binary switching behavior of the devices. Note, however, that the catenane switch has an ON/OFF ratio of 2–3: in the rotaxane, it goes up to 10. By contrast, control devices based on dumbbell components and non-redox-active molecules did not display any switching behavior. These data are consistent with the existence (Fig. 19.9) of a similar nanoelectromechanical mechanism involving voltage-gated control over the tetracationic cyclophane’s location operating in the amphiphilic bistable [2]rotaxanes within the devices, just as was proposed for the switchable [2]catenane 244+. The only significant difference is that the cyclophane undergoes a linear mechanical movement rather than a circumrotational one. The physical basis behind the changes in the measured conductance between the ON and OFF states is attributed to the switching of the TTF unit’s molecular energy levels into and out of resonance with electron tunneling pathways at the energies that are preset by the reading voltages. This hypothesis has been evaluated by first-principles computation. In these stud-

Fig. 19.9 Graphical representation of the proposed nanoelectromechanical switching mechanism of the amphiphilic bistable rotaxane 274+ in devices

509

510

19 From Cyclophanes to Molecular Machines

ies, the current was calculated as a function of voltage by simulating a crossbar device with a model system based on a single molecule held between electrodes that were represented by clusters of three gold atoms. The computational results confirm [47] that the ON and OFF states correspond to the co-conformers with the DNP ring system and TTF unit encircled by cyclophane, respectively. 19.7.3

Memory Devices

A simple crossbar provides the basic element from which to construct [38] a molecular random access memory (RAM) chip. An 8 ´ 8 crosspoint structure, comprised in total of 64 crossbars containing a monolayer of two-station [2]rotaxanes, was constructed [39] to demonstrate a 64-bit memory. In order to write to just one crosspoint, one of the leads is set to +1 V and the other –1 V, thus defining +2 V at the address of interest. With all the other leads grounded, voltage magnitudes of only +1 V or –1 V are sensed at all the other nonaddressed crosspoints. This procedure allows for selective addressing without any ‘cross-talk’ or ‘half-select’. Although eight of the bits failed to work, 56 bits operated, allowing the acronyms DARPA (Fig. 19.10), SRC and CNSI to be written successfully into and read out of the chip in ASCII code. Contrary to standard RAM, where the memory addresses need to be continually rewritten every tens of milliseconds, the metastability of the ON state in these molecular analogs displays half-lives of 15–60 min. 19.7.4

Miniaturization of the Crossbar

The ultimate and inevitable evolution of molecular electronics finds its expression in a device defined by a single molecule that switches the conduction between two nanowires. Consequently, the cationically charged analog to pyrene, diazapyrenium (DAP2+) [48], provided rapid access (Fig. 19.11 a) to a switchable [2]catenane 284+ that, when spread onto a water surface using DMPA– as counterions, formed stable Langmuir monolayers. Devices were fabricated [49] on the nanoscale using

Fig. 19.10 Demonstration of DARPA written in ASCII code into an 8 ´ 8 64-bit molecular RAM chip

19.7 Electronic Devices Containing Molecular Switches

Fig. 19.11 a) Structural formula and graphical representation of the bistable [2]catenane 284+; b) the device constructed on a carbon nanotube; c) binary switching behavior in the switchable and degenerate [2]catenanes, 284+ and 294+, respectively

a modification of the general method discussed in Section 19.7.1 for catenanes. In this case, a single semiconducting SWNT, which was generated on top of an insulating silicon oxide substrate, was identified and wired for electrical connectivity. LB Monolayers of 284+ were transferred, and topped (Fig. 19.11 b) with a monolayer of DMPA– counterions for protection from the Ti/Al top electrodes. In keeping with the control studies undertaken on the catenane and rotaxane devices discussed in Sections 19.7.1 and 19.7.2, two additional controls with respect to 284+ were introduced. One control utilized just the DAP2+-containing cyclophane with no interlocking counterpart. Another control (294+) retains the catenane constitution, but utilizes a macrocyclic polyether that does not possess any switchable character. Only the switchable catenane produces a hysteretic remnant molecular signature, which bears the same form as the one recorded for all of the other devices based on switchable catenanes and rotaxanes. A minor shift in the threshold voltages to ± 2.5 V for writing the device was required (Fig. 19.11 c) to guarantee binary switching activity. The same write-read cycles applied to the degenerate catenane 294+ produced no changes in the current, displaying a flat featureless trace. In the context of a single-molecule transistor, break junction devices, coupled with a third gate electrode, have been fabricated [50] using [2]rotaxanes and their dumbbell components. These devices allow the electrical transport through a sin-

511

512

19 From Cyclophanes to Molecular Machines

gle molecule to be recorded [51] between two electrodes as a function of the applied bias and gate voltage. In these types of molecular systems, the molecule behaves as a structurally flexible quantum dot, resulting in a resonant tunneling transistor-type device. A rotaxane 304+ (Fig. 19.12) with disulfide tethers for attachment to gold was prepared for these measurements in order to bridge across the break junction. The differential conductance measurements revealed that the signatures observed are dominated by the interface between the rotaxanes and the gold electrodes. This interpretation follows from the observed symmetry in the differential conductance measurements. For 304+, a symmetric signature is observed. However, in the comparison compound 314+, in which two different tethering groups are utilized – the disulfide for chemisorption and the tetraaryl stopper for physisorption – the signature is asymmetric. If, however, the differential conductance was dominated by the asymmetric disposition of the molecular components, TTF and DNP, in terms of their location along the dumbbell and electronic character, then, the signature is expected to be asymmetrically displayed for both 304+ and 314+. This result is markedly different from all other electronic devices based on switchable catenanes and rotaxanes, where the observed conductance measurements directly follow the constitution and mechanical movements within the molecules and not from the nature of the stoppers. The observation of this contrasting behavior emphasizes [50] the importance of selecting the electrode materials based not only on device fabrication requirements, but also on matching the work functions of the molecules and the electrodes. While the blueprint for putting some of the world’s tiniest molecular machines to work in electronic devices was formed by the interplay of synthetic chemists with device builders, the same theme holds equally well for any advanced technologies to emerge out of fundamental nanoscience in the future. It was this multidisciplinary interaction that saw the development of mechanically operating molecular switches become the focal point for attention and provide for opportunities to broaden their applicability. Whereas the next steps in molecular electronics are

Fig. 19.12 Structural formulas of the “symmetrical” rotaxane 304+ and the “unsymmetrical” rotaxane 314+ used in single-molecule transistors

19.8 Mechanical Devices with Molecular Machines

likely to be focused on logic [52], the operation of artificial molecular machines is expected to progress on into the realm of mechanical manipulation.

19.8

Mechanical Devices with Molecular Machines

The switchable catenanes and rotaxanes, incorporated in electronic devices that operate as molecular switches, offer the potential to have the switching harnessed in wholly mechanical devices such as actuators [53] and molecular valves [54]. Taking a lesson from biomolecular motors [55] such as myosin and actin in muscle fiber, the production of mechanical force can be extracted from artificial molecular machines that are self-organized on surfaces in order for their cooperative mechanical movements to be amplified and harnessed in nanoelectromechanical systems (NEMS). The first demonstration relies on verifying that the tetracationic cyclophane moves linearly from one donor unit to the other in closely-packed condensed phases of bistable rotaxanes on solid substrates. Langmuir monolayers of bistable rotaxanes at the air-water interface and LB monolayers transferred to Si substrates have been investigated [56]. Langmuir monolayers are formed from amphiphilic molecules that are capable of self-organization at the air-water interface [57]. For this purpose, an amphiphilic bistable rotaxane 324+ was prepared (Fig. 19.13 a). The long p-terphenyl spacer was introduced in order to tune the shuttling barrier and to enhance the rigidity of the rotaxane. This amphiphilic system was found to form Langmuir monolayers readily at the air-water interface [56]. Prior to proceeding with the challenges of switching rotaxanes in compressed monolayers, the switching properties [27] that are expected to occur in solution were verified [36] by 1H NMR and UV/Vis spectroscopies. Now that the relative motions of the components in these molecular machines have been demonstrated, the [2]rotaxane molecules were transferred to 7 and constrained within two dimensions at 7 the air-water interface. The amphiphilic rotaxane and its dumbbell were studied using Langmuir isotherm techniques in order to establish their ability to form stable monolayers and for their capacity to switch in this type of condensed phase. The simplest approach [45 a] relies on the comparison (Fig. 19.13 b) of Langmuir layers prepared on neutral and oxidizing subphases. The Langmuir isotherm of the [2]rotaxane 324+ was recorded with water as the subphase and then again with the standard oxidant, Fe(ClO4)3 premixed into the subphase. The two isotherms are different. In general, the oxidized molecules in the monolayer appear to occupy a larger mean molecular area compared to the unoxidized form, at almost every pressure. Control Langmuir isotherm experiments performed on the dumbbells display a negligible difference upon oxidation, and hence this behavior, which contrasts with that of the rotaxane 324+, supports the interpretation that the redox-driven mechanical movement of the tetracationic cyclophane causes the change observed in the rotaxane’s isotherm. In order to provide independent spectroscopic verification for the movement of the cyclophane as a consequence of the oxidizing conditions in the subphase,

513

514

19 From Cyclophanes to Molecular Machines

Fig. 19.13 a) Structural formulas and graphical representations of the amphiphilic rotaxanes 324+ and 344+, and a dumbbell 334+; b) in situ Langmuir switching of the monolayers display changes in the c) XPS spectra

that correlate with the movement of the tetracationic cyclophane; d) Mechanical movement of the cyclophane in a LB monolayer was correlated with changes

and, consequently, the influence of the cyclophane’s movement on the switched isotherm, X-ray photoelectron spectroscopy (XPS) of the LB monolayers was conducted [56]. The intensity of the photoemisson from the 1s orbital on the nitrogen atoms was recorded for the unswitched and switched LB monolayers transferred to a Si substrate. It was found (Fig. 19.13 c) that the relative intensity of the nitrogen signal, and therefore the cyclophane’s signal, increased when the oxidant was added to the subphase. The intensity increase was calculated to be equal to a 44% change in height with respect to the monolayer’s thickness. The percentage

19.9 Conclusions

height change is consistent with the cyclophane moving 3.7 nm upwards from the TTF unit to the DNP ring system. By contrast, the dumbbell 33 displayed no signal and an amphiphilic [2]rotaxane 344+ in which the order of the TTF and DNP units were reversed with respect to the SiO2 substrate, displayed consistent spectroscopic changes. The ultimate test of mechanical motion of the tetracationic cyclophane in a closely packed monolayer was performed on LB layers transferred to a SiO2 substrate. For these purposes, a LB double layer of 344+was prepared in order to display a hydrophilic outer surface to the aqueous solution of the oxidant Fe(ClO4)3, into which the LB double layer was immersed. The XPS data (Fig. 19.13 d) display a signal increase that is consistent with the cyclophane’s movement, following exposure to the oxidizing conditions. This experiment provides spectroscopic evidence that, even when amphiphilic bistable [2]rotaxanes are closely packed in the condensed phase of a monolayer, the cyclophane undergoes mechanical movement following oxidation of the TTF unit. The verification of the linear mechanical movement of the cyclophane within rotaxanes in a LB monolayer provides strong support for developing NEMS devices that incorporate switchable rotaxanes. An alternative usage for the mechanical movements of 14+ is in the fabrication [54 b] of a molecular valve based on a surface-bound pseudorotaxane. The DNPbased thread 26, when bound to a silica surface, and complexed with 14+, displays [54 a] reversible light-driven dethreading and rethreading. From this initial demonstration, the surface-bound pseudorotaxane [126]4+ can be localized around the edges of silica nanopores that serve as reservoirs for an iridium-based luminescent dye. Chemically-driven dethreading of 14+ unblocks the pore’s opening, leading to a release of the dye and hence the observation of its luminescent signal from the surrounding solution.

19.9

Conclusions

The p-electron-deficient tetracationic cyclophane, cyclobis(paraquat-p-phenylene) is undoubtedly the central compound and component responsible for the development of a big and important class of molecular machines. The extensive host– guest studies of complexes of the cyclophane provided an understanding of the thermodynamic factors that allow us to control the relative locations of molecular components in both catenanes and rotaxanes. In other words, such a high level of positional security, resolved down to the subnanometer scale clears the decks for the construction of the linear and rotary motors with the machines’ parts fixed in their starting positions. When a second donor unit is introduced, thermodynamic factors are identified that make it possible for the mechanical motions of molecular components to be controlled with enhanced precision. That is, the physical movements back and forth, or around and around, within different types of motors can be controlled to perform their activities at any instant. Whereas relative

515

516

19 From Cyclophanes to Molecular Machines

movements can be stimulated using chemical, electrochemical or photochemical means, it was voltage-gated movements that were utilized in the fabrication of the 64-bit molecular RAM. And so, the practical utility of this unique class of machinery was vindicated and their position, on the shelves of the modern nanomechanic, secured. Finally, the challenges that lie ahead revolve around their potential to be deployed in mechanical devices where their machinations may be liberated from the academic straitjacket of experimental science for the purposes of doing real hard work. Such labor will not only be operable from the machines’ origins at the nanoscale, but also with options to move other objects around across the scales that reach up to, and possibly into, the world of their macroscopic counterparts – thus bringing their realization back to one of the original sources of inspiration leading to their creation. 19.10

References 1

2

3

4

5

B. Odell, M. V. Reddington, A. M. Z. Slawin, N. Spencer, J. F. Stoddart, D. J. Williams, Angew. Chem. Int. Ed. Engl. 1988, 27, 1547–1550. a) L. Grubert, D. Jacobi, K. Buck, W. Abraham, C. Mugge, E. Krause, Eur. J. Org. Chem. 2001, 20, 3921–3932; b) K.-C. Zhang, L. Liu, T.W. Mu, Q.-X. Guo, Chem. Phys. Lett. 2001, 333, 195–198; c) D. Damgaard, M. B. Nielsen, J. Lau, K. B. Jensen, R. Zubarev, E. Levillain, J. Becher, J. Mater. Chem. 2000, 10, 2249–2258; d) G. A. Kaminski, W. L. Jorgensen, J. Chem. Soc., Perkin Trans. II 1999, 11, 2365–2375; e) M. Lahav, A. N. Shipway, I. Willner, J. Chem. Soc., Perkin Trans. II 1999, 9, 1925–1931; f) M. Lahav, R. Gabai, A. N. Shipway, I. Willner, Chem. Comm. 1999, 19, 1937–1938; g) A. B. Kharitonov, A. N. Shipway, E. Katz, I. Willner, Rev. Anal. Chem. 1999, 18, 255–260; h) S. Capobianchi, G. Doddi, G. Ercolani, J. W. Keyes, P. Mencarelli, J. Org. Chem. 1997, 62, 7015– 7017; i) T. Lu, L. Zhang, G. W. Gokel, A. E. Kaifer, J. Am. Chem. Soc. 1993, 115, 2542–2543. a) C. A. Hunter, J. K. M. Sanders, J. Am. Chem. Soc. 1990, 112, 5525–5534; b) C. A. Hunter, Angew. Chem. Int. Ed. Engl. 1993, 32, 1584–1586. a) G. R. Desiraju, Acc. Chem. Res. 1991, 24, 290–296; b) G. R. Desiraju, Acc. Chem. Res. 1996, 29, 441–449; c) K. N. Houk, S. Menzer, S. P. Newton, F. M. Raymo, J. F. Stoddart, D. J. Williams, J. Am. Chem. Soc. 1999, 121, 1479–1487; d) F. M. Raymo, M. D. Bartberger, K. N. Houk, J. F. Stoddart, J. Am. Chem. Soc. 2001, 123, 9264–9267; e) G. R. Desiraju, Acc. Chem. Res. 2002, 35, 565–573. a) V. Balzani, A. Credi, F. M. Raymo, J. F. Stoddart, Angew. Chem. Int. Ed. 2000, 39, 3348– 3391; b) V. Balzani, M. Venturi, A. Credi, Molecular Devices and Machines – a Journey into the Nanoworld, Wiley-VCH, Weinheim, 2003.

6

7

8

9

10

11 12

13

14 15

a) W. Geuder, S. Hünig, A. Suchy, Angew. Chem. Int. Ed. Engl. 1983, 22, 489–490; b) W. Geuder, S. Hünig, A. Suchy, Tetrahedron 1986, 42, 1665–1677; c) M. Buhner, W. Geuder, W. K. Gries, S. Hünig, M. Koch, T. Poll, Angew. Chem. Int. Ed. Engl. 1988, 27, 1553–1556. P. R. Ashton, B. Odell, M. V. Reddington, A. M. Z. Slawin, J. F. Stoddart, D. J. Williams, Angew. Chem. Int. Ed. Engl. 1988, 27, 1550–1553. a) M. Nishio, Y. Umezawa, M. Hirota, Y. Takeuchi, Tetrahedron 1995, 51, 8665–8701; b) M. Nishio, Y. Umezawa, M. Hirota, Y. Takeuchi, The C–H


p Interaction, Wiley-VCH, New York, 1998. a) M. V. Reddington, A. M. Z. Slawin, N. Spencer, J. F. Stoddart, C. Vicent, D. J. Williams, J. Chem. Soc., Chem. Commun. 1991, 630–634; b) P. R. Ashton, D. Philp, N. Spencer, J. F. Stoddart, D. J. Williams, J. Chem. Soc., Chem. Commun. 1994, 181–184. E. Córdova, R. A. Bissell, N. Spencer, P. R. Ashton, J. F. Stoddart, A. E. Kaifer, J. Org. Chem. 1993, 58, 6550–6552. A. Mirzoian, A. E. Kaifer, J. Org. Chem. 1995, 60, 8093–8095. P.-L. Anelli, M. Asakawa, P. R. Ashton, R. A. Bissell, G. Clavier, R. Gorski, A. E. Kaifer, S. J. Langford, G. Mattersteig, S. Menzer, D. Philp, A. M. Z. Slawin, N. Spencer, J. F. Stoddart, M. S. Tolley, D. J. Williams, Chem. Eur. J. 1997, 3, 1113–1135. T. T. Goodnow, M. V. Reddington, J. F. Stoddart, A. E. Kaifer, J. Am. Chem. Soc. 1991, 113, 4335–4337. A. Bernardo, J. F. Stoddart, A. E. Kaifer, J. Am. Chem. Soc. 1992, 114, 10624–10631. a) S. A. Staley, B. D. Smith, Tetrahedron Lett. 1996, 37, 283–286; b) M. A. Lipton, Tetrahedron Lett. 1996, 37, 287–290.

19.10 References 16 17

18

19

20

21

22

23

24 25

F. Wudl, G. M. Smith, E. J. Hufnagel, J. Chem. Soc., Chem. Commun. 1970, 1453–1454. D. Philp, A. M. Z. Slawin, N. Spencer, J. F. Stoddart, D. J. Williams, J. Chem. Soc., Chem. Commun. 1991, 1584–1586. a) P.-L. Anelli, P. R. Ashton, R. Ballardini, V. Balzani, M. Delgado, M. T. Gandolfi, T. T. Goodnow, A. E. Kaifer, D. Philp, M. Pietraszkiewicz, L. Prodi, M. V. Peddington, A. M. Z. Slawin, N. Spencer, J. F. Stoddart, C. Vicent, D. J. Williams, J. Am. Chem. Soc. 1992, 114, 193– 218; b) R. Castro, K. R. Nixon, J. D. Evanseck, A. E. Kaifer, J. Org. Chem. 1996, 61, 7298–7303. M. B. Nielsen, J. O. Jeppesen, J. Lau, C. Lomholt, D. Damgaard, J. P. Jacobsen, J. Becher, J. F. Stoddart, J. Org. Chem. 2001, 66, 3559–3563. a) C. L. Brown, D. Philp, J. F. Stoddart, Synlett 1991, 462–464; b) M. Asakawa, W. Dehaen, G. L’abbé, S. Menzer, J. Nouwen, F. M. Raymo, J. F. Stoddart, D. J. Williams, J. Org. Chem. 1996, 61, 9591–9595. a) Templated Organic Synthesis (eds: F. Diederich, P. J. Stang), Wiley-VCH, Weinheim, 1999; b) J.-C. Chambron, C. O. Dietrich-Buchecker, J.-P. Sauvage, Top. Curr. Chem. 1993, 165, 131– 162; c) H. W. Gibson, H. Maraud, Adv. Mater. 1993, 5, 11–21; d) D. B. Amabillino, J. F. Stoddart, Chem. Rev. 1995, 95, 2725–2828; e) D. Philp, J. F. Stoddart, Angew. Chem. Int. Ed. Engl. 1996, 35, 1154–1196; f) R. Jäger, F. Vögtle, Angew. Chem. Int. Ed. Engl. 1997, 36, 930–944; g) T. J. Hubin, D. H. Busch, Coord. Chem. Rev. 2000, 200–202, 5–52; h) L. Raehm, D. G. Hamilton, J. K. M. Sanders, Synlett 2002, 1743–1761. P. R. Ashton, T. Goodnow, A. E. Kaifer, M. V. Reddington, A. M. Z. Slawin, N. Spencer, J. F. Stoddart, C. Vicent, D. J. Williams, Angew. Chem. Intl. Ed. Engl. 1989, 28, 1396–1399. P. R. Ashton, C. T. Brown, E. J. T. Chrystal, T. T. Goodnow, A. E. Kaifer, K. P. Parry, D. Philp, A. M. Z. Slawin, N. Spencer, J. F. Stoddart, D. J. Williams, J. Chem. Soc., Chem. Commun. 1991, 634–639. D. B. Amabilino, P. R. Ashton, J. F. Stoddart, Supramol. Chem. 1995, 5, 5–8. For a few examples, see: a) P. R. Ashton, M. Blower, D. Philp, N. Spencer, J. F. Sotddart, M. S. Tolley, R. Ballardini, M. Ciano, V. Balzani, M. T. Gandolfi, L. Prodi, C. H. McLean, New J. Chem. 1993, 17, 689–695; b) D. B. Amabilino, P. R. Ashton, G. R. Brown, W. Hayes, J. F. Stoddart, M. S. Tolley, D. J. Williams, J. Chem. Soc., Chem. Commun. 1994, 2475–2478; c) D. B. Amabilino, P.-L. Anelli, P. R. Ashton, G. R. Brown, E. Cordova, L. A. Godinez, W. Hayes, A.E. Kaifer, D. Philp, A. M. Z. Slawin, N. Spencer, J. F. Stoddart, M. S. Tolley, D. J. Williams, J. Am. Chem. Soc. 1995, 117, 11142– 11170; d) M. J. Gunter, M. R. Johnson, Tetrahedron Lett. 1990, 31, 4801–4804; e) M. J. Gunter, M. R. Johnson, B. W. Skelton, A. H. White, J. Chem. Soc., Perkin Trans. I 1994, 1009–1018.

26

27

28

29

30

31

32 33 34

35 36

37

a) D. B. Amabilino, P. R. Ashton, C. L. Brown, E. Cordova, L. A. Godinez, T. T. Goodnow, A. E. Kaifer, S. P. Newton, P. Pietraszkiewicz, D. Philp, F. M. Raymo, A. S. Reder, M. T. Rutland, A. M. Z. Slawin, N. Spencer, J. F. Stoddart, D. J. Williams, J. Am. Chem. Soc. 1995, 117, 1271– 1293; b) P. R. Ashton, M. A. Blower, S. Iqbal, C. H. McLean, J. F. Stoddart, M. S. Tolley, D. J. Williams, Synlett 1994, 1059–1063; c) P. R. Ashton, M. A. Blower, C. H. McLean, J. F. Stoddart, M. S. Tolley, Synlett 1994, 1063–1066. a) M. Asakawa, P. R. Ashton, V. Balzani, A. Credi, C. Hamers, G. Mattersteig, M. Montalti, A. N. Shipway, N. Spencer, J. F. Stoddart, M. S. Tolley, M. Venturi, A. J. P. White, D. J. Williams, Angew. Chem. Int. Ed. 1998, 37, 333–337; b) R. Ballardini, V. Balzani, J. Becher, A. Di Fabio, M. T. Gandolfi, G. Mattersteig, M. B. Nielsen, F. M. Raymo, S. J. Rowan, J. F. Stoddart, A. J. P. White, D. J. Williams, J. Org. Chem. 2000, 65, 4120–4126. P. R. Ashton, C. L. Brown, E. J. T. Chrystal, T. T. Goodnow, A. E. Kaifer, K. P. Parry, A. M. Z. Slawin, N. Spencer, J. F. Stoddart, D. J. Williams, Angew. Chem. Int. Ed. Engl. 1991, 30, 1039–1042. a) D. B. Amabilino, P. R. Ashton, A. S. Reder, N. Spencer, J. F. Stoddart, Angew. Chem. Int. Ed. Engl. 1994, 33, 1286–1290; b) D. B. Amabilino, P. R. Ashton, V. Balzani, S. E. Boyd, A. Credi, J. Y. Lee, S. Menzer, J. F. Stoddart, M. Venturi, D. J. Williams, J. Am. Chem. Soc. 1998, 120, 4295–4307. D. B. Amabilino, P. R. Ashton, S. E. Boyd, J. Y. Lee, S. Menzer, J. F. Stoddart, D. J. Williams, Angew. Chem. Int. Ed. Engl. 1997, 36, 2070–2072. P. R. Ashton, M. Grognuz, A. M. Z. Slawin, J. F. Stoddart, D. J. Williams, Tetrahedron Lett. 1991, 32, 6235–6238. P.-L. Anelli, N. Spencer, J. F. Stoddart, J. Am. Chem. Soc. 1991, 113, 5131–5133. R. A. Bissell, E. Córdova, A. E. Kaifer, J. F. Stoddart, Nature 1994, 369, 133–137. a) J. O. Jeppesen, J. Perkins, J. Becher, J. F. Stoddart, Angew. Chem. Int. Ed. 2001, 40, 1216–1221; b) J. O. Jeppesen, K. A. Nielsen, J. Perkins, S. A. Vignon, A. Di Fabio, R. Ballardini, M. T. Gandolfi, M. Venturi, V. Balzani, J. Becher, J. F. Stoddart, Chem. Eur. J. 2003, 9, 2982–3007. H.-R. Tseng, S. A. Vignon, J. F. Stoddart, Angew. Chem. Int. Ed. 2003, 42, 1491–1495. R. Ballardini, V. Balzani, M. T. Gandolfi, L. Prodi, M. Venturi, D. Philp, H. G. Ricketts, J. F. Stoddart, Angew. Chem. Int. Ed. Engl. 1993, 32, 1301–1303. P. R. Ashton, R. Ballardini, V. Balzani, S. E. Boyd, A. Credi, M. T. Gandolfi, M. Go´mezLo´pez. S. Iqbal, D. Philp, J. A. Preece, L. prodi, H. G. Ricketts, J. F. Stoddart, M. S. Tol-

517

518

19 From Cyclophanes to Molecular Machines

38

39

40 41

42

43

44

45

ley, M. Venturi, A. J. P. White, D. J. Williams, Chem. Eur. J. 1997, 3, 152–170. a) C. P. Collier, G. Mattersteig, E. W. Wong, Y. Luo, K. Beverly, J. Sampaio, F. M. Raymo, J. F. Stoddart, J. R. Heath, Science 2000, 289, 1172–1175; b) C. P. Collier, J. O. Jeppesen, Y. Luo, J. Perkins, E. W. Wong, J. R. Heath, J. F. Stoddart, J. Am. Chem. Soc. 2001, 123, 12632– 12641; c) E. W. Wong, C. P. Collier, M. Behloradsky, F. M. Raymo, J. F. Stoddart, J. R. Heath, J. Am. Chem. Soc. 2000, 122, 5831–5840; d) C. P. Collier, E. W. Wong, M. Belohradsky, F. M. Raymo, J. F. Stoddart, P. J, Kuekes, R. S. Williams, J. R. Heath, Science 1999, 285, 391– 394. Y. Luo, C. P. Collier, J. O. Jeppesen, K. A. Nielsen, E. Delonno, G. Ho, J. Perkins, H. R. Tseng, T. Yamamoto, J. F. Stoddart, J. R. Heath, ChemPhysChem 2002, 3, 519–525. A. Aviram, M. A. Ratner, Chem. Phys. Lett. 1974, 29, 277–283. a) J. Chen, M. A. Reed, A. M. Rawlett, J. M. Tour, Science 1999, 286, 1550–1552; b) C. Li, D. H. Zhang, X. L. Liu, S. Han, T. Tang, C. W. Zhou, W. Fan, J. Koehne, J. Han, M. Meyyappan, A. M. Rawlett, D. W. Price, J. M. Tour, Appl. Phys. Lett. 2003, 82, 645–647; c) M. A. Reed, C. Zhou, C. J. Muller, T. P. Burgin, J. M. Tour, Science 1997, 278, 252–254; d) N. A. Melosh, A. Boukai, F. Diana, B. Gerardot, A. Badolato, P. M. Petroff, J. R. Heath, Science 2003, 300, 112–115. a) Q. L. Li, S. Surthi, G. Mathur, S. Gowda, V. Misra, T. A. Sorenson, R. C. Tenent, W. G. Kuhr, S. Tamaru, J. S. Lindsey, Z. M. Liu, D. F. Bocian, Appl. Phys. Lett. 2003, 83, 198– 200; b) K. M. Roth, A. A. Yasseri, Z. M. Liu, R. B. Dabke, V. Malinovskii, K. H. Schweikart, L. H. Yu, H. Tiznado, F. Zaera, J. S. Lindsey, W. G. Kuhr, D. F. Bocian, J. Am. Chem. Soc. 2003, 125, 505–517; c) C. R. Kagan, D. B. Mitzi, C. D. Dimitrakopoulos, Science 1999, 286, 945–947; d) C. P. Husband, S. M. Husband, J. S. Daniels, J. M. Tour, IEEE Trans. Electron Dev. 2003, 50, 1865–1875. a) J. G. Kushmerick, J. Naciri, J. C. Yang, R. Shashidhar, Nano Lett. 2003, 3, 897–900; b) J. K. N. Mbindyo, T. E. Mallouk, J. B. Mattzela, I. Kratochvilova, B. Razavi, T. N. Jackson, T. S. Mayer, J. Am. Chem. Soc. 2002, 124, 4020–4026. a) Y. Cui, Q. Q. Wei, H. K. Park, C. M. Lieber, Science 2001, 293, 1289–1292; b) Y. Huang, X. F. Duan, Q. Q. Wei, C. M. Lieber, Science 2001, 291, 630–633; c) Y. Huang, X. F. Duan, Y. Cui, L. J. Lauhon, K. H. Kim, C. M. Lieber, Science 2001, 294, 1313–1317; d) A. Javey, J. Guo, Q. Wang, M. Lundstrom, H. J. Dai, Nature 2003, 424, 654– 657. a) M. Asakawa, M. Higuchi, G. Mattersteig, T. Nakamura, A. R. Pease, F. M. Raymo, T.

46 47 48

49

50

51

52

53

54

55

56

57

Shimizu, J. F. Stoddart, Adv. Mater. 2000, 12, 1099–1102; b) R. C. Ahuja, P. L. Caruso, D. Mobius, G. Wildburg, H. Ringsdorf, D. Philp, J. A. Preece, J. F. Stoddart, Langmuir 1993, 9, 1534–1544. H.-R. Tseng, D. Wu, N. Fang, X. Zhang, J. F. Stoddart, ChemPhysChem 2004, 5, 111–116. A. H. Flood, R. J. A. Ramirez, W.-Q. Deng, et al., Aust. J. Chem. 2004, 57, 301–322. a) V. Balzani, A. Credi, S. J. Langford, F. M. Raymo, J. F. Stoddart, M. Venturi, J. Am. Chem. Soc. 2000, 122, 3542–3543; b) P. R. Ashton, S. E. Boyd, A. Brindle, S. J. Langford, S. Menzer, L. Pere´z-García, J. A. Preece, F. M. Raymo, N. Spencer, J. F. Stoddart, A. J. P. White, D. J. Williams, New J. Chem. 1999, 23, 587–602. M. R. Diehl, D. W. Steuerman, H.-R. Tseng, S. A. Vignon, P. C. Celestre, J. F. Stoddart, J. R. Heath, ChemPhysChem 2003, 4, 1335– 1339. H. Yu, Y. Luo, K. Beverly, H.-R. Tseng, J. F. Stoddart, J. R. Heath, Angew. Chem. Int. Ed. 2003, 42, 5706–5711. a) H. Park, J. Park, A. K. L. Lim, E. H. Anderson, A. P. Alivisators, P. L. McEuen, Nature 2000, 407, 57–60; b) J. W. Park, A. N. Pasupathy, J. I. Goldsmith, A. V. Soldatov, C. Chang, Y. Yaish, J. P. Sethna, H. D. Abruna, C. C. Ralph, P. L. McEuen, Thin Solid Films 2003, 438, 457–461. Y. Chen, G. Y. Jung, D. A. A. Ohlberg, X. M. Li, D. R. Stewart, J. O. Jeppesen, K. A. Nielsen, J. F. Stoddart, R. S. Williams, Nanotechnol. 2003, 14, 462–468. a) M. Lahav, C. Durkan, R. Gabai, E. Katz, I. Willner, M. E. Welland, Angew. Chem. Int. Ed. 2001, 40, 4095–4097; b) R. H. Baughman, C. Cui, A. A. Zakhidov, Z. Iqbal, J. N. Barisci, G. M. Spinks, G. G. Wallace, A. Mazzoldi, D. De Rossi, A. G. Rinzler, O. Jaschinski, S. Roth, M. Kertesz, Science 1999, 284, 1340– 1344; c) R. Pelrine, R. Kornbluh, Q. Pei, J. Joseph, Science, 2000, 287, 836–839. a) S. Chia, J. Cao, J. F. Stoddart, J. I. Zink, Angew. Chem. Int. Ed. 2001, 40, 2447–2451; b) R. Hernandez, H.-R. Tseng, J. W. Wong, J. F. Stoddart, J. I. Zink, J. Am. Chem. Soc. 2004, 126, 3370–3371. a) D. S. Goodsell, Our Molecular Nature: The Body’s Motors, Machines, and Message, Copernicus, USA, 1996; b) P. D. Boyer, J. Biol. Chem. 2002, 277, 39045–39061. T. J. Huang, H.-R. Tseng, L. Sha, W. Lu, B. Brough, A. H. Flood, B.-D. Yu, P. C. Celestre, J. P. Chang, J. F. Stoddart, C.-M. Ho, Nano Lett. 2004, in press. A. Ulman, An Introduction to Ultrathin Organic Films from Langmuir-Blodgett to Self-Assembly, Academic Press, Inc, San Diego, 1991.

463

18

Vapor-Based Polymerization of Functionalized [2.2]Paracyclophanes: A Unique Approach towards Surface-Engineered Microenvironments Doris Klee, Norbert Weiss, and Jörg Lahann

18.1

Introduction

For several high-tech. applications, there is a strong demand for thin-film polymer coatings. Among those are advanced biomaterials [1], insulating layers in integrated circuits [2] and thin-film transistors [3], light-emitting diodes [4, 5], optical or microelectrical–mechanical systems (MEMS) [6], lasers [7, 8], waveguides [9], and photodiodes [10]. Poly(p-xylylenes) belong to a small group of special polymers that are under investigation for thin-film applications. Some members of this polymer family have gained commercial acceptance because they exhibit high solvent resistance, low dialectical constants and good barrier properties [11]. Although poly(p-xylylenes) have been prepared conventionally, e.g. via an electrochemical method [12], chemical vapor deposition (CVD) polymerization of [2.2]paracyclophanes is their preferred preparation method when their intended use is as a surface coating [13]. Generally, CVD polymerization is a room-temperature process, which requires no catalyst, solvent or initiator, and ideally no byproducts are created. Well-defined and usually chemically robust polymer films are prepared by this method. Beside [2.2]paracyclophanes, a,a'-dihydroxy-p-xylylenes [14], a,a'-dibromo-p-xylylenes [15], and a,a'-diacetoxy-p-xylylenes [16] were alternatively chosen as monomers for CVD polymerization and resulted in poly(p-xylylenes) with comparable properties. In addition, CVD polymerization of [2.2]paracyclophane was used in an elegant approach to growing patterned films of poly(p-xylylene) [17]. Although CVD polymerization has been known for more than 30 years [18], the exploitation of [2.2]paracyclophanes for generating functionalized surfaces has only recently been realized [19]. Similarly, germanium- and siliconsubstituted poly(p-xylylenes) were prepared [20, 21]. We have shown that biomolecules can be covalently attached to CVD polymers patterned by lCP [22].

Modern Cyclophane Chemistry. Edited by Rolf Gleiter, Henning Hopf Copyright © 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 3-527-30713-3

464

18 Vapor-Based Polymerization of Functionalized [2.2]Paracyclophanes

18.2

Synthesis of Functionalized [2.2]Paracyclophanes

Fig. 18.1 summarizes synthesis routes resulting in [2.2]paracyclophanes with functional groups that may be of interest for targeted surface modification, such as the immobilization of biomolecules. Friedel-Crafts acylation of [2.2]paracyclophane (1) with trifluoroacetic acid anhydride, using an excess of AlCl3, resulted in 4-trifluoroacetyl[2.2]paracyclophane (2) in 92% yield. Subsequent hydrolysis of (2) under basic aqueous conditions led to 4-carboxy[2.2]paracyclophane (3) (93% yield). Acid-catalyzed esterification with methanol resulted in [2.2]paracyclophane-4-carboxylic acid methyl ester (4) in yields of 60%. [2.2]Paracyclophane-4-carboxylic acid pentafluorophenolester (5) was then synthesized by conversion of 3 with pentafluorophenol trifluoroacetate [23]. As shown in Fig. 18.1, 3 also served as a substrate in the synthesis of 4-hydroxymethyl[2.2]paracyclophane (6) and its derivatives. Reduction of 3 with lithium aluminum hydride in tetrahydrofuran yielded 95% of 6. Esterification of 6 under mild conditions with either acetic acid anhydride, trifluoroacetic acid anhydride (0 8C) or triflic acid anhydride (0 8C) resulted in (methylcarbonyloxy)[2.2]paracyclophane (7), (trifluoromethylcarbonyloxy)[2.2]paracyclophane (8), and (trifluoromethylsulfoxyloxy)[2.2]paracyclophane (9) respectively. 4-Methoxymethyl[2.2]paracyclophane (10) can be synthesized from 6 by etherification. Trimethyloxonium tetrafluoroborate in combination with the bulky tertiary base bis(dimethyl amino)naphthalene as catalyst proved itself as an efficient methylation agent under gentle reaction conditions (yield: 69%). Although amino-substituted polymers are interesting polymers for biomedical applications, their synthesis by CVD is limited by the lack of facile and effective syntheses of amino-functionalized [2.2]paracyclophanes. 4-Amino[2.2]paracyclophane (13) is commonly synthesized from 1 by a five-step synthesis [24] via [2.2]paracyclophanecarboxylic acid and Curtius rearrangement. This synthesis provides poor yields of 13, and pseudo-meta-diamino[2.2]paracyclophane (14) has yet to be synthesized following this approach. Alternatively, 13 was recently prepared from 4-bromo[2.2]paracyclophane by metalation with butyllithium and successive amination in 46% yield [25]. The direct synthesis of amino[2.2]paracyclophanes by nitration and subsequent reduction of the nitro[2.2]paracyclophanes suffers from the poor resistance of [2.2]paracyclophanes to harsh oxidation conditions and a tendency to polymerization. In early studies, Cram et al. reported the synthesis of nitroparacyclophanes by treatment of [2.2]paracyclophane with mixtures of nitric acid and sulfuric acid resulting in 26% mono 2 a and 8% dinitro product 12 [26]. In addition to poor yields, purification of the products from polymeric byproducts was difficult and limited the broad application of this route. Alternatively, treatment of anhydrous nitric acid with trifluoromethanesulfonic acid delivers free nitronium ions [27] which exhibit high nitration power even at low temperatures. We found that 1 is completely nitrated at temperatures as low as –78 8C. The low temperatures and short reaction times mean that side reactions, like oxidation or polymerization of 1, are not favored and were not observed. As a result, yields of 12 were increased from 8% to 93%.

18.2 Synthesis of Functionalized [2.2]Paracyclophanes

i: ii: iii: iv:

AlCl3, TFAA; aq. KOH, reflux methanol, aq. HCl, reflux pentafluorophenol trifluoroacetate, pyridine v: LAH, reflux vi: acetic acid anhydride, pyridine 130 8C vii: TFAA, 0 8C viii: triflic acid anhydride, dichloromethane, 0 8C

465

ix: bis(dimethylamino) naphthalene, trimethyloxonium tetrafluoroborate x: superacidic condition, HNO3 xi: Mg, propargylic bromide xii: propargylic bromide CuCl, H xiv: acetylenedicarboxylic acid dimethylester, H xv: anhyd. H2SO4 xvi: NaBH4

Fig. 18.1 Synthesis of functionalized [2.2]paracyclophanes [28]

These nitration conditions could be adjusted to synthesize either 11 or 12 selectively. Using the superacidic ion exchange resin Nafion®/nitric acid, 11 is obtained in 95% yield. Synthesis of dinitro[2.2]paracyclophane 12 is best carried out with stirring for 30 min at –78 8C and an additional 2 h at –20 8C. Only traces of 11 were found under these conditions. 12 was determined to be mainly pseudopara dinitro[2.2]paracyclophane, as it usually contains around 25% of the pseudometa isomer. Other isomers were produced in less than 2% yield, as shown by GC analysis. This ratio was not affected by the subsequent reduction of the nitro[2.2]paracyclophanes to mono- and diamino[2.2]paracyclophane. Reduction was carried out with tri-iron dodecacarbonyl under toluene/aqueous KOH phase transfer conditions in the presence of [18]crown-6 as phase transfer catalyst [28]. 4,5,12,13-tetrakis(methoxycarbonyl)[2.2]paracyclophane (15) was obtained by Diels–Alder reaction of acetylenedicarboxylic acid dimethylester with hexatetraene, which was previously generated by the coupling of propargylic bromide with its Grignard compound allenylic magnesium bromide [29]. Treatment of 15 with conc. sulfuric acid resulted in [2.2]paracyclophane 4,5,12,13-tetracarboxylic acid dianhydride (16). [2.2]Paracyclophane 4,5,12,13-tetracarboxylic c-butyryl lactone (17) was then synthesized from 16 by cautious reduction with sodium borohydride. Using 4-amino[2.2]paracyclophane (13) as a CVD substrate for amino functional poly-p-xylylene coatings resulted in polymers with aromatic amino groups. Their reactivity towards electrophiles resembled aniline, and straightforward conversion

466

18 Vapor-Based Polymerization of Functionalized [2.2]Paracyclophanes

Fig. 18.2 Synthesis of 4-aminomethyl[2.2]paracyclophane 20

is typically only observed with activated reactants. In terms of polymer analog surface modifications, CVD polymers containing more nucleophilic amino groups were desirable. In an attempt to functionalize surfaces with CVD polymers presenting aliphatic amino groups, 4-aminomethyl[2.2]paracyclophane was selected as being the structurally simplest [2.2]paracyclophane derivative with an aliphatic amino function (Fig. 18.2). 4-Hydroxymethyl[2.2]paracyclophane (6) served as a starting material for the three-step synthesis. First, the conversion of 6 with phosphorus tribromide resulted in 4-bromomethyl[2.2]paracyclophanine (18) in 97 % yield. 4-Azidomethyl[2.2]paracyclophane (19) was then obtained by conversion of 18 with sodium azide. When carried out at 70 8C in dimethylformamide (DMF), quantitative reaction occurred. Finally, 4-aminomethyl[2.2]paracyclophane (20) could be obtained in high yields (99%) by catalytic hydrogenation of 19. The reaction was carried out in methanol using a Pd/carbon catalyst. The structures of the [2.2]paracyclophanes 18 to 20 were identified using 1H and 13C nuclear magnetic resonance (NMR), infrared spectroscopy (IR), and matrix-assisted laser-desorption/ionization time-offlight mass spectrometry (MALDI-TOF-MS) [30].

18.3

CVD Polymerization of Functionalized [2.2]Paracyclophanes

Chemically reactive polymer films can be prepared by chemical vapor deposition (CVD) polymerization, a method that constitutes an essentially substrate-independent platform for surface modification [19, 31]. Thus, CVD coating was utilized for the bioactive modification of implant materials using functionalized [2.2]paracyclophanes. If the chemical integrity of functional groups is not compromised during CVD polymerization, chemical starting points for subsequent modification of the material surface are created.

18.3 CVD Polymerization of Functionalized [2.2]Paracyclophanes

The basic mechanism of CVD polymerization has been known for more than 30 years, and certain poly(p-xylylenes) find commercial application as barrier layers registered under the Parylene trademark. In the CVD process, originally developed by Gorham to prepare non-functionalized poly-p-xylylenes (Fig. 18.4), the dimer [2.2]paracyclophane is transferred into a pyrolysis zone after its sublimation. Control of polymerization parameters allows selective cleavage of the C–C single bonds resulting in the corresponding quinodimethanes [32, 33]. Subsequently; the monomer is transferred to a cooler deposition chamber where it spontaneously polymerizes (see Fig. 18.3). Optimized polymerization conditions are summarized for each [2.2]paracyclophane in Tab. 18.1. Since the temperature of the substrate is relatively low (< 25 8C), even temperature-sensitive substrates can be employed for coating without affecting the substrate material. The starting material is sublimed under reduced pressure into the sublimation zone that was heated between 220 and 355 8C depending on the individual [2.2]paracyclophane. The sublimed material is then transferred to the pyrolysis zone and is exposed to temperatures between 600 and 750 8C using pressures between 0.07 and 0.2 mbar to ensure cleavage of the C–C bonds resulting in the corresponding para-quinodimethanes (monomers). In the last step, the monomers polymerize on the substrate at temperatures typically below 25 8C. Exploitation of functionalized [2.2]paracyclophanes for CVD polymerization is generally limited by the requirement to preserve integrity of the functional groups under the conditions of para-quinodimethane creation in the pyrolysis zone. CVD polymerization of a specific [2.2]paracyclophane therefore requires optimum adjustment of the reaction conditions to minimize decomposition. Thus, the experimental set-up is designed to individually control several polymerization parameters, such as pyrolysis temperature, pressure, gas flow, substrate temperature, and temperature and heating rate of the sublimation zone. The pyrolysis temperature is critical for the quality of the reactive coatings: For sensitive polymers such as esters, ketones, or triflates, pyrolysis is best conducted at temperatures below 670 8C rather than the 750 8C used for the non-functionalized [2.2]paracyclophane. On the other hand, amine 20, or lactone 17 were deposited as homogeneous films with excellent properties at temperatures above 720 8C. Furthermore, flow rate and pressure are related by a delicate balance that is also critical for optimum polymerization results. An increase in argon flow rate at a constant pressure reduces the concentration of the precursor in the gas phase. This reduces interactions between [2.2]paracyclophanes and diminishes intermolecular side reactions. On the other hand, a decrease in pressure at a constant argon flow rate causes decreased sublimation and results in lower monomer concentrations in the gas phase. In this case, [2.2]paracyclophane reaction time inside the pyrolysis chamber is extended, resulting in both enhanced cleavage of the C–C bonds and possible decomposition of functional groups. Generally, low pressures between 0.12 and 0.2 mbar were used. However, if pyrolysis temperatures above 720 8C are employed, pressures can be as low as 0.05 mbar. In addition, argon was used as carrier gas and the argon flow rate varied substantially between polymer prepara-

467

Polymer (#)

2a 2b 2c 2d 2e 2f 2g 2h 2i 2j 2k 2l 2m 2n

Functionality

Alcohol Methylether Acetate Methylester Tetramethylester Anhydride Lactone Amine Diamine Pentafluorophenol ester Trifluoroacetate Trifluoromethylketone Triflate Unsubstituted

1f 1j 1g 1d 1o 1p 1q 1m 1n 1e 1h 1b 1i 1a

Precursor (#)

100 99 100 100 100 95 98 100 95 99 95 99 100 100

Purity of precursor (%) 56 30 55 50 50 36 30 59 39 50 70 60 60 20

Amount of precursor (mg)

Tab. 18.1 Representative parameters used for polymer preparation by CVD polymerization

220 240 260 275 325 340 355 270 280 230 248 240 340 270

Sublimation temp. ( 8C) 750 720 610 670 670 620 750 700 620 600 720 650 650 750

Pyrolysis temp. ( 8C) 20 20 15 15 15 12 12 20 15 12 15 20 12 20

Substrate temp. ( 8C) 0.08 0.1 0.2 0.11 0.12 0.2 0.05 0.2 0.12 0.12 0.10 0.12 0.07 0.07

System pressure (mbar)

10.0 0.5 1.2 2.0 2.0 4.5 5.0 15.0 2.0 2.0 8.0 2.0 0.9 7.0

Argon mass flow (sccm)

468

18 Vapor-Based Polymerization of Functionalized [2.2]Paracyclophanes

18.3 CVD Polymerization of Functionalized [2.2]Paracyclophanes

tions. Highest flow rates were used when high pyrolysis temperatures and moderate pressures were combined. In the case of polymer PPX-PFE, carefully purified [2.2]paracyclophane 5 is sublimed under a reduced pressure of 0.2 mbar at temperatures between 120 and 130 8C. Sublimed 5 was transferred to the pyrolysis zone, which was heated to 600 8C to ensure cleavage of the C–C bonds resulting in the corresponding quinodimethanes. In the last step, monomers were adsorbed on the substrate at temperatures around 45 8C and spontaneously polymerized (see Fig. 18.3). CVD polymerization of 5 results in transparent and topologically uniform polymer films of thicknesses between 90 and 600 nm. The film thickness is mainly determined by the amount of 5 used for polymerization. The thickness of a film produced by the deposition of 30 mg of 5 was examined by means of spectroscopic ellipsometry (SE) and was determined to be 190.0(± 5.8) nm. Atomic force microscopy was used to characterize the surface topology: The root-mean square roughness was determined to be 0.4 nm (1 lm2 spot). The reactive coating shows excellent chemical stability in a dry air environment. Adhesion of the reactive coating to the gold substrate was examined by gently pressing a 1 cm2 area of a Scotch tape onto the polymer coating. After subsequently peeling off the tape, the sample was examined by optical microscopy and IR spectroscopy and was mechanically and chemically intact. Poly(p-xylylenes) are typically optically anisotropic polymers and show optical birefringence [34]. The polymer chains are aligned in a preferred orientation and exhibit strong anisotropic molecular polarizability caused by their main-chain phenyl groups. Benzene is a planar molecule with a delocalized p-electron system that has an anisotropic molecular polarizability of Da = 5.62 Å3 [35]. Substitution of the benzene rings in the main-chain of poly-p-xylylenes alters the electronic properties of the polymers and changes its anisotropic molecular polarizability. Therefore, the electronic nature of the functional groups can be expected to directly influence

Fig. 18.3 Mechanism of CVD polymerization of functionalized [2.2]paracyclophanes; scheme of the Gorham process for CVD polymerization

469

470

18 Vapor-Based Polymerization of Functionalized [2.2]Paracyclophanes

the optical properties of the polymer films. In fact, uniaxial behavior was reflected by determination of extraordinary and ordinary indices of refraction. Experimental data were compared to data generated by a uniaxial Cauchy dispersion model and showed excellent correlation. The observed refractive indices showed a distinct variation with electronic properties of the functional groups. Similarly, optical birefringence depended strongly upon the nature of the functional group as significant variation in optical birefringence was found among functionalized poly(p-xylylenes). With respect to immobilization of biomolecules, amino groups are particularly interesting since they can be used as nucleophilic linking groups for the covalent immobilization of substances to enduringly improve the interface properties of a substrate. Amines 13 and 20 were CVD-polymerized to prepare well-adherent polymeric films on planar substrates with variable chemical nature, such as silicon, stainless steel, titanium and polyethylene (Low Density Polyethylene, LDPE). The resulting polymers were insoluble in water as well as in organic solvents including ethanol, isopropanol, acetone, methylenchloride, chloroform, hexane, and toluene. The chemical composition of the CVD-coating based on CVD-coated LDPEfilms were characterized using X-ray photoelectron spectroscopy (XPS). Tab. 18.2 summarizes the theoretical elementary compositions of amino-ppx and aminomethyl-ppx in comparison with those detected by XPS. The theoretical values were calculated assuming the composition of substrates and polymer are identical, respectively. The experimental results for both surfaces correspond well with the expected composition of the polymers. However, in both cases a slightly reduced portion of nitrogen was detected indicating an enrichment of unsubstituted monomer units in the polymer. The enhanced nitrogen content may be caused by the different deposition temperatures of substituted and unsubstituted monomer units. Since the deposition temperatures of substituted xylylene units are generally higher compared to unsubstituted units, they display a stronger tendency of deposition at warmer areas of the polymerization chamber. Thus on cooled areas, its fraction may be slightly decreased.

Tab. 18.2 Theoretical and experimental elementary compositions using XPS of aminomethyl-ppx

and amino-ppx in atom-% Binding energy in eV

Carbon (1s) C-C C-N, C-O

– 285.0 286.5

Aminomethyl-ppx

Amino-ppx

Calculated

Experimental a)

Calculated

Experimental a)

94.4 88.9 5.6

92.8 86.9 5.9

94.1 88.2 5.9

92.9 85.5 7.4

Oxygen (1s)

532.5

0

2.5

0

2.0

Nitrogen (1s)

399.7

5.6

4.7

5.9

5.1

a) Error margin: ± 0.5 atom-%.

18.3 CVD Polymerization of Functionalized [2.2]Paracyclophanes

Furthermore, small amounts of oxygen were detected for both polymers. Accordingly, the detected carbon species ratio of aminomethyl-ppx bound to more electron-negative elements like oxygen or nitrogen is slightly higher than the theoretical value. Electrochemical surface properties are critical for biomaterials’ interaction with biological media. The considerable influence of the surface charges on the initial protein adsorption, thrombogenicity, and activation of the complement system is undisputed [36–42]. Electrochemical characteristics of a surface can be considerably influenced by surface functionalities, in particular if they contain ionizable groups. Conversely, the electrochemical properties of a surface may give valuable insights into its acid-base behavior and, therefore, provide qualitative information concerning their chemical functionalities [43–45]. Streaming potential measurements were used to study pH-depending zeta potentials of LDPE-films coated by CVD polymerization of [2,2]paracyclophanes 1, 13, and 20 (see Fig. 18.4). Surprisingly, the pH dependence of the zeta potential determined for polymer X resemble that observed for unsubstituted poly-p-xylylene. Even below pH 4, where quasi-complete protonation of the weakly basic, aromatic amino groups may be assumed (cf. pKs (aniline) = 4.63), no significantly higher zeta potential values were found. In contrast, zeta potentials measured for aminomethyl-ppx showed a clear shift towards positive values. The difference is similarly manifested in a drastic shift in the x-axis intercepts indicating the isoelectric point (IEP) of a surface. The IEP is defined as the pH value, at which the zeta potential, and therefore, the surface potential and the net surface charge, respectively, amount to zero. Aminomethyl-ppx with an IEP of 6.1 shows a base shift of 1.6 pH units as compared to amino-ppx (IEP = 4.5). This effect indicates the basic character of aliphatic aminomethyl groups. Although a more thorough discussion would benefit from assessment of the pKa values of the polymer surfaces, the amount of basic groups obviously caused an increase of the zeta potential compared to unsubstituted ppx.

Fig. 18.4 Zeta potentials of CVD-coated LDPE-foils depending on the pH-value (adjusted by HCl or KOH) using the streaming potential method; conductive electrolyte: 0.001 M KCl; n = 6

471

472

18 Vapor-Based Polymerization of Functionalized [2.2]Paracyclophanes

18.4

Immobilization of Bioactive Substances via Functionalized Poly(p-xylylenes) 18.4.1

Introduction

CVD-based polymer coatings are of interest as interfaces for biomedical applications because they have potential to incorporate functional groups. These functional groups can then be used to conjugate biomolecules such as proteins, antigens or cell receptors to implant surfaces [46]. The resulting biomimetic coatings provide interfaces that may allow control of the interactions between biomaterials and organisms to improve the interfacial biocompatibility durably (see Fig. 18.5). There is an increasing demand for bioactive surfaces, particularly in rapidly growing fields of advanced biomaterials, bio- and nanotechnology. CVD-coatings based on amino functionalized polymers like amino- and aminomethyl-ppx are interesting candidates for functionalized surface coatings. In combination with bivalent spacers, covalent binding between the surface and the coupled substance can be achieved. Corresponding to its length, the resulting bridging group can additionally function as spacer molecule in order to generate a steric distance to the surface. This is of particular importance for bioactive substances, whose efficacy requires a sufficient accessibility, e.g. for cell membrane receptors. The NHS-activated ester method is often used for covalent binding of bioactive substances to surfaces. This method is adapted from peptide chemistry [47] and can also be used as polymer analogous modification of amino functionalized polymer surfaces. A biomolecule to be immobilized (carboxyl component) is transformed to a reactive ester (activated ester) and reacts with the amino component engaging in a stable carboxylamide linkage. N-hydroxylsuccylimidylesters (NHSesters) serve as ester component characterized by a high reactivity against primary and secondary amino groups (see Fig. 18.6). To determine the amount of amino groups reactive to NHS-esters, silicon wafers, CVD-coated with amino-ppx and amino methyl-ppx, were treated with sulfosuccinimidyl-4-O(4,4'-dimethoxytrityl)-butyrate (Sulfo-SDTB). After cleavage of the surface-bound DTB the amount of released 4,4'-dimethoxytrityl cation was quantified calorimetrically.

Fig. 18.5 Strategy for surface biomimetics as a tool towards enhanced biocompatibility via spacer chemistry

18.4 Immobilization of Bioactive Substances via Functionalized Poly(p-xylylenes)

Fig. 18.6 Reaction of NHS-ester with a primary amino group

As shown in Fig. 18.7 the treatment with the active ester leads to a high amount of surface-bound DTB on the amino-functionalized surfaces compared with the surfaces without functional groups. Treatment with nonactive DTB led to lower amounts of released 4,4'-dimethoxytrityl cation. In case of the amino-ppx, there is a distinct indication of covalent bonding of DTB using the reactive ester on the amino-functionalized surfaces. The fivefold higher release from the amino methyl-ppx indicates the higher reactivity of the aliphatic amino groups. Although amino-functionalized interfaces were successfully exploited in some applications, the additional activation step not only limits the feasibility of microengineering, but also causes the contamination of the substrate with organic sol-

Fig. 18.7 Release of 4,4'-dimethoxytrityl cation from untreated and CVD-coated silicon wafer surfaces

473

474

18 Vapor-Based Polymerization of Functionalized [2.2]Paracyclophanes

Fig. 18.8 Fluorescence micrographs of (a) a gold substrate coated with ppx-PPFE and patterned with antibody 23 and (b) a PTFE substrate coated with ppx-anhydride and patterned with antibody 23. Both samples were immunoassayed with a fluorescein-labeled secondary antibody against 23. The width of each individual line is 50 lm

vents and volatile compounds. These contaminations reduce crucial advantages of CVD coatings, such as low intrinsic cytotoxicity due to the absence of harmful solvents, initiators, or accelerators during polymerization. Therefore, an one-step coating procedure that provides linkable reactive groups is highly desirable. For this reason, the technological platform was extended by introduction of so-called reactive coatings, that is poly(p-xylylene carboxylic acid pentafluorophenolester-co-pxylylene) [48] (PPX-PPFE) and poly(p-xylylene-2,3-dicarboxylic acid anhydride) [23]. Without the need for further activation, the high chemical reactivity of their functional groups supported rapid conversion with biological ligands or proteins and was used for surface patterning using microcontact printing [49]. (+)-Biotinyl-3,6,9-trioxaundecanediamine (21) was used for lCP of different patterns on gold substrates coated with ppx-PPFE (Fig. 18.8). Biotin-based ligands were chosen, since biotin is a prototype of a small ligand. Its interaction with streptavidin is characterized by strong noncovalent interaction and allows the facile patterning of streptavidin on the surface. Streptavidin is a widely used immobilization protein that has two pairs of binding sites on opposite faces and therefore represents a universal platform for further patterning of biotin-labeled biomolecules. Fluorescein-labeled streptavidin can be used to examine the micropatterns.

18.4 Immobilization of Bioactive Substances via Functionalized Poly(p-xylylenes)

18.4.2

Surfaces for Blood Contact

Although the use of metallic stents is a widely accepted alternative therapy to percutaneous transluminal coronary angioplasty, the main complications are believed to be caused by properties of the metal surface, such as high thrombogenicity and high free surface energy [50]. An interesting approach to overcome these limitations is to cover the metal surface with a bioactive surface. This approach corresponds to a recent trend in biomaterials’ research combining desirable mechanical properties of an existing material with improved biocompatibility of another. In order to prevent unwanted reactions, one approach consists of the immobilization of anticoagulants on the surface of the biomaterial. An anticoagulant gaining clinical acceptance is r-hirudin, a recombinant protein consisting of 65 amino acid residues (6964 Da). r-Hirudin is the strongest known thrombin inhibitor forming a non-covalent complex with a 1 : 1 molar ratio [51]. The mimicry of the metal surface is based on a two-step procedure. During a preliminary step, the metal coronary implants are coated with a functionalized polymer interface by means of CVD polymerization. The anchor groups available on the polymer surface are subsequently used for the conjugation of the bioactive molecule r-hirudin. These polymer coatings were shown to fulfill the chemical and physical requirements being necessary for the application as a stent coating. These polymer coatings can then be used as interfaces that provide functional groups for the conjugation of biomolecules. The feasibility of the concept was proven by the bioactive immobilization of r-hirudin. Surface bound r-hirudin was shown to improve the blood compatibility of the bioactively coated metallic coronary implants as shown in vitro by a decreased thrombogenicity and a strongly reduced platelet adhesion (see Fig. 18.9).

Fig. 18.9 Influence of the r-hirudin coating on platelet adhesion as shown by comparison of a hydroxy methyl-ppx-coated stainless steel foil (top) with the r-hirudin equipped and hydroxy methyl-ppx-coated foil after r-hirudin immobilization (bottom)

475

476

18 Vapor-Based Polymerization of Functionalized [2.2]Paracyclophanes

18.4.3

Surfaces for Tissue Contact

Deficient biocompatibility of medical implants is commonly caused by their surface properties. The poor integration of an implant is often due to the lack of specific interactions between its surface and the cells of the surrounding tissue. Such interactions can be induced by the immobilization of bioactive substances onto the surface. Surfaces with tissue-integrative properties are of particular interest. This effect can be achieved by a covalent immobilization of cell adhesion or proliferation stimulating substances. The positive effect e.g. of covalent-bound fibronectin, collagen or synthetic peptides like GRGDS and YRGDS containing the fibronectin cell adhesion sequence RGD is verified in in-vitro cell culture tests [52, 53]. Recombinant human insulin was chosen as bioactive substance to be immobilized covalently. The tethered proteohormone was expected to show a cell growth promoting activity [54, 55]. In order to achieve proliferative effects on an implant surface, insulin might represent a more convenient alternative compared to real growth factors. For its immobilization the insulin molecule presents three amino groups located at the sequence positions A1, B1 and B29 (see Fig. 18.10). Among those, the B1 amino group is most distant from the receptor binding site. Therefore, aside from an unselective coupling, it was also intentional to achieve a selective B1tethering by use of a proper protective group chemistry. For surface modification of titanium or stainless steel with insulin samples were amino-functionalized by chemical vapour deposition (CVD)-polymerization of 13 [56, 57, 46]. The resulting ultra-thin layers of polyamino-p-xylylene-co-poly-pxylylene (amino-ppx) were surface activated by means of the homobifunctional linker hexamethylenediisocyanate (HDI). The samples were subsequently treated with insulin or A1, B29-Msc2-insulin (1 mg/ml) in slightly alkaline-buffered solu-

Fig. 18.10

Structure of insulin

18.4 Immobilization of Bioactive Substances via Functionalized Poly(p-xylylenes)

Fig. 18.11 Surface modification steps of the immobilization of insulin on metal surfaces

tion (see Fig. 18.11). For the removal of Msc-groups the samples were finally treated with 10% piperidine in water. The amounts of remaining protein on insulin-treated surfaces were investigated by means of ELISA and 125I-radiolabeling experiments. HDI-activated amino-ppx samples showed a significantly higher surface concentration of insulin than the non-activated ones. Furthermore, surface MALDI-TOF-MS measurements gave evidence of covalently immobilized protein on preactivated amino-ppx surfaces. Nonactivated amino-ppx samples showed a distinct signal of desorbable insulin, whereas no detection could be obtained from insulin remaining on HDI-activated amino-ppx samples. Cell culture experiments were carried out with human umbilical vein endothelial cells (HUVECs). Results of a preliminary cell test with HUVECs are illustrated in Fig. 18.12. The amounts of remaining protein on insulin-treated surfaces were investigated by means of ELISA and 125I-radiolabelling experiments. HDI-activated amino-ppx samples showed a significantly higher surface concentration of insulin than the non-activated ones. Furthermore, surface MALDI-TOF-MS measurements gave evidence of covalently immobilized protein on preactivated amino-ppx surfaces.

Fig. 18.12 HE-staining after 24 h incubation of HUVECs on amino-ppx-modified stainless steel (left) and on amino-ppx/HDI/insulin-modified stainless steel (right)

477

478

18 Vapor-Based Polymerization of Functionalized [2.2]Paracyclophanes

Non-activated amino-ppx samples showed a distinct signal of desorbable insulin, whereas no detection could be obtained from insulin remaining on HDI-activated amino-ppx samples. Cell culture experiments were carried out with human umbilical vein endothelial cells (HUVECs). Results of a preliminary cell test with HUVECs are illustrated in Fig. 18.12. No cells were attached on amino-ppx samples without insulin. Only insulinmodified surfaces showed a high density of well spread cells. Surface-bound insulin was shown to improve both, cell adhesion and cell proliferation indicating improved tissue compatibility. 18.4.4

Surface Engineering of Microfluidic Devices

Miniaturized cell assays are of interest in the evaluation of pharmacologically active molecules including molecules that affect cell proliferation and adhesion. For fabrication, a PDMS stamp was casted from a photolithographically produced silicon master [58]. Prior to use, the PDMS stamp was oxidized by means of an oxygen plasma. Ligand 21 was printed via PDMS stamp on the substrate surface coated with ppx-PFE with a contact time of 60 s. Subsequently, the remaining pentafluorophenol ester groups were reacted with 2-(aminoethoxy)ethanol (22) to passivate non-printed areas of the surface. Self-assembly of streptavidin was studied on chemically distinct substrates such as PTFE, PE, gold, silicon, stainless steel, and glass, all coated with reactive coating ppx-PFE. Fig. 18.6 shows PTFE and gold substrates after lCP of ligand 21 and subsequent incubation with fluorescein-conjugated streptavidin. Successful patterning was verified by fluorescence microscopy and high accuracy with respect to contrast and pattern formation was found. Within the range of substrates in this study, the formation of patterns was independent from the substrate material. Alternatively, patterns of streptavidin were formed by immediately printing streptavidin to the reactive coating. In this case, primary amino groups of streptavidin were linked to the reactive coating. Fig. 18.13 shows the AFM image of a surface that was modified by lCP of streptavidin onto a silicon substrate coated with polymer 11. Again, the created patterns were characterized by high accuracy. The quality of the generated patterns compare well with those generated by lCP of ligand 21 and subsequent self-assembly of streptavidin (Fig. 18.14). We next assessed whether the driving force of this system to self-assemble can be used to manipulate nanometer-sized objects that are chemically conjugated to streptavidin. To explore a biologically relevant system with dimensions on the micron-scale, interaction of patterned surfaces with endothelial cells was studied. Our motivation was to assess whether the established platform allows spatially defined selfassembly of antibodies that bind to features expressed on a cell surface and whether those antibodies could be used to guide cells to defined locations on the substrate.

18.4 Immobilization of Bioactive Substances via Functionalized Poly(p-xylylenes)

Fig. 18.13 AFM image of a gold substrate coated with ppx-PFE after lCP with ligand 21 and layer-by-layer self-assembly of streptavidin and biotin-derived antibody 23

It is well known that cell adhesion is a specific process [59] that involves interactions between cell adhesion mediators (fibronectin, laminin, collagens etc.) and cell surface receptors (CSRs) comprising cadherins, integrins, immunoglobulins, or selectins [60, 61]. CSRs possess specific binding sites for extracellular matrix proteins and their expression varies with cell type. While some CSRs are expressed almost universally (e.g. laminin receptors), others are specific for a cell type. The fibronectin receptor VLA-5 is expressed only by a few cell types including endothelial cells, epithelial cells, platelets, and fibroblasts. VLA-5 is formed on the cell surface by association of b1-integrin with a5-integrin being itself a dimer of 135/25 kDa. a5-Integrin was chosen as target for its role in cell adhesion and its relative specifity for endothelial cells. Our strategy based on the hypothesis that a surface-bound antibody against the VLA-5 receptor fragment a5-integrin (human anti-a5-integrin) would induce attachment of endothelial cells. To verify this hypothesis, adsorption of biotin-conjugated human anti-a5-integrin (23) on a pre-structured surface was investigated. In this case, we altered our strategy slightly, by first allowing streptavidin to bind to the surfaces modified by lCP with biotin-ligand 21 (see Fig. 18.14). To monitor the result of this adsorption study, surface-bound antibody 23 was marked with a fluorescein-conjugated secondary antibody that recognizes the heavy chain of antibody 23. Fig. 18.15 shows fluorescence micrographs of two substrates after spatially controlled immobilization of antibody 23 and subsequent association with the secondary antibody. The fluorescein-labeled secondary antibody was found to bind only at areas patterned with antibody 23. Fig. 18.15 reveals sharp contrast between binding and non-binding regions. The micrograph also demonstrates homogeneous and reproducible distribution of antibody 23 on the biotin-terminated areas of the surface. We then used this method for confinement of endothelial cells in the lumen of a microchannel and studied the in vitro activity of echistatin, a potent disintegrin

479

480

18 Vapor-Based Polymerization of Functionalized [2.2]Paracyclophanes Fig. 18.14 Schematic representation of the surface modification steps that lead to a microdevice with a biologically active surface; PDMS is first modified with a reactive coating, which is then used to bind biotin ligands and to self-assemble streptavidin. Biotin-labeled antibody 23 is then bound to the modified PDMS surface and used to study cell surface receptor activity

and cell adhesion inhibitor. Immobilization of an antibody that specifically captures a5-integrin (HAI) to a biotin-modified, ppx-PFE-coated microdevice was used to confine endothelial cells inside the microfluidic system. Suspensions of endothelial cells containing increasing concentrations of echistatin were perfused into the microchannels presenting a5-integrin-capturing microchannels. After 4 h seeding time, non-adherent cells were separated by thorough rinsing and the dose-dependent activity of echistatin was assessed by determining the number of attached endothelial cells after trypsin treatment (Fig. 18.16). Fibronectin-coated tissue culture polystyrene (physisorbed fibronectin) was used as reference surface. The

Fig. 18.15 Fluorescence micrograph of the microfluidic device coated with ppx-PFE after immobilization of amino-terminated biotin and self-assembly of fluorescein-conjugated streptavidin

18.5 Conclusions

Fig. 18.16 Dose-dependent activity of the cell adhesion inhibitor echistatin. Echistatin’s biological activity is based on tight binding to a5-integrin and decreases cell adhesion

minimum concentration for biological activity was determined to be 0.1 lg ml–1 (18.4 lM) and was equal for both, antibody-coated and fibronectin-coated surfaces. Above this minimum inhibition of cell adhesion increased logarithmically (R2 = 0.962) with the concentration of echistatin. For higher concentrations, the inhibition was higher on the antibody-coated surfaces than on the reference surfaces revealing a higher selectivity, with which cells bind to the antibody-coated surfaces. These results provide first evidence of the applicability of the above described systems as cell-based biosensors for chemical or biological agents [63] or microfluidic-based cell assays [64].

18.5

Conclusions

CVD polymerization of functionalized [2.2]paracyclophanes establishes a general, but simple protocol for preparation of polymer films. The simplicity in providing a wide range of functional groups, the excellent adhesion to various substrates, and its applicability to devices with three-dimensional geometries are key advantages when compared to polymers deposited by solvent-based methods. A one-step coating procedure that provides linkable reactive groups without requiring posttreatment of the deposited films with harsh chemicals is highly desirable for assay development. The reactive coatings have anchor groups with reactivity pattern that enable selective binding of biomolecules using aqueous chemistry. This eliminates contaminations and guarantees a biocompatible process being free of harmful solvents, initiators, or accelerators. These reactive coatings can improve the interfacial biocompatibility of implant surfaces or can be compatible with complex

481

482

18 Vapor-Based Polymerization of Functionalized [2.2]Paracyclophanes

biological features as they represented a designable interlayer stable under the conditions of the bioassay. Insulin was covalently immobilized on CVD-coated surfaces to enhance attachment and growth of cells in vitro. The presented surface modification is widely applicable, because the procedure can be performed on various implant materials, metals as well as polymers. By creating a microfluidic device with a biotin-presenting interior surface a wide range of bioassays can potentially be addressed. While overcoming restrictions associated with conventional PDMS-based microfluidic devices, the methodology retains PDMS-intrinsic advantages, e.g. processability by rapid prototyping, broad availability, and low costs. The variability in functional groups that can be prepared by CVD polymerization allows the application-driven surface engineering of microfluidic devices. Reactive functional groups, such as those in PPX-PFE, allow for immobilization of a wide variety of biomolecules; amino- or carboxylic acid groups may control surface charges and electro-osmotic flows; and alkyl groups may provide hydrophobic interfaces for electrochromatographic applications.

18.6

References 1

2 3 4

5 6

7 8

9

J. Lahann, D. Klee, H. Thelen, H. Bienert, D. Vorwerk, H. Höcker, J. Mater. Sci.: Mater. Med. 1999, 10, 443. N. Majid, S. Dabral, J. F. L. McDonald, Electron. Mater. 1988, 18, 301. H. Stirringhausen, N. Tessler, R. H. Friend, Science 1998, 280, 1741. J. H. Burroughes, D. D. C. Bradley, A. R. Brown, R. N. Marks, K. Macay, R. H. Friend, P. L. Burns, A. B. Holmes, Nature 1990, 347, 539. D. Braun, A. J. Heeger, Appl. Phys. Lett. 1991, 58, 1982. M. A. Burns, B. N. Johnson, S. N. Brahmasandra, K. Handique, J. R. Webster, M. Krishnan, T. S. Sammarco, P. M. Man, D. Jones, D. Heldsinger, C. H. Mastrangelo, D. T. Burke, Science 1998, 282, 484. N. Tessler, G. J. Denton, R. H. Friend, Nature 1991, 382, 695. F. Hide, M. A. Diaz-Garcia, B. J. Schwartz, M. R. Anderson, Q. Pei, A. J. Heeger, Science 1996, 273, 1833. J. Gaynor, S. B. Desu, J. Mater. Res. 1994, 9, 3125.

10

11 12 13 14

15

16

17

18 19 20

G. T. Yu, J. Gao, J. C. Hummelen, F. Wudl, A. J. Heeger, Science, 1995, 270, 1789. A. Greiner, Trends Polym. Sci. 1997, 5, 12. J. H. P. Ultley, Y. Gao, J. Gruber, R. Lines, J. Mater. Chem. 1995, 5, 1297. W. F. Gorham, J. Polym. Sci. Part A–1 1966, 4, 3027. P. Simon, S. Mang, A. Hasenhindl, W. Gronski, A. Greiner, Macromolecules 1998, 31, 8775. L. You, G. R. Yang, C. I. Lang, P. Wu. J. A. Moore, J. F. McDonald, T. M. J. Lu, Vac. Sci. Technol. A 1993, 11, 3047. C. Schmidt, V. Stuempfen, J. H. Wendorff, A. Hasenhindl, W. Gronski, M. Ishaque, A. Greiner, Acta Polymer. 1998, 49, 232. K. M. Vaeth, R. J. Jackman, A. J. Black, G. M. Whitesides, K. F. Jensen, Langmuir 2000, 16, 8495. W. F. Gorham, Y. L. Yeh, J. Org. Chem. 1969, 34, 2366. J. Lahann, D. Klee, H. Höcker, Macromol. Rapid Commun. 1998, 19, 441. H. Hopf, G. N. Gerasimov, S. A. Chvalun, V. L. Rozenberg, E. L. Popova, E. V.

18.6 References

21

22

23

24

25

26 27 28 29 30 31 32 33 34 35 36 37 38 39 40

41

Nikolaeva, E. I. Grigoriev, S. A. Zavjalov, L. I. Trakhtenberg, Adv. Mater., Chem. Vap. Deposition 1997, 3, 197. G. N. Gerasimov, E. L. Popova, E. V. Nikolaeva, S. N. Chvalun, E. I. Grigoriev, L. I. Trakhtenberg, V. I. Rozenberg, H. Hopf, Macromolecul. Chem. Phys. 1998, 199, 2179. J. Lahann, I. S. Choi, J. Lee, K. F. Jensen, R. Langer, Angew. Chem. Int. Ed. Engl. 2001, 40, 3166. J. Lahann, M. Balcells, T. Rodon, J. Lee, I. S. Choi, K. F. Jensen, R. Langer, Langmuir 2002, 18, 3632. J. F. Waters, J. K. Sutter, M. A. B. Meador, L. J. Baldwin, M. A. Meador, J. Polym. Sci. Part A: Polym. Chem. 1991, 29, 1917–1924. A. Cipiciani, F. Fringuelli, V. Mancini, O. Piermatti, F. Pizzo, R. Ruzziconi, J. Org. Chem. 1997, 62, 3744–3747. H. J. Reich, D. J. Cram, J. Am. Chem. Soc. 1969, 91, 3527–3533. G. A. Olah, S. C. Narang, Synthesis 1978, 9, 690–691. J. Lahann, R. Langer, Macromolecules 2002, 35(11), 4380–4386. H. Hopf, Angew. Chem. 1972, 84, 471. N. Weiss, PhD Thesis, RWTH Aachen (2003) to be published. J. Lahann, W. Plüster, D. Klee, H. Höcker, Biomaterials 2001, 22, 817. W. F. Gorham, J. Polym. Sci. Part A-1 1966, 4, 3027. A. Greiner, Trends Polym. Sci. 1997, 5, 12. J. J. Senkevich, S. B. Desu, V. Simkovic, Polymer 2000, 41, 2379. M. J. S. Dewar, J. J. Stewart, Chem. Phys. Lett. 1984, 111, 416. K. Fujimoto, H. Inoue, Y. Ikada, J. Biomed. Mater. Res. 1993, 27, 1559. C. Causserand, M. Nystrom, P. Aimar, J. Membr. Sci. 1994, 88, 211. A. Krajewski, R. Malavolti, A. Piancastelli, Biomaterials 1996, 17, 53. W. Norde, Macromol. Symp. 1996, 10, 35. W. Norde, J. Lyklema in The Vroman effect, eds. C. H. Bramford, S. L. Cooper, T. Tsuruta, VSP, Utrecht, Tokyo 1992. W. Norde, J. Lyklema in Surface and interfacial aspects of biomedical polymers, Vol.

42

43 44 45

46 47 48

49 50 51

52

53 54 55 56 57

58 59

60 61

2: Protein adsorption, ed. J. D. Andrade, Plenum Press, New York 1985. C. Werner, H-J. Jacobasch, G. Reichelt, J. Biomater. Sci. Polym. Ed. 1995, 7, 61. R. H. Ottewill, J. N. Shaw, Disc. Faraday Soc. 1966, 42, 154. T. Healy, R. L. White, Adv. Coll. Interf. Sci. 1978, 9, 303. H.-J. Jacobasch, F. Simon, C. Werner, C. Bellmann, Technisches Messen 1996, 63, 12. J. Lahann, D. Klee, H. Höcker, Materialwiss. Werkstofftech. 1999, 30, 763. G. W. Anderson, J. E. Zimmerman, F. M. Callahan, J. Am. Soc. 1964, 86, 1839. J. Lahann, I. S. Choi, J. Lee, K. F. Jensen, R. Langer, Angew. Chem. Int. Ed. Engl. 2001, 40, 3166. J. Lahann, R. Langer, Polymer Prepints 2001, 42, 113. J. C. Palmaz, Am. J. Roentgenol. 1993, 160, 613–618. T. J. Rydel, K. G. Ravichandran, A. Tulinsky, W. Bode, R. Huber, C. Roitsch, J. W. Fenton II, Science 1990, 249, 277– 280. Y. Hirano, M. Okuno, T. Hayashi, K. Goto, A. Nakajima, J. Biomater. Sci., Polym. Edn 1993, 4, 235–243. N. Weiss, D. Klee, H. Höcker, Biomaterialien, 2001, 2, 81–86. Y. Ito, S. Q. Liu, Y. Imanishi, Biomaterials 1991, 12, 449–453. E. J. Kim, I.-K. Kang, M. K. Jang, Y. B. Park, Biomaterials 1998, 19, 239–249. D. Klee, H. Höcker, Spektrum der Wissenschaft 1999, 1, 76–78. J. Lahann, D. Klee, H. Thelen, H. Bienert, D. Vorwerk. H. Höcker, J. Mater. Sci. Mater. Med. 1999, 10, 443–448. Y. Xia, G. M. Whitesides, Angew. Chem. Int. Ed. Engl. 1998, 37, 550. E. Genui, D. Granchi, G. Cipetti, E. Verri, D. Cavedagna, S. Gamberini, M. Cervellati, A. DiLeo, A. Pizzoferrato, Biomaterials 1997, 18, 489. P. S. Frenette, D. D. Wagner, N. Engl. J. Med. 1996, 334, 1526. A. Huttenlocher, R. R. Sandborg, A. F. Horwitz, Curr. Opin. Cell Bio. 1995, 7, 697.

483

484

18 Vapor-Based Polymerization of Functionalized [2.2]Paracyclophanes G. O. Delwel, A. Sonnenberg, Adhes. Recept. Ther. Targets 1996, 9. 63 S. Takayama, E. Ostuni, P. LeDuc, K. Naruse, D. E. Ingber, G. M. Whitesides, Nature 2001, 411, 1016. 62

64

J. Lahann, M. Balcells, H. Lu, T. Rodon, K. F. Jensen, R. Langer, Anal. Chem. 2003, 75, 2117–2122.

519

20

Molecular Recognition Studies with Cyclophane Receptors in Aqueous Solutions François Diederich

20.1

Introduction

The vast majority of biomimetic molecular recognition studies in aqueous and other polar–protic solutions have been conducted with two classes of receptors, the natural cyclodextrins [G1] and water-soluble cyclophanes. Inclusion complexation in the hydrophobic cavities of large cyclophanes has been intensively investigated over the past two decades, and these studies have been extensively reviewed [2, 3]. Since the early 1990s, efficient water-soluble cyclophane-type receptors have also been constructed by metal-ion-mediated self-assembly [4]. Molecular recognition studies with cyclophanes have made numerous contributions to an enhanced understanding of the intermolecular forces determining chemical and biological association processes. A true highlight has certainly been the discovery of cation–p interactions in host–guest studies with cyclophanes by Dougherty and coworkers [5, 6]. This finding has profoundly influenced biology: cation–p interactions are increasingly observed in biostructural work and first applications in rational lead development and optimization in medicinal chemistry are being described [7]. This chapter reviews molecular recognition studies with water-soluble cyclophane hosts conducted by the Diederich group over the past two decades. After briefly summarizing some important aspects of earlier work on the complexation of aromatic substrates, which has already been reviewed [3, 8], steroid binding studies are presented [9]. Subsequently, the functionalization of the receptors to yield catalytically active cyclophanes is illustrated with the example of an efficient mimic of the enzyme pyruvate oxidase. The article finishes with the description of the recognition properties of dendritically encapsulated cyclophanes (dendrophanes) in which the binding cavity at the core is shielded by dendritic arms of various generations. In order to leave space for an in-depth discussion of the most important molecular recognition results, the preparation of the cyclophane receptors is not described in this chapter and the reader is referred to the cited literature for synthetic protocols.

Modern Cyclophane Chemistry. Edited by Rolf Gleiter, Henning Hopf Copyright © 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 3-527-30713-3

520

20 Molecular Recognition Studies with Cyclophane Receptors in Aqueous Solutions

20.2

Complexation of Aromatic Solutes 20.2.1

Polar Effects in Cyclophane Complexation

Cyclophane 1 (Fig. 20.1) forms axial inclusion complexes with 2,7-disubstituted naphthalenes 2 a–i (Tab. 20.1) in CD3OD (303 K) [10]. The analysis of characteristic complexation-induced changes in chemical shift of the 1H NMR resonances of both binding partners in the formed 1 : 1 host–guest complexes support the preferential formation of two degenerate, rapidly exchanging complex conformers (A and B in Fig. 20.1). We investigated whether polar host–guest effects influence complexation strength. With its electron-rich trialkylanisol moieties, 1 was expected to preferentially complex electron-poor substrates in its molecular box-type cavity. Thus, both the p–p stacking interactions between the guest and the “sandwiching” electron-rich aromatic rings of the host and the intermolecular edge-to-face (C–H(guest)


phost)) aro-

Fig. 20.1 Inclusion complex formed by cyclophane 1 and naphthalene guests 2 a–i (for substituents X and Y, see Tab. 20.1) in CD3OD [10]. The analysis of the complexation-induced changes in chemical shift observed for the resonances of both host and guest in the 1H NMR spectra support the pre-

ferred formation of the two degenerate complex conformations A and B. Characteristic shifts Dd [ppm] of the guest protons: H–C(1,4,5,8): –2 to –3 (“–” = upfield shift), H–C(3,7): – 0.5 to –1.0; of the host protons: H–C(3',9): +0.2 to + 0.5 (“+” = downfield shift), H–C(2,3): *–0.5 to –1.0

20.2 Complexation of Aromatic Solutes Tab. 20.1 Association constants Ka (±10%) and free enthalpies of complexation –DG0 for the 1 : 1 complexes formed between cyclophane 1 and the naphthalene derivatives 2 a–i in CD3OD, T = 303 K

Guest

X

Y

Ka (L mol–1)

–DG0 (kcal mol–1)

(a) Donor donor guests 2a OH 2b NH2 2c OMe

OH NH2 OMe

20 30 50

1.9 2.1 2.3

(b) Donor–acceptor guests 2d NH2 2e OMe 2f OMe

NO2 NO2 CN

100 110 120

2.8 2.8 2.9

(c) Acceptor–acceptor guests 2g COOMe 2h NO2 2i CN

COOMe NO2 CN

190 210 280

3.1 3.2 3.4

matic–aromatic interactions should become enhanced with increasing electron-acceptor strength of the guest substituents. The experimental findings from 1H NMR binding titrations corresponded to these expectations (Tab. 20.1). Acceptor–acceptor-substituted naphthalenes were found to form the most stable complexes, followed by donor–acceptor and then by donor–donor-substituted guests. Thus, the complex of 2,7-dicyanonaphthalene (2 i; Ka = 280 M–1, –DG0 = 3.4 kcal mol–1) is substantially more stable (by DG0 = 1.1 kcal mol–1) than the complex of 2,7-dimethoxybenzene (2c; Ka = 50 M–1, –DG0 = 2.3 kcal mol–1). Clearly, relative complexation strength is controlled by polar effects, which could also be viewed as electron donor–acceptor interactions. This conclusion was further supported by a linear Hammett plot of the complexation free enthalpies as a function of the substituent constant r+p. Interestingly, such correlation could not be observed for similar host–guest systems in water, where specific substituent solvation effects seem to dominate relative complexation strength [11]. Polar contributions to both p–p and edge-to-face aromatic interactions have been extensively and quantitatively investigated in a variety of model studies since the 1990s, and the reader is referred to the relevant literature [6, 12, 13]. 20.2.2

Complexation of Polycyclic Aromatic Hydrocarbons (PAHs) in Water

Cyclophane 3 (Fig. 20.2) with C5-bridges between the two diphenylmethane spacers that provide the necessary preorganization of the cavity binding site is a powerful receptor for polycyclic aromatic hydrocarbons (PAHs) in water [14, 15]. Perylene, pyrene, and other similarly-sized PAHs are extremely insoluble in water.

521

522

20 Molecular Recognition Studies with Cyclophane Receptors in Aqueous Solutions

Fig. 20.2 Cyclophane 3 and the preferred geometry of the inclusion complex with pyrene in D2O, as revealed by characteristic complexation-induced changes in chemical shift Dd [ppm] of host and guest resonances in the 1 H NMR spectrum of a solution with [3] = 5.5 ´ 10–3 M and [pyrene] = 2.8 ´ 10–3 M.

Dd (pyrene protons): H–C(1): –1.02, H–C(2): –0.44, H–C(4): –1.25. Dd (host protons): H–C(2): –0.81, H–C(2'): –0.49, H–C(2''): + 0.07, H–C(3): –0.52, H–C(3'): –0.77, H–C(3'') + 0.16, H–C(10): + 0.31, Aryl–CH3: + 0.02, N(1')-CH3: –0.20, N(1'')-CH3: + 0.02

By complexation with 3, the solubility of these molecules in H2O is dramatically increased. Thus, the maximum solubility of pyrene in H2O is only 8 ´ 10–7 M. By solid–liquid extraction with a 5.5 ´ 10–3 M aqueous solution of 3, a solution with a total pyrene concentration of 2.8 ´ 10–3 M is obtained, with almost all of the PAH present in complexed form. Similarly, by liquid–liquid extraction, pyrene can be extracted from a hexane solution into the aqueous host phase. Both extraction protocols have been used to determine the association constants of the complexes formed in homogeneous aqueous phase (Tab. 20.2) [14]. The association constants of the PAH complexes of complementary size to the cyclophane cavity are very high. Perylene binds best, since its volume matches best the one of the host cavity. Upon complexation, a large number of host–guest

20.2 Complexation of Aromatic Solutes Tab. 20.2 Association constants Ka (± 20%) and free enthalpies of complexation –DG0 for the 1 : 1 complexes formed between cyclophane 3 and polycyclic aromatic hydrocarbons in D2O, T = 293–295 K a)

Guest

Ka (L mol–1)

–DG0 (kcal mol

Perylene Fluoranthene Pyrene Biphenyl Azulene Naphthalene Durene

1.6 ´ 107 1.8 ´ 106 1.8 ´ 106 2.2 ´ 104 2.1 ´ 104 1.2 ´ 104 1.9 ´ 103

9.6 8.4 8.4 5.8 5.8 5.5 4.4

a)

–1

)

For the determination of the association constants by solid–liquid and liquid–liquid extraction, see [14].

van der Waals contacts are established and many of the (apolar) surface-solvating water molecules are released into the bulk. Highly stable complexes are also formed by fluoranthene and pyrene whereas the smaller substrates azulene, naphthalene, and durene undergo weaker association. Again, 1H NMR studies provided useful information on the inclusion geometry which is shown in Fig. 20.2. As many as 13 host and guest resonances show specific complexation-induced up- and down-field shifts. The piperidinium ions attached to the C5-bridges approach the encapsulating substrate, adding to the gain in dispersion and desolvation energy (see Section 20.2.4). Furthermore, cation–p interactions may contribute to complex stability. Cyclophane 3 also acts as a carrier and mediates the passive transport of PAHs such as pyrene from a hexane donor phase across an aqueous solution (containing 3) into a hexane acceptor phase. This process represents the inverse of the carrier-mediated ion transport across lipid membranes and was observed for the first time in this investigation. 20.2.3

The Combination of Apolar Binding and Ion Pairing Leads to Very High Substrate Selectivity

The complexation of neutral naphthalene derivatives in water by 3 occurs with complexation-free enthalpies of ca. 5.5 kcal mol–1 (Tab. 20.3) [16]. In contrast, the driving force for inclusion of naphthalene derivatives bearing anionic substituents is higher by ca. 2–3 kcal mol–1. In the complexes, the quaternary piperidinium rings at the C5-chains approach the included guest molecule, leading to efficient ion pairing at the periphery of the cavity (Fig. 20.3). In agreement with this explanation, ammonium ion-substituted naphthalenes do hardly bind at all, as a consequence of charge repulsion. Thus, substrates with identical geometric complementarity to the cavity site show differences in binding-free enthalpy of as much as

523

524

20 Molecular Recognition Studies with Cyclophane Receptors in Aqueous Solutions

Fig. 20.3 Schematic drawing of the geometry of the complex between 3 and naphthalene2,6-disulfonate in D2O according to 1H NMR analysis. Observed complexation-induced

changes in chemical shift Dd [ppm] for the guest protons in a solution with [host] = [guest] = 4 ´ 10–3 M: H–C(1): –1.36, H–C(3): –0.94, H–C(4): –3.09

Tab. 20.3 Association constants Ka (± 10%) and free enthalpies of complexation –DG0 for the

1 : 1 complexes formed between cyclophane 3 and naphthalene derivatives in D2O, T = 293–295 K Guest

(a) Naphthalene derivatives with neutral substituents 1,5-Dimethylnaphthalene 2,6-Dimethylnaphthalene 2,7-Naphthalenediol 1,3-Naphthalenediol 1,5-bis(Dimethylamino)naphthalene a) 1-(Dimethylamino)naphthalene (b) Naphthalene derivatives with ionic substituents ANS b) 2,6-Naphthalenedisulfonate 1,5-Naphthalenedisulfonate 1-Naphthalenesulfonate 5-Dimethylamino-1-naphthalenesulfonate 1-(Trimethylammonio)naphthalenefluorosulfonate 1,5-bis(Dimethylammonio)naphthalenebis(deuteriochloride) c) a) b) c)

1.5 ´ 10–2 M K2CO3, pH&11. ANS = 8-phenylamino-1-naphthalenesulfonate. D2O/DCl/KCl, pD = 1.2, ionic strength I = 0.27 M.

Ka (L mol–1)

–DG0 (kcal mol–1)

3.3 ´ 104 2.6 ´ 104 1.9 ´ 104 9.8 ´ 103 9.7 ´ 103 9.3 ´ 103

6.0 5.9 5.7 5.4 5.4 5.3

3.2 ´ 106 > 106 4.4 ´ 105 3.5 ´ 105 1.4 ´ 105 1.7 ´ 103 *< 10

8.7 > 8.0 7.6 7.4 6.9 4.3 *< 1.3

20.2 Complexation of Aromatic Solutes –1

6–7 kcal mol , depending on whether charge–charge attraction or repulsion are effective in the complex (Tab. 20.3) [2 a, 17]. 20.2.4

Solvent Dependency of Cyclophane-Arene Complexation and the Nonclassical Hydrophobic Effect: Enthalpic Driving Forces in Aqueous Solution

When we studied by microcalorimetry the thermodynamics for inclusion complexation between cyclophane 4 and p-disubstituted benzene derivatives in water (Fig. 20.4), we did not obtain the thermodynamic quantities characteristic for binding driven by the classical hydrophobic effect [18], namely (a) a large favorable complexation entropy TDS0, (b) a small complexation enthalpy DH0, and (c) a large negative change in the heat capacity DCp0. Rather, the inclusion complexes were characterized by a large enthalpic driving force, partially compensated by an unfavorable entropic term. This enthalpic driving force is drastically reduced upon changing from water to methanol (Tab. 20.4) [19]. Today, it is well established by numerous investigations, that tight apolar complexation in chemistry and biology involving small solutes as substrates is mostly driven by a large favorable enthalpic change, partially compensated by an unfavorable entropic term. Examples are the inclusion complexation of aromatic solutes by other cyclophane receptors [20] and cyclodextrins [1] and, in biology, enzyme– substrate and enzyme–inhibitor binding, antibody recognition, DNA base-pair intercalation and minor groove binding, protein–protein binding, and interactions of proteins with DNA or lipids. The latter examples from biology are summarized in a recent review [6]. To shed more light onto the enthalpic driving force for apolar complexation and the role of solvent, we investigated the solvent dependency of the formation of pyrene complex 5 by a cryptand-like macrobicyclic cyclophane which features remarkable solubility in solvents of all polarities (Fig. 20.5) [21, 22]. Upon changing from water, the most polar, to carbon disulfide, the least polar of the considered solvents, the complexation-free enthalpy decreases from DG0 = –9.4 kcal mol–1 to DG0 = –1.3 kcal mol–1 (303 K, Tab. 20.5). Thus, the apolar binding strength increases consistently from apolar to dipolar aprotic solvents, to protic solvents, and to water. Since complex 5 has the same geometry in all solvents, the measured changes in

Fig. 20.4 Cyclophane 4 forms axial inclusion complexes with p-disubstituted benzene derivatives in water

525

526

20 Molecular Recognition Studies with Cyclophane Receptors in Aqueous Solutions Tab. 20.4 Thermodynamic characteristics from 1H NMR binding titrations and microcalorimetry for the complexation between cyclophane 4 and p-disubstituted benzene derivatives in water and methanol a)

DG0293K (kcal mol–1)

DH0293K (kcal mol–1)

DC0p (cal mol–1 K–1)

TDS0293K (kcal mol –1)

(a) in D2O or H2O Dimethyl p-benzenedicarboxylate p-Nitrotoluene p-Tolunitrile p-Nitrophenol p-Dimethoxybenzene p-Xylene p-Dicyanobenzene p-Dinitrobenzene p-Cresol Hydroquinone

–6.8 –6.0 –6.0 –5.9 –5.4 –5.3 –5.2 –5.2 –4.7 –3.7

–11.8 –8.1 –8.1 –10.5 –10.0 –7.2 –10.3 –9.8 –10.6 –10.3

–60 –50 –70 –130 –20 –20 –30 –40 –110 –60

–5.0 –2.1 –2.1 –4.6 –4.6 –1.9 –5.1 –4.6 –5.9 –6.6

(b) in CD3OD or CH3OH p-Dicyanobenzene p-Dimethoxybenzene

–1.9 –1.20

–4.2 –3.7

– –

– –2.5

Guest

a)

For details of the study see [19].

Fig. 20.5 The pyrene complex 5 is soluble in solvents of all

polarities, allowing investigation of the solvent dependency of apolar inclusion complexation

association strength are mostly solvent-modulated, a conclusion that was further confirmed by free-energy simulations [23]. Calorimetric investigations revealed that the formation of complex 5 is enthalpy-driven in most solvents [19, 24]. Complexation in alcohols is the most exothermic and, in general, the enthalpic driving force decreases from polar protic, to dipolar aprotic, and to apolar solvents. Based on these findings and work by others [25, 26] we proposed that favorable changes in solvent cohesive interactions and a gain in dispersion interactions are the main components of the enthalpic driving force for apolar complexation in

20.3 Steroid Recognition by Cyclophane Receptors Tab. 20.5 Thermodynamic characteristics from 1H NMR binding titrations and microcalorimetry for the formation of complex 5 in solvents of differing polarity a)

Solvent

DG0303K (kcal mol–1)

DH0303K (kcal mol–1)

TDS0303K (kcal mol–1)

Water 2,2,2-Trifluoroethanol Ethylene glycol Methanol Formamide Ethanol N-Methylacetamide N-Methylformamide N,N-Dimethylacetamide Acetone Dimethyl sulfoxide N,N-Dimethylformamide Dichloromethane Tetrahydrofuran Chloroform Benzene Carbon disulfide

–9.4 –7.8 –7.3 –6.4 –6.2 –6.1 –5.8 –5.1 –4.4 –4.3 –3.9 –2.9 –2.9 –2.7 –2.3 –1.5 –1.3

– –20.0 – –12.0 – –11.0 –9.0 –5.6 –2.0 –6.6 –6.4 –3.7

– –12.2 – –5.6 – –4.9 –3.2 –0.5 +2.4 –2.3 –2.5 –0.8

–3.0 –3.1 –0.8 –

–0.3 –0.8 +0.7 –

a)

For details of the study see [19, 22].

aqueous solution. Enthalpically driven apolar complexation in aqueous solution is now widely known as the non-classical hydrophobic effect [22]. For more details on this subject, which is relevant to both chemical and biological recognition processes, the reader is referred to other review articles [6, 8, 27].

20.3

Steroid Recognition by Cyclophane Receptors 20.3.1

The Search for Cyclophanes with Larger Preorganized Cavity-Binding Sites

Cyclophanes such as 1, 3, and 4 with cavities shaped by two dioxydiphenylmethane spacers (distance between the two terminal O-atoms in one spacer ca. 8.5 Å) [28] are too small for full incorporation of bulky aliphatic substrates, and only the formation of “nesting” complexes is observed [28]. We therefore searched for larger spacers to obtain more spherical binding sites targeting efficient inclusion complexation of steroids [9]. In one approach, new tricyclic spacers, readily available through fourfold Mannich reaction of substituted dibenzyl ketones, were introduced into the macrocyclic skeleton [30]. The X-ray crystal structure of one of the representatives of the resulting new class of water-soluble cyclophanes, the tri-

527

528

20 Molecular Recognition Studies with Cyclophane Receptors in Aqueous Solutions

topic receptor 6 (Fig. 20.6) with one potential neutral molecule and two cation binding sites, showed a large rectangular open cavity with dimensions of roughly 9 ´ 14 Å and a distance of 9.7 Å between the O-atoms of the two convergent C=O groups. The distance between the two O-atoms of the tricyclic spacer was found to be 11.3 Å which, according to molecular model examinations, should be suitable for shaping spherical cavities for steroid inclusion. However, the binding properties of 6 and related compounds were rather modest, presumably since the guests are unable to make short contacts with the aromatic rings surrounding the cavity due to steric hindrance by the two convergent C=O groups. Another approach towards the construction of large cavity binding sites for the incorporation of alicyclic guests involved the bridging of the major grooves of 2,2',6,6'- and 2,2,7,7'-tetraoxy-1,1'-binaphthalene spacers; however, the resulting optically active cyclophanes such as (S,S)-7 and (R,R)-8 again displayed only modest

Fig. 20.6 Cyclophanes 6, (S,S)-7, and (R,R)-8 possess large cavities but display only modest complexation properties

20.3 Steroid Recognition by Cyclophane Receptors

Fig. 20.7 First-generation steroid receptors 9 and 10

complexation properties which we tentatively explain by poor contacts between the incorporated substrate and the aromatic rings lining the binding site [31]. We subsequently decided to expand the dioxydiphenylmethane moieties, that had been so effective in shaping arene-binding sites in hosts such as 1, 3, and 4, and prepared the water-soluble cyclophanes 9 and 10 (Fig. 20.7) incorporating two dioxy-naphthylphenylmethane spacers [32, 33]. The new macrocycles indeed featured binding cavities suitably shaped and preorganized for incorporation of large guests such as steroids, adamantane derivatives, and even [2.n]paracyclophanes (n = 2–4) [33]. X-ray crystallographic study of a cyclophane related to 9 and 10 later showed that the distance between the two O-atoms in the dioxynaphthylphenylmethane spacers amounts to 11.2–11.8 Å and that these spacers are effective in providing the desired open cavity-binding sites [34]. 20.3.2

Steroid Complexation by Cyclophanes 9 and 10

Cyclophane 9 was found to be an efficient binder of steroids such as 11–18 in aqueous solutions [35]. Complexation strength (–DG0) in D2O/CD3OD 1 : 1 was determined in 1H NMR titrations by monitoring the changes in chemical shift of the steroidal methyl group resonances, which move upfield (Dd –0.40 to –1.49 ppm) upon inclusion into the cyclophane cavity [33]. The thermodynamic quantities DH0 and TDS0 were obtained by variable-temperature 1H NMR titrations followed by van’t Hoff linear regression analysis (Tab. 20.6). Tab. 20.6 shows that binding affinity in the series of bile acid derivatives 11–15 decreases with an increasing number of polar HO-groups in the guests. Thus the complex of lithocholic acid (15) with 9 is 2.2 kcal mol–1 more stable than the complex of cholic acid (11) with two additional HO-groups attached to the steroidal skeleton. Upon inclusion complexation [36] with an axial-type orientation of the guest in the cavity (the long axis of the steroid orients approximately along the C2-axis of 9 passing through the cavity), the HO-groups on the central B and C rings of the guest become partially desolvated without regaining new H-bonding interactions in the complex. This unfavorable polar group desolvation contributes to the observed substantial bind-

529

530

20 Molecular Recognition Studies with Cyclophane Receptors in Aqueous Solutions Tab. 20.6 Thermodynamic characteristics from 1H NMR binding titrations for the formation of 1 : 1 inclusion complexes by cyclophane 9 with steroids 11–18 in D2O/CD3OD 1 : 1

Guest

DG0 (kcal mol–1)

DH0 [c] (kcal mol–1)

TDS0 [c] (kcal mol–1)

11 12 13 14 15 16 17 18

–2.9 a, b) –3.2 a, b) –3.9 a, b) –4.1 b, c) –5.1 b, c) –3.8 c) –4.0 c) –4.7 c)

– – – –13.5 –8.7 –12.6 –13.7 –12.0

– – – –9.4 –3.6 –8.8 –9.7 –7.3

a) b) c)

Determined at 293 K [33 a]. In the presence of 0.1 M Na2CO3. Determined at 298 K [33 b].

ing selectivity. An additional contribution results from the increase in lipophilicity upon changing from cholic acid (11) to lithocholic acid (15). With increasing lipophilicity, the partitioning of the steroids from the polar aqueous phase into the less polar binding cavity becomes more favorable. Thus, the observed selectivity does not originate from differences in attractive host–guest interactions. As in the case of cyclophane–arene binding in protic solvents, the inclusion complexation of the steroidal substrates is strongly enthalpy-driven (Tab. 20.6),

20.3 Steroid Recognition by Cyclophane Receptors

with the enthalpic driving force being partially compensated by an unfavorable entropic term. A plot of DH0 vs. TDS0 yielded a strong linear enthalpy–entropy compensation, which is nearly universally observed for chemical and biological recognition processes [37]. The highly charged cyclophane 10 was sufficiently soluble to undertake 1 H NMR and microcalorimetric binding titrations in pure water (containing 0.1 M Na2CO3) with steroids 14 and 19–24 (Tab. 20.7). These steroids were selected as guests due to their elevated critical micelle concentration, which ensures that the stoichiometric inclusion complexation in water is not perturbed by competing self-association equilibria [14]. Tab. 20.7 Thermodynamic characteristics from 1H NMR and microcalorimetric binding titra-

tions for the formation of 1 : 1 inclusion complexes by cyclophane 10 with steroids 14 and 19–24 in pure water (containing 0.1 M Na2CO3, T = 298 K) [33 b] DH0 [c] (kcal mol–1)

Guest

DG0 (kcal mol–1)

19 a) 20 a) 14 a) 21 a) 22 b) 23 a) 24 b)

–4.8 –3.1 –5.7 +0.7 –5.9 –4.0 –6.0 –2.1 –6.3 – –6.8 –1.5 no measurable complexation

a) b)

Microcalorimetric titration in H2O. H NMR titration in D2O.

1

TDS0 [c] (kcal mol–1) +1.7 +6.4 +1.9 +3.9 – +5.3

531

532

20 Molecular Recognition Studies with Cyclophane Receptors in Aqueous Solutions

A series of interesting results was obtained: (i)

Upon changing the solvent from methanol/water 1 : 1 to pure water, the binding free enthalpy –DG0 increases by ca. 2 kcal mol–1, in agreement with previous solvent-dependent studies of cyclophane–arene complexation [8, 11]. The relative stability of the formed complexes again correlates with the number and orientation of the functional groups attached to the steroidal skeleton. Thus dehydrocholic acid (19) and hyocholic acid (20), with three polar groups attached to the steroidal skeleton, form weaker complexes than hyodeoxycholic acid (23) with only two such groups. (ii) The data afford good evidence that ion pairing between the ammonium groups attached to the cavity walls of cyclophane 10 and the negatively charged side chains of the steroids provides additional stabilization to the complexes. On the other hand, charge repulsion prevents any measurable complexation of the zwitterionic substrate 24. Ion pairing is also indicated by the thermodynamic characteristics which strongly differ from those obtained for the complexation by cyclophane 9 lacking the ammonium side chains. Whereas binding in methanol/water 1 : 1 by 9 was found to be enthalpy-driven (Tab. 20.6), inclusion of the anionic steroids by host 10 with four ammonium centers is accompanied by a favorable change in entropy (Tab. 20.7). One contributing factor to the favorable entropy change could be the classical hydrophobic effect, i.e. the release of apolar surface-solvating water molecules into the bulk [18, 38]. The dominating contribution, however, probably is the entropy gain due to the release of ordered water molecules from the solvation shells of the ionic centers of host and guest that undergo ion pairing in the complex. Ion complexation and ion pairing are often accompanied by large favorable entropy changes resulting from desolvation of the ions [39]. (iii) Differences in inclusion geometry may substantially alter the thermodynamic characteristics for the complexation process. Whereas the complexation of hyodeoxycholate (23) is mainly entropy-driven (DH0 = –1.5 kcal mol–1, TDS0 = +5.3 kcal mol–1), the inclusion of ursodeoxycholate (14) is more enthalpically governed (DH0 = –4.0 kcal mol–1, TDS0 = +1.9 kcal mol–1). Differences in ion pairing – as a result of different complex geometries – could account for these observations. According to molecular model examinations, the carboxylate of hyodeoxycholate (23) can interact with one of the pendant ammonium groups of 10 if both the steroidal B and C rings are complexed (Fig. 20.8 A). On the other hand, the presence of the 7a-HO-group on the B ring of ursodeoxycholate (14) seems to enforce a different inclusion geometry, hindering the charged groups to approach each other close enough for the formation of a strong, partially desolvated salt bridge (Fig. 20.8 B), thereby reducing the entropic gain. It is noteworthy that complexation by 10 in water again obeys a linear enthalpy–entropy compensation: in this case, an increase in binding entropy is compensated by an increasingly less negative enthalpic term.

20.3 Steroid Recognition by Cyclophane Receptors Fig. 20.8 Representation of two possible inclusion geometries for the complexes of cyclophane 10 (schematically drawn) with hyodeoxycholate (23, A) and ursodeoxycholate (14, B). To avoid unfavorable inclusion and desolvation of the equatorial 7aHO-group, 14 probably adopts a complex geometry which does not allow efficient ion pairing with the pendant ammonium side arms of the receptor

20.3.3

Double-decker Cyclophanes for Efficient Steroid Complexation: Dissolution of Cholesterol in Water

Solid cholesterol (25) is the main component of atherosclerotic plaque, one of the leading causes of death in the industrialized world. To dissolve and complex cholesterol, we developed the D2-symmetrical macrotricyclic double-decker cyclophanes (±)-26 and (±)-27 (Fig. 20.9) [40, 41]. The key step in the synthesis of (±)-26 represents a remarkable, completely diastereoselective four-fold Stille macrocyclization between two dibrominated cyclophane and two bis(tributylstannyl)acetylene molecules. Optical resolution of (±)-26 was accomplished by high-performance liquid

Fig. 20.9 The double-decker cyclophanes (±)-26 and (±)-27 are efficient steroid binders and dissolve cholesterol in water through complexation

533

534

20 Molecular Recognition Studies with Cyclophane Receptors in Aqueous Solutions

chromatography (HPLC) on a chiral stationary phase. Force-field energy minimizations of the two receptors with encapsulated cholesterol were undertaken. They revealed a very high degree of preorganization for both receptors, but also substantial differences in cavity size that have a profound influence on steroid-binding selectivity (see below). The cavity of the smaller receptor (±)-26 (cavity opening: 11 ´ 8 Å, cavity depth: 11 Å) is less wide and also less deep than that of the larger system (±)-27 (cavity opening: 12 ´ 9 Å, cavity depth: 13 Å). Cholesterol has a rather flat A/B ring segment and, according to modeling, was expected to bind better to (±)26. On the other hand, the deeper cavity of (±)-27 should be capable of encapsulating substantial parts of the steroidal side chain attached to ring D. The computational predictions were nicely validated by the experiments. Both double-decker cyclophanes form stable complexes with cholesterol (25) in water, and their aqueous solutions are capable of extracting and solubilizing the solid steroid. The more stable complex is formed by (±)-26, with an association constant Ka = 1.1 ´ 106 L mol–1 (–DG0 = 8.2 kcal mol–1) as determined by solid–liquid extraction at 295 K (Tab. 20.8). Solubilization of cholesterol is efficient: whereas the steroid has a solubility in water of only 4.7 ´ 10–6 M–1, a 1 mM aqueous solution of (±)-26 allows the dissolution of 0.85 equivalents of cholesterol, which corresponds to an increase in solubility by a factor of 180. Biological study further showed that the double-decker cyclophanes are very efficient in shuttling cholesterol between cells and serum lipoproteins [42]. Progesterone (28) and testosterone (18) were also dissolved in water by solid–liquid extraction, although their complexes are less stable than the complex formed between (±)-26 and cholesterol. Since accurate and comparative 1H NMR binding studies in D2O were hampered by broad signals resulting from exchange kinetics on the 1H NMR time scale, titrations were performed in CD3OD at 298 K, where complexation remains

Tab. 20.8 Association constants Ka (L mol–1) and binding free enthalpies –DG0 (kcal mol–1) for

1 : 1 complexes of steroids with double-decker cyclophanes (±)-26 and (±)-27 in water as determined at 295 K by solid–liquid extraction a) Steroid

Receptor

Ka (L mol–1)

–DG0 (kcal mol–1)

Cholesterol (25) Cholesterol (25) Progesterone (28) Testosterone (18)

(±)-26 (±)-27 (±)-27 (±)-26

1.1 ´ 106 1.5 ´ 105 1.5 ´ 105 6.8 ´ 104

8.2 7.1 7.1 6.5

a)

Reproducibility in DG0: ± 0.4 kcal mol–1.

20.3 Steroid Recognition by Cyclophane Receptors Tab. 20.9 Association constants Ka (± 10%) and binding-free enthalpies DG0 from 1H NMR titrations for steroid complexes of double-decker cyclophanes (±)-26 and (±)-27 in CD3OD at 298 K

Steroid

5a-Cholestane (29) Progesterone (28) Cholesteryl acetate (30) 5-Cholestene (31) Pregnenolone acetate (32) Dihydrocholesterol (33) Cholesterol (25) 5a-Androstane (34) Testosterone (18) b-Estradiol (35)

Receptor (±)-26

Receptor (±)-27

Ka (L mol–1)

–DG0 (kcal mol–1)

Ka (L mol–1)

–DG0 (kcal mol–1)

870 – 4800 3200 – – 1500 500 2100 390

4.0 – 5.0 4.8 – – 4.3 3.7 4.5 3.5

2700 2600 2300 2300 2100 1200 900 370 200 170

4.7 4.7 4.6 4.6 4.5 4.2 4.1 3.5 3.1 3.0

remarkably strong (Tab. 20.9) [40 b, 41]. The best evaluated signals during the titrations were those of the various steroidal CH3-groups, such as CH3 (18) (for the numbering see the formula of progesterone 28) which shifted upfield upon axial encapsulation of the substrate into the cavity of the double-decker cyclophanes. The following results were obtained in binding studies with steroids 18, 25, and 28–35: (i)

The narrower cavity of (±)-26 prefers encapsulation of steroids that possess a double bond in their B-ring and, therefore, are flattened relative to their fully aliphatic analogs. This is nicely seen in the comparison between 5a-choles-

535

536

20 Molecular Recognition Studies with Cyclophane Receptors in Aqueous Solutions

tane (29), which fits better into (±)-27, and cholestene, which forms the more stable complex with (±)-26. (ii) In addition to shape differences, functional group desolvation upon inclusion complexation is responsible for selectivity effects. Thus, cholesteryl acetate (30) is bound more strongly than cholesterol (25) by both receptors. Also, progesterone (28) binds better than testosterone (18) to (±)-27. It clearly costs more to partially desolvate a hydroxy group in an inclusion complex than an acetoxy or a keto group. (iii) The side chain of cholesterol (25) and related steroids has a remarkable effect on the stability of the complexes formed by the deeper receptor (±)-27. Thus, 5a-cholestane (29) binds by 1.2 kcal mol–1 better than 5a-androstane (34) which lacks this side chain. Steroidal side chain incorporation into the cavity is nicely supported by strong complexation-induced upfield shifts of its Me-group resonances in the 1H NMR spectrum. (iv) Both receptors discriminate between aliphatic and aromatic guests, with b-estradiol (35) only forming a much weaker complex than the aliphatic steroids. Steroids with polar groups on the B/C rings, such as cholic acid derivatives, do not show any measurable binding affinity. To establish more detailed structure-activity relationships, the inclusion complexation of another series of 30 steroids by double-decker cyclophane (±)-27 in CD3OD was investigated in collaboration with researchers from the Schering company [43]. This study clearly showed that (±)-27 does not only provide a favorable lipophilic phase for the steroid but that it also acts as a specific receptor. Although steroids with a high log P value (logarithm of partition constant of steroid between octanol and water) tend to form more stable complexes than those with low log P values, no general correlation between the thermodynamic driving force for complexation –DG0 and the partition coefficient was observed.

20.4

Catalytic Cyclophanes

By elaborate chemical functionalization, the cyclophane receptors for arenes described in Section 20.2 were transformed into enzyme-like catalysts [44]. Examples include artificial esterases, such as 36 [45] and 37 [46] or efficient cytochrome P450 mimics such as the porphyrin-bridged cyclophane 38 (Fig. 20.10) [47]. The bis(imidazole)-functionalized macrocycle 37 forms stoichiometric inclusion complexes with nitronaphthyl acetates in aqueous phosphate buffer (pH 8) and catalyzes the hydrolysis of the bound substrates under turnover conditions [46]. In the presence of iodosylbenzene as oxygen transfer reagent in CF3CH2OH, cytochrome P-450 mimic 38 was found to catalyze with turnover the oxidation of cavity-bound acenaphthylene to acenaphthen-1-one [47]. The best results in our investigations with catalytic cyclophanes, however, were obtained with the bis(coenzyme) system 39 (Fig. 20.11) which efficiently catalyzes

20.4 Catalytic Cyclophanes

Cyclophanes as artificial enzymes. Compounds 36 and 37 show esterase activity whereas the porphyrin conjugate acts as a cytochrome P-450 mimic and oxidizes acenaphthylene to acenaphthen-1-one in the presence of iodosylbenzene as oxygen transfer agent

Fig. 20.10

537

538

20 Molecular Recognition Studies with Cyclophane Receptors in Aqueous Solutions

The flavo-thiazolio-cyclophane 39 is a model system for pyruvate oxidase

Fig. 20.11

the oxidation of aromatic aldehydes to the corresponding esters in a biomimetic way [48, 49]. By its function and mechanism of action, 39 mimics pyruvate oxidase, a flavin- and thiamine diphosphate-dependent enzyme found in lactobacteria which catalyzes the transformation of pyruvate into acetyl phosphate [50]. The mechanisms proposed for the biological and the biomimetic transformation are shown in Fig. 20.12. Cyclophane 39 was shown by 1H NMR titrations to complex aromatic aldehydes such as benzaldehyde or 2-naphthaldehyde both in D2O/CD3OD as well as in pure CD3OD. It was subsequently shown to catalyze the oxidation of 2-naphthaldehyde to methyl 2-naphthoate in MeOH in the presence of D2O under Ar atmosphere ([39] = 0.5 mM and [2-naphthaldehyde] = 2–50 mM, T = 303 K). In initial rate studies, saturation kinetics was observed and the Michaelis-Menten parameters determined as kcat = 0.24 s–1, one of the highest turnover numbers reported for artificial enzymes, and KM = 23 mM. A comparison of the apparent bimolecular rate constant kcat/KM for catalysis by 39 with the calculated second-order rate constants for catalysis by simple thiazolium ion and flavin derivatives lacking any binding sites revealed that the rate acceleration was up to 670-fold. In the absence of thiazolium ion or flavin, no product was produced. The large rate acceleration with 39 results from the intramolecularity of all key steps in the catalytic cycle. We subsequently used 39 as a catalyst on a preparative scale, which required the regeneration of the oxidized flavin moiety. Attempts to reoxidize the reduced flavin by running the reaction under aerobic conditions (in analogy to pyruvate oxidase) led largely to the oxidative destruction of the catalyst. On the other hand, electrochemical regeneration of the oxidized flavin form was successful. By applying the conditions displayed in Fig. 20.13, the flavo-thiazolio-cyclophane 39 could be employed as a redox mediator for the efficient one-pot electrochemical oxida-

20.4 Catalytic Cyclophanes

Fig. 20.12 Catalytic cycles proposed for the transformation of pyruvate to acetyl phosphate catalyzed by the enzyme pyruvate oxidase and for the oxidation of aromatic aldehydes to the corresponding carboxylic acids and esters in the presence of the biomimetic catalyst 39

Fig. 20.13 Preparative-scale oxidation of aromatic aldehydes to the corresponding methyl esters

tion of aromatic aldehydes. The isolated yields of the corresponding carboxylic esters after a reaction time of 16 h were 78% (for naphthaldehyde) and 48% (for benzaldehyde). No ester was formed in the absence of catalyst. Later attempts to further improve the catalytic activity of thiazolio-cyclophanes by inserting them as cores into large dendrimers were not very successful [51, 52].

539

540

20 Molecular Recognition Studies with Cyclophane Receptors in Aqueous Solutions

Fig. 20.14 Two series of dendrophanes of first (G-1), second (G-2), and third (G-3) generation for apolar inclusion complexation in aqueous solution

20.5 Dendritically Encapsulated Cyclophanes (Dendrophanes)

20.5

Dendritically Encapsulated Cyclophanes (Dendrophanes)

Microenvironments at the core of dendrimers [53] resemble the local environments generated within protein superstructures [54, 55]. To investigate how such microenvironments may modulate the molecular recognition properties of cyclophanes encapsulated within the branched architecture, we prepared two series of water-soluble dendrophanes (dendritically encapsulated cyclophanes), 40–42 [56, 57] and 43–45 [57, 58] with poly(ether amide) branching and 12 (generation 1, G1), 36 (G-2), and 108 (G-3) terminal carboxy groups, respectively (Fig. 20.14). Compounds 40–42 feature an arene-binding cavity whereas the larger cyclophane core in compounds 43–45 is suitable for the complexation of steroidal guests (see Sections 20.2 and 20.3). The third-generation dendrimers 42 and 45 have molecular weights Mr around 18 000 Daltons (Da) and were shown by NMR and mass spectrometry to be of high purity and monodispersity. The binding properties of 40–45, in comparison with those of the core cyclophanes, were investigated in basic aqueous buffers by using 1H NMR and fluorescence titrations. A series of interesting results were obtained: (i)

The cyclophane recognition sites at the cores are accessible for substrate binding at all dendritic generation levels, and the stability of the inclusion association is nearly independent from the dendritic generation. For example, the 1 : 1 host–guest complexes of the fluorescent probe 6-(p-toluidino)naphthalene-2-sulfonate (TNS, 46) with 40–42 had formation free enthalpies varying from DG0 = –5.5 (G-1) to –5.3 (G-2), and to –5.1 kcal mol–1 (G-3) (300 K). Also, the complexes of testosterone (18) and 43–45 had nearly similar stability [DG0 = –3.9 to –4.1 kcal mol–1; at 300 K in borate-buffered D2O (pD 10.5)/CD3OD 1 : 1]. Most importantly, the substrates in all complexes are exclusively located in the central cyclophane cavities, as revealed by the 1 H NMR titration data, and nonspecific incorporation into fluctuating voids in the dendritic shell is negligible.

(ii) Dendritic encapsulation affects the micropolarity around the cyclophane core: with increasing dendritic generation, the micropolarity is significantly reduced. Thus, fluorescence titrations with TNS (46), whose emitting properties are highly sensitive to environmental polarity, in aqueous phosphate buffer (pH 8.0) showed that the micropolarity at the core of the G-3 dendrophane 42 is similar to that of ethanol. (iii) Complexation and decomplexation kinetics are remarkably fast (kdecompl > 102 to 103 s–1) even at the third-generation level, as shown by 1H NMR titrations and

541

542

20 Molecular Recognition Studies with Cyclophane Receptors in Aqueous Solutions

Fig. 20.15 The threading of two molecules of dendrophane 44 onto the rigid molecular rod 48 in 0.15 M aqueous phosphate buffer (pH 8.7)/methanol 1:1 (298 K) leads to the

well-defined supermolecule 47. The free enthalpies for 1 : 1 and 2 : 1 complexation are DG01:1 = –6.8 kcal mol–1 and D G02:1 = –4.9 kcal mol–1

fluorescence relaxation measurements with fluorescent steroids. Since the four dendritic wedges in the dendrophanes are attached to large nanometer-sized cyclophane cores, they do not tightly pack together and, therefore, leave apertures through which substrates can rapidly enter or leave the binding cavity. We subsequently used the dendrophanes to assemble multinanometer-sized superstructures in aqueous solution [59]. The supermolecule 47 with a molecular weight Mr of 14 714 Da was obtained in aqueous methanol (pH 8.7) by threading two dendrophanes 44 onto the 5.5-nm long molecular rod 48 (Fig. 20.15). This rod features a central benzene ring bearing two solubilizing quaternary ammonium side chains. Two testosterone termini are linked to this central benzene ring via rigid oligo(phenylacetylene) spacers. 1H NMR analysis clearly showed that the two threaded dendrophane molecules preferentially encapsulate the steroidal termini. This assembly process is driven by a combination of apolar forces, hydro-

20.7 Acknowledgments

phobic desolvation, and ion pairing, and depends strongly on the correct length of the oligo(phenylacetylene) spacers in the rod. Much larger supramolecular aggregates with molecular weights approaching 100 000 Da can be envisaged by threading four G-3 dendrophanes 45 onto four steroidal termini attached to suitably sized spacers departing from a tetrahedral focal point.

20.6

Conclusions

This short account illustrates some of the prominent roles that water-soluble cyclophanes have played in molecular recognition studies over the past two decades. They have been shown to be powerful receptors for nearly all types of apolar substrates – arenes, steroids, heterocycles (not discussed in this article) – rivaling in many cases enzymes and other protein receptors in their binding potency. The analysis of the thermodynamics of cyclophane inclusion complexation in solvents of all polarity including water has been essential for establishing that tight apolar complexation in aqueous solution is enthalpy-driven, which is now widely accepted as the non-classical hydrophobic effect. Double-decker cyclophanes provided the first synthetic molecular systems, capable of efficiently dissolving solid cholesterol in water. Further functionalization led to efficient catalysts displaying some of the highest turnover numbers known for artificial enzymes. Finally, dendritic encapsulation of cyclophane cores nicely mimics the protein environment around biological active sites and the resulting dendrophanes were shown to thread onto rigid molecular rods leading to high-molecular weight assemblies with multi-nanometer dimensions. Water-soluble cyclophanes will continue to play a prominent role in molecular recognition studies but it can be safely predicted that the majority of future cyclophane-type receptors will be constructed by programmed, modular self-assembly, which requires much less synthetic effort compared with the synthesis of the systems described in this account.

20.7

Acknowledgments

The author thanks the Fonds der Chemischen Industrie, Hoffmann–La Roche, and the ETH for supporting this work.

543

544

20 Molecular Recognition Studies with Cyclophane Receptors in Aqueous Solutions

20.8

References 1 2.

3

4

5

6

M. V. Rekharsky, Y. Inoue, Chem. Rev. 1998, 98, 1875–1917. a) H.-J. Schneider, A. Yatsimirsky, Principles and Methods in Supramolecular Chemistry, Wiley, Chichester, 1999; b) F. Vögtle, Cyclophan-Chemie, Teubner, Stuttgart, 1990; c) C. Seel, F. Vögtle, Angew. Chem. 1992, 104, 542–563; Angew. Chem. Int. Ed. 1992, 31, 528–549; T. H. Webb, C. S. Wilcox, Chem. Soc. Rev. 1993, 22, 383–395; d) Comprehensive Supramolecular Chemistry, Vol. 2, ed. F. Vögtle, Elsevier-Pergamon, Oxford, 1996. This volume contains the following chapters on complexation with cyclophane receptors: H.-J. Schneider, A. K. Mohammad-Ali, pp. 69–101; A. Pochini, R. Ungaro, pp. 103–142; K. Odashima, K. Koga, pp. 143–194; D. A. Dougherty, pp. 195– 209; F. Vögtle, C. Seel, P.-M. Windscheif, pp. 211–265; C. A. Hunter, pp. 267–277; Y. Aoyama, pp. 279–307; B. J. Whitlock, H. W. Whitlock, pp. 309– 324; A. Collet, pp. 325–365; Y. Murakami, O. Hayashida, pp. 419–438. a) F. Diederich, Angew. Chem. 1988, 100, 372–396; Angew. Chem. Int. Ed. 1988, 27, 362–386; b) F. Diederich, Cyclophanes, The Royal Society of Chemistry, Cambridge, 1991. a) M. Fujita, S. Nagao, M. Iida, K. Ogata, K. Ogura, J. Am. Chem. Soc. 1993, 115, 1574–1576; b) T. Kusukawa, M. Yoshizawa, M. Fujita, Angew. Chem. 2001, 113, 1931–1936; Angew. Chem. Int. Ed. 2001, 40, 1879–1884. a) J. C. Ma, D. A. Dougherty, Chem. Rev. 1997, 97, 1303–1324; b) D. A. Dougherty, Science 1996, 271, 163–168; c) D. A. Dougherty, D. A. Stauffer, Science 1990, 250, 1558–1560; d) D. A. Stauffer, D. A. Dougherty, Tetrahedron Lett. 1988, 29, 6039–6042. For a recent review on interactions involving aromatic rings, including cation– p interactions, in chemical and biological systems, see: E. A. Meyer, R. K. Castellano, F. Diederich, Angew. Chem. 2003, 115, 1244–1287; Angew. Chem. Int. Ed. 2003, 42, 1210–1250.

7

8

9

10

11

12

13 14 15

16 17

a) N. Zacharias, D. A. Dougherty, Trends Pharmacol. Sci. 2002, 23, 281–287; b) for a recent example on efficient cation–p interactions in enzyme–inhibitor complexes, see: A. Niedzwiecka, J. Marcotrigiano, J. Stepinski, M. Jankowska-Anyszka, A. Wyslouch-Cieszynska, M. Dadlez, A.- C. Gingras, P. Mak, E. Darzynkiewicz, N. Sonenberg, S. K. Burley, R. Stolarski, J. Mol. Biol. 2002, 319, 615–635. F. Diederich, D. B. Smithrud, E. M. Sanford, T. B. Wyman, S. B. Ferguson, D. R. Carcanague, I. Chao, K. N. Houk, Acta Chem. Scand. 1992, 46, 205–215. P. Wallimann, T. Marti, A. Fürer, F. Diederich, Chem. Rev. 1997, 97, 1567– 1608. S. B. Ferguson, F. Diederich, Angew. Chem. 1986, 98, 1127–1129; Angew. Chem. Int. Ed. 1986, 25, 1127–1129. S. B. Ferguson, E. M. Sanford, E. M. Seward, F. Diederich, J. Am. Chem. Soc. 1991, 113, 5410–5419. a) F.-G. Klärner, U. Burkert, M. Kamieth, R. Boese, J. Benet-Buchholz, Chem. Eur. J. 1999, 5, 1700–1707; b) C. A. Hunter, J. K. M. Sanders, J. Am. Chem. Soc. 1990, 112, 5525–5534; c) F. J. Carver, C. A. Hunter, P. S. Jones, D. J. Livingstone, J. F. McCabe, E. M. Seward, P. Tiger, S. E. Spey, Chem. Eur. J. 2001, 7, 4854–4862; d) E. Kim, S. Paliwal, C. S. Wilcox, J. Am. Chem. Soc. 1998, 120, 11192–11193; e) F. Cozzi, J. S. Siegel, Pure Appl. Chem. 1995, 67, 683–689. M. L. Waters, Curr. Opin. Chem. Biol. 2002, 6, 736–741. F. Diederich, K. Dick, J. Am. Chem. Soc. 1984, 106, 8024–8036. For another receptor for PAHs, see: F. Vögtle, W. M. Müller, U. Werner, H.W. Losensky, Angew. Chem. 1987, 99, 930–932; Angew. Chem. Int. Ed. 1987, 26, 901–903. F. Diederich, D. Griebel, J. Am. Chem. Soc. 1984, 106, 8037–8046. For free enthalpy increments for ion pairing interactions, see a) H.-J. Schneider, Chem. Soc. Rev. 1994, 23, 227–234;

20.8 References

18

19

20 21 22 23

24

25

26

27

28

29

b) H.-J. Schneider, Angew. Chem. 1991, 103, 1419–1439; Angew. Chem. Int. Ed. 1991, 30, 1417–1436. N. T. Southall, K. A. Dill, A. D. J. Haymet, J. Phys. Chem. B 2002, 106, 521–533 and refs. cited. D. B. Smithrud, T. B. Wyman, F. Diederich, J. Am. Chem. Soc. 1991, 113, 5420–5426. E. L. Piatnitski, R. A. Flowers II, K. Deshayes, Chem. Eur. J. 2000, 6, 999–1006. F. Diederich, K. Dick, D. Griebel, J. Am. Chem. Soc. 1986, 108, 2273–2286. D. B. Smithrud, F. Diederich, J. Am. Chem. Soc. 1990, 112, 339–343. T. Z. Mordasini-Denti, W. F. van Gunsteren, F. Diederich, J. Am. Chem. Soc. 1996, 118, 6044–6051. For other examples of enthalpy-controlled arene complexation by cyclophanes in water, see: a) I. K. Chun, M. H. Lee, J. Korean Pharm. Sci. 1989, 19, 71–75; b) R. M. Izatt, K. Pawlak, J. S. Bradshaw, R. L. Bruening, Chem. Rev. 1995, 95, 2529–2586. For the role of solvent cohesive interactions, see: a) W. Saenger, Angew. Chem. 1980, 92, 343–361; Angew. Chem. Int. Ed. 1980, 19, 344–362; b) O. Sinanoglu in Molecular Associations in Biology, ed. B. Pullman, Academic Press, New York, 1968, pp. 427– 445; c) C. Carey, Y.-K. Cheng, P. J. Rossky, Chem. Phys. 2000, 258, 415–425; d) N. T. Southall, K. A. Dill, J. Phys. Chem. B 2000, 104, 1326–1331. a) A. Fersht, Enzyme Structure and Mechanism, 2nd edn, Freeman, New York, 1985, pp. 293–310; b) A. Fersht, C. Dingwall, Biochemistry 1979, 18, 1245–1249. D. B. Smithrud, E. M. Sanford, I. Chao, S. B. Ferguson, D. R. Carcanague, J. Evanseck, K. N. Houk, F. Diederich, Pure Appl. Chem. 1990, 62, 2227–2236. a) C. Krieger, F. Diederich, Chem. Ber. 1985, 118, 3620–3631; b) S. B. Ferguson, E. M. Seward, F. Diederich, E. M. Sanford, A. Chou, P. Inocencio-Szweda, C. B. Knobler, J. Org. Chem. 1988, 53, 5593–5595. F. Diederich, K. Dick, Chem. Ber. 1985, 118, 3817–3829.

30

31 32

33

34

35

36

37

38

39

40

41

a) D. R. Carcanague, C. B. Knobler, F. Diederich, J. Am. Chem. Soc. 1992, 114, 1515–1517; b) F. Diederich, D. R. Carcanague, Helv. Chim. Acta 1994, 77, 800–818. P. P. Castro, F. Diederich, Tetrahedron Lett. 1991, 32, 6277–6280. a) H. Kawakami, O. Yoshino, K. Odashima, K. Koga, Chem. Pharm. Bull. 1985, 33, 5610–5613; b) K. Koga, K. Odashima, J. Inclusion Phenom. Mol. Recognit. Chem. 1989, 7, 53–60. a) D. R. Carcanague, F. Diederich, Angew. Chem. 1990, 102, 836–838; Angew. Chem. Int. Ed. 1990, 29, 769–771; B. R. Peterson, P. Wallimann, D. R. Carcanague, F. Diederich, Tetrahedron 1995, 51, 401–421. P. Wallimann, S. Mattei, P. Seiler, F. Diederich, Helv. Chim. Acta 1997, 80, 2368–2390. For a review on steroid complexation in chemistry and biology, see 9. For studies with other steroid-binding cyclophanes in protic solvents, see: a) ref 32; b) S. Kumar, H.-J. Schneider, J. Chem. Soc. Perkin Trans 2 1989, 245–250; c) Y. Murakami, O. Hayashida, T. Ito, Y. Hisaeda, Pure Appl. Chem. 1993, 65, 551–556; d) T. H. Webb, H. Suh, C. S. Wilcox, J. Am. Chem. Soc. 1991, 113, 8554–8555. For an analysis of complex geometries by pseudo-Monte Carlo liquid-phase simulation, see: H. A. Carlson, W. L. Jorgensen, Tetrahedron 1995, 51, 449–472. a) Y. Inoue, Y. Liu, L. H. Tong, B. J. Shen, D. S. Jin, J. Am. Chem. Soc. 1993, 115, 10637–10644; b) J. D. Dunitz, Chem. Biol. 1995, 2, 709–712. W. Blokzijl, J. B. F. N. Engberts, Angew. Chem. 1993, 105, 1610–1648; Angew. Chem. Int. Ed. 1993, 32, 1545–1579. a) J. M. Lehn, Struct. Bonding (Berlin) 1973, 16, 1–69; b) L. Sebo, B. Schweizer, F. Diederich, Helv. Chim. Acta 2000, 83, 80–92. a) B. R. Peterson, F. Diederich, Angew. Chem. 1994, 106, 1688–1690; Angew. Chem. Int. Ed. 1994, 33, 1625–1628; b) B. R. Peterson, T. Mordasini-Denti, F. Diederich, Chem. Biol. 1995, 2, 139–146. T. Marti, B. R. Peterson, A. Fürer, T. Mordasini-Denti, J. Zarske, B. Jaun, F.

545

546

20 Molecular Recognition Studies with Cyclophane Receptors in Aqueous Solutions

42

43

44

45 46 47

48

49

Diederich, V. Gramlich, Helv. Chim. Acta 1998, 81, 109–144. A. E. Christian, H.-S. Byun, N. Zhong, M. Wanunu, T. Marti, A. Fürer, F. Diederich, R. Bittman, G. H. Rothblat, J. Lipid Res. 1999, 40, 1475–1482. A. Fürer, T. Marti, F. Diederich, H. Künzer, M. Brehm, Helv. Chim. Acta 1999, 82, 1843–1859. For reviews on artificial enzymes, see: a) A. J. Kirby, Angew. Chem. 1996, 108, 771– 790; Angew. Chem. Int. Ed. 1996, 35, 707– 724; b) Y. Murakami, J.-i. Kikuchi, Y. Hisaeda, O. Hayashida, Chem. Rev. 1996, 96, 721–758; J. K. M. Sanders, Chem. Eur. J. 1998, 4, 1378–1383. F. Diederich, G. Schürmann, I. Chao, J. Org. Chem. 1988, 53, 2744–2757. I. Chao, F. Diederich, Recl. Trav. Chim. Pays-Bas 1993, 112, 335–338. a) D. R. Benson, R. Valentekovich, F. Diederich, Angew. Chem. 1990, 102, 213–216; Angew. Chem. Int. Ed. 1990, 29, 191–193; b) D. R. Benson, R. Valentekovich, S.-W. Tam, F. Diederich, Helv. Chim. Acta 1993, 76, 2034–2060. a) P. Mattei, F. Diederich, Angew. Chem. 1996, 108, 1434–1437; Angew. Chem. Int. Ed. 1996, 35, 1341–1344; b) P. Mattei, F. Diederich, Helv. Chim. Acta 1997, 80, 1555–1588. For earlier studies with thiazolium ionsubstituted cyclophanes, see: a) F. Diederich, H. D. Lutter, J. Am. Chem. Soc. 1989, 111, 8438–8447; b) S.-W. Tam, L. Jimenez, F. Diederich, J. Am. Chem. Soc. 1992, 114, 1503–1505; c) S.-W. TamChang, L. Jimenez, F. Diederich, Helv. Chim. Acta 1993, 76, 2616–2639.

50

51

52

53

54

55

56

57

58 59

a) B. Sedewitz, K. H. Schleifer, F. Götz, J. Bacteriol. 1984, 160, 273–278; b) Y. A. Muller, G. E. Schulz, Science 1993, 259, 965–967; c) Y. A. Muller, G. Schumacher, R. Rudolph, G. E. Schulz, J. Mol. Biol. 1994, 237, 315–335. T. Habicher, F. Diederich, V. Gramlich, Helv. Chim. Acta 1999, 82, 1066– 1095. For a water-soluble flavo-cyclophane with redox-switchable binding properties, see: E. M. Seward, R. B. Hopkins, W. Sauerer, S.-W. Tam, F. Diederich, J. Am. Chem. Soc. 1990, 112, 1783–1790. G. R. Newkome, C. N. Moorefield, F. Vögtle, Dendrimers and Dendrons. Concepts, Syntheses, Applications, Wiley-VCH, Weinheim, 2000. a) D. K. Smith, F. Diederich, Chem. Eur. J. 1998, 4, 1353–1361; b) D. K. Smith, F. Diederich, Top. Curr. Chem. 2000, 210, 183–227; c) F. Diederich, B. Felber, Proc. Natl. Acad. Sci. USA 2002, 99, 4778–4781. S. Hecht, J. M. J. Fréchet, Angew. Chem. 2001, 113, 76–94; Angew. Chem. Int. Ed. 2001, 40, 74–91. S. Mattei, P. Seiler, F. Diederich, V. Gramlich, Helv. Chim. Acta 1995, 78, 1904–1912. S. Mattei, P. Wallimann, B. Kenda, W. Amrein, F. Diederich, Helv. Chim. Acta 1997, 80, 2391–2417. P. Wallimann, P. Seiler, F. Diederich, Helv. Chim. Acta 1996, 79, 779–788. B. Kenda, F. Diederich, Angew. Chem. 1998, 110, 3357–3361; Angew. Chem. Int. Ed. 1998, 37, 3154–3158.

547

Subject Index a absorption spectra 276 acetylene-bridged cyclophanes 23 4-acetyl[2.2]paracyclophane 439 – stereoselective allylboration 439 actuators 513 advanced biomaterials 472 aldol reactions 446 a-alkyl-a-amino acids 441 – stereoselective synthesis 441 alkylation 448 – allylic 448 alkyne metathesis method 3 [34]allenophane 406 – 13C NMR spectrum 406 – configuration 406 – 1H NMR spectrum 406 all-homocalix[4]arene 46 allylic acetates 448 – stereoselective alkylation 448 allylic alcohols 439 – asymmetric epoxidation 439 aminomethyl[2.2]paracyclophane 466 4-amino[2.2]paracyclophane 464 [n](N6,9)-aminopurinophanes 294 – UV absorption spectrum 294 [n](N6,9)-6-aminopurinophanes 294 – conformational distortion 294 – electronic excitation 294 – UV/Vis spectra 294 amphiphilic macrocycles 22 5a-androstane 535 anionic polymerization 424 – silicon-bridged [1]ferrocenophanes 424 [14]annulene 7 [16]annulene 7 [18]annulene 7, 9, 206 f. – perethynylated 9 [24]annulene 9

– perethynylated 9 ansa(vinylene)[2]ferrocenophane 136 – polymerization 136 – ROP 136 ansa-metallocenes 160 ansamitocin P-3 51 ansatrienine A (mycotrienine) 50 anthracenophanes 191 – photocyclization 191 [6](1,4)anthracenophanes 90, 93 – photo-dimerization 93 – preparation 90 [6](9,10)anthracenophanes 90 – preparation 90 anti-[2.2]metacyclophanediene 196 anti[22](1,3)cyclophane-1,9-diene 287 – UV absorption spectrum 287 anti[22](1,3)cyclophane 2287 – UV absorption spectrum 287 antibody recognition 525 anticoagulants 475 antikekulene 5 apolar binding 523 apolar complexation 527 [m.n]arenophanes 288 aromatic aldehydes 539 – oxidation 539 aromatic interactions 521 – edge-to-face 521 – p-p 521 aromaticity 6 aromatics 337 – tubular 337 artificial enzymes 538, 543 – turnover number 538 artificial esterases 536 [24]arylenophanetetraene anions 407 – 13C NMR spectra 407

Modern Cyclophane Chemistry. Edited by Rolf Gleiter, Henning Hopf Copyright © 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 3-527-30713-3

548

Subject Index – 1H NMR spectra 406 aryne 193 f. ASCII code 510 association constants 489 f., 521 ff., 534 f. – naphthalene derivatives 524 asymmetric addition 443 – diethylzinc to imines 443 asymmetric hydrogenation 451 atherosclerotic plaque 533 axial inclusion complexes 525 axial orientation 529 aza[3]ferrocenophanes 368 – electrochemistry 368 aza[n](1,4)naphthalenophanes 386 – activation parameters 386 – conformational behaviour 386 – VT NMR data 386 4-azidomethyl[2.2]paracyclophane 466

b Bamford-Stevens reaction 120 b band 278 b' band 278 barrelene 193 ff. – bridged 193 – derivative 193 beltenes 337, 341 – fully-conjugated 341 belt-like phanes 111 2-benzal[3]ferrocenophane-1,3-dione 139 – molecular structure 139 – preparation 139 17-(1,4)benzena-1,8-diazabicyclo[6.6.5]nonadecaphane-4,11-diyne 180 – complex with Ag+ 180 – molecular structure 180 benzene 318 ff. – boat-type deformed 319 – dipole transition moments 318 – orbital correlation diagram 320 – oscillator strengths 318 benzene fragments 319 – interacting 319 benzenophanes 45 benzoannulenes 342 – belt-like 342 m-benzyne 225 – matrix-isolation 225 – IR absorption 225 – synthesis 225 o-benzyne 224 – synthesis 224 bifunctional ligands 456

bile acid derivatives 529 – binding affinity 529 binaphthalene spacers 528 binding cavities 529 binding studies 293 – cylobis(paraquat-p-phenylene) 293 – tetrathiafulvalene (TTF) 293 biocompatibility 476, 481 biomaterials 463 biomolecular motors 513 biomolecules 464, 470 – immobilization 464, 470 biosensors 481 biotin 474 bis[benzene]chromium disilabridged 169 – molecular structure 169 – synthesis 169 bis(bipyridino)phane 406 – 1H NMR spectrum 406 – VT NMR data 406 1,1'-bis(x-cyanoalkyl)ferrocene 147 – cyclization 147 biscyclobutadieno superphanes 114 biscyclopentadienono superphane 117 – synthesis 117 bis(Dewar benzene) isomers 98 – valence isomerization 98 1,4-bis(p-ethylphenyl)butane 284 – UV/Vis spectra 284 (bishomo-calixpyridine) 43 1,1'-bis(a-hydroxy-a-phenylethyl)ferrocene 142 – acid-catalyzed reaction 142 bis(indacenyl)diiron 151 – syn and anti 151 – synthesis 151 1,2-bismethylene[2.2]paracyclophane 223 bis(g6-[2.2]paracyclophane)ruthenium(II) 371 (E)- and (Z)-2,3-bis(4'-[2.2.2]paracyclophanyl)but-2-ene 179 – complex with Ag+ 179 – molecular structure 179 1,1'-bis(phenylethynyl)ferrocene 134 bistable rotaxane 508 1,1'-bis((trimethylsilyl)ethynyl)ferrocene 141 – cyclization 141 blood contact 475 boat-type deformation 314 break junction 511 [34]-bridged syn-tricyclo[4.2.0.02.5]octa-3,7diene 116 4-bromo[2.2]paracyclophane 200, 464

Subject Index Buchwald/Hartwig amination 450 1,1'-(1,3-butadienylene)ferrocene 142 – ROMP 142 butadiyne-bridged cyclophynes 23 – association constants 23 6-tert-butyl-9-methoxy-[3.3]metacyclophane2,11-diones 399 – 1H NMR spectra 399 – syn/anti-conformer ratio 399 1,1'-(1-tert-butyl-1,3-butadienylene)ferrocene 143 – ROMP 143

c C60 33 – generation 33 cage hydrocarbon 192, 202 – molecular work benches 202 (calix[4]furane) 43 (calix[6]pyrrole) 43 carba[2.2]metacyclophanes 55 – planar chiral 55 carbaphanes 45 (–)-4-carbomethoxy[2.2]paracyclophane 214 – racemization 214 carbon anions 2 carbon nanotubes 337 carbon tubes 2 carcerand 67 carrier 523 catalytic cyclophanes 536 catenanes 27, 49, 485, 492, 499, 502, 513 – smectic mesophases 27 – switchable 499, 502, 513 – thermotropic nematic mesophase 27 [2]catenane 493, 502, 506 f., 511 – bistable 507, 511 – chemical switching 502 – crystal structure 493 – electrochemical switching 502 – electronic device 506 – template-directed synthesis 493 – switchable 506 [3]catenane 495 – self-assembly 495 [5]catenane 496 [6]catenane 496 cation-p interactions 519 cavity 485, 528 C60Cl6 33 – generation 33 CD spectrum 56 f., 312 – calculations 57

– charge transfer band 56 – measurements 57 – theoretical prediction 312 CD-measurements 58 – in gas phase 58 cell adhesion 476, 479 cesium effect 54 C60H6 33 – generation 33 charge resonance interactions 281 charge resonance 291 charge transfer 276, 503 charge transfer (CT) transition 277 charge transfer interactions 485, 488 chenodeoxycholic acid 530 chiral [2.2]paracyclophanes 454 – building blocks 454 chiral plane 436 chirality-inducing step 54 chiroptical properties 56 5a-cholestane 535 5-cholestene 535 cholesterol 533, 535 – dissolution 533 – water 533 cholic acid 530 [2.2]chrysenophanes 395 – conformational behaviour 389 – 1H NMR spectra 395 circular dichroism 311 circumrotary motion 503 circumrotational processes 494, 508 clipping 497 C58N2 33 – generation 33 cocondensation 168 collective electronic oscillators 281 complexation 520, 523 – naphthalene 523 conductance 506, 509, 512 conjugated belts 355 control of motion 504 coronary angioplasty 475 cortisone 530 CpCo-cyclobutadiene complexes doubly bridged 171 – molecular structures 171 – synthesis 171 (CpCo)cyclobutadieno[14]annulene 10 (CpCo)cyclobutadieno[18]annulene 10 CpCo-cyclopentadienone complexes doubly bridged 173 – molecular structures 173

549

550

Subject Index – synthesis 173 cryptophane 67 CT band 305 – change of intensity 305 CT interactions 277 cucurbituril 48 CVD polymerization 463, 469 – mechanism 469 cyclacene 337, 341, 354 – Möbius systems 354 cyclic voltammetry 115, 360 – superphanes 115 [2+2]cycloadditions 111, 205 cyclo[36]carbon 340 cyclo[n]carbon 341 [2.2.1]-p,p,m-cyclophane 177 2,5-cyclohexadiene-1,4bis(ylideneketene) 223 – formation 223 – complex with Ag+ 177 [2n]cyclophane 287 – r-p interaction 287 – low energy singlet 287 – triplet states 287 [2.2.1]-p,p,p-cyclophane 177 – complex with Ag+ 177 [2.2.2](1,2,4)cyclophane 199 – synthesis 199 [2.2.2](1,4)cyclophane 176 – complex with Ag+ 176 – molecular structure 176 – quantum chemical calculations 176 [2.2.2]-p,p,m-cyclophane 177 – complex with Ag+ 177 [22](1,4)cyclophane 287 – UV absorption spectrum 287 [23](1,3,5)cyclophane 286 – cyclophane band 286 – UV/Vis spectrum 286 [24](1,2,4,5)cyclophane 122 – Birch reduction 122 [25](1,2,3,4,5)cyclophane 286 – cyclophane band 286 – UV/Vis spectrum 286 [26](1,2,3,4,5,6)cyclophane 105 [26](1,2,3,4,5,6)cyclophane 372 – Ru(II) complex 372 [26](1,2,4,5)cyclophane 176 – complex with Ag+ 176 – molecular structure 176 – NMR spectrum 176 [33](1,3,5)cyclophane 123, 192 [34](1,2,4,5)cyclophane 192

[36](1,2,3,4,5,6)cyclophane 123, 125 – molecular structure 125 – TCNE complex 126 – TCNQF4 complex 125 – UV/Vis-spectrum 126 [34](1,2,4,5)cyclophane 192 [23](1,2,4)cyclophane 215 – thermolysis 215 [23](1,3,5)cyclophane 215 – thermolysis 215 [n]cyclophane 277, 323, 364 – bathochromic shifts 277 – electrochemistry 364 – UV absorption spectra 277 [22](1,4)cyclophane-1,9-diene 286 – UV absorption spectrum 286 [26](1,2,3,4,5,6)cyclophane-1-ene 122 [16.16.16](1,3,5)cyclophanehexaenoctadecayne 32 – preparation 32 [7.7.7](1,3,5)cyclophanehexayne 32 – preparation 32 [23](1,3,5)cyclophane-1,9,17-triene 287 – UV absorption spectrum 287 cyclophyne – amphiphilic substituents 24 – nematic mesophase 24 – thermotropic mesophase 24 [46]cyclophyne 30 – butadiyne-bridged 30 [32]cyclopropenonoacetylenophanes 106 [42]cyclopropenonoacetylenophanes 106 [52]cyclopropenonoacetylenophanes 106 [n2]cyclopropenonophanes 106 cyclo[8]pyrrole 44 cyclobis(paraquat-p-phenylene) 374 cyclobis(paraquat-p-phenylene) 485, 487, 515 – X-ray 487 cyclobutadieno annulenes 10, 348 – synthesis 10 cyclobutadieno “bow tie” 10 – synthesis 10 cyclobutadieno superphanes 109, 111 – molecular structure 111 cyclobutadieno “three-quarters square” 10 – synthesis 10 cyclodeca-1,6-diyne 106, 272 – PE spectra 272 cyclodextrins 519, 525 cyclodiyne dimerization 112 – one-pot synthesis 112 cyclododeca-1,7-diyne 106 cyclophane band 280, 285

Subject Index cyclophane complexation 520 cyclophane recognition 541 cyclophane – in natural product synthesis 196 cyclophane-arene complexation 525, 532 cyclophane-Ru-complexes 370 – electrochemical reduction 370 cyclophanes 111, 189 f., 201, 205 f., 283, 288, 311, 359, 436 f., 515, 519, 536 f., 543 – artificial enzymes 537 – belt-structured 207 – Cahn-Ingold-Prelog (CIP) system 436 – catalytic 536 – catalytically active 519 – conformation 189 – double decker 543 – electrochemistry 359 – electronic circular dichroism 311 – electronic interactions 359 – intra-annular distance 189 – intramolecular reactions 189 – intramolecular (or intradeck) ring formation 189 – mechanical movement 515 – molecular workbenches 201 – multibridged 206 – oxabridged 205 – [2+2]photoaddition 205 – photoisomerization 190 – photophysics 283 – pseudo-gem substitutions 189 – reactions between benzene rings 190 – redox behavior 359 – stereoselective synthesis 437 – water-soluble 519, 543 – with silyl bridges 111 cyclophanes of pyrene (as donor) and 1,8;4,5 naphthalenetetracarboxydiimide (as acceptor) 305 – UV absorption spectrum 305 f. cyclophanes with unsaturated bridges 221 – cleavage 221 cyclophynes 1, 6, 16, 194, 341 – binaphthol-based 16 – double-helical 16 – pentaphenylene-based 6 – triphenylene-based 6 cyclophynes with carbazole rings 27 – redox active 27 cyclopropanation 446, 455 – asymmetric 446 cyclopropenone 271 – PE spectra 271

cyclopropenone rings 108 cyclopropeno superphanes 106 cyclopropenyliophane 109 – superbridged 109 cyclotetradeca-1,8-diyne 106 cyclotrimerization 6 – CpCo-catalyzed 6 cylobis(paraquat-p-phenylene) 293 5-cylindrophane 181 – complexes with Ag+, Cu+ 181 cytochrome P-450 mimics 536

d d-p* back-donation 178 dehydro[12]annulene 5 – hexa-ethynyl derivatives 5 – structures 5 – transition metal complexes 5 dehydrobenzo[12]annulene 3 – hyperpolarizabilities 3 dehydrobenzo[14]annulene 7 – derivatives 7 – dibenzo 7 – 1H NMR chemical shifts 7 – monobenzo 7 – monothiopheno 7 dehydrobenzo[18]annulene 9 – derivatives 9 – donor-acceptor substituted derivatives 9 – hyperpolarizabilities 9 – organometallic 9 dehydrobenzoannulenes 1 dehydrocholic acid 531 4,5-dehydro[2.2]paracyclophane 194 dendrimers 454 dendritic encapsulation 541 dendrophanes 519, 540 ff. density functional theory (DFT) 313 – application to cyclophanes 313 deoxycholic acid 530 dethreading/rethreading 486, 505 1,1'-diallylferrocene 144 – ring closing metathesis (RCM) 144 diamino[2.2]paracyclophane 465 diazabicyclophanes 409 – energy barrier to ring rotation 409 – temperature-dependent 1H NMR spectra 409 – VT NMR data 409 1,10-diazacyclophane 54 5,12-diaza[24](1,2,4,5)cyclophane 292 – UV absorption spectra 292 5,15-diaza[24](1,2,4,5)cyclophane 292

551

552

Subject Index – UV absorption spectra 292 4,7-diaza[2.2]paracyclophane 291 – absorption and emission spectra 291 – solvent dependence 291 2,11-diaza[3.3](3,5)pyridinophane 397 – barrier for bridge inversion 397 – VT 1H NMR data 397 11,16-(1,4)dibenzena-1,8-diazabicyclo[6.5.5] octadecaphane-4-yne 180 – complex with Ag+ 180 – molecular structure 180 dibenzo[12]dehydroannulene 6 – perethynyl-substituted 6 dibenzoequinene 190 dichloro[1.1]metacyclophane 89 dichloro[n2]cyclopropenyliophanes 106 1,1'-dicinnamoylferrocenes 139 – molecular structure 139 dicyclohexano[18]C-6 47 Diels-Alder addition 447, 458 – 4-vinyl[2.2]paracyclophane 458 Diels-Alder reactions 446 – laterally substituted 448 diethylzinc addition 455 f. diethynylterphenyl 14 – dimerization 14 – intramolecular cyclization 14 dihetera[2.2]metacyclophane 53 dihydrocholesterol 535 diisopropyl-zinc 440 – chiral addition to paracyclophanes 440 dilithioferrocene 416 1,4-dimethoxybenzene 488 – molecular recognition 488 12,15-dimethoxy-4,7-diaza[2.2]paracyclophanes 291 – absorption and emission spectra 291 – solvent dependence 291 (S)-14,17-dimethyl[2](1,3)azuleno[2]paracyclophane 333 – comparison of experimental and theoretical CD spectra 333 14,17-dimethyl[2](1,3)azuleno[2]paracyclophane 333 – CD spectra 333 p,p'-dimethyldibenzyl 214 8,12-dimethyl-2,5-dithia[6]metacyclophane 58 p,p'-dimethylenedibenzyl diradicals 214 – intermediates 214 p,p-dimethylenedibenzyl-diradical 218 – observation 218

1,2-dimethyl[3](1,1')[3](3,3')ferrocenophane 133 2,6-dimethylnaphthalene 288 – UV absorption spectra 288 9,12-dimethyl-4-oxa[7]paracyclophane 58 (R)-9,12-dimethyl-4-oxa[7]paracyclophane 324 – comparison of experimental and theoretical CD spectra 324 dinactine 47 dinitro[2.2]paracyclophane 465 N6,9-dinonyladenine 294 – UV absorption spectrum 294 dioxa[2.2]phanes 57 1,7-dioxa[7](2,7)pyrenophane 97 – degree of the overall bend 97 1,n-dioxa[n](2,7)pyrenophanes 278 1,14-dioxo[3.3]ferrocenophane-2,15diene 139 – preparation 139 1,4-dipentylbenzene 278 – UV absorption data 278 1,1'-diphospha[2]ferrocenophane 135 – preparation 135 dipole transition moments 318 1,10-dithia[2.2]metacyclophane 54 2,11-dithia[3.3]metacyclophane 396 – bridge wobbling 396 – 13C NMR spectrum 396 – 1H NMR spectrum 396 – syn/anti interconversion 396 – VT NMR data 396 di-p-xylylene 212 – generation 212 – resonance structures 212 – p-xylene 212 – gas phase pyrolysis 212 disodium dexamethasone-21-phosphate 531 dispersion interactions 526 distyryl[2.2]paracyclophanes 281 – UV absorption/emission data 281 – photophysics 281 distyrylbenzene 283 dithia[3.3]metacyclophanes 397 f. – conformation 398 – internally substituted 398 – – conformational preferences 398 – NOE measurements 398 – protonation studies 397 dithia[3.3]metaparaphane 397 – 1H NMR chemical shifts calculations 397 dithia[7.7]cyclophantetrayne 31

Subject Index dithia[8]paracyclophanes 56 – molecular structures synthesis 56 dithia[9]corannulenophane 388 – conformation 388 – 1H NMR spectrum 388 – shielding of bridge protons 388 dithia[n]metacyclophanes 57 dithiacorannulenophane 400 – 1H NMR shielding 400 dithiafluorenophanes 1,8-bridged 398 – ring flipping barrier 398 – VT 1H NMR data 398 divinyl[2.2]paracyclophanes 281 – UV absorption emission data 281 DNA base-pair intercalation 525 r-donation 178 donor-acceptor cyclophanes 268 – PE spectra 268 donor-acceptor cyclophanes 298 – UV/Vis spectra 298 donor-acceptor distances 299 donor-acceptor interactions 268 donor-acceptor-substituted [n.n]cyclophanes 305 – extensive p-electron systems 305 donor-acceptor substituted [n.n]paracyclophanes 298 – UV/Vis spectra 298 double helix 295 – hypochromism 295 – UV absorption 295 double-decker cyclophanes – steroid complexation 533 dumbbell-shaped compound 501

e E,Z-bis[3]ferrocenophane-1-ylidenes 137 – preparation 137 echistatin 479 Eglinton-coupling 18 electrochromic materials 428 electron transfer 359 electron withdrawing substituents 268 electronic devices 505, 507, 513 endohedral cobalt phanes 171 – synthesis 171 endohedral rhodium phanes 171 – synthesis 171 enzyme-inhibitor binding 525 enzyme-like catalysts 536 enzyme-substrate binding 525 epoxidation 446 – asymmetric 446

8,16-epoxy[2.2]metacyclophane 204 equinenes 191 b-estradiol 535 2-ethoxymethyl[5]ferrocenophane1,5-dione 144 – synthesis 1441 O-ethylcobalticinium salts 119 – synthesis 119 ethynylene-butadiynylene macrocycles 13 o-(ethynylphenyl)butadiyne derivatives 9 excimer 276 – pyrene 277 exciplexes 276 excited states 314 – cyclophanes 314 exciton interaction 281 exciton-resonance 291 experimental CD spectra 54 extended Hückel calculations 263 exterior electron density model 265

f ferrocene 131 – conformation 131 ferroceno “half square” 10 – synthesis 10 ferroceno[14]annulene 10 ferrocenophane 117, 132, 368, 370, 410, 422 – bridged 117 – 31P chemical shifts 410 – 13C NMR spectra 409 – electrochemistry 368 – 1H NMR spectra 409 – mononuclear 132 – multiply-bridged 132 – nomenclature 132 – penta-bridged 117 – ring-opening polymerization 422 – silicon-bridged 422 1,2-(1,1'-ferrocenylene)-3,4,5,6tetraphenylbenzene 134 [0.0]ferrocenophane 150 – mixed valence system 150 – molecular structure 150 – synthesis 150 [1]ferrocenophane 133, 411, 415 ff., 430 – 57Fe NMR data 411 – chalcogen-bridged 430 – germanium-bridged 418 – group 13 bridged 417 – group 14 bridged 417 – group 15 bridged 419

553

554

Subject Index – – – – – – – – – –

group 16 bridged 419 phosphorus-bridged 422 relative strain 133 silicon-bridged 418, 424 sulfur-bridged 420 synthesis 416 structure 416 tilt angle 133 tin-bridged 423 transition metal-catalyzed polymerization 425 [1n]ferrocenophane 152 – n = 2, 3, 4, and 5 152 [1.1]ferrocenophane 151 – synthesis 151 [1.1]ferrocenophane-1,12-dione 152 – synthesis 152 [2]ferrocenophane 411, 415, 420 – 57Fe NMR data 411 – synthesis 420 [2.2]ferrocenophane-1,13-diene 153 – preparation 153 – synthesis 153 [3]ferrocenophane 140, 411 – 13C NMR spectra 411 – 1,3-trans-disubstituted 140 – 1H NMR spectra 411 – 31P NMR spectra 411 – preparation 140 – relative signs of coupling constants 411 – 119Sn NMR data 411 [3]ferrocenophane-1,3-dione 138 – preparation 138 [3]ferrocenophane-1-ene 138 – preparation 138 [3]ferrocenophane-1-one 136 ff. – condensation reactions 138 – McMurry coupling 137 – molecular structure 137 – preparation 136 [3]ferrocenophane-2-one 138 – preparation 138 [3][3]ferrocenophane 147 – synthesis 147 [3.3]ferrocenophane-1,15-diene3,14-dione 153 – preparation 153 – spectral data 153 [3][3](1,1';2,2')ferrocenophane 147 – synthesis 147 [3][3](1,1';3,3')ferrocenophane 147 – molecular structure 147 – synthesis 147

[3][3][3]ferrocenophane 148 – molecular structure 148 – preparation 148 [4]ferrocenophane-1,4-dione 142 – synthesis 142 [4]ferrocenophane-1-one 140 – preparation 140 [4][3][3](1,1';3,3';4,4')ferrocenophane 149 – synthesis 149 [4][3][3][3][3]ferrocenophane 149 – synthesis 149 [4][3][3][3]ferrocenophane 149 – synthesis 149 [4.4]ferrocenophane-1,3,15,17-tetrayne 153 – CV-data 153 – molecular structure 153 – synthesis 153 – UV/Vis spectrum 153 [5]ferrocenophane-1-dione 145 – UV/Vis spectrum 145 [5]ferrocenophane-1,5-dione 144 – synthesis 144 [5.5]ferrocenophane-1,4,16,19-tetraene3,18-dione 153 – preparation 153 – spectral data 153 [5.5]ferrocenophane-1,3,17,19-tetraene5,16-dione 153 – preparation 153 – spectral data 153 [m]ferrocenophane 132 [m]nferrocenophane 132 b-ferrocenylpropionic acid 137 – Friedel-Crafts cyclization 137 – preparation 137 FHPC 441, 444 f., 454, 458 – allylboration 458 flavin 296 – cyclophanes as models for active site complexes 296 – UV absorption spectra 296 flavin-flavin interaction 293 flavoenzymes 296 – cyclophanes as models for active sites 296 – UV absorption spectra 296 [2.2](2,7)fluorenophane 277 fluorescence 276 fluorinated cyclophanes 290 – UV/Vis spectra 290 5-fluoro-m-benzyne 226 fluoro[33](1,3,5)cyclophanes 190 – ab initio MO calculations 290

Subject Index – blue shifts of p-p* absorption bands 290 4-fluoro-[2.2]paracyclophane 327 – CD spectra 327 (R)-4-fluoro-[2.2]paracyclophane 327 f. – calculated CD 328 – calculated structures 327 – comparison of experimental and theoretical CD spectra 328 force field calculations 217 5-formyl-4-hydroxy[2.2]paracyclophane (FHPC) 441 Fries rearrangement 448 fullerene 337, 355 functionalized [2.2]paracyclophanes 466 – CVD polymerization 466 functionalized cyclophanes 363 – molecular electrochemistry 363 functionalized surfaces 463 [2.2]furanophane 217 – photolysis 217

g glycochenodeoxycholic acid 531 graphdiyne 3 graphyne 3 – nonlinear optical properties 3 – semiconductive properties 3 group IVB ansa-metallocenes 161 f. – singly-boron-bridged synthesis 162 – singly-phosphorous-bridged synthesis 162 – singly-silicon-bridged synthesis 161 group IVB metallocenes 160, 163, 165 – doubly-carbon-bridged synthesis 163 – doubly-silicon-bridged synthesis 163 – singly-carbon-bridged synthesis 160 group IVB metallocenes doubly bridged 166 – dihedral angles 166 – interplanar angles 166 – molecular structures 166 group IVB metals 170 – bridged metallocenophanes 170

h [22]hafnocenophandichloride 165 Hay-coupling 18 helical chiral phanes 51 – circular dichroism 51 [2.2]helicenophanes 392 – 2D NMR data 392 – 1H NMR data 392

13

– C NMR data 392 – NOE stereochemistry 393 – regiochemistry 392 [n](1,6)heptalenophanes 387 – 1H NMR spectra 387 [n](1,6)heptalenophane dianions 387 – 1H NMR 387 – shielding of bridge protons 387 hetera[3]ferrocenophanes 368 – electrochemistry 368 hetera[n]metacyclophanes 56 – planar chiral 56 hetera[n]paracyclophanes 56 – planar chiral 56 heteraphanes 42 – definitions 42 heterocyclophanes 270, 290 – PE spectra 270 – SiH2-SiH2 bridges 270 – UV/Vis spectra 290 [n]heterometacyclophanes 383, 385 – activation parameters 383, 385 – aromatic ring flipping 385 – bridge-flipping 383 – conformational behaviour 383 – 1H NMR data 383, 385 – 13C NMR data 383 – VT NMR data 383, 385 hex-3-ene-1,4-diyne 224 – formation 224 hexa-ethynylbenzene 6 hexaoxa[18.3]paracyclophane quinhydrone 304 hexaprismanes 192 high-dilution conditions 54 higher order [n]rotaxanes 60 – non-ionic template effect 60 HIV protease inhibitor 449 1 H NMR binding studies 534 1 H NMR titrations 529 HOMO of [2n]cyclophanes 122 homoconjugation 259, 272 host-guest chemistry 488 host-guest complexation 486 hydrocarbon cyclophanes 361 – molecular electrochemistry 361 hydrocholic acid 531 hydrocortisone 530 hydrogenation 449 hydrophobic effect 527 b-hydroxy-a-amino acids 441 – diastereoselective synthesis 441 1,1'-(4-hydroxy-1-butenylene)ferrocene 141

555

556

Subject Index – synthesis 141 1-hydroxy-2-methyl[5]ferrocenophane 133 hyodeoxycholic acid 531 hyperconjugation 280

i implant materials 466 inclusion complexation 519, 525, 530, 536, 540 inclusion complexes 485, 520 inclusion geometries 533 increase of donor-acceptor distances 300 indolophane 195 insulin 476 f., 482 – immobilization 476 f., 482 p-interactions 259 r-p interaction 280 interfaces 473 – amino-functionalized 473 internal conversion 283 intramolecular CT interactions 277 intramolecular reaction 201 ion pairing 523 [3.3](3,10)isoalloxazinophane 292 – CT absorptions 292 – flavin bands 292 – p-p interactions 292 [n](3,5)isoxazolophanes 384 – activation parameters 384 – conformational behaviour 384 – VT NMR data 384

k Kammermeierphane 350 Kekuléne 49 b-ketoesters 450 – asymmetric reduction 450 keudomycine 48 Koopmans’ approximation 263

l ladderanes 203 Lalezari procedure 113 lamellar structures 427 Langmuir monolayers 510, 513 Langmuir switching 514 large cavity binding sites 528 lateral substitution 448 laulimalide 47 layered cyclophanes 297 – bathochromism 297 – broadening 297 – hyperchromism 297

N,O-ligands 454, 457 N,P-ligands 454 light-emitting diodes 463 linalool 452 f. – asymmetric synthesis 453 linear movement 515 lipid membranes 523 lipoproteins 534 liquid-liquid extraction 522 lithocholic acid 530 location control 499 low energy excimer states 287 low temperature spectroscopy 211

m macrocyclic polyethers 492 Manzacidin 449 matrix isolation spectroscopy 221 maytansine 50 McMurry coupling 203 mechanical devices 513 mechanical movements 512, 514 mechanically interlocked molecules 499 medicinal chemistry 519 medium-sized cyclic diyne 272 – PE spectra 272 memory 510 memory devices 510 meso-a,a,a,a-tetrakis(2-hydroxy-1-naphthyl)porphyrin 236 – p-p interactions 236 – molecular structure 236 – tetramethoxyquinone complex 236 [n]metacyclophane 81, 382, 384 – acrylonitrile 85 – addition of a phosphinidene complex 86 – deformation angles 83 – deformation of the aromatic ring 83 – dichlorocarbene addition 85 – Diels-Alder reactions 85 – DMAD 85 – 1H NMR spectrum 382 f. – NMR data of dibenzoannulated 384 – molecular structure 86 – photochemical reactions 84 – strain energy 83 – thermal reactions 84 [n.1]metacyclophane 404 – conformations 404 – 1H NMR data 404 – VT NMR data 404 [2.2]metacyclophandienes 197 – photoisomerization 197

Subject Index [2.2]metacyclophane 41, 52, 195 f., 331, 361, 389 – ab initio calculations 389 – calculation of 1H NMR shieldings 389 – CD spectra 321 – chirality 41 – 1H NMR of [2]metacyclo[2](2,7)pyrenophane and reduction product 394 – intra-annular ring closure 196 – oxidation 361 – pyrene 196 [3.2]metacyclophanes 400 – anti-anti'-interconversion 400 – VT 1H NMR data 400 [3.3]metacyclophane 203 [3.3]metacyclophane-2,11-dione 397, 399 – anti-conformation 399 – VT NMR data 399 [3.3]metacyclophane quinhydrone dimethylethers 303 – donor-acceptor orientation 303 – stereoisomers 303 [4]metacyclophane 82 [5]metacyclophane 81, 86, 382 ff. – ab initio calculations 383 – acid-catalyzed rearrangement 86 – activation parameters 382 – bridge-flipping 382 – calculation of 1H NMR shielding 383 – conformation 382 – 1H NMR spectra 382 f. – Meisenheimer complex 87 – NOE results 382 – nucleophilic substitutions 87 – reaction with hydroxide ion 88 – VT NMR data 382 [6]metacyclophane 81, 383 – activation parameters 383 – bridge-flipping 383 – conformational behaviour 383 – 13C NMR spectra 383 – 1H NMR spectra 383 – NOE results 383 – spin-spin coupling constants 383 – VT NMR data 383 [7]metacyclophane 384 – activation parameters 384 – bridge-flipping 384 – conformational behaviour 384 – VT NMR data 384 metacyclophanediynes 272 – PE spectra 272 [n.2]metacyclophane-enes and -ynes 404

– conformations 404 – 1H NMR data 404 – VT NMR data 404 [n.n]metacyclophane quinhydrones 303 – UV/Vis spectra 303 metacyclophynes 17, 19, 25 – alkyne metathesis 19 – complex with C60 fullerene 25 – complex formation with C60 29 – complexes with ruthenium porphyrin 26 – double-decker complex 26 – guest-binding ability 17 – hexamethylbenzene 29 – host-guest chemistry 25 – inclusion complex 29 – nanobox 26 – porphyrin-containing 19 – self-association 17 – solid phase synthesis 19 – strained 19 – supramolecular structures 17, 26 [26]metacyclophynes 19, 22 – acetylene-bridged 19 – alkoxy substituents 22 – nematic liquid crystalline phases 22 metalla(cyclo)metallocenophanes 369 – electrochemistry 369 metallametallocenophanes 367 – electrochemistry 367 metallaphanes 69 [1]metalloarenophanes 419 – silicon-bridged 419 metallocenophanes 367 – electrochemistry 367 metallocephanes 415 – heteroatom-bridged 415 metallocyclophanes 415 metallophanes 44 – belt-shaped 44 metathesis 349 5,12-methano[2.2.2]paracyclophane 178 – complex with Ag+ 178 5-methoxy-m-benzyne 226 1,1'-(1-methoxy-1,3-butadienylene)ferrocene 141 4-methoxymethyl[2.2]paracyclophane 464 5-methyl-m-benzyne 226 (methylcarbonyloxy)[2.2]paracyclophane 464 1-methyl[1.1](1,1'';1',1''')ferrocenophane 133 1-methyl[4]ferrocenophane-1-one 140 – preparation 140 (R)-4-methyl[2.2]paracyclophane 330

557

558

Subject Index – calculated CD spectra 330 – comparison of experimental and theoretical CD spectra 330 4-methyl[2.2]paracyclophane 329 – CD spectra 329 Michael addition 446, 447 Michaelis-Menten parameters 538 microcalorimetry 525 f. microcontact printing 474 microdevice 480 microenviroments 463, 541 microfluidic devices 478 – surface engineering 478 micropeptin T-20 48 miniaturization 510 mixed superphane 113, 117 – synthesis 117 mixed valence compounds 370 Möbius annulene 355 Möbius belts 354 Möbius coronene 354 molecular electrochemistry 359 f. – methods 360 molecular electronics 510 molecular machine 485 f., 498, 515 – definition 485 – design 486 molecular memories 485, 505 molecular recognition 488, 519, 543 molecular shuttle 498, 500 f. – amphiphilic 501 – controllable 500 – template-directed synthesis 498 molecular signature 506 molecular switch 196 molecular trefoil knot 61 – synthesis 61 molecular tube 68 molecular valves 513, 515 monactine 47 monohetera [2.2]metacyclophane 53 monolayers 506 monomer state 283 monothial[3.2]cyclophanes 400 – anti-conformation 400 – protonation studies 400 motion control 504 multibridged [2n]cyclophanes 285 – UV/Vis-spectra 285 multi-bridged cyclophanes 201 multi-layered cyclophanes 297 – CT complex 297 – UV/Vis spectra 297

muscle fiber 513

n nakadomarin A 50 nanoelectromechanical switching 509 nanoelectromechanical systems 513 nanofibers 427 nanomechanic 515 nanotubes 337 – vector notation 337 nanowires 429 [3.2]naphthalenophane 277 – excimer 277 – intramolecular p-p interaction 277 [2.2]naphthalenophane 277, 288 – excimer 277 – excited states 288 – fluorescence and phosphorescence spectra 288 – intramolecular p-p interaction 277 [2.2](2,6)naphthalenophane 288 f. – UV absorption 289 – UV absorption spectra 288 [3.2]naphthalenophane 288 – excited states 288 – fluorescence and phosphorescence spectra 288 [3.3]naphthalenophane 277, 288 – excimer 277 – intramolecular p-p interaction 277 – excited states 288 – fluorescence and phosphorescence spectra 288 [3.3](2,6)naphthalenophane 289 – UV absorption spectra 289 [5](1,3)naphthalenophane 82 – preparation 82 [5](1,4)naphthalenophane 93 – valence isomerization into Dewar form 93 [6](1,4)naphthalenophane 90, 93 – preparations 90 – valence isomerization into Dewar form 93 [m.n]naphthalenophane 288 – UV/Vis spectra 288 [2]naphthalene[2]paracyclophanes 277 NHS-esters 472 f. nicotinamide 296 – cyclophanes as models for active site complexes 296 – UV absorption spectra 296 nitro[2.2]paracyclophane 464

Subject Index NMR titrations 521 nonactine 47 nonyl-(9-nonyl-9H-purin-6-yl)amine 294 – UV absorption spectrum 294 nuclear independent chemical shift (NICS) 6

o octafluoro[2.2]paracyclophane 213, 287 – flash vacuum pyrolysis 213 – UV/Vis spectra 287 oligooxa[m.n]paracyclophane quinhydrones 304 – CT absorption 304 oligooxa[3n.3]paracyclophane quinhydrones 304 – CT absorption 304 – interactions with alkali-, alkaline earth-, mercury(II)-ions 304 – UV absorption spectra 304 oligooxa[n.n]paracyclophane quinhydrones 304 – CT absorption 304 – interactions with alkali-, alkaline earth-, mercury(II)-ions 304 – UV absorption spectra 304 oligo(phenylacetylene) 542 olympiadane 495 f. orbital correlation diagram 266, 315 organometallic cyclophanes 367 – molecular electrochemistry 367 organozinc derivatives 442 – stereoselective addition 442 [24]orthocyclophanetetrayne 13 – structure 13 orthocyclophynes 2 – aromaticity 2 orthoparacyclophyne 30 – preparation 30 oscillator strengths 318 outer-valence Green function calculations 265 2-oxa-1(1,3),3(1,4)-dibenzenacyclodecaphane6-ol and -one 404 – hindered rotation 404 1-oxa[2.2]metacyclophane 54, 57 – calculation of CD spectra 57 1-oxa-10-aza[2.2]metacyclophanes 54 oxaza[2.2]phanes 57 oxazolins 453 oxidative coupling 16 – Cu(II)-templated 16 oxidative demetallation 116

– CpCo-cyclobutadiene-complexes 116 1,1'-(4-oxo-1-butenylene)ferrocene 141 – reduction 141 1-oxo-3-phenacyl[3]ferrocenophane 139 – preparation 139 oxygen trapping 217 o-oxylene intermediate 193

p paddlanes 205 (+)-pamamycine-607 51 [2]paracyclo[2]paracyclo[2]ferrocenophane1,9,17-triene 147 – synthesis 146 [2]paracyclo[2]paracyclo[2]ferrocenophane9-ene-1,17-diyne 147 – synthesis 147 [4]paracyclophandiene 95 – reaction with cyclopentadiene 95 paracyclophane 449 – pseudo-ortho-disubstituted 449 paracyclophyne 28 [0,0]paracyclophane 344 [05]paracyclophane 354 – calculated structures 354 [06]paracyclophane 354 – calculated structures 354 [0n]paracyclophane 344 [1.1]paracyclophane 98 ff., 203, 279 – absorption bands 101 – calculated energies 101 – geometrical structures 100 – kinetic stabilization 100 – molecular structure 101 – preparation 98 – stability 99 – strain energy 100 – thermodynamic stability 100 – transannular interatomic distance 101 – UV/absorption spectra 279 – UV/Vis absorption 100 [2.2]paracyclophane 122, 182, 194 f., 200, 211, 215, 217, 220, 263 ff., 280, 287, 297, 321, 325 f., 361, 363, 372, 389, 391 f., 395, 437 ff., 449, 463 f., 457, 467 – activation parameters 391 – additional bridges 200 – Birch reduction 122, 194 – bis-isocyanate 201 – calculated CD spectra 326 – catalytic processes 439 – CD spectra 325 – chemical vapor deposition 463

559

560

Subject Index – chiral nomenclature and stereochemical assignment 437 – chiral disubstituted inductors 440 – 13C NMR spectra 389 – conformational behaviour 389 – conformational behaviour of the bridges 389 – conjugating substituents 266 – CVD 463 – diamine 201 – enthalpy profile for ring-opening 215 – epoxidation 195 – fluorination 287 – fluorine atoms 265 – frontier orbitals 321 – functionalized 464 – 1H NMR data 389, 391 – 1H NMR spectra of carbazolophanes 395 – 1H NMR spectra of 1,8-fluorenophane 394 – 1H NMR spectra of [2]paracyclo[2](2,7) pyrenophane 394 – 19F NMR spectra 389 – inductive effect 265 – influence of p-p conjugation 267 – r-p interaction 287 – methyl-substituted 267 – r,p mixing 265 – modification of the ethano bridges 287 – monosubstituted 437 – NOE results 389 – optimized polymerization conditions 467 – oxidation 361 – oxidative cleavage 220 – Penning ionization (PI) 264 – PE spectrum 263 – photolysis 217 – polarized absorption spectra 297 – resolution 437 – resolved derivatives 438 – single electron transfer 220 – spin-spin coupling constants 389, 391 – stereoselective hydrogenation 449 – stereoselective nucleophilic addition 457 – thermolysis 211 – thiophenyl derivatives, electrochemistry 363 – through-bond interaction 265 – through-space spin-spin coupling constants 389 – – VT NMR data 391 – tri-ruthenium complex 372

– unsaturated bridges 265 – unusual electronic structure 280 – UV/Vis spectra 280 – vapor-based polymerization 463 [2.2]paracyclophane derivatives 281 – UV/Vis spectra 281 [2.2]paracyclophan-1-ene 221 – thermolysis 221 [2.2]paracyclophane-4,5,12,13-tetracarboxylic 465 [2.2]paracyclophane with one tetracyanobenzene unit 298 – CT absorptions 298 [2.2]paracyclophane-4,7-quinone 456, 458 [2.2.2]paracyclophaneketone 204 (g12-[2.2]paracyclophane)chromium(0) 168 – synthesis 168 (g12-[3.3]paracyclophane)chromium(0) 168 – conformation 168 – molecular structure 168 – radical cation 168 – synthesis 168 [3.2]paracyclophane 1,8-naphthylenebridged 401 – 13C NMR spectrum 401 – 1H NMR spectrum 401 [3.3]paracyclophane 182 [3.3]paracyclophane quinhydrones 300 f. – CT absorptions 300 [3.3]paracyclophane with one tetracyanobenzene unit 298 – CT absorptions 298 [3.3]paracyclophane with TMPD as donor and TCNQ as acceptor 302 [4]paracyclophane 89, 95 f., 279, 314 – diatropicity 96 – matrix isolation 89 – extensive pyramidalization-rehybridization 96 – kinetic stabilization 95 – lifetime in solution 95 – strain energy 96 – UV absorption spectra 279 [4.2]paracyclophane 281 – UV/Vis spectra 281 [4.4]paracyclophane 402 – L-p-boronophenylalanine dimer 402 – 1H NMR data 402 – VT NMR data 402 [4.4]paracyclophane with one tetracyanobenzene unit 298 – CT absorptions 298 [5]paracyclophane 89, 94, 279

Subject Index 1

– H NMR spectra 92 – UV/Vis spectra 92, 279 – reaction with methanol and bromine 94 [6]paracyclophane 89, 94 f., 271, 385 – addition of BuLi 95 – calculation of 1H NMR chemical shifts 385 – 1H NMR spectra 92 – homolysis at benzylic carbon-carbon bonds 92 – PE spectra 271 – oxidation with mCPBA 95 – reaction with acids 94 – reaction with dichlorocarbene 95 – reaction with dienophiles 94 – UV/Vis spectra 92 [7]paracyclophane 271 – PE spectra 271 [8]paracyclophane 271 – PE spectra 271 [13]paracyclophane 386 – activation parameters 386 – aromatic ring flipping 386 – conformational behaviour 386 – 13F NMR data 386 – 1H NMR data 386 – VT NMR data 386 [m.n]paracyclophane 284 f. – UV/Vis spectra 284, 285 [n]paracyclophane 83, 278, 385 f. – deformation angles 83 – UV absorption data 278 – high-pressure 1H NMR spectrum 386 – NMR spectrum 385 – valence isomerization into the Dewar form 93 [6]paracyclophane-8-carboxylic acid 324 – CD spectra 324 [12.12]paracyclophanedodecayne 32 [8.8]paracyclophaneoctayne 31 [6]paracyclophane preparations 90 – by rearrangement from 1,4-Dewar benzenes 90 – by ring contraction 90 [n.n]paracyclophane quinhydrones 300 – UV/Vis spectra 300 [6.6]paracyclophanetetrayne 31 [2.2]paracyclophanyl units 454 [2n]paracyclophyne 28 – 13C NMR spectra 28 – preparation 28

[63]paracyclophyne 30 – hexatriyne-bridged 30 [2]paracyclo[2](2,5)pyridinophane 291 f. – pseudo-geminal 291 – pseudo-meta 291 – pseudo-para 291 – pseudo-ortho 291 – n-p* transitions 292 – UV absorption data 291 paraphenylacetylenes 341 paraquat 487 parylene 467 passive transport 523 patterned surfaces 478 pentakisdehydro[14]annulene 7 pentaoxa[15.3]paracyclophane quinhydrone 304 perfect square 10 [2.2]PHANEPHOS 449 [2.2]phanes 389 – NMR spectra 389 [3.3]phanes 396 – NMR spectra 396 [4.3]phanes 401 – flexibility 401 – 1H NMR spectra 401 [m.n]phanes (m>2, n³2) 400 – NMR spectra 400 [mn]phanes 406 – NMR spectra 406 [n]phanes 387 – NMR spectra 387 phane state 283 phane-specific changes 299 – absorption spectra 299 – broadening 299 – loss of vibrational structure 299 PHANOL 452 [2.2]phenanthrenophanes 392 – 13C NMR spectra 392 – 2D NMR spectra 392 – 1H NMR spectra 392 – identification of stereoisomers 392 [3.2](2,7)phenanthrenophane 401 – conformation 401 – 1H NMR spectra 401 [n.n]phenothiazine bipyridinium cyclophanes (n=3,4,8,11,14,17,20) 405 – conformations 404 – NOE results 404 – ring rotation 405 3-phenyl[5]ferrocenophane-1,5-dimethylene 146

561

562

Subject Index – polymerization 146 3-phenyl[5]ferrocenophane-1,5-dione 146 – polymerization 146 phospha[1]ferrocenophanes 410 – 13C NMR spectra 410 – 1H NMR spectra 410 – 15N NMR spectra 410 – 31P NMR, isotope-induced shifts by 13C and 15N 410 – 195Pt NMR spectra 410 7k3-phosphanorbornadiene 86 photochromic 196 photoluminescence 283 photonic ink 428 picotube 351 ff. – Friedel-Crafts alkylation 353 – pyrolysis 352 – reduction 353 – tetra-anion 353 pilot atom 436 planar chiral phanes 51 – circular dichroism 51 planar chirality 435 platelet adhesion 475 pocket porphyrinophane 251 – molecular structure 251 poly(ferrocenylsilanes) 133 – thermal ROP 133 poly(p-xylylene-2,3-dicarboxylic acid anhydride) 474 poly(p-xylylenes) 463, 469 ff. – immobilization of bioactive substances 472 – optical birefringence 470 – properties 469 – zeta potentials 471 polycyclic aromatic hydrocarbons 521 – complexation 521 polyferrocenylgermanes 430 polyferrocenylphosphines 430 polyferrocenylsilanes 424, 425 ff. – crosslinking 428 – cyclic voltammetric studies 428 – electrical conductivity 428 – functionalization 428 – hole transport properties 428 – liquid crystals 428 – preceramic polymers 429 – pyrolysis 429 polyphenylene 344 porphyrin cyclophanes 403 – conformations 403 – 1H NMR spectra 402

porphyrin cyclophanes phenylspacered 245 – molecular structures 245 – singly bridged by pyridine derivatives 245 porphyrinophane 249 – bridged by anthracene 249 – bridged by pyridine 249 – “capped” by benzene ring 249 – molecular structure 249 porphyrinophane benzene-spacered 243 – molecular structures 243 – singly bridged by pyrene 243 porphyrinophane naphthalene-spacered 235, 238 – molecular structures 235, 238 – singly bridged by dichloro-p-benzoquinone 235 – singly bridged by dimethoxybenzene 238 porphyrinophane phenyl-spacered 230, 232, 240 f., 247 – biphenyl strapped 247 – doubly bridged by p-benzoquinone 230 – interplanar distance 232 – molecular structures 232, 240 f., 247 – singly bridged by 1,4-dialkoxybenzene 240 – singly bridged by dimethoxy-p-benzoquinone 232 – singly bridged by p-benzoquinone 232 – singly bridged by pyromellitic diimide 241 – twist of phenyl rings 230 – – crystal structure 230 – – deviations from planarity 230 pregnenolone acetate 535 p-prismand 182 – complex with Ga+ 182 – molecular structure 182 progesterone 534 propano bridged cyclophanes 125 – pinwheel conformation 125 propella[34]cubane 116 [4.2.2]propellane derivatives 91 – isomerization 90 protein superstructures 541 pseudo-gem derivatives 452 – chiral 452 pseudo-gem effect 198 f. pseudo-geminal orientation 301 pseudo-ortho orientation 301 pseudorotaxanes 486, 492, 499, 503, 505 – photochemically switchable 503

Subject Index – switchable 499 purinophanes 295 – hypochromism 295 – stacking geometries 295 – stacking interactions of nucleic acid base 295 – absorption bands 278 pyrene 278, 522, 526 – extraction 522 pyrenophane 337 – bent 337 [2.2](2,7)pyrenophane 289 – UV absorption spectrum 289 [3.3]pyrenophane 289 – UV absorption spectrum 289 [4.4]pyrenophane 290 – UV absorption spectrum 290 [n.n]pyrenophane 289 – UV/Vis spectra 289 [n](2,7)pyrenophane 97, 387 – addition of PTAD 97 – reaction with TCNE 97 – 1H NMR data 387 – shielding of bridge protons 387 [n](2,7)pyrenophane (n=7–12, 2–7) 278 – absorption spectra 278 [n](2,7)pyrenophane (n=7–9, 8–10) 278 – absorption spectra 278 [23](2,4)pyrenophanetriene dimer 408 – 1H NMR deshielding effects 408 [n](2,7)pyrenophane, unexpected reduction product 388 – 13C NMR spectrum 388 – 1H NMR spectrum 388 – shielding of bridge protons 388 [3](3,6)pyridazino[3](1,3)indolophane 397 – ring conformation 397 [2.2](2,5)pyridinophanes 291 – pseudo-geminal 291 – pseudo-meta 291 – pseudo-para 291 – pseudo-ortho 291 – UV absorption data 291 pyruvate oxidase 538 – cyclophane model 538

q quadruply-layered cyclophanes 297 – absorption spectra 297 quantum yield 202 quinhydrones of [2.2]- to [6.6]paracyclophanes 300

– dependence of CT absorption on distance 300 – mutual orientation 300 quinodimethanes 467 quinolinophane 455

r receptors 519 recognition sites 505 rectangular 528 redox mediator 538 redox potentials 359 redox process 360 – reversibility 360 reversible 501 – switching 501 r-hirudin 475 Rieche formylation 120 rifamycin S 50 ring-opened polyferrocenes 426 – applications 426 – properties 426 ring-opening polymerization 415, 423 – thermal 423 ring-opening polymerization (ROP) 133 rotary motors 515 rotatory strength 311 rotaxanes 49, 59, 374, 485, 497, 499, 503, 513 – self-assembly 497 – switchable 499, 503, 513 – template synthesis 59 [1]rotaxanes 62 – sulfonamide-based 62 [2]rotaxane 497, 500 f., 508 f. – amphiphilic 509 – bistable 508 – electronic devices 508 – slow-shuttling 501 – switchable 500, 509 [3]rotaxanes 61 – anionic template 61 – non-ionic template 61 – synthesis 61 [6]rotaxane 60 – penta-wheeled 60 (Rp)-[2.2]paracyclophan-[4,5-d]-1,3-oxazol-2(3H)-one 446 ruthenocenophanes 416 [n]ruthenocenophanes 170 – synthesis 170 [n2]ruthenocenophanes 170 – synthesis 170

563

564

Subject Index

s salen ligands 444 ff. sanjoinine G1 50 sapphyrine 44 second “phane” excited state 283 self-assembly 488, 495, 519 – metal-ion-mediated 519 sensors 428 separation of enantiomers 57 sexipyridine 49 shuttling process 499 single-molecule transistor 511 f. Sonogashira reactions 3 special pair 375 spectra of cyclophanes 323 – theoretical and experimental 323 spherical binding sites 527 stability constant 488 p,p-stacking 485, 488, 520 state correlation diagram 316 – boat-type deformations of benzene 316 – paracyclophanes 316 p-p* states 317 Stephens-Castro reaction 3 stepwise superphane synthesis 113 stereoselective hydrogenation 449 stereoselective reactions 457 – side chain 457 steroid binding 519, 533 steroid complexation 527, 529 steroid complexes 530 f. – thermodynamic characteristics 530 f. steroid receptors 529 steroid recognition 527 steroids 535 – binding studies 535 stilbene 283 stilbene “dimers” 281 Stille macrocyclization 533 strained [n]paracyclophanes 91 f. – calculated structural parameters 91 – longest wavelength absorption bands 92 – predicted deformation angles 91 – pyramidalization 91 – strain energy 92 – structure 91 streptavidin 474 strongly helical heteraphanes 58 structure-chiroptic rules 312 structure-chiroptics relationships 52 strychnine 195 styrene 362, 455

– cyclic voltammograms 362 – cyclopropanation 455 – polymerization 362 sulfur substituents 108 superferrocenophane 372 superphane 105, 121, 286, 372 f. – Birch reduction 122 – charge transfer complex 122 – cyclophane band 286 – molecular geometry 121 – photoelectron spectrum 121 – UV/Vis spectrum 286 [36]superphane 192 superstolide A 50 superstructures 542 switching 500, 507 – nanoelectromechanical 507 syn-[2.2]metacyclophandienes 196

t template-directed synthesis 491 terpyridine-based macrocycles 24 – layered structures 24 terthiophenophyne 19 – butadiyne-bridged 19 testosterone 530, 534 tetracationic cyclophane 490, 492 – catenones 492 – variation of donors 490 tetradehydrodianthracene 345 ff. – dimerization 351 – electronic structure 346 – metathesis 350 – photochemical reaction 348 – reaction 347 – reaction with pyridazine 349 – reaction with a-pyrone 349 2,3',5,6'-tetrahydro[2.2]paracyclophane 264 – PE spectrum 264 tetrakisdehydro[12]annulene 7 4,5,12,13-tetrakis(methoxycarbonyl)[2.2]paracyclophane 465 tetramethyl[2.2]paracyclophane 297 – absorption spectra 297 tetranactine 47 tetraoxa[4.4]metacyclophane 402 – conformational analysis 402 – 1H NMR data 402 – VT NMR data 402 1,1',7,7' tetraselena[7,7']cyclopropenonophane 108 tetrathiacyclodiynes 108 tetrathiafulvalene (TTF) 293, 489

Subject Index tetrathiafulvalenophanes 365 – electrochemistry 365 thermal copolymerization 424 – ferrocenophanes 424 1-thia-10-aza[2.2]metacyclophane 54, 331 – CD spectra 332 (M)-thia-10-aza[2.2]metacyclophane 332 – comparison of experimental and theoretical CD spectra 332 1-thia[2.2]metacyclophanes 57, 331 – calculation of CD spectra 57 – CD spectra 331 (M)-1-thia[2.2]metacyclophane 331 – comparison of experimental and theoretical CD spectra 331 thin-film polymer coatings 463 thin-film transistors 463 thiopheno[18]annulene 9 thiophene formation 119 – zirconocene dichloride mediated 119 thiopheno cyclobutadieno superphane 119 – stepwise synthesis 119 [24](2,3,4,5)thiophenophane 117 – synthesis 117 – transannular interaction 117 thiophenophynes 15 – twisted 15 – cyclophyne 15 – twisted 15 threading 497 threshold voltages 506 through-bond interactions 259, 287, 334 through-space contributions 287 – near-cancellation 287 through-space interaction 268, 275 tilt-angle 416, 421 tilted structures 415 tilting of rings 416 time-dependent density functional theory 334 time-resolved spectroscopy 211 tissue contact 476 [22]titanocenophanedichloride 163 (p-tolylsulfonyl)methyl isocyanide (TosMIC) 123 topochemical reaction control 202 transannular aldol condensation 123 transannular effect 291 transannular electronic interactions 280 trans-cinnamic 202 – photolysis 202 transition dipole moments 313

translational isomers 494 trans-pyramidalization 348 trapping experiments 211 trefoil knot 63 – synthesis 63 4,10,15-(1,4)tribenzena-1,7-diazabicyclo[5.5.5] heptadecaphane 180 – complex with Ag+ 180 – molecular structure 180 tribenzene-gallium+ 183 – frontier orbitals 183 tricyclic spacers 527 tricyclic tetrathiacyclopropenonophanes 108 tridentate ligands 443 2,4,5-trimethylbenzyl chloride 120 – gas phase pyrolysis 120 5,7,9-trimethyl[3.3.3](1,3,5)cyclophane 409 – barrier to bridge flipping 409 – conformational properties 409 – VT NMR data 409 trimethylsilylcyanation 445 – asymmetric 445 trinactine 47 triple-layered [2.2][2.2]naphthalenophane 297 – CT absorption 297 – tetracyanoethylene-complex 297 triplet states 288 triplet states of the cyclophanes 288 tri-ruthenium complex 372 tris(g6-benzene)gallium 181 – quantum chemical calculations 181 b-truxinic acid 202 twisted macrocycles 14 two-wing propeller molecules 53

u ultracyclic heteraphane 51 ursodeoxycholic acid 530 UV/Vis absorption 276 – classification 276 UV/Vis absorption spectra of arenes 276 – classification by Clar 276 – classification by Platt 276 UV/Vis fine structure 281 UV/Vis spectra 275 – cyclophanes 275 UV spectrum 117

v vancomycin 50 vibrational fine structure 282

565

566

Subject Index 4-vinyl[2.2]paracyclophane 362 – cyclic voltammograms 362 – polymerization 362 Vögtle belt 337

w woodrosine 1 48 write-read cycles 511

x o-xylylene 120 – dimerization reaction 120 p-xylylene 223 xylyl-PHANEPHOS 450

z zeta potentials 471 Zn-porphyrines phenyl-spacered 246 – including H2O, imidazole, benzimidazole 246 – molecular structures 246 – strapped by phenanthroline 246

Zn-porphyrinophane 253 – fullerene (C60) host 253 – molecular structure 253 Zn-porphyrinophane naphthalenespacered 241 – molecular structures 241 – single bridged by 1,4-dialkoxybenzene 241 Zn-porphyrinophane phenyl-spacered 234, 238, 244, 248 – bridged by diphenyl-methane derivative 248 – inclusion of paraquat 244 – molecular structures 234, 238, 244, 248 – single bridged by bis(dimethylamino) benzene 238 – single bridged by 1,5-dialkoxynaphthalene 244 – single bridged by dimethyl-p-benzoquinone 234 – single bridged by tetramethoxybenzene 238