Hansen Solubility Parameters: A User's Handbook

7248_C000.fm Page i Thursday, May 24, 2007 1:40 PM

HANSEN SOLUBILITY PARAMETERS A User’s Handbook Second Edition

7248_C000.fm Page ii Thursday, May 24, 2007 1:40 PM

7248_C000.fm Page iii Thursday, May 24, 2007 1:40 PM

HANSEN SOLUBILITY PARAMETERS A User’s Handbook Second Edition

Charles M. Hansen

7248_C000.fm Page iv Thursday, May 24, 2007 1:40 PM

CRC Press Taylor & Francis Group 6000 Broken Sound Parkway NW, Suite 300 Boca Raton, FL 33487-2742 © 2007 by Taylor & Francis Group, LLC CRC Press is an imprint of Taylor & Francis Group, an Informa business No claim to original U.S. Government works Printed in the United States of America on acid-free paper 10 9 8 7 6 5 4 3 2 1 International Standard Book Number-10: 0-8493-7248-8 (Hardcover) International Standard Book Number-13: 978-0-8493-7248-3 (Hardcover) This book contains information obtained from authentic and highly regarded sources. Reprinted material is quoted with permission, and sources are indicated. A wide variety of references are listed. Reasonable efforts have been made to publish reliable data and information, but the author and the publisher cannot assume responsibility for the validity of all materials or for the consequences of their use. No part of this book may be reprinted, reproduced, transmitted, or utilized in any form by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying, microfilming, and recording, or in any information storage or retrieval system, without written permission from the publishers. For permission to photocopy or use material electronically from this work, please access www.copyright.com (http:// www.copyright.com/) or contact the Copyright Clearance Center, Inc. (CCC) 222 Rosewood Drive, Danvers, MA 01923, 978-750-8400. CCC is a not-for-profit organization that provides licenses and registration for a variety of users. For organizations that have been granted a photocopy license by the CCC, a separate system of payment has been arranged. Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. Library of Congress Cataloging-in-Publication Data Hansen solubility parameters : a user’s handbook. -- 2nd ed. / edited by Charles Hansen. p. cm. Rev. ed. of: Hansen solubility parameters / Charles M. Hansen. c2000. Includes bibliographical references and index. ISBN 0-8493-7248-8 (alk. paper) 1. Solution (Chemistry) 2. Polymers--Solubility. 3. Thin films. I. Hansen, Charles M. II. Hansen, Charles M. Hansen solubility parameters. QD543.H258 2007 547’.70454--dc22 Visit the Taylor & Francis Web site at http://www.taylorandfrancis.com and the CRC Press Web site at http://www.crcpress.com

2006051083

7248_C000.fm Page v Thursday, May 24, 2007 1:40 PM

Contributors Dr. John Durkee Consultant in Critical and Metal Cleaning Hunt, Texas U.S.A. Dr. techn. Charles M. Hansen Consultant Hoersholm, Denmark Prof. Georgios M. Kontogeorgis Technical University of Denmark Department of Chemical Engineering Lyngby, Denmark Prof. Costas Panayiotou Department of Chemical Engineering University of Thessaloniki Thessaloniki, Greece Tim S. Poulsen Sr. Research Scientist Molecular Pathology Glostrup, Denmark

Dr. rer. nat. Hanno Priebe Sr. Research Scientist Chemical Development – Process Research GE Healthcare Amersham Health AS Oslo, Norway Per Redelius Research Manager Nynas Bitumen Product Technology Nynashamn, Sweden Prof. Laurie L. Williams Department of Physics & Engineering Fort Lewis College Durango, Colorado U.S.A.

7248_C000.fm Page vi Thursday, May 24, 2007 1:40 PM

7248_C000.fm Page vii Thursday, May 24, 2007 1:40 PM

Preface to the First Edition My work with solvents started in Denmark in 1962 when I was a graduate student. The major results of this work were the realization that polymer film formation by solvent evaporation took place in two distinct phases and the development of what has come to be called Hansen solubility (or cohesion) parameters, abbreviated in the following by HSP. The first phase of film formation by solvent evaporation is controlled by surface phenomena such as solvent vapor pressure, wind velocity, heat transfer, etc., and the second phase is controlled by concentration-dependent diffusion of solvent molecules from within the film to the air surface. It is not controlled by the binding of solvent molecules to polymer molecules by hydrogen bonding as was previously thought. My solubility parameter work was actually started to define affinities between solvent and polymer to help predict the degree of this binding which was thought to control solvent retention. This was clearly a futile endeavor as there was absolutely no correlation. The solvents with smaller and more linear molecular structure diffused out of the films more quickly than those with larger and more branched molecular structure. HSP were developed in the process, however. HSP have been used widely since 1967 to accomplish correlations and to make systematic comparisons which one would not have thought possible earlier. The effects of hydrogen bonding, for example, are accounted for quantitatively. Many of these correlations are discussed later, including polymer solubility, swelling, and permeation; surface wetting and dewetting; solubility of inorganic salts; and biological applications including wood, cholesterol, etc. The experimental limits on this seemingly universal ability to predict molecular affinities are apparently governed by the limits represented by energies of the liquid test solvents themselves. There had/has to be a more satisfactory explanation of this universality than just “semiempirical” correlations. I decided to try to collect my experience for the purpose of a reference book, both for myself and for others. At the same time, a search of the major theories of polymer solution thermodynamics was undertaken to explore how the approaches compared. A key element in this was to explain why the correlations all seemed to fit with an apparently “universal” 4 (or 0.25 depending on which reference is used). This is described in more detail in Chapter 2 (Equation 2.5 and Equation 2.6). My present view is that the “4” is the result of the validity of the geometric mean rule to describe not only dispersion interactions but also permanent dipole–permanent dipole and hydrogen bonding (electron interchange) interactions in mixtures of unlike molecules. The Hildebrand approach uses this and was the basis of my earliest approach. The Prigogine corresponding states theory yields the “4” in the appropriate manner when the geometric mean rule is adopted (Chapter 2, Equation 2.11). Any other kind of averaging gives the wrong result. Considering these facts and the massive amount of data that has been correlated using the “4” in the following, it appears proven beyond a reasonable doubt that the geometric mean assumption is valid not only for dispersion-type interactions (or perhaps more correctly in the present context those interactions typical of aliphatic hydrocarbons) but also for permanent dipole–permanent dipole and hydrogen bonding as well. For those who wish to try to understand the Prigogine theory, I recommend starting with an article by Donald Patterson.1 This article explains the corresponding states/free volume theory of Prigogine and coworkers in a much simpler form than in the original source. Patterson2 has also reviewed in understandable language the progression of developments in polymer solution thermodynamics from the Flory–Huggins theory, through that of Prigogine and coworkers, to the so-called “New Flory Theory.”3 Patterson also has been so kind as to aid me in the representations of the earlier theories as they are presented here (especially Chapter 2). All of the previous theories and their extensions also can be found in a more recent book.4 For this reason, these more classical

7248_C000.fm Page viii Thursday, May 24, 2007 1:40 PM

theories are not treated extensively as such in this book. The striking aspect about all of this previous work is that no one has dared to enter into the topic of hydrogen bonding. The present quantitative treatment of permanent dipole–permanent dipole interactions and hydrogen bonding is central to the results reported in every chapter in this book. An attempt to relate this back to the previous theories is given briefly here and more extensively in Chapter 2. This attempt has been directed through Patterson,1 which may be called the Prigogine–Patterson approach, rather than through the Flory theory, as the relations with the former are more obvious. I strongly recommend that studies be undertaken to confirm the usefulness of the “structural parameters” in the Prigogine theory (or the Flory theory). It is recognized that the effects of solvent molecular size, segment size, and polymer molecular size (and shapes) are not fully accounted for at the present time. There is hope that this can be done with structural parameters. The material presented here corresponds to my knowledge and experience at the time of writing, with all due respect to confidentiality agreements, etc. I am greatly indebted to many colleagues and supporters who have understood that at times one can be so preoccupied and lost in deep thought that the present just seems not to exist. Charles M. Hansen October 19, 1998

REFERENCES 1. Patterson, D., Role of Free Volume Changes in Polymer Solution Thermodynamics, J. Polym. Sci. Part C, 16, 3379–3389, 1968. 2. Patterson, D., Free Volume and Polymer Solubility. A Qualitative View, Macromolecules, 2(6), 672–677, 1969. 3. Flory, P. J., Thermodynamics of Polymer Solutions, Discussions of the Faraday Society, 49, 7–29, 1970. 4. Lipatov, Y. S. and Nesterov, A. E., Polymer Thermodynamics Library, Vol. 1, Thermodynamics of Polymer Blends, Technomic Publishing Co., Inc., Lancaster, PA, 1997.

7248_C000.fm Page ix Thursday, May 24, 2007 1:40 PM

Preface to the Second Edition When the question about a second edition of this handbook was posed, I was not in doubt that several additional authors were necessary to meet the demands it would require. The writings of the five contributors that were chosen speak for themselves. There is theoretical impact in Chapter 3 (Costas Panayiotou) and in Chapter 4 (Georgios M. Kontogeorgis). Chapter 3 introduces statistical thermodynamics to confirm the division of cohesive energy into three parts enabling separate calculation of each. Chapter 4 describes how the Hansen solubility parameters (HSP) fit into other theories of polymer solutions. The practical applications and understanding provided in Chapter 9 (Per Redelius) related to asphalt, bitumen, and crude oil should accelerate new thinking in this area and emphasize that simple explanations of seemingly complex phenomena are usually the right ones. The thermodynamic treatment of carbon dioxide given in Chapter 10 (Laurie L. Williams) is a model for similar work with other gases and emphatically confirms the usefulness of Hansen solubility parameters for predicting the solubility behavior of gases in liquids and therefore also in polymers. Chapter 11 (John Durkee) goes through the process of demonstrating how “designer” solvents can be used in cleaning operations to replace, or partly replace, ozone-depleting solvents, in spite of the problem of their HSP not being sufficiently close to the HSP of the soils that are to be removed. I have added two chapters because of apparent need. Chapter 14 discusses environmental stress cracking (ESC). ESC is a major cause of unexpected and sometimes catastrophic failure of plastics. The recent improved understanding provided by HSP seemed appropriate for inclusion in this context. Chapter 16 discusses absorption and diffusion in polymers. Many of the HSP correlations presented in this handbook cannot stand on HSP alone but must include consideration of absorption and diffusion of chemicals in polymers. These effects are often disguised by use of a molecular volume, as molecular size/volume correlates reasonably well with diffusion coefficients, especially at low concentrations. Polymer surface layers are often significantly different from the bulk polymer. Surface mobility of polymer chain segments plays an important role in surface dewetting, ESC, and resistance and/or delays to the absorption of chemicals. This chapter tries to unify the effects of a verifiable surface resistance and verifiable concentration-dependent diffusion coefficients. Solutions to the diffusion equation simultaneously considering these two effects explain the “anomalies” of absorption and also correctly model desorption phenomena, including the drying of a lacquer film from start to finish. Each of the chapters in the first edition has been reviewed and added to where this was felt appropriate without increasing the number of pages unduly. There is still a lack of significant activity in the biological area, in controlled release applications, and in other areas discussed in Chapter 18, such as nanotechnology. The relative affinity of molecules or segments of molecules for each other can be predicted and in many cases controlled in self-assembly with the understanding provided by HSP. Chapter 15 treating biological materials has been expanded more than the others included in the first edition. This was done with the help of Tim Svenstrup Poulsen. Perhaps the most surprising of the additions in Chapter 15 is a HSP correlation for the (noncovalent) solvent interactions with DNA. The δD;δP;δH values of 19.0;20.0;11.0 for DNA, all in MPa1/2, clearly show that hydrogen bonding interactions (H) contribute much less to the noncovalent interactions that determine the structure of the DNA than the dispersion (D) and dipolar interactions (P). Only about 14% of the cohesion energy involved in what is commonly called “hydrogen bonding” derives from hydrogen bonding.

7248_C000.fm Page x Thursday, May 24, 2007 1:40 PM

Table Appendix A.1 is greatly expanded both in number and in information. The latter is due to the generous help of Hanno Priebe, the extent of which is clearly evident for those familiar with the first edition. There are close to 1200 entries in this table vs. the approximately 860 in the first edition. However, please be advised that most of these are calculated and not experimental values as indicated in the comments to the table. Table Appendix A.2 is not greatly expanded. There have been too many restrictions on what may be published to allow any major expansion of this table. The majority of my work as a consultant has usually involved agreements that prohibit or severely limit publication of results paid for by private sources. I have also included Appendix A.3 with the original solubility data on which the division of the energy was based. I have regularly found this more specific data of considerable interest. Once more resources and timing have not been conducive to do a complete literature search to provide additional explanations of phenomena that should have had Hansen solubility parameters included in their interpretation. In view of the large expansion in the number of pages over the first edition it is hoped that the principles, both theoretical and practical, are well illuminated. For those who still lack information in a given situation I can suggest a search using the key words “Hansen solubility parameters” followed by additional key words as required. This is true both for Internet searches as well as for searches in the more traditional literature. It has been satisfying to see how much can be interpreted with very simple observations and calculations. If it cannot be done simply, then rethink. I want to once more thank those who have contributed to this second edition. Let us hope others will take up the effort and relate their findings for the benefit of all. Charles M. Hansen

7248_C000.fm Page xi Thursday, May 24, 2007 1:40 PM

The Author Charles M. Hansen consults on the topics covered by this book. He works from his home in Hoersholm, 22 kilometers north of Copenhagen, Denmark. He received a BChE from the University of Louisville and an MS degree from the University of Wisconsin. After being awarded the Dr. techn. degree from the Technical University of Denmark in 1967, he held leading positions with PPG Industries in Pittsburgh, and as director of the Scandinavian Paint and Printing Ink Research Institute in Hoersholm, Denmark. Dr. Hansen dealt with polymers at FORCE Technology, Broendby, Denmark, for the 17 years prior to the start of the current state of semi-retirement. Dr. Hansen is perhaps best known for his extension of the Hildebrand solubility parameter to what are now called Hansen solubility parameters. These have been found mutually confirming with the I. Prigogine corresponding states theory of polymer solutions and can be used to directly calculate the Flory–Huggins interaction coefficient. The statistical thermodynamics approach developed by Costas Panayiotou and coworkers, which is reported in Chapter 3 of this second edition, also confirms the viability of the division of the cohesion energy into separate parts, and allows their independent calculation. Dr. Hansen has published widely in the fields of polymer solubility, diffusion and permeation in polymers and films, surface science, and coatings science. He is currently vice president of the Danish Society for Polymer Technology, having recently completed a 5-year period as president. He frequently reviews papers for leading journals, and is on the editorial board of Progress in Organic Coatings, as well as being a member of the Danish Academy of Technical Sciences (ATV).

7248_C000.fm Page xii Thursday, May 24, 2007 1:40 PM

7248_C000.fm Page xiii Thursday, May 24, 2007 1:40 PM

Key to Symbols Note: The symbols used in Chapters 3 and 16 are so numerous and different that they have been placed in these chapters, respectively. A12 D D DM ED EP EH ΔEv G G ΔGM ΔGMnoncomb H ΔHv ΔHM KH L P P P Q P* R Ra RA RM Ro RED S ΔSM T T Tb Tc Tr V

Energy difference defined by Chapter 2, Equation 2.12 Diffusion coefficient in Chapter13 Dispersion cohesion (solubility) parameter — in tables and computer printouts Dipole moment — debyes Dispersion cohesion energy Polar cohesion energy Hydrogen bonding cohesion energy Energy of vaporization (=) cohesion energy Number of “good” solvents in a correlation, used in tables of correlations Gibbs Energy in Chapter 4 Molar free energy of mixing Noncombinatorial molar free energy of mixing Hydrogen bonding cohesion (solubility) parameter — in tables and computer printouts Molar heat of vaporization Molar heat of mixing Henry’s law constant in Equation 10.5 Ostwald coefficient in Equation 10.6 Permeation coefficient in Chapter 13 Polar cohesion (solubility) parameter — in tables and computer printouts Pressure in Chapter 10 Solvent quality number Total pressure, atm. (Chapter 13, Figures 13.4 and 13.5) Gas constant (1.987 cal/mol K) Distance in Hansen space, see Chapter 1, Equation 1.9 or Chapter 2, Equation 2.5 Distance in Hansen space, see Chapter 2, Equation 2.7 Maximum distance in Hansen space allowing solubility (or other “good” interaction) Radius of interaction sphere in Hansen space Relative energy difference (Chapter 1, Equation 1.10) Solubility coefficient in Chapter 13 Molar entropy of mixing Absolute temperature “Total” number of solvents used in a correlation as given in tables (Normal) boiling point, degrees K Critical temperature, degrees K Reduced temperature, Chapter 1, Equation 1.12 Molar volume, cm3/gram molecular weight

7248_C000.fm Page xiv Thursday, May 24, 2007 1:40 PM

V Vf V* VW VM a ai bi b c c ci f fi fi0 i k k n n nD p pi pis r r ts x y H Ω ΩI ∞ Σ ΔT α α β β δD δH δP δt δ ε ε γ γ

Total volume in Chapter 4 Free volume (Equation 4.2) Hard core or close packed volume in Equation 4.2 van der Waals volume Volume of mixture Constant in van der Waals equation of state (Chapter 4) Activity coefficient of the “i”th component in Appendix 10.A.1 Coefficients in Equations 10.17 and 10.19 Constant in van der Waals equation of state (Chapter 4) Dispersion cohesion energy density from Chapter 1, Figure 1.2 or Figure 1.3 Concentration in Chapter 8, Equation 8.4 Coefficients (state constants) in Equations 10.17 and 10.19 Fractional solubility parameters, defined by Chapter 5, Equations 5.1 to 5.3 Fugacity of the “i”th component in Appendix 10.A.1 Fugacity at standard state in Appendix 10.A.1 Component “i” in a mixture Constant in Equation 6.1 Constant in Equations 10.21–10.23 Coefficient in Equation 10.13 Coefficient in Equaitons 10.21, 10.22, and 10.23 Index of refraction in Equation 10.25 Partial pressure (of carbon dioxide) in Chapter 10 Partial pressure of the “i”th component in Appendix 10.A.1 Saturation pressure of the “i”th component in Appendix 10.A.1 Number of segments in a given molecule, Chapter 2 Ratio of polymer volume to solvent volume (Chapter 4) Sedimentation time, see Chapter 7, Equation 7.1 Mole fraction in liquid phase (Chapter 13, Figures 13.4 and 13.5, and Chapter 10) Mole fraction in vapor phase (Chapter 13, Figures 13.4 and 13.5, and Chapter 10) Ratio of cohesive energy densities; Chapter 2, Equation 2.6 Bunsen coefficient (Equation 10.6) Infinite dilution activity coefficient Summation Lydersen critical temperature group contribution Thermal expansion coefficient Constant in Equation 4.15 Constant in Chapter 2, Equation 2.1 Compressibility in Chapter 10 Dispersion cohesion (solubility) parameter Hydrogen bonding cohesion (solubility) parameter Polar cohesion (solubility) parameter Total (Hildebrand) cohesion (solubility) parameter Prigogine normalized interaction parameter, Chapter 2, Equation 2.8 Cohesive energy for a polymer segment or solvent in Chapter 2 Dielectric constant in Equation 10.25 Surface free energy of a liquid in air or its own vapor Activity coefficient in Chapter 4

7248_C000.fm Page xv Thursday, May 24, 2007 1:40 PM

η ηs ηo [η] [η]N ϕi μ ν Θ Θa Θr ρ ρ ρ ρp ρs σ χ χ12 χc χlit χs 1 2 D P H d p h

Viscosity of solvent, Chapter 7, Equation 7.1 Viscosity of solution Viscosity of solvent Intrinsic viscosity, see Chapter 8, Equation 8.4 Normalized intrinsic viscosity Volume fraction of component “i” Dipole moment Interaction parameter, see Chapter 2, Equation 2.11 Contact angle between liquid and surface Advancing contact angle Receding contact angle Prigogine parameter for differences is size in polymer segments and solvent, Chapter 2, Equation 2.10 Density in Chapter 7, Equation 7.1 Density in Chapter 10 Particle density in Chapter 7, Equation 7.1 Solvent density in Chapter 7, Equation 7.1 Prigogine segmental distance parameter, Chapter 2, Equation 2.10 Polymer–liquid interaction parameter (Flory–Huggins), Chapter 2 Interaction parameter — “New Flory Theory” Critical polymer–liquid interaction parameter, Chapter 2 Representative χ value from general literature Entropy component of χ (Subscript) indicates a solvent (Subscript) indicates a polymer (or second material in contact with a solvent) (Subscript) dispersion component (Subscript) polar component (Subscript) hydrogen bonding component (Subscript) dispersion component (Subscript) polar component (Subscript) hydrogen bonding component

7248_C000.fm Page xvi Thursday, May 24, 2007 1:40 PM

7248_C000.fm Page xvii Thursday, May 24, 2007 1:40 PM

Table of Contents Chapter 1

Solubility Parameters — An Introduction ...................................................................1

Abstract ..............................................................................................................................................1 Introduction ........................................................................................................................................1 Hildebrand Parameters and Basic Polymer Solution Thermodynamics ...........................................2 Hansen Solubility Parameters ............................................................................................................4 Methods and Problems in the Determination of Partial Solubility Parameters ...............................6 Calculation of the Dispersion Solubility Parameter δD ...................................................................13 Calculation of the Polar Solubility Parameter δP ............................................................................16 Calculation of the Hydrogen Bonding Solubility Parameter δH .....................................................17 Supplementary Calculations and Procedures ..................................................................................17 Temperature Dependence .......................................................................................................18 Some Special Effects Temperature Changes .........................................................................19 Effects of Solvent Molecular Size .........................................................................................19 Computer Programs................................................................................................................20 Hansen Solubility Parameters for Water .........................................................................................21 Conclusion........................................................................................................................................22 References ........................................................................................................................................24 Chapter 2

Theory — The Prigogine Corresponding States Theory, χ12 Interaction Parameter, and Hansen Solubility Parameters ...........................................................27

Abstract ............................................................................................................................................27 Introduction ......................................................................................................................................27 Hansen Solubility Parameters (HSP)...............................................................................................28 Resemblance between Predictions of Hansen Solubility Parameters and Corresponding States Theories...............................................................................................30 The χ12 Parameter and Hansen Solubility Parameters.....................................................................32 Comparison of Calculated and Experimental χ12 Parameters .........................................................34 Polybutadiene .........................................................................................................................35 Polyisobutylene.......................................................................................................................36 Polystyrene .............................................................................................................................38 Polyvinylacetate......................................................................................................................39 Polyacrylonitrile .....................................................................................................................39 General Discussion ..........................................................................................................................39 Postscript ..........................................................................................................................................40 Conclusion........................................................................................................................................41 References ........................................................................................................................................42 Chapter 3

Statistical Thermodynamic Calculations of the Hydrogen Bonding, Dipolar, and Dispersion Solubility Parameters..........................................................45

Key words ........................................................................................................................................45 Abstract ............................................................................................................................................45 Introduction ......................................................................................................................................45

7248_C000.fm Page xviii Thursday, May 24, 2007 1:40 PM

Theory ..............................................................................................................................................46 The Equation-of-State Framework.........................................................................................46 The Contribution from Dipolar Forces ..................................................................................50 Applications .....................................................................................................................................52 Discussion and Conclusions ............................................................................................................59 Acknowledgments ............................................................................................................................62 List of Symbols Special to this Chapter..........................................................................................63 References ........................................................................................................................................64 Appendix 3.I: The Acid Dimerization.............................................................................................65 Appendix 3.II: An Alternative Form of the Polar Term..................................................................66 Appendix 3.III: A Group-Contribution Method for the Prediction of δ and δD.............................66 Chapter 4

The Hansen Solubility Parameters (HSP) in Thermodynamic Models for Polymer Solutions ......................................................................................................75

Abstract ............................................................................................................................................75 Group Contribution Methods for Estimating Properties of Polymers ............................................76 The Group-Contribution Principle and Some Applications (Density, Solubility Parameters) ................................................................................................76 GC Free-Volume-Based Models for Polymers (Entropic-FV, Unifac-FV) ...........................77 The Free-Volume Concept .........................................................................................77 The UNIFAC-FV Model ............................................................................................77 The Entropic Model ...................................................................................................78 The Flory–Huggins Model and the Regular Solution Theory ..............................................80 Rules of Thumb and Solvent Selection Using the Flory–Huggins Model and Solubility Parameters ..................................................................................81 Activity Coefficients Models Using the HSP..................................................................................82 Flory–Huggins Models Using Hildebrand and Hansen Solubility Parameters (HSP) .........82 The FH/Hansen Model vs. the GC Methods .............................................................84 Applications............................................................................................................................85 Solvent Selection for Paints (Activity Coefficients at Infinite Dilution) ..................85 Mixed Solvent–Polymer Phase Equilibria .................................................................88 Conclusions and Future Challenges ................................................................................................90 List of Abbreviations........................................................................................................................91 Symbols in this Chapter...................................................................................................................92 Appendix 4.I: An Expression of the Flory–Huggins Model for Multicomponent Mixtures .........92 References ........................................................................................................................................93 Chapter 5

Methods of Characterization — Polymers ................................................................95

Abstract ............................................................................................................................................95 Introduction ......................................................................................................................................95 Calculation of Polymer HSP ...........................................................................................................97 Solubility — Examples....................................................................................................................98 Swelling — Examples ...................................................................................................................106 Melting Point Determinations — Effect of Temperature..............................................................106 Environmental Stress Cracking......................................................................................................107 Intrinsic Viscosity Measurements ..................................................................................................107 Other Measurement Techniques ....................................................................................................109 Conclusion......................................................................................................................................109 References ......................................................................................................................................110

7248_C000.fm Page xix Thursday, May 24, 2007 1:40 PM

Chapter 6

Methods of Characterization — Surfaces................................................................113

Abstract ..........................................................................................................................................113 Introduction ....................................................................................................................................113 Hansen Solubility Parameter Correlations with Surface Tension (Surface Free Energy)............113 Method to Evaluate the Cohesion Energy Parameters for Surfaces.............................................114 A Critical View of the Critical Surface Tensions..........................................................................116 A Critical View of the Wetting Tension ........................................................................................117 Additional Hansen Solubility Parameter Surface Characterizations and Comparisons ...............118 Self-Stratifying Coatings................................................................................................................120 Maximizing Physical Adhesion .....................................................................................................122 Conclusion......................................................................................................................................122 References ......................................................................................................................................122 Chapter 7

Methods of Characterization for Pigments, Fillers, and Fibers ..............................125

Abstract ..........................................................................................................................................125 Introduction ....................................................................................................................................125 Methods to Characterize Pigment, Filler, and Fiber Surfaces ......................................................126 Discussion — Pigments, Fillers, and Fibers .................................................................................127 Hansen Solubility Parameter Correlation of Zeta Potential for Blanc Fixe.................................131 Carbon Fiber Surface Characterization .........................................................................................131 Controlled Adsorption (Self-Assembly) ........................................................................................132 Conclusion......................................................................................................................................134 References ......................................................................................................................................134 Chapter 8

Applications — Coatings and Other Filled Polymer Systems................................137

Abstract ..........................................................................................................................................137 Introduction ....................................................................................................................................137 Solvents ..........................................................................................................................................137 Techniques for Data Treatment......................................................................................................142 Solvents and Surface Phenomena in Coatings (Self-Assembly) ..................................................144 Polymer Compatibility...................................................................................................................145 Hansen Solubility Parameter Principles Applied to Understanding Other Filled Polymer Systems ..................................................................................................................147 Conclusion......................................................................................................................................147 References ......................................................................................................................................148 Chapter 9

Hansen Solubility Parameters of Asphalt, Bitumen, and Crude Oils .....................151

Abstract ..........................................................................................................................................151 Symbols Special to Chapter 9 .......................................................................................................151 Introduction ....................................................................................................................................151 Models of Bitumen ........................................................................................................................152 Asphaltenes ....................................................................................................................................154 Molecular Weight .................................................................................................................154 Polarity..................................................................................................................................155 Solubility Parameters of Bitumen..................................................................................................155 Testing of Bitumen Solubility........................................................................................................156 Hildebrand Solubility Parameters ..................................................................................................156 Hansen Solubility Parameters (HSP).............................................................................................158

7248_C000.fm Page xx Thursday, May 24, 2007 1:40 PM

The Solubility Sphere ....................................................................................................................159 Computer Program for Calculation and Plotting of the Hansen 3D Pseudosphere .....................161 Components of Bitumen ................................................................................................................164 Bitumen and Polymers...................................................................................................................166 Crude Oil........................................................................................................................................169 Turbidimetric Titrations .................................................................................................................170 BISOM Test ...................................................................................................................................170 Conclusion......................................................................................................................................173 References ......................................................................................................................................174 Chapter 10 Determination of Hansen Solubility Parameter Values for Carbon Dioxide ..........177 Abstract ..........................................................................................................................................177 Introduction ....................................................................................................................................177 Methodology ..................................................................................................................................178 One-Component Hildebrand Parameter as a Function of Temperature and Pressure..................187 Three-Component (Hansen) Solubility Parameters — Pure CO2 .................................................189 Temperature and Pressure Effects on HSPs: δd.............................................................................190 Temperature and Pressure Effects on HSPs: δp.............................................................................191 Temperature and Pressure Effects on HSPs: δh.............................................................................191 Conclusion......................................................................................................................................196 Acknowledgments ..........................................................................................................................196 Chapter 10 Addendum ...................................................................................................................196 Symbols Special to this Chapter....................................................................................................197 References ......................................................................................................................................197 Appendix 10.A.1: Ideal Solubility of Gases in Liquids and Published CO2 Solubility Data .....199 Ideal Solubility of Gases in Liquids..............................................................................................199 References ......................................................................................................................................201 Chapter 11 Use of Hansen Solubility Parameters to Identify Cleaning Applications for “Designer” Solvents .................................................................................................203 Abstract ..........................................................................................................................................203 Introduction ....................................................................................................................................203 A Variety of Solvents.....................................................................................................................204 Pathology of Soils ..........................................................................................................................204 HSP of Multiple-Component Soils................................................................................................204 Method for Calculating HSP of Composites (Soils or Solvents) .................................................205 More Realistic View about Evaluating HSP of Composite Soils .................................................206 Method for Choice of Suitable Solvents .......................................................................................206 Reference Soils for Comparison....................................................................................................208 Identification of Designer Solvents ...............................................................................................208 An Open Question — Answered ...................................................................................................208 Limiting RA Value for Expected Good Cleaning Performance ....................................................210 Application of HSP Methodology to Cleaning Operations ..........................................................212 Analysis of Capability of Designer Solvents ................................................................................213 Conclusions ....................................................................................................................................215 Notes ..............................................................................................................................................227

7248_C000.fm Page xxi Thursday, May 24, 2007 1:40 PM

Chapter 12 Applications — Chemical Resistance......................................................................231 Abstract ..........................................................................................................................................231 Introduction ....................................................................................................................................231 Chemical Resistance — Acceptable-or-Not Data .........................................................................232 Effects of Solvent Molecular Size.................................................................................................232 Chemical Resistance — Examples ................................................................................................233 Tank Coatings .......................................................................................................................233 PET Film Coating ................................................................................................................234 Acceptable or Not — Plastics..............................................................................................234 Tensile Strength ....................................................................................................................237 Special Effects with Water.............................................................................................................238 Conclusion......................................................................................................................................239 References ......................................................................................................................................240 Chapter 13 Applications — Barrier Polymers............................................................................243 Abstract ..........................................................................................................................................243 Introduction ....................................................................................................................................243 Concentration-Dependent Diffusion ..............................................................................................244 Solubility Parameter Correlations Based on Permeation Phenomena ..........................................245 Solubility Parameter Correlations of Breakthrough Times .................................................245 Solubility Parameter Correlation of Permeation Rates .......................................................248 Solubility Parameter Correlation of Polymer Swelling ................................................................250 Solubility Parameter Correlation of Permeation Coefficients for Gases ......................................251 Laminates..............................................................................................................................253 General Considerations ..................................................................................................................255 Conclusion......................................................................................................................................256 References ......................................................................................................................................257 Chapter 14 Applications — Environmental Stress Cracking in Polymers ................................259 Abstract ..........................................................................................................................................259 Introduction ....................................................................................................................................259 ESC Interpreted Using HSP ..........................................................................................................260 ESC with Nonabsorbing Stress Cracking Initiators......................................................................263 Discussion ......................................................................................................................................264 Conclusion......................................................................................................................................267 References ......................................................................................................................................267 Chapter 15 Hansen Solubility Parameters — Biological Materials...........................................269 Abstract ..........................................................................................................................................269 Introduction ....................................................................................................................................270 Hydrophobic Bonding and Hydrophilic Bonding (Self-Association)...........................................271 DNA ..............................................................................................................................................273 Cholesterol .....................................................................................................................................275 Lard ................................................................................................................................................277

7248_C000.fm Page xxii Thursday, May 24, 2007 1:40 PM

Human Skin....................................................................................................................................277 Proteins — Blood Serum and Zein ...............................................................................................279 Chlorophyll and Lignin..................................................................................................................279 Wood Chemicals and Polymers .....................................................................................................279 Urea ..............................................................................................................................................283 Water ..............................................................................................................................................289 Surface Mobility ............................................................................................................................290 Chiral Rotation, Hydrogen Bonding, and Nanoengineering.........................................................290 Conclusion......................................................................................................................................291 References ......................................................................................................................................291 Chapter 16 Absorption and Diffusion in Polymers ....................................................................293 Abstract ..........................................................................................................................................293 List of Symbols Used in This Chapter..........................................................................................293 Introduction ....................................................................................................................................294 Steady State Permeation ................................................................................................................296 The Diffusion Equation..................................................................................................................296 Constant Diffusion Coefficients ...........................................................................................296 Concentration Dependent Diffusion Coefficients ................................................................297 Surface Resistance .........................................................................................................................298 Mathematical Background....................................................................................................298 Surface Resistance in Absorption Experiments ...................................................................300 Surface Resistance in Permeation Experiments ..................................................................301 Surface Resistance — A Discussion....................................................................................302 Side Effects ....................................................................................................................................304 Measuring Diffusion Coefficients with Surface Resistance and Concentration Dependence.......................................................................................304 Film Formation by Solvent Evaporation .......................................................................................305 Anomalous Diffusion (Case II, Super Case II).............................................................................306 General Comments.........................................................................................................................308 Conclusion......................................................................................................................................308 References ......................................................................................................................................309 Chapter 17 Applications — Safety and Environment ................................................................311 Abstract ..........................................................................................................................................311 Introduction ....................................................................................................................................311 Substitution.....................................................................................................................................311 Alternative Systems .......................................................................................................................312 Solvent Formulation and Personal Protection for Least Risk.......................................................313 The Danish Mal System — The Fan.............................................................................................313 Selection of Chemical Protective Clothing ...................................................................................315 Uptake of Contents by a Plastic Container ...................................................................................315 Skin Penetration .............................................................................................................................316 Transport Phenomena.....................................................................................................................316 Conclusion......................................................................................................................................317 References ......................................................................................................................................318

7248_C000.fm Page xxiii Thursday, May 24, 2007 1:40 PM

Chapter 18 The Future ................................................................................................................321 Abstract ..........................................................................................................................................321 Introduction ....................................................................................................................................321 Hansen Solubility Parameter Data and Data Quality....................................................................324 Group Contribution Methods.........................................................................................................328 Polymers as Points — Solvents as Spheres ..................................................................................328 Characterizing Surfaces .................................................................................................................330 Materials and Processes Suggested for Further Attention ............................................................332 Surface Active Agents ..........................................................................................................332 Surface Mobility (Self-Assembly) .......................................................................................333 Water.....................................................................................................................................334 Gases.....................................................................................................................................336 Organic Salts ........................................................................................................................337 Inorganic Salts ......................................................................................................................337 Organometallic Compounds .................................................................................................338 Aromas and Fragrances........................................................................................................338 Absorption of Chemicals in Plastics....................................................................................339 Chemical Resistance.............................................................................................................339 Controlled Release................................................................................................................339 Nanotechnology....................................................................................................................340 Theoretical Problems Awaiting Future Resolution........................................................................341 Polymer Solubility................................................................................................................341 Surface Phenomena ..............................................................................................................342 Conclusion......................................................................................................................................342 References ......................................................................................................................................342 Appendix A: Comments to Table A.1 ...........................................................................................345 References ......................................................................................................................................346 Table A.1 ........................................................................................................................................347 Appendix A: Comments to Table A.2 ...........................................................................................485 References ......................................................................................................................................490 List of Trade Names and Suppliers ...............................................................................................491 Table A.2 ........................................................................................................................................493 Appendix A: Comments to Table A.3 ...........................................................................................507 Table A.3 ........................................................................................................................................508 Index...............................................................................................................................................511

7248_C000.fm Page xxiv Thursday, May 24, 2007 1:40 PM

7248_C001.fm Page 1 Monday, April 23, 2007 1:56 PM

Parameters — 1 Solubility An Introduction Charles M. Hansen ABSTRACT Solubility parameters have found their greatest use in the coatings industry to aid in the selection of solvents. They are used in other industries, however, to predict compatibility of polymers, chemical resistance, and permeation rates, and even to characterize the surfaces of pigments, fibers, and fillers. Liquids with similar solubility parameters will be miscible, and polymers will dissolve in solvents whose solubility parameters are not too different from their own. The basic principle has been “like dissolves like.” More recently, this has been modified to “like seeks like,” as many surface characterizations have also been made, and surfaces do not (usually) dissolve. Solubility parameters help put numbers into this simple qualitative idea. This chapter describes the tools commonly used in Hansen solubility parameter (HSP) studies. These include liquids used as energy probes and computer programs to process data. The goal is to arrive at the HSP for interesting materials either by calculation or, if necessary, by experiment and preferably with agreement between the two.

INTRODUCTION The solubility parameter has been used for many years to select solvents for coatings materials. A lack of total success has stimulated further research. The skill with which solvents can be optimally selected with respect to cost, solvency, workplace environment, external environment, evaporation rate, flash point, etc., has improved over the years as a result of a series of improvements in the solubility parameter concept and widespread use of computer techniques. Most commercial suppliers of solvents have computer programs to help with solvent selection. One can now easily predict how to dissolve a given polymer in a mixture of two solvents, neither of which can dissolve the polymer by itself. Unfortunately, this book cannot include discussion of all the significant efforts leading to our present knowledge of the solubility parameters. An attempt is made to outline developments, provide some background for a basic understanding, and give examples of uses in practice. The key factor is to determine those affinities that the important components in a system have for each other. For many products this means evaluating or estimating the relative affinities of solvents, polymers, additives, pigment surfaces, filler surfaces, fiber surfaces, and substrates. It is noteworthy that the concepts presented here have developed toward not just predicting solubility that requires high affinity between solvent and solute, but for predicting affinities between different polymers, leading to compatibility, and affinities to surfaces to improve dispersion and adhesion. In these applications the solubility parameter has become a tool, using well-defined liquids as energy probes, to measure the similarity, or lack of the same, of key components. Materials with widely different chemical structures may be very close in affinities. Only those materials that interact differently with different solvents can be characterized in this manner. It can be expected that many inorganic materials, such as fillers, will not interact differently with these energy probes

1

7248_C001.fm Page 2 Monday, April 23, 2007 1:56 PM

2

Hansen Solubility Parameters: A User’s Handbook

as their energies are very much higher. An adsorbed layer of water on the high-energy surface can also play an important role. Regardless of these concerns, it has been possible to characterize pigments, both organic and inorganic, as well as fillers like barium sulfate, zinc oxide, etc., and also inorganic fibers (see Chapter 7). Changing the surface energies by various treatments can lead to a surface that can be characterized more readily and often interacts more strongly with given organic solvents. When the same solvents that dissolve a polymeric binder are those which interact most strongly with a surface, it can be expected that the binder and the surface have high affinity for each other. Solubility parameters are sometimes called cohesion energy parameters as they are derived from the energy required to convert a liquid to a gas. The energy of vaporization is a direct measure of the total (cohesive) energy holding the liquid’s molecules together. All types of bonds holding the liquid together are broken by evaporation, and this has led to the concepts described in more detail later. The term cohesion energy parameter is more appropriately used when referring to surface phenomena.

HILDEBRAND PARAMETERS AND BASIC POLYMER SOLUTION THERMODYNAMICS The term solubility parameter was first used by Hildebrand and Scott.1,2 The earlier work of Scatchard and others was contributory to this development. The Hildebrand solubility parameter is defined as the square root of the cohesive energy density: δ = (E/V)1/2

(1.1)

Where V is the molar volume of the pure solvent, and E is its (measurable) energy of vaporization (see Equation 1.15). The numerical value of the solubility parameter in MPa1/2 is 2.0455 times larger than that in (cal/cm3)1/2. The solubility parameter is an important quantity for predicting solubility relations, as can be seen from the following brief introduction. Thermodynamics requires that the free energy of mixing must be zero or negative for the solution process to occur spontaneously. The free energy change for the solution process is given by the relation: ΔGM = ΔHM – ΔTSM

(1.2)

where ΔGM is the free energy of mixing, ΔHM is the heat of mixing, T is the absolute temperature, and ΔSM is the entropy change in the mixing process. Equation 1.3 gives the heat of mixing as proposed by Hildebrand and Scott: ΔHM = ϕ1ϕ2VM(δ1 – δ2)2

(1.3)

The φ1 and φ2 are volume fractions of solvent and polymer, and VM is the volume of the mixture. Equation 1.3 is not correct, and it has often been cited as a shortcoming of this theory in that only positive heats of mixing are allowed. It has been shown by Patterson, Delmas, and coworkers that ΔGMnoncomb is given by the right-hand side of Equation 1.3 and not ΔGM. This is discussed more in Chapter 2. The correct relation is3–8: ΔGMnoncomb = ϕ1ϕ2VM(δ1 – δ2)2

(1.4)

The noncombinatorial free energy of solution, ΔGMnoncomb, includes all free energy effects other than the combinatorial entropy of solution that results by simply mixing the components. Equation

7248_C001.fm Page 3 Monday, April 23, 2007 1:56 PM

Solubility Parameters — An Introduction

3

1.4 is consistent with the Prigogine corresponding states theory (CST) of polymer solutions (see Chapter 2) and can be differentiated to give expressions3,4 predicting both positive and negative heats of mixing. Therefore, both positive and negative heats of mixing can be expected from theoretical considerations and have been measured accordingly. It has been clearly shown that solubility parameters can be used to predict both positive and negative heats of mixing. Previous objections to the effect that only positive values are allowed in this theory are incorrect. This discussion clearly demonstrates why the solubility parameter should be considered as a free energy parameter. This is more in agreement with the use of the solubility parameter plots to follow. These use solubility parameters as axes and have experimentally determined boundaries of solubility defined by the fact that the free energy of mixing is zero. The combinatorial entropy enters as a constant factor in the plots of solubility in different solvents, for example, as the concentrations are usually constant for a given study. It is important to note that the solubility parameter, or rather the difference in solubility parameters for the solvent–solute combination, is important in determining the solubility of the system. It is clear that a match in solubility parameters leads to a zero change in noncombinatorial free energy, and the positive entropy change (the combinatorial entropy change), found on simple mixing to result in a disordered mixture compared to the pure components, will ensure that a solution is possible from a thermodynamic point of view. The maximum difference in solubility parameters that can be tolerated where the solution still occurs is found by setting the noncombinatorial free energy change equal to the combinatorial entropy change: ΔGMnoncomb = TΔSMcomb

(1.5)

This equation clearly shows that an alternate view of the solubility situation at the limit of solubility is that it is the entropy change that dictates how closely the solubility parameters must match each other for the solution to occur. It will be seen in Chapter 2 that solvents with smaller molecular volumes will be thermodynamically better than larger ones having identical solubility parameters. A practical aspect of this effect is that solvents with relatively low molecular volumes, such as methanol and acetone, can dissolve a polymer at larger solubility parameter differences than might be expected from comparisons with other solvents with larger molecular volumes. An average solvent molecular volume is usually taken as about 100 cc/mol. The converse is also true. Larger molecular species may not dissolve, even though solubility parameter considerations might predict they would. This can be a difficulty in predicting the behavior of plasticizers solely based on data for lower molecular weight solvents. These effects are also discussed elsewhere in this book, particularly in Chapter 2, Chapter 12, Chapter 13, and Chapter 16. A shortcoming of the earlier solubility parameter work is that the approach was limited to regular solutions, as defined by Hildebrand and Scott,2 and does not account for association between molecules, such as those that polar and hydrogen-bonding interactions would require. The latter problem seems to have been largely solved with the use of multicomponent solubility parameters; however, the lack of accuracy with which the solubility parameters can be assigned will always remain a problem. Using the difference between two large numbers to calculate a relatively small heat of mixing, for example, will always be problematic. A more detailed description of the theory presented by Hildebrand, and the succession of research reports that have attempted to improve on it, can be found in Barton’s extensive handbook.9 The slightly older, excellent contribution of Gardon and Teas10 is also a good source of related information, particularly for coatings and adhesion phenomena. The approach of Burrell,11 who divided solvents into hydrogen bonding classes, has found numerous practical applications; the approach of Blanks and Prausnitz12 divided the solubility parameter into two components, “nonpolar” and “polar.” Both are worthy of mention, however, in that the first has found wide use and the second greatly influenced the author’s earlier activities. The Prausnitz article, in particular, was

7248_C001.fm Page 4 Monday, April 23, 2007 1:56 PM

4

Hansen Solubility Parameters: A User’s Handbook

farsighted in that a corresponding states procedure was introduced to calculate the dispersion energy contribution to the cohesive energy. This is discussed in more detail in Chapter 2. It can be seen from Equation 1.2 that the entropy change is beneficial to mixing. When multiplied by the temperature, this will work in the direction of promoting a more negative free energy of mixing. This is the usual case, although there are exceptions. Increasing temperature does not always lead to improved solubility relations. Indeed, this was the basis of the pioneering work of Patterson and coworkers,3–8 to show that subsequent increases in temperature can predictably lead to insolubility. Their work was done in essentially nonpolar systems. Increasing temperature can also lead to a nonsolvent becoming a solvent and, subsequently, a nonsolvent again with still further increase in temperature. Polymer solubility parameters do not change much with temperature, but those of a liquid frequently decrease rapidly with temperature. This situation allows a nonsolvent, with a solubility parameter that is initially too high, to pass through a soluble condition to once more become a nonsolvent as the temperature increases. These are usually “boundary” solvents on solubility parameter plots. The entropy changes associated with polymer solutions will be smaller than those associated with liquid–liquid miscibility, for example, as the “monomers” are already bound into the configuration dictated by the polymer they make up. They are no longer free in the sense of a liquid solvent and cannot mix freely to contribute to a larger entropy change. This is one reason polymer–polymer miscibility is difficult to achieve. The free energy criterion dictates that polymer solubility parameters match extremely well for mutual compatibility, as there is little to be gained from the entropy contribution when progressively larger molecules are involved. However, polymer–polymer miscibility can be promoted by the introduction of suitable copolymers or comonomers that interact specifically within the system. Further discussion of these phenomena is beyond the scope of the present discussion; however, see Chapter 5.

HANSEN SOLUBILITY PARAMETERS A solubility parameter approach proposed by the author for predicting polymer solubility has been in wide use. The basis of these so-called HSPs is that the total energy of vaporization of a liquid consists of several individual parts.13–17 These arise from (atomic) dispersion forces, (molecular) permanent dipole–permanent dipole forces, and (molecular) hydrogen bonding (electron exchange). Needless to say, without the work of Hildebrand and Scott1,2 and others not specifically referenced here, such as Scatchard, this postulate could never have been made. The total cohesive energy, E, can be measured by evaporating the liquid, i.e., breaking all the cohesive bonds. Thus the total cohesive energy is considered as being identical to the energy of vaporization. It should also be noted that these cohesive energies arise from interactions of a given solvent molecule with another of its own kind. The basis of the approach is, therefore, very simple, and it is surprising that so many different applications have been possible since 1967 when the idea was first published. A rather large number of applications are discussed in this book. Others are found in the works of Barton.9 A lucid discussion by Barton18 enumerates typical situations where problems occur when using solubility parameters. These appear most often where the environment causes the solvent molecules to interact, with or within themselves, differently from the way they do in situations where they make up their own environment, i.e., as pure liquids. Several cases are discussed where appropriate in the following chapters. Materials with similar HSP have high affinity for each other. The extent of the similarity in a given situation determines the extent of the interaction. The same cannot be said of the total or Hildebrand solubility parameter.1,2 Ethanol and nitromethane, for example, have similar total solubility parameters (26.1 vs. 25.1 MPa1/2, respectively), but their affinities are quite different. Ethanol is water soluble, whereas nitromethane is not. Indeed, mixtures of nitroparaffins and alcohols were demonstrated in many cases to provide synergistic mixtures of two nonsolvents that dissolved

7248_C001.fm Page 5 Monday, April 23, 2007 1:56 PM

Solubility Parameters — An Introduction

5

polymers.13 This could never have been predicted by Hildebrand parameters, whereas the HSP concept readily confirms the reason for this effect. There are three major types of interactions in common organic materials. The most general are the nonpolar interactions. These are derived from atomic forces and have also been called dispersion interactions in the literature. As molecules are built up from atoms, all molecules contain those types of attractive forces. For the saturated aliphatic hydrocarbons, for example, these are essentially the only cohesive interactions, and the energy of vaporization is assumed to be the same as the dispersion cohesive energy, ED. Finding the dispersion cohesive energy as the cohesion energy of the homomorph, or hydrocarbon counterpart, is the starting point for calculating the three Hansen parameters for a given liquid. As discussed in more detail later, this is based on a corresponding states calculation. The permanent dipole–permanent dipole interactions cause a second type of cohesion energy, the polar cohesive energy, EP. These are inherently molecular interactions and are found in most molecules to one extent or another. The dipole moment is the primary parameter used to calculate these interactions. A molecule can be mainly polar in character without being water soluble, hence there is a misuse of the term polar in the general literature. The polar solubility parameters referred to here are well-defined, experimentally verified, and can be estimated from molecular parameters as described later. As noted previously, the most polar of the solvents include those with relatively high total solubility parameters that are not particularly water soluble, such as nitroparaffins, propylene carbonate, and tri-n-butyl phosphate. Induced dipoles have not been treated specifically in this approach but are recognized as a potentially important factor, particularly for solvents with zero dipole moments (see the Calculation of the Polar Solubility Parameter section). The third major cohesive energy source is hydrogen bonding, EH. This can be called more generally an electron exchange parameter. Hydrogen bonding is a molecular interaction and resembles the polar interactions in this respect. The basis of this type of cohesive energy is attraction among molecules because of the hydrogen bonds. In this perhaps oversimplified approach, the hydrogen bonding parameter has been used to more or less collect the energies from interactions not included in the other two parameters. Alcohols, glycols, carboxylic acids, and other hydrophilic materials have high-hydrogen-bonding parameters. Other researchers have divided this parameter into separate parts — for example, acid and base cohesion parameters — to allow both positive and negative heats of mixing. These approaches will not be dealt with here but are described in Barton’s handbook9 and elsewhere.19–21 The most extensive division of the cohesive energy has been done by Karger et al.,22 who developed a system with five parameters: dispersion, orientation, induction, proton donor, and proton acceptor. As a single parameter, the Hansen hydrogen bonding parameter has served remarkably well in the experience of the author and keeps the number of parameters to a level that allows ready practical usage. It is clear that there are other sources of cohesion energy arising in various types of molecules from, for example, induced dipoles, metallic bonds, electrostatic interactions, or whatever type of separate energy can be defined. The author stopped with the three major types found in organic molecules. It has been recognized that additional parameters could be assigned to separate energy types. For example, the description of organometallic compounds could be an intriguing study. This would presumably parallel similar characterizations of surface-active materials, where each part of the molecule requires separate characterization for completeness. The Hansen parameters have mainly been used in connection with solubility relations, mostly, but not exclusively, in the coatings and related industries. Solubility and swelling have been used to confirm the solubility parameter assignments of many of the liquids. Group contribution methods and suitable equations based on molecular properties were then derived from these. They make possible estimates of the three parameters for additional liquids. The goal of a prediction is to determine the similarity or difference of the cohesion energy parameters. The strength of a particular type of hydrogen bond or any other bond is important only to the extent that it influences the cohesive energy density.

7248_C001.fm Page 6 Monday, April 23, 2007 1:56 PM

6

Hansen Solubility Parameters: A User’s Handbook

HSPs do have direct applications in other scientific disciplines, such as surface science, where they have been used to characterize the wettability of various surfaces and the adsorption properties of pigment surfaces,10,14,16,23–26 and have even led to systematic surface treatment of inorganic fibers so that they could be readily incorporated into polymers of low-solubility parameters such as polypropylene27 (see also Chapter 7). Many widely different applications have been discussed by Barton9 and Gardon.28 Surface characterizations have not been given the attention deserved in terms of a unified similarity-of-energy approach. The author can certify that thinking in terms of similarity of energy, whether surface or cohesive energies, can lead to rapid decisions and plans of action in critical situations that lack data. In other words, the everyday industrial crisis situation often can be reduced in scope by appropriate systematic approaches based on similarity of energy. The success of the HSPs for surface applications are not surprising in view of the similarity of predictions offered by these, and the Prigogine corresponding states theory of polymer solutions discussed in Chapter 2. Flory also emphasized that it is the surface of molecules that interact to produce solutions,29 so the interactions of molecules residing in surfaces should clearly be included in any general approach to interactions among molecules. Surface mobility and surface rotation are important factors in environmental stress cracking (Chapter 14), certain biological phenomena (Chapter 15), the wetting of surfaces, and in other important phenomena relating to nanotechnology (Chapter 18). The basic equation governing the assignment of Hansen parameters is that the total cohesion energy, E, must be the sum of the individual energies that make it up. E = ED + EP + EH

(1.6)

Dividing this by the molar volume gives the square of the total (or Hildebrand) solubility parameter as the sum of the squares of the Hansen D, P, and H components. E/V = ED/V + EP/V + EH/V

(1.7)

δ2 = δ2 D + δ2 P + δ2 H

(1.8)

To sum up this section, it is emphasized that HSPs quantitatively account for the cohesion energy (density). Up to this point of time, an experimental latent heat of vaporization has been considered a more reliable method to arrive at a cohesion energy rather than using molecular orbital calculations or other calculations based on potential functions. Indeed, the goal of such extensive calculations for polar and hydrogen bonding molecules should be to accurately arrive at the energy of vaporization. The statistical thermodynamics approach of Panayiotou and coworkers reported in Chapter 3 may have changed this. An alternative method of calculating the three parameters has been presented, but full evaluation of this new information has not been possible as yet.

METHODS AND PROBLEMS IN THE DETERMINATION OF PARTIAL SOLUBILITY PARAMETERS The best method to calculate individual HSPs depends to a great extent on what data are available. The author originally adopted an essentially experimental procedure and established values for 90 liquids based on solubility data for 32 polymers.13 This procedure involved calculation of the nonpolar parameter according to the procedure outlined by Blanks and Prausnitz.12 This calculation procedure is still in use and is considered the most reliable and consistent one for this parameter. It is outlined in the following section. The division of the remaining cohesive energy between the polar and hydrogen bonding interactions was initially done by trial and error to fit experimental polymer solubility data. A key to parameter assignments in this initial trial-and-error approach was

7248_C001.fm Page 7 Monday, April 23, 2007 1:56 PM

Solubility Parameters — An Introduction

7

that mixtures of two nonsolvents could be systematically and synergistically (but predictably) found to dissolve given polymers. This meant that these had parameters placing them on opposite sides of the solubility region, a spheroid. Using a large number of such predictably synergistic systems as a basis, reasonably accurate divisions into the three energy types were possible. Using the experimentally established, approximate, δP and δH parameters, Hansen and Skaarup15 found that the Böttcher equation (Equation 10.25) could be used to calculate the polar parameter quite well, and this led to a revision of the earlier values to those now accepted for the same liquids. These values were also consistent with the experimental solubility data for 32 polymers available at that time and with Equation 1.6. Furthermore, Skaarup developed the equation for the solubility parameter “distance,” Ra, between two materials based on their respective partial solubility parameter components: (Ra)2 = 4(δD2 – δD1)2 + (δP2 – δP1)2 + (δH2 – δH1)2

(1.9)

This equation was developed from plots of experimental data where the constant “4” was found convenient and correctly represented the solubility data as a sphere encompassing the good solvents (see Chapter 5). When the scale for the dispersion parameter is doubled, in comparison with the other two parameters essentially spherical, rather than spheroidal, regions of solubility are found. This greatly aids two-dimensional plotting and visualization. There are, of course, boundary regions where deviations can occur. These are most frequently found to involve the larger molecular species as being less effective solvents compared to their smaller counterparts that define the solubility sphere. Likewise, smaller molecular species such as acetone, methanol, nitromethane, and others often appear as outliers, in that they dissolve a polymer even though they have solubility parameters placing them at a distance greater than the experimentally-determined radius of the solubility sphere, Ro. This dependence on molar volume is inherent in the theory developed by Hildebrand, Scott, and Scatchard discussed previously. Smaller molar volume favors lower ΔGM, as discussed in Chapter 2. This in turn promotes solubility. Such smaller-molecular-volume species that dissolve “better” than predicted by comparisons, based on solubility parameters alone, should not necessarily be considered outliers. The molar volume is frequently and successfully used as a fourth parameter to describe the effects of molecular size. For example, these are especially important in correlating diffusional phenomena with HSP (see Chapter 12, Chapter 13, and Chapter 16). The author has preferred to retain the three, well-defined partial-solubility parameters with a fourth, separate, molar volume parameter, rather than multiplying the solubility parameters by the molar volume raised to some power to redefine them. The reason for the experimentally determined constant 4 in Equation 1.9 will be discussed in more detail in Chapter 2. It will be noted here, however, that the constant 4 is theoretically predicted by the Prigogine corresponding states theory of polymer solutions when the geometric mean is used to estimate the interaction in mixtures of dissimilar molecules.30 The constant 4 differentiates between atomic and molecular intereactions. This is exceptionally strong evidence that dispersion, permanent dipole–permanent dipole, and hydrogen bonding interactions all follow the geometric mean rule. Patterson and coworkers have been especially instrumental in relating the Prigogine theory to solubility parameters and to the Flory–Huggins theory of polymer solutions.3–8 The HSP approach of dividing the cohesive energy into parts derived from different types of cohesive forces has been confirmed both by experimental studies, as well as the Prigogine theory. The use of the geometric mean is basic to this agreement between the HSP approach and that of Prigogine (see Chapter 2). The approach of optimizing solubility data to spheres is still very much in use. Plotting regions of solubility based on experimental solubility data, or computer-optimizing boundaries of solubility by locating the maximum difference in solubility parameters allowed by Equation 1.9 are both used. The total free energy of mixing, ΔGM, is equal to zero on the boundary. It should be recognized

7248_C001.fm Page 8 Monday, April 23, 2007 1:56 PM

8

Hansen Solubility Parameters: A User’s Handbook

that using the solubility parameters relating to ΔGMnoncomb in Equation 1.4 differs from this by the combinatorial entropy of mixing. Another promising approach to arrive at the HSP for materials based on experimental data is to use multivariable analysis of one type or another, as discussed in Chapter 5. This type of approach has not been attempted by the author, but it clearly has advantages in some cases. The author’s preferred approach of locating the polymer HSP as the center of a sphere has a problem in that it is, in reality, the poor solvents or nonsolvents located near the boundary of the sphere that fix the boundary (and center) rather than the best solvents in the middle. This may present problems for smaller sets of data, but it is an advantage when extrapolating into regions of HSP higher than those of any liquid that can be used in testing. This is discussed in more detail in Chapter 5 and the definition of the limited segment of the boundary of the HSP sphere derivable from such correlations is based on Equation 1.9. Equation 1.9 is readily used on a computer (or on a hand calculator), and supplementary relations allow easier scanning of large sets for data. It is obvious that solubility, or high affinity, requires that Ra be less than Ro. The ratio Ra/Ro has been called the RED number, reflecting the relative energy difference. RED = Ra/Ro

(1.10)

A RED number of 0 is found for no energy difference, RED numbers less than 1.0 indicate high affinity; RED equal to or close to 1.0 is a boundary condition; and progressively higher RED numbers indicate progressively lower affinities. Scanning a computer output for RED numbers less than 1.0, for example, rapidly allows location of the most interesting liquids for a given application. Parenthetically, it should be noted that the ratio of Ra to Ro is really a ratio of quantities having the same units as the solubility parameter. The ratio (Ra/Ro)2 = (RED)2 is a ratio of cohesion energies. The latter quantity is important for relating the HSP approach to that of Huggins and Flory, as discussed in Chapter 2. The revised set of parameters for the 90 original solvents was the basis for group contribution procedures developed (most notably) by van Krevelen,31 Beerbower,32 and Hansen and Beerbower,17 who also used Fedors’ work.33 These various developments have been summarized by Barton,9 although Beerbower’s latest values have only appeared in the National Aeronautics and Space Administration (NASA) document.32 Table 1.1 is an expanded table of Beerbower group contributions, which was distributed among those who were in contact with Beerbower in the late 1970s. The majority of the data in this table, as well as Table 1.2, have also appeared in Reference 34. Beerbower also developed a simple equation for the polar parameter,17 which involved only the dipole moment and the square root of the molar volume. This is also given later (Equation 1.13) and has been found quite reliable by Koenhen and Smolders.35 This equation has been found reliable by the author as well, giving results generally consistent with Equation 1.6 to Equation 1.8, which, again, is the basis of the whole approach. Koenhen and Smolders also give correlation coefficients for other calculation procedures to arrive at the individual Hansen parameters. The group contributions in Table 1.1 have been used extensively to arrive at the collection of HSP data in Appendix Table A.1. Most of the chemicals of primary interest for which full data are available are presumably already in this table. The trend has been to calculate HSP for larger and still larger molecules. Many of these have multiple groups, and it becomes more and more difficult to make decisions as to how to treat them best. At times the HSP for the larger molecules can be estimated from the HSP of larger segments that make them up. Rather than expanding Table 1.1 with additional data, except as noted briefly later, the usual practice has been to locate chemicals with similar groups and to use their HSP values in a group contribution-type calculation. The procedure has developed to the point where its principle features can be identified in the following table. If a boiling point is available, the procedures for calculating δD have been used. If a boiling point is not available, the similarity with related molecules has been used. If a dipole

7248_C001.fm Page 9 Monday, April 23, 2007 1:56 PM

Solubility Parameters — An Introduction

9

moment is available, the procedures given here were used in preference to group contributions. If necessary, group contributions can be derived from similar molecules to the one in question, when dipole moments are available for these, and not for the molecule in question. There is often a change in the group contribution as a function of molecular size. This is the main reason for the lack of expansion of Table 1.1. It is thought best that the uncertainty be clear to the user. For example, it has been found that the group contribution for the polar component of aliphatic esters should not be less than 300 cal/mol as given in Table 1.1. This is necessary to prevent δP for materials like plasticizers from being clearly too low, based on their compatibility with, for example, polyvinyl chloride. Sulfur containing compounds have also been somewhat difficult in this respect, with major changes in estimated group contributions depending on molecular weight of the chemical in question. Group contributions for sulfur, amides, and other groups not found in Table 1.1 can be easily derived from the data on similar compounds reported in Appendix Table A.1. The same is true of the δH component. The problem with this procedure is obvious: any error or distortion of value for a class of compounds is perpetuated. This has been recognized and dealt with to the extent possible, but there are limits to what can be done with limited data. The scope of this situation has been beyond the resources available for its fully satisfactory resolution. The extensive list of group contributions at the end of Chapter 3 provides what may be a partial replacement and/or a supplement for Table 1.1. This requires some experience with the techniques involved. A sizable number of materials have been assigned HSPs using the procedures described here. Many of these have not been published. Exxon Chemical Corporation36,37 has indicated a computer program with data for over 500 solvents and plasticizers, 450 resins and polymers, and 500 pesticides. The author’s files contain the three parameters for about 1200 chemicals (See Appendix Table A.1), although several of them appear with two sets of possible values awaiting experimental confirmation. In some cases, this is due to questionable physical data, for example, for latent heats of vaporization, or wide variations in reported dipole moments. Another reason is that some liquids are chameleonic,38 as defined by Hoy, in that they adopt configurations depending on their environment. Hoy38 cites the formation of cyclic structures for glycol ethers with (nominally) linear structure. The formation of hydrogen-bonded tetramers of alcohols in a fluoropolymer has also been pointed out.39 The term compound formation can be found in the older literature, particularly where mixtures with water were involved and structured species were postulated to explain phenomena based on specific interactions among the components of the mixtures. Barton has discussed some of the situations where cohesion parameters need a more careful use, and points out that Hildebrand or Hansen parameters must be used with particular caution where the extent of donor–acceptor interactions, especially, hydrogen bonding within a compound, is very different from that between compounds.18 Amines, for example, are known to associate with each other. Pure component data cannot be expected to predict the behavior in such cases. Still another reason for difficulties is the large variation of dipole moments reported for the same liquid. The dipole moment for some liquids depends on their environment, as discussed later. A given solvent can be listed with different values in files to keep these phenomena in mind. Large data sources greatly enhance a search for similar materials and the locating of new solvents, as an example, for a polymer for which there are limited data. Unfortunately, different authors have used different group contribution techniques, and there is a proliferation of different “Hansen” parameters for the same chemicals in the literature. This would seem to be an unfortunate situation, but may ultimately provide benefits. In particular, partial solubility parameter values found in Hoy’s extensive tables9,40 are not compatible with the customary Hansen parameters reported here. Hoy has provided an excellent source of total solubility parameters. He independently arrived at the same type of division of cohesion energies as Hansen, although the methods of calculation were quite different. Many solvent suppliers have also presented tables of solvent properties and/or use computer techniques with these tables in their technical service. Partial solubility parameters not taken directly from earlier well-documented sources should be used with caution. The Hoy dispersion parameter,

33.5 16.1 –1.0 –19.2 28.5 13.5 –5.5 — 16 16 18.0 40.0 66.0 24.0 52.0 81.9 30.0 62.0 97.2 31.5 66.6 111.0 3.8 10.8 (23.2) 18.0 28.5 10.0 Same Same Same Same Same Same Same 71.4 — Same 22.0 48.0 78.0 28.0 60.0 73.9 34.0 70.0 109.2 35.5 74.6 123.0 Same Same (31.4) Same Same Same

Aliphatic Aromatic 1,125 1,180 820 350 850 ± 100 875 ± 100 800 ± 100 — — — 0 0 0 1,400 ± 100 3,650 ± 160 4,750 ± 300c 1,950 ± 300c 4,300 ± 300c 5,800 ± 400c 2,350 ± 250c 5,500 ± 300c ? 0 —e 950 ± 300 —f 3,350 ± 300 1,770 ± 450

Alkane Same Same Same Same ? ? ? — 250 250 0 0 0 ? ? ? 1,500 ? ? 2,200 ? ? 0 2,350 ? ? 3,550 1,370 ± 250 ± 500

± 400

± 250c

± 175

Cyclo Same Same Same Same ? ? ? 7,530 — 250 0 0 0 1,300 ± 100 3,100 ± 175c ? 1,650 ± 140 3,500 ± 300c ? 2,000 ± 250c 4,200 ± 300c ? 0 2,800 ± 325 550 ± 275 —f 3,600 ± 400 1,870 ± 600

Aromatic

London Parameter, ΔVδD2 (cal/mol)

0 0 0 0 25 ± 10 18 ± 5 60 ± 10 — 0 0 1,000 ± 150 700 ± 250c ? 1,250 ± 100 800 ± 150 300 ± 100 1,250 ± 100 800 ± 250c 350 ± 150c 1,250 ± 100 800 ± 250c ? 500 ± 150 (15, 000 ± 7%)/V 2,100 ± 200 (56,000 ± 12%)/V 500 ± 150 700 ± 200

Alkane 0 0 0 0 ? ? ? — 0 0 ? ? ? 1,450 ± 100 ? ? 1,700 ± 150 ? ? 1,350 ± 100 ? ? 600 ± 150 1,000 ± 300 3,000 ± 500 ? 300 ± 50 1,100 ± 300

Cyclo Aromatic 0 0 0 0 ? ? ? 50 ± 25 — 0 700 ± 100 500 ± 250c ? 800 ± 100 400 ± 150c ? 800 ± 100 400 ± 150c ? 575 ± 100 400 ± 150c ? 450 ± 150 950 ± 300 2,750 ± 200 (338,000 ± 10%)/V 750 ± 350 800 ± 150

Polar Parameter, ΔVδP2 (cal/mol

0 0 0 0 180 ± 75 180 ± 75 180 ± 75 — 0 0 0 0 0 100 ± 20c 165 ± 10c 350 ± 250c 500 ± 100 825 ± 200c 1,500 ± 300c 1,000 ± 200c 1,650 ± 250c ? 450 ± 25 800 ± 250d 1,000 ± 200 1,250 ± 150 2,750 ± 250 4,650 ± 400

Aliphatic 0 0 0 0 ? ? ? 50 ± 50c — 0 0 0 0 Same 180 ± 10c ? 500 ± 100 800 ± 250c ? 1,000 ± 200c 1,800 ± 250c ? 1,200 ± 100 400 ± 125c 750 ± 150 475 ± 100c 2,250 ± 250c 4,650 ± 500

Aromatic

Electronic Transfer Parameter, ΔVδH2 (cal/mol)

1,125 1,180 820 350 1,030 1,030 1,030 — 250 250 1,000 1,700 1,650 2,760 4,600 5,400 3,700 5,900 7,650 4,550 8,000 11,700 800 4,150 (4,050) 4,300 6,600 7,120

Aliphatic

Same Same Same Same Same Same Same 7630 — 250 800b 1,360b 1,315b 2,200b 3,670b 4,300b 2,960b 4,700b 6,100b 3,600b 6,400b 9,350b (1,650 ± 150) Same Same Same Same Same

Aromatic

Total Parametera ΔVδ2 (cal/mol)

10

CH3 CH2< –CH< >C< CH2 = olefin –CH = olefin >C = olefin PhenylC-5 ring (saturated) C-6 ring –F F2 twinf F3 tripletf –Cl Cl2 twinf Cl3 tripletf –Br Br2 twinf Br3 tripletf –I I2 twine I3 triplete –O– ether >CO ketone –CHO –COO-ester –COOH –OH

Functional Group

Molar Volume,a ΔV (cm3/mol)

TABLE 1.1 Group Contributions to Partial Solubility Parameters

7248_C001.fm Page 10 Monday, April 23, 2007 1:56 PM

Hansen Solubility Parameters: A User’s Handbook

Same Same 32.0 Same Same Same Same

26.0

24.0 24.0 19.2 4.5 (6.7) 28.0 1,600 3,000 1,050 1,150 ? —e

0 ± ± ± ±

850c 600 300 225 ? ? 1,050 ± 450c ? ? ?

? 0 2,550 ± 125 150 ± 150c ? ? ?

? 4,000 ± 800c 3,600 ± 600 600 ± 200 100 ± 50 ? (81,000 ± 10%)/V

1,500 ± 100 ? ? 600 ± 350c ? ? ?

? 3,750 ± 300c 1,750 ± 100 800 ± 200 ? ? ?

? 500 ± 200d 400 ± 50d 1,350 ± 200 750 ± 200 2,700 ± 550c 3,000 ± 500

9,000 ± 600 400 ± 125c 350 ± 50c 2,250 ± 200d ? ? ?

9,300 ± 600 4,150 7,000 3,000 2,000 (5,850) (7,000)

10,440 Same (4,400) Same Same Same Same

Same

Source: From Hansen, C. M., Paint Testing Manual, Manual 17, Koleske, J. V., Ed., American Society for Testing and Materials, Philadelphia, 1995, 388. Copyright ASTM. Reprinted with permission.

b

Data from Fedors, R.F., A method for estimating both the solubility parameters and molar volumes of liquids, Polym. Eng. Sci., 14(2), 147–154, 472, 1974. With permission. These values apply to halogens attached directly to the ring and also to halogens attached to aliphatic double-bonded C atoms. c Based on very limited data. Limits shown are roughly 95% confidence; in many cases, values are for information only and not to be used for computation. d Includes unpublished infrared data. e Use formula in ΔVδ 2 column to calculate, with V for total compound. P f Twin and triplet values apply to halogens on the same C atom, except that ΔVδ 2 also includes those on adjacent C atoms. P

a

(OH)2 twin or adjacent –CN –NO2 –NH2 amine >NH2 amine –NH2 amide PO4 ester

7248_C001.fm Page 11 Monday, April 23, 2007 1:56 PM

Solubility Parameters — An Introduction 11

7248_C001.fm Page 12 Monday, April 23, 2007 1:56 PM

12

Hansen Solubility Parameters: A User’s Handbook

TABLE 1.2 Lydersen Group Constants Group

Aliphatic, ΔT

Cyclic, ΔT

ΔPT

Aliphatic, ΔP

Cyclic, ΔP

CH3 CH2 >CH– >C< CH2 CH– C< CH aromatic CH aromatic

0.020 0.020 0.012 0.000 0.018 0.018 0.000 — —

— 0.013 0.012 –0.007 — 0.011 0.011 — —

0.0226 0.0200 0.0131 0.0040 0.0192 0.0184 0.0129 0.0178 0.0149

0.227 0.227 0.210 0.210 0.198 0.198 0.198 — —

— 0.184 0.192 0.154 — 0.154 0.154 — —

–O– >O epoxide –COO– >CO –CHO –CO2O

0.021 — 0.047 0.040 0.048 —

0.014 — — 0.033 — —

0.0175 0.0267 0.0497 0.0400 0.0445 0.0863

0.16 — 0.47 0.29 0.33 —

0.12 — — 0.02 — —

–OH→ –H→ –OH primary –OH secondary –OH tertiary –OH phenolic

— — 0.082 — — 0.035

— — — — — —

0.0343 –0.0077 0.0493 0.0440 0.0593 0.0060

0.06 — — — — –0.02

— — — — —

–NH2 –NH– >N– –CN

0.031 0.031 0.014 0.060

— 0.024 0.007 —

0.0345 0.0274 0.0093 0.0539

0.095 0.135 0.17 0.36

— 0.09 0.13 —

–NCO HCON< –CONH– –CON< –CONH2 –OCONH–

— — — — — —

— — — — — —

0.0539 0.0546 0.0843 0.0729 0.0897 0.0938

— — — — — —

— — — — — —

–S– –SH

0.015 0.015

0.008 —

0.0318 —

0.27 —

0.24 —

–Cl 1° –Cl 2° Cl1 twin Cl aromatic

0.017 — — —

— — — —

0.0311 0.0317 0.0521 0.0245

0.320 — — —

— — — —

–Br –Br aromatic

0.010 —

— —

0.0392 0.0313

0.50 —

— —

–F –I

0.018 0.012

— —

0.006 —

0.224 0.83

— —

7248_C001.fm Page 13 Monday, April 23, 2007 1:56 PM

Solubility Parameters — An Introduction

13

TABLE 1.2 (CONTINUED) Lydersen Group Constants Group

Aliphatic, ΔT

Cyclic, ΔT

ΔPT

Aliphatic, ΔP

Cyclic, ΔP

Conjugation cis double bond trans double bond

— — —

— — —

0.0035 –0.0010 –0.0020

— — —

— — —

4 5 6 7

— — — —

— — — —

0.0118 0.003 –0.0035 0.0069

— — — —

— — — —

Ortho Meta Para

— — —

— — —

0.0015 0.0010 0.0060

— — —

— — —

Bicycloheptyl Tricyclodecane

— —

— —

0.0034 0.0095

— —

— —

Member Member Member Member

ring ring ring ring

Source: Hansen, C.M., Solubility parameters, in Paint Testing Manual, Manual 17, Koleske, J.V., Ed., American Society for Testing and Materials, Philadelphia, PA, 1995, 383–404. Reprinted with permission.

in particular, is consistently lower than that found by Hansen. Hoy subtracts estimated values of the polar and hydrogen-bonding energies from the total energy to find the dispersion energy. This allows for more calculational error and underestimates the dispersion energy, as the Hoy procedure does not appear to fully separate the polar and hydrogen-bonding energies. The van Krevelen dispersion parameters appear to be too low. The author has not attempted these calculations, being completely dedicated to the full procedure based on corresponding states described here, but values estimated independently using the van Krevelen dispersion parameters are clearly low. A comparison with related compounds or the similarity principle gives better results than those found from the van Krevelen dispersion group contributions. In the following, calculation procedures and experience are presented according to the procedures most reliable for the experimental and/or physical data available for a given liquid.

CALCULATION OF THE DISPERSION SOLUBILITY PARAMETER δD The δD parameter is calculated according to the procedures outlined by Blanks and Prausnitz.12 Figure 1.1 to Figure 1.3 can be used to find this parameter, depending on whether the molecule of interest is aliphatic, cycloaliphatic, or aromatic. These figures have been inspired by Barton,9 who converted earlier data to Standard International (SI) units. All three of these figures have been straight-line extrapolated into a higher range of molar volumes than that reported by Barton. Energies found with these extrapolations have also provided consistent results. As noted earlier, the solubility parameters in SI units (MPa1/2) are 2.0455 times larger than (ca1/cc)1/2 in the older cgs centimeter gram second (cgs) system, which still finds extensive use in the U.S., for example. The figure for the aliphatic liquids gives the dispersion cohesive energy, ED, whereas the other two figures directly report the dispersion cohesive energy density, c. The latter is much simpler to use, as one need only take the square root of the value found from the figure to find the respective partial solubility parameter. Barton also presented a similar figure for the aliphatic solvents, but it

7248_C001.fm Page 14 Monday, April 23, 2007 1:56 PM

14

Hansen Solubility Parameters: A User’s Handbook

Δ ED kJ/mol 70

60 = Tr

50

0 0.4 5 0.4 0 0.5 5 0.5 0.60 0.65

40

0.70

30

20

10

0 50

100

150

200

250

V, cm3/mol

FIGURE 1.1 Energy of vaporization for straight chain hydrocarbons as a function of molar volume and reduced temperature. (From Hansen, C. M., Paint Testing Manual, Manual 17, Koleske, J. V., Ed., American Society for Testing and Materials, Philadelphia, 1995, 389. Copyright ASTM. Reprinted with permission.)

400

350 c, MPa

Tr = 0.40 0.45

300

0.50 0.55 0.60 0.65 0.70

250

50

60

70

80

90

100

110

120

130

3

V, cm /mol FIGURE 1.2 Cohesive energy density for cycloalkanes as a function of molar volume and reduced temperature. (From Hansen, C. M., Paint Testing Manual, Manual 17, Koleske, J. V., Ed., American Society for Testing and Materials, Philadelphia, 1995, 389. Copyright ASTM. Reprinted with permission.)

7248_C001.fm Page 15 Monday, April 23, 2007 1:56 PM

Solubility Parameters — An Introduction

15

400

c, MPa

Tr = 0.40 350 0.45 0.50

300

0.55 0.60 0.65 0.70

250

80

90

100

110

120 130 V, cm3/mol

140

150

160

170

FIGURE 1.3 Cohesive energy density for aromatic hydrocarbons as a function of molar volume and reduced temperature. (From Hansen, C. M., Paint Testing Manual, Manual 17, Koleske, J. V., Ed., American Society for Testing and Materials, Philadelphia, 1995, 389. Copyright ASTM. Reprinted with permission.)

is inconsistent with the energy figure and in error. Its use is not recommended. When substituted cycloaliphatics or substituted aromatics are considered, simultaneous consideration of the two separate parts of the molecule is required. The dispersion energies are evaluated for each of the molecules involved, and a weighted average is taken for the molecule of interest based on the number of significant atoms. For example, hexyl benzene would be the arithmetic average of the dispersion energies for an aliphatic and an aromatic liquid, each with the given molar volume of hexyl benzene. Liquids such as chlorobenzene, toluene, and ring compounds with alkyl substitutions that have only two or three carbon atoms have been considered only as cyclic compounds. Such weighting has been found necessary to satisfy Equation 1.6. The critical temperature, Tc, is required to use the dispersion energy figures. If the critical temperature cannot be found, it must be estimated. A table of the Lydersen group contributions,41 ΔT , as given by Hoy40 for calculation of the critical temperature is included as Table 1.2. In some cases, the desired groups may not be in the table, which requires some educated guessing. The end result does not appear too sensitive to these situations. The normal boiling temperature, Tb, is also required in this calculation. This is not always available and must be estimated by similarity, group contribution, or some other technique. The Lydersen group contribution method involves the use of Equation 1.11 and Equation 1.12 as follows: Tb/Tc = 0.567 + ΣΔT – (ΣΔT)2

(1.11)

Tr = T/Tc

(1.12)

and

where T has been taken as 298.15 K. The dispersion parameter is based on atomic forces. The size of the atom is important. It has been found that corrections are required for atoms significantly larger than carbon, such as chlorine, sulfur, bromine, etc., but not for oxygen or nitrogen that have a similar size. The carbon atom in hydrocarbons is the basis of the dispersion parameter in its present form. These corrections are applied by first finding the dispersion cohesive energy from the appropriate figure. This requires multiplication by the molar volume for the cyclic compounds using data from the figures here, as these figures give the cohesive energy densities. The dispersion cohesive energy is then increased by adding on the correction factor. This correction factor for chlorine, bromine, and sulfur has been

7248_C001.fm Page 16 Monday, April 23, 2007 1:56 PM

16

Hansen Solubility Parameters: A User’s Handbook

taken as 1650 J/mol for each of these atoms in the molecule. Dividing by the molar volume and then taking the square root gives the (large atom corrected) dispersion solubility parameter. The need for these corrections has been confirmed many times, both for interpretation of experimental data and to allow Equation 1.6 to Equation 1.8 to balance. Research is definitely needed in this area. The impact of these corrections is, of course, larger for the smaller molecular species. Taking square roots of the larger numbers involved with the larger molecular species reduces the errors involved in these cases, as the corrections are relatively small. It can be seen from the dispersion parameters of the cyclic compounds that the ring has an effect similar to increasing the effective size of the interacting species. The dispersion energies for cycloaliphatic compounds are larger than their aliphatic counterparts, and they are higher for aromatic compounds than their corresponding cycloaliphatics. Similar effects also appear with the ester group. This group appears to act as if it were, in effect, an entity that is larger than the corresponding compound containing only carbon (i.e., its homomorph), and it has a higher dispersion solubility parameter without any special need for corrections. The careful evaluation of the dispersion cohesive energy may not have a major impact on the value of the dispersion solubility parameter, because square roots of rather large numbers are taken. Larger problems arise because of Equation 1.6. Energy assigned to the dispersion portion cannot be reused when finding the other partial parameters using Equation 1.6 (or Equation 1.8). This is one reason group contributions are recommended in some cases, as discussed later.

CALCULATION OF THE POLAR SOLUBILITY PARAMETER δP The earliest assignments of a “polar” solubility parameter were given by Blanks and Prausnitz.12 These parameters were, in fact, the combined polar and hydrogen bonding parameters as used by Hansen, and they cannot be considered polar in the current context. The first Hansen polar parameters13 were reassigned new values by Hansen and Skaarup according to the Böttcher equation (Equation 10.25).15 This equation requires the molar volume, the dipole moment (DM), the refractive index, and the dielectric constant. These are not available for many compounds, and the calculation used is more difficult than the much simpler equation developed by Hansen and Beerbower17: δP = 37.4(DM)/V1/2

(1.13)

The constant 37.4 gives this parameter in SI units. Equation 1.13 has been consistently used by the author over the past years, particularly in view of its reported reliability.35 This reported reliability appears to be correct. The molar volume must be known or estimated in one way or another. This leaves only the dipole moment to be found or estimated. Standard reference works have tables of dipole moments, with the most extensive listing still being McClellan.42 Other data sources also have the same, as well as other relevant parameters, and data such as latent heats and critical temperatures. The Design Institute for Physical Property Research (DIPPR)43 database has been found useful for many compounds of reasonably common usage, but many interesting compounds are not included in the DIPPR. When no dipole moment is available, similarity with other compounds, group contributions, or experimental data can be used to estimate the polar solubility parameter. It must be noted that the fact of zero dipole moment in symmetrical molecules is not basis enough to assign a zero polar solubility parameter. An outstanding example of variations of this kind can be found with carbon disulfide. The reported dipole moments are mostly 0 for gas phase measurements, supplemented by 0.08 in hexane, 0.4 in carbon tetrachloride, 0.49 in chlorobenzene, and 1.21 in nitrobenzene. There is a clear increase with increasing solubility parameter of the media. The latter and the highest value has been found experimentally most fitting for correlating permeation through a fluoropolymer film used for chemical protective clothing.44 Many fluoropolymers

7248_C001.fm Page 17 Monday, April 23, 2007 1:56 PM

Solubility Parameters — An Introduction

17

have considerable polarity. The lower dipole moments seem to fit in other instances. Diethyl ether has also presented problems as an outlier in terms of dissolving or not and permeating rapidly or not. Here, the reported dipole moments42 vary from 0.74 to 2.0, with a preferred value of 1.17 and 1.79 in chloroform. Choosing a given value seems rather arbitrary. The chameleonic cyclic forms of the linear glycol ethers would also seem to provide a basis for altered dipole moments in various media.38 When Equation 1.13 cannot be used, the polar solubility parameter has been found using the Beerbower table of group contributions, by similarity to related compounds and/or by subtraction of the dispersion and hydrogen bonding cohesive energies from the total cohesive energy. The question in each case is, “Which data are available and judged most reliable?” New group contributions can also be developed from related compounds whose dipole moments are available. These new polar group contributions then become supplementary to the Beerbower table. For large molecules, especially those with long hydrocarbon chains, the accurate calculation of the relatively small polar (and hydrogen bonding) contributions present special difficulties. The latent heats are not generally available with sufficient accuracy to allow subtraction of two large numbers from each other to find a very small one. In such cases, the similarity and group contribution methods are thought best. Unfortunately, latent heats found in a widely used handbook45 are not clearly reported as to the reference temperature. There is an indication that these are 25°C data, but checking indicated many of the data to be identical with boiling point data reported elsewhere in the literature. Subsequent editions of this handbook46 have a completely different section for the latent heat of evaporation. Again, even moderate variations in reported heats of vaporization can cause severe problems in calculating the polar (or hydrogen bonding) parameter when Equation 1.6 or Equation 1.8 are strictly adhered to.

CALCULATION OF THE HYDROGEN BONDING SOLUBILITY PARAMETER δH In the earliest work, the hydrogen bonding parameter was almost always found by subtracting the polar and dispersion energies of vaporization from the total energy of vaporization. This is still widely used where the required data are available and reliable. At this stage, however, the group contribution techniques are considered reasonably reliable for most of the required calculations and, in fact, more reliable than estimating several other parameters to ultimately arrive at the subtraction step just mentioned. Therefore, in the absence of reliable latent heat and dipole moment data, group contributions are judged to be the best alternative. Similarity to related compounds can also be used, of course, and the result of such a procedure should be essentially the same as for using group contributions. The above paragraph is not changed from the first edition of this handbook. This is to emphasize the importance of the work of Panayiotou and coworkers reported in Chapter 3. It now appears possible not only to calculate the hydrogen bonding parameter independently, but also to arrive at all three parameters by statistical mechanics. This is clearly a major step forward. Whether or not one understands all of the equations and methodology of Chapter 3, the procedure in itself confirms the need for (at least) three cohesion energy parameters, and similar results are found by the approach of the first paragraph as well as with statistical thermodynamics.

SUPPLEMENTARY CALCULATIONS AND PROCEDURES The procedures listed previously are those most frequently used by the author in calculating the three partial solubility parameters for liquids when some data are available. There are a number of other calculations and procedures that are also helpful. Latent heat data at 25°C have been found consistently from those at another temperature, using the relation given by Fishtine.47

7248_C001.fm Page 18 Monday, April 23, 2007 1:56 PM

18

Hansen Solubility Parameters: A User’s Handbook

ΔHv(T1)/ΔH(T2) = [(1 – Tr1)/(1 – Tr2)]0.38

(1.14)

This is done even if the melting point of the compound being considered is higher than 25°C. The result is consistent with all the other parameters, and to date no problems with particularly faulty predictions have been noted. It appears as if the predictions are not significantly in error when experimental data are available for checking. When the latent heat at the boiling point is given in cal/mol, Equation 1.14 is used to estimate the latent heat at 25°C. RT equal to 592 cal/mol is then subtracted from this according to Equation 1.15, to find the total cohesion energy, E, in cgs units at this temperature: E = ΔEv = ΔHV – RT

(1.15)

where R is the gas constant and T is the absolute temperature. A computer program has been developed by the author to assign HSP to solvents, based on experimental data alone. This has been used in several cases where the parameters for the given liquids were desired with a high degree of accuracy. The procedure is to enter solvent quality, good or bad, into the program for a reasonably large number of polymers where the solubility parameters, and appropriate radius of interaction for the polymers are known. The program then locates that set of δD, δP , and δH parameters for the solvent that best satisfies the requirements of a location within the spheres of the appropriate polymers, that have good solvent quality, and outside the appropriate spheres where the solvent quality is bad. An additional aid in estimating HSP for many compounds is that these parameters can be found by interpolation or extrapolation, especially for homologous series. The first member may not necessarily be a straight-line extrapolation, but comparisons with related compounds should always be made where possible to confirm assignments. Plotting the parameters for homologous series among the esters, nitroparaffins, ketones, alcohols, and glycol ethers has aided in finding the parameters for related compounds.

TEMPERATURE DEPENDENCE Only very limited attempts have been made to calculate solubility parameters at a higher temperature prior to the second edition of this handbook. The inclusion of Chapter 3 and Chapter 10 in this handbook helps by providing a more accurate treatment of temperature dependence when the situation warrants it. Solubility parameter correlations of phenomena at higher temperatures have generally been found satisfactory when the established 25°C parameters have been used. Recalculation to higher temperatures is possible but has not generally been found necessary. In this direct but approximate approach, it is assumed that the parameters all demonstrate the same temperature dependence, which, of course, is not the case. It might be noted in this connection that the hydrogenbonding parameter, in particular, is the most sensitive to temperature. As the temperature increases, more and more hydrogen bonds are progressively broken or weakened, and this parameter will decrease more rapidly than the others. The gas-phase dipole moment is not temperature dependent, although the volume of a fluid does change with the temperature, which will also change its cohesive energy density. The change of the δD, δP , and δH parameters for liquids with temperature, T, can be estimated by the following equations, where α is the coefficient of thermal expansion17: dδD/dT = –1.25αδD

(1.16)

dδP/dT = –0.5αδP

(1.17)

dδH/dT = –δH(1.22 × 10–3 + 0.5α)

(1.18)

7248_C001.fm Page 19 Monday, April 23, 2007 1:56 PM

Solubility Parameters — An Introduction

19

Higher temperature means a general increase in rate of solubility/diffusion/permeation, as well as larger solubility parameter spheres. δD, δP , and δH decrease with increased temperature, as can be seen by a comparison of Equation 1.16, through Equation 1.18. This means that alcohols, phenols, glycols, and glycol ethers become better solvents for polymers of lower solubility parameters as the temperature increases. Thus, increasing the temperature can cause a nonsolvent to become a good solvent, a fact that is often noted in practice. As mentioned earlier, it is possible that a boundary solvent can be a good solvent at a given temperature, but turn bad with either an increase or a decrease in temperature. These phenomena are discussed in great detail by Patterson and coworkers.3,4 They can be explained either by the change in solubility parameter with temperature or more completely by the Prigogine CST. The effects of temperature changes on solubility relations are most obvious with systems having a high hydrogen-bonding character. Examples are given in the next section for some special situations involving water and methanol.

SOME SPECIAL EFFECTS TEMPERATURE CHANGES Water (and methanol) uptake in most polymers increases with increasing temperature. This is because the solubility parameters of the water and the polymer are closer at higher temperatures. The δH parameter of water (and methanol) falls with increasing temperature, whereas that of most polymers remains reasonably constant. Water is also well known as an exceptionally good plasticizer because of its small molecular size. The presence of dissolved water not only softens (reduces the glass transition temperature) a polymer as such, but it also means diffusion rates of other species will be increased. The presence of water in a film can also influence the uptake of other materials, such as in solubility parameter studies or resistance testing, with hydrophilic materials being more prone to enter the film. This can cause blistering on rapid cooling as discussed in Chapter 12 and in Reference 48 (see Chapter 8 and Chapter 12). Figure 8.3 shows how rapid cooling from a water-saturated state at higher temperature can lead to blistering. Figure 12.3 and Figure 12.4 show how this effect can be measured experimentally with an increase in water content above the equilibrium value when temperature cycling is encountered. This leads to premature failure of polymeric products used in such environments. A related problem has been encountered with methanol. It was intended to follow the rate of uptake of methanol in an epoxy coating at room temperature by weighing coated-metal panels periodically on an analytical balance. Blistering was encountered in the coating near the air surface shortly after the experiment began. The methanol that had been absorbed into the coating near the surface became insoluble as the temperature of the coating near the surface was lowered by the evaporation of excess methanol during the handling and weighing of the panels. This is a rather extreme case, and, as mentioned earlier, use of the HSP (determined at 25°C) at elevated temperatures can most often be done without too much trouble from a practical point of view. One should be aware that the changes in the δH parameter would be larger than those in the other parameters, and this effect would be most significant for those liquids with larger δH values.

EFFECTS

OF

SOLVENT MOLECULAR SIZE

The size of both solvent and solute molecules is important for solubility, permeation, diffusion, and chemical resistance phenomena. Smaller molecules tend to be more readily soluble than larger ones. As stated previously, the Hildebrand solubility parameter theory also points to smaller molar volume solvents as being better than those with larger molar volumes, even though they may have identical solubility parameters.1,2 This fact of expected improved solvency for smaller molecules is also known from the Flory–Huggins theory of polymer solutions.29 Solvents with smaller molecular size have also been repeatedly noted as being superior to those with larger molecular size, when highly crystalline polymers or solids are being tested for solubility. So it is not surprising

7248_C001.fm Page 20 Monday, April 23, 2007 1:56 PM

20

Hansen Solubility Parameters: A User’s Handbook

that solvent molecular size can be an important fourth parameter in solubility and, in some cases, in chemical resistance. Specific examples are given in Chapter 5 and Chapter 12. The size and shape of the solvent molecule are also very important for kinetic phenomena such as diffusion, permeation, and attainment of equilibrium. Smaller and more linear molecules diffuse more rapidly than larger and more bulky ones. The diffusion coefficient may be so low that equilibrium is not attained for hundreds of years at room temperature. This was demonstrated in common solvent exposures of rigid polymers like polyphenylene sulfide (PPS) with thicknesses of several millimeters.49 Likewise, the second stage in the two-stage drying process in polymer film formation by solvent evaporation can last for many years.16,50 Polymer samples used for solubility parameter or other testing may well retain solvent or monomer for many years, and this may affect the evaluations. Attempts to include the molecular volume in new composite solubility and size parameters have not been particularly successful.20,21 This may be because the size effect is most often not caused due to the thermodynamic considerations on which the solubility parameters are based, but rather through a kinetic effect of diffusion rates or other free volume considerations. The similarities in the HSP approach and the Prigogine theory, discussed in Chapter 2, indicate a remarkably close, if not identical, relation between the Prigogine ρ (segment size parameter) and the δD parameter, suggesting that molecular size differences are at least partially accounted for in the δD parameter. The Prigogine theory also has a parameter to describe “structural effects,” including size of polymer molecules, but this has not been explored in relation to the present discussion. The increase of δD with increasing molecular size among the aliphatic hydrocarbons, the higher δD values for the larger units represented by cycloaliphatic and aromatic rings, and the need for corrections for larger atoms discussed earlier tend to support the molecular size differences. Sorting output data according to the molecular volume of the test solvents in a computer analysis helps to discover whether solvent molecular size is indeed an additional significant factor in a given correlation or testing program.

COMPUTER PROGRAMS The author has used two computer programs extensively in his own studies and in collecting material for this book. These are called SPHERE and SPHERE1. They are very similar, the only difference being that SPHERE optimizes the polymer (or other material, of course) parameters based on all the data, whereas SPHERE1 considers data for only those solvents considered as “good.” It neglects the nonsolvent data. SPHERE1 has been most useful in correlations with pigments, fillers, and fibers, as described in Chapter 7. The data input is by solvent number followed by an indication of the quality of interaction with that solvent. A “1” indicates a “good” solvent, whereas a “0” is used for a “bad” solvent. What is considered good or bad varies according to the level of interaction being studied. This can be solution or not, a given percentage of swelling or uptake, breakthrough time being less than a given interval, permeation coefficients higher than a given value, long-time suspension of a pigment, etc. The program systematically evaluates the input data using a quality-of-fit function called the desirability function.51 This suggestion was made by a reputed statistician many years ago as the most appropriate statistical treatment for this type of problem. It has been in use since the late 1960s. The function has the form: DATA FIT = (A1 * A2 *...An)1/n

(1.19)

where n is the number of solvents for which there is experimental data in the correlation. The DATA FIT approaches 1.0 as the fit improves during an optimization and reaches 1.0 when all the good solvents are included within the sphere and all the bad ones are outside the sphere.

7248_C001.fm Page 21 Monday, April 23, 2007 1:56 PM

Solubility Parameters — An Introduction

21

Ai = e–(ERROR DISTANCE)

(1.20)

The Ai quotient for a given good solvent within the sphere and a bad solvent outside the sphere will be 1.0. The error distance is the distance of the solvent in error to the sphere boundary. It could either denote being good and outside the sphere or being bad and inside the sphere. Ro is the radius of the sphere, and Ra is the distance from a given solvent point to the center of the sphere. For a good solvent outside the sphere, an error enters the DATA FIT according to: Ai = e+(Ro – Ra)

(1.21)

Such errors are often found for solvents having low molecular volumes. For a bad solvent inside the sphere, the contribution to the DATA FIT is Ai = e+(Ra – Ro)

(1.22)

Such errors can sometimes be found for larger molecular species such as plasticizers. This is not unexpected for the reasons mentioned earlier. Solvents with large and/or small molecules that give the “errors” can sometimes be (temporarily) disregarded by generating a new correlation; this gives an excellent DATA FIT for an abbreviated range of molecular volumes. There is a special printout with the solvents arranged in order of molecular volume that helps to analyze such situations. The computer printouts all include a column for the RED number. The program assumes a starting point, based on an average of each of the HSP for the good solvents only. The program then evaluates eight points at the corners of a cube, with the current best values as center. Different radii are evaluated at each of these points in the optimization process. When better fits are found among the eight points, the point with the best fit is taken as a new center, and eight points around it are evaluated in a similar manner. This continues until the DATA FIT cannot be improved upon. The length of the edge of the cube is then reduced in size to finetune the fit. The initial length of the cube is 1 unit, which is reduced to 0.3 unit, and finally to 0.1 unit in the last optimization step. Experimental data for the solvents are entered with solvent number (comma) and a “1” for a good solvent, or a “0” for a bad one. Errors in the correlations are indicated with an “*” in the SOLUB column where the experimental input data are indicated. As stated above, systematic errors can sometimes be seen in the molar volume printout. This may suggest a new analysis of the data. Nonsystematic errors may be real, such as for reactions or some extraneous effect not predictable by the solubility parameter. They may also be bad data, and rechecking data indicated with an “*” in the output has become a routine practice. The output of this program is for the least radius allowing the maximum DATA FIT. An example is found in Table 5.4. Results from the SPHERE program reported in this book generally include the HSP, given as D (δD), P (δP), and H (δH), respectively, and Ro for the correlation in question, as well as the DATA FIT, the number of good solvents (G), and the total solvents (T) in the correlation. This latter information has not always been recorded and may be lacking for some correlations, especially the older ones.

HANSEN SOLUBILITY PARAMETERS FOR WATER Water is such an important material that a special section is dedicated to its HSP at this point. The behavior of water often depends on its local environment, which makes general predictions very difficult. Water is still so unpredictable that its use as a test solvent in solubility parameter studies

7248_C001.fm Page 22 Monday, April 23, 2007 1:56 PM

22

Hansen Solubility Parameters: A User’s Handbook

TABLE 1.3 HSP Correlations Related to Water Correlation Water Water Water Water a b

— Single molecule — >1% soluble ina — Total miscibility 1a — Total miscibility “1b”

δD

δP

δH

Ro

FIT

G/T

15.5 15.1 18.1 18.1

16.0 20.4 17.1 12.9

42.3 16.5 16.9 15.5

— 18.1 13.0 13.9

— 0.856 0.880 1.000

— 88/167 47/166 47/47

Based on SPHERE program. Based on SPHERE1 Program.

is not recommended. This is true of water as a pure liquid or in mixtures. Table 1.3 includes data from various HSP analyses of the behavior of water. The first set of data is derived from the energy of vaporization of water at 25°C. The second set of data is based on a correlation of the solubility of various solvents in water, where “good” solvents are soluble to more than 1% in water. “Bad” solvents dissolve to a lesser extent. The third set of data is for a correlation of total miscibility of the given solvents with water. The second and third entries in Table 1.3 are based on the SPHERE program where both good and bad solvents affect the DATA FIT and hence the result of the optimization. The last entry in Table 1.3 is for an analysis using the SPHERE1 program. The HSP data are for the minimum sphere that encompasses only the good solvents. The bad solvents are simply not considered in the data processing. This type of comparison usually results in some of the parameters being lower than when all the data are included. A frequent problem is that a considerable portion of the HSP spheres, such as in the case for water, covers such high energies that no liquid can be found. The cohesion energy is so high as to require solids. The constant 4 in the correlations (Equation 1.9) is still used for these correlations, primarily based on successes at lower levels of cohesion energies, but this is also supported by the comparison with the Prigogine cst of polymer solutions, discussed at some length in Chapter 2. The HSP for water as a single molecule, based on the latent heat at 25°C is sometimes used in connection with mixtures with water to estimate average HSP. More recently, it has been found in a study involving water, ethanol, and 1,2-propanediol that the HSP for water indicated by the total water solubility correlation could be used to explain the behavior of the mixtures involved. The averaged values are very questionable as water can associate, and water has a very small molar volume as a single molecule. It almost appears to have a dual character. The data for the 1% correlation,52 as well as for the total water miscibility, suggest that about six water molecules associate into units. Traditionally, solvents are considered as points. This is practical and almost necessary from an experimental point of view as most solvents are so miscible as to not allow any experimental characterization in terms of a solubility sphere. An exception to this is the data for water reported in Table 1.3. The HSP reported here are the center points of HSP spheres where the good solvents are either those that are completely miscible or those that are miscible to only 1% or more, as discussed previously. It should also be mentioned that amines were a major source of outliers in these correlations. No solids were included. Their use to predict solubility relations for amines and for solids must therefore be done with caution.

CONCLUSION This chapter has been dedicated to describing the tools with which different HSP characterizations can be made and some of the pitfalls that may be encountered in the process. The justification for

7248_C001.fm Page 23 Monday, April 23, 2007 1:56 PM

Solubility Parameters — An Introduction

Chlorinated

23

Alcohols

Esters

Glycols

Ethers

Amides

Amines

Phenols

Methyl ethyl ketone

Chloroform

Cyclohexanone

Dichloromethane

THF

Nitriles

Toluene

Dimethylsulfoxide

Isooctane Nitromethane

Bold type indicates relatively high δD 24

20

N

Nitriles A S

δP, Polar Parameter

16

Amides F

Alcohols

B

12

M

A Ketones

E

M E

8

C

Chlorinated

Glycols

Phenols

E T

P C

B Esters

4

P

B

O

Ethers E P

Tol

Amines

0 0

I

4

8

12

16

20

24

28

δH, Hydrogen Bonding Parameter

FIGURE 1.4 δp vs. δH plot showing the location of various common solvents. The glycols are ethylene glycol and propylene glycol. The alcohols include methanol (M), ethanol (E), 1-butanol (B), and 1-octanol (O). The amides include dimethyl formamide (F) and dimethyl acetamide (A). The nitriles are acetonitrile (A) and butyronitrile (B). The esters are ethyl acetate (E) and n-butyl acetate (B). The amines are ethyl amine (E) and propyl amine (P). The phenols are phenol (P) and m-cresol (C). The ethers are symbolized by diethyl ether.

the tools is further confirmed in Chapter 2 and Chapter 3, and their use is demonstrated in all the subsequent chapters. Figure 1.4 is included to show where many common solvents are located on a δp vs. δH plot.

7248_C001.fm Page 24 Monday, April 23, 2007 1:56 PM

24

Hansen Solubility Parameters: A User’s Handbook

REFERENCES 1. Hildebrand, J. and Scott, R.L., The Solubility of Nonelectrolytes, 3rd ed., Reinhold, New York, 1950. 2. Hildebrand, J. and Scott, R.L., Regular Solutions, Prentice-Hall, Englewood Cliffs, NJ, 1962. 3. Patterson, D. and Delmas, G., New aspects of polymer solution thermodynamics, Off. Dig. Fed. Soc. Paint Technol., 34(450), 677–692, 1962. 4. Delmas, D., Patterson, D., and Somcynsky, T., Thermodynamics of polyisobutylene-n-alkane systems, J. Polym. Sci., 57, 79–98, 1962. 5. Bhattacharyya, S.N., Patterson, D., and Somcynsky, T., The principle of corresponding states and the excess functions of n-alkane mixtures, Physica, 30, 1276–1292, 1964. 6. Patterson, D., Role of free volume changes in polymer solution thermodynamics, J. Polym. Sci. Part C, 16, 3379–3389, 1968. 7. Patterson, D.D., Introduction to thermodynamics of polymer solubility, J. Paint Technol., 41(536), 489–493, 1969. 8. Biros, J., Zeman, L., and Patterson, D., Prediction of the C parameter by the solubility parameter and corresponding states theories, Macromolecules, 4(1), 30–35, 1971. 9. Barton, A.F.M., Handbook of Solubility Parameters and Other Cohesion Parameters, CRC Press, Boca Raton, FL, 1983; 2nd ed., 1991. 10. Gardon, J.L. and Teas, J.P., Solubility parameters, in Treatise on Coatings, Vol. 2, Characterization of Coatings: Physical Techniques, Part II, Myers, R.R. and Long, J.S., Eds., Marcel Dekker, New York, 1976, chap. 8. 11. Burrell, H., Solubility parameters for film formers, Off. Dig. Fed. Soc. Paint Technol., 27(369), 726–758, 1972; Burrell, H., A solvent formulating chart, Off. Dig. Fed. Soc. Paint Technol., 29(394), 1159–1173, 1957; Burrell, H., The use of the solubility parameter concept in the United States, VI Federation d’Associations de Techniciens des Industries des Peintures, Vernis, Emaux et Encres d’Imprimerie de l’Europe Continentale, Congress Book, 21–30, 1962. 12. Blanks, R.F. and Prausnitz, J.M., Thermodynamics of polymer solubility in polar and nonpolar systems, Ind. Eng. Chem. Fundam., 3(1), 1–8, 1964. 13. Hansen, C.M., The three dimensional solubility parameter — key to paint component affinities I, J. Paint Technol., 39(505), 104–117, 1967. 14. Hansen, C.M., The three dimensional solubility parameter — key to paint component affinities II, J. Paint Technol., 39(511), 505–510, 1967. 15. Hansen, C.M. and Skaarup, K., The three dimensional solubility parameter — key to paint component affinities III, J. Paint Technol., 39(511), 511–514, 1967. 16. Hansen, C.M., The Three Dimensional Solubility Parameter and Solvent Diffusion Coefficient, Doctoral dissertation, Danish Technical Press, Copenhagen, 1967. 17. Hansen, C.M. and Beerbower, A., Solubility parameters, in Kirk-Othmer Encyclopedia of Chemical Technology, Suppl. Vol., 2nd ed., Standen, A., Ed., Interscience, New York, 1971, pp. 889–910. 18. Barton, A.F.M., Applications of solubility parameters and other cohesion energy parameters, Polym. Sci. Technol. Pure Appl. Chem., 57(7), 905–912, 1985. 19. Sørensen, P., Application of the acid/base concept describing the interaction between pigments, binders, and solvents, J. Paint Technol., 47(602), 31–39, 1975. 20. Van Dyk, J.W., Paper presented at the Fourth Chemical Congress of America, New York, August 25–30, 1991. 21. Anonymous [Note: This was, in fact, Van Dyk, J.W., but this does not appear on the bulletin], Using Dimethyl Sulfoxide (DMSO) in Industrial Formulations, Bulletin No. 102, Gaylord Chemical Corp., Slidell, LA, 1992. 22. Karger, B.L., Snyder, L.R., and Eon, C., Expanded solubility parameter treatment for classification and use of chromatographic solvents and adsorbents, Anal. Chem., 50(14), 2126–2136, 1978. 23. Hansen, C.M. and Wallström, E., On the use of cohesion parameters to characterize surfaces, J. Adhes., 15, 275–286, 1983. 24. Hansen, C.M., Characterization of surfaces by spreading liquids, J. Paint Technol., 42(550), 660–664, 1970. 25. Hansen, C.M., Surface dewetting and coatings performance, J. Paint Technol., 44(570), 57–60, 1972.

7248_C001.fm Page 25 Monday, April 23, 2007 1:56 PM

Solubility Parameters — An Introduction

25

26. Hansen, C.M. and Pierce, P.E., Surface effects in coatings processes, Ind. Eng. Chem. Prod. Res. Dev., 13(4), 218–225, 1974. 27. Hennissen, L., Systematic Modification of Filler/Fiber Surfaces to Achieve Maximum Compatibility with Matrix Polymers, Lecture for the Danish Society for Polymer Technology, Copenhagen, February 10, 1996. 28. Gardon, J.L., Critical review of concepts common to cohesive energy density, surface tension, tensile strength, heat of mixing, interfacial tension and butt joint strength, J. Colloid Interface Sci., 59(3), 582–596, 1977. 29. Flory, P.J., Principles of Polymer Chemistry, Cornell University Press, New York, 1953. 30. Prigogine, I. (with the collaboration of Bellemans, A. and Mathot, A.), The Molecular Theory of Solutions, North-Holland, Amsterdam, 1957, chap. 16, 17. 31. van Krevelen, D.W. and Hoftyzer, P.J., Properties of Polymers: Their Estimation and Correlation with Chemical Structure, 2nd ed., Elsevier, Amsterdam, 1976. 32. Beerbower, A., Environmental Capability of Liquids, in Interdisciplinary Approach to Liquid Lubricant Technology, NASA Publication SP-318, 1973, 365–431. 33. Fedors, R.F., A method for estimating both the solubility parameters and molar volumes of liquids, Polym. Eng. Sci., 14(2), 147–154, 472, 1974. 34. Hansen, C.M., Solubility parameters, in Paint Testing Manual, Manual 17, Koleske, J.V., Ed., American Society for Testing and Materials, Philadelphia, PA, 1995, pp. 383–404. 35. Koenhen, D.N. and Smolders, C.A., The determination of solubility parameters of solvents and polymers by means of correlation with other physical quantities, J. Appl. Polym. Sci., 19, 1163–1179, 1975. 36. Anonymous, Brochure: Co-Act — A Dynamic Program for Solvent Selection, Exxon Chemical International Inc., 1989. 37. Dante, M.F., Bittar, A.D., and Caillault, J.J., Program calculates solvent properties and solubility parameters, Mod. Paint Coat., 79(9), 46–51, 1989. 38. Hoy, K.L., New values of the solubility parameters from vapor pressure data, J. Paint Technol., 42(541), 76–118, 1970. 39. Myers, M.M. and Abu-Isa, I.A., Elastomer solvent interactions III — effects of methanol mixtures on fluorocarbon elastomers, J. Appl. Polym. Sci., 32, 3515–3539, 1986. 40. Hoy, K.L., Tables of Solubility Parameters, Union Carbide Corp., Research and Development Dept., South Charleston, WV, 1985; 1st ed. 1969. 41. Reid, R.C. and Sherwood, T.K., Properties of Gases and Liquids, McGraw-Hill, New York, 1958 (Lydersen Method — see also Reference 31). 42. McClellan, A.L., Tables of Experimental Dipole Moments, W.H. Freeman, San Francisco, 1963. 43. Tables of Physical and Thermodynamic Properties of Pure Compounds, American Institute of Chemical Engineers Design Institute for Physical Property Research, Project 801, Data Compilation, Danner, R.P. and Daubert, T.E., Project Supervisors, DIPPR Data Compilation Project, Department of Chemical Engineering, Pennsylvania State University, University Park. 44. Hansen, C.M., Selection of Chemicals for Permeation Testing Based on New Solubility Parameter Models for Challenge 5100 and Challenge 5200, under contract DTCG50-89-P-0333 for the U.S. Coast Guard, June 1989, Danish Isotope Centre, Copenhagen. 45. Weast, R.C., (Editor-in-Chief), CRC Handbook of Chemistry and Physics, 65th ed., CRC Press, Boca Raton, FL, 1988–1989, pp. C-672–C-683. 46. Majer, V., Enthalpy of vaporization of organic compounds, in Handbook of Chemistry and Physics, 72nd ed., Lide, D.R., (Editor-in-Chief), CRC Press, Boca Raton, FL, 1991–1992, pp. 6-100–6-107. 47. Fishtine, S.H., Reliable latent heats of vaporization, Ind. Eng. Chem., 55(4), 20–28, 1963; Ind. Eng. Chem., 55(5), 55–60; Ind. Eng. Chem., 55(6), 47–56. 48. Hansen, C.M., New developments in corrosion and blister formation in coatings, Prog. Org. Coat., 26, 113–120, 1995. 49. Hansen, C.M., Solvent Resistance of Polymer Composites — Glass Fiber Reinforced Polyphenylene Sulfide, Centre for Polymer Composites (Denmark), Danish Technological Institute, Taastrup, 1993, 1–62, ISBN 87-7756-286-0. 50. Hansen, C.M., A mathematical description of film drying by solvent evaporation, J. Oil Color Chem. Assoc., 51(1), 27–43, 1968.

7248_C001.fm Page 26 Monday, April 23, 2007 1:56 PM

26

Hansen Solubility Parameters: A User’s Handbook 51. Harrington, E.C., Jr., The desirability function, Ind. Qual. Control, 21(10), 494–498, April 1965. 52. Hansen, C.M. and Andersen, B.H., The affinities of organic solvents in biological systems, Am. Ind. Hyg. Assoc. J., 49(6), 301–308, 1988.

7248_C002.fm Page 27 Wednesday, May 9, 2007 8:25 AM

— The Prigogine 2 Theory Corresponding States Theory, χ 12 Interaction Parameter, and Hansen Solubility Parameters Charles M. Hansen ABSTRACT Patterson has shown that the χ12 interaction parameter can be estimated from the corresponding states theory (CST) of Prigogine. Correlations using Hansen solubility parameters (HSP) confirm the usage of the term cohesive energy difference proposed in the Prigogine CST. Therefore, the HSP approach can be expected to be useful to predict the Flory interaction coefficient, χ12. Equations for this purpose are presented and discussed based on comparisons of calculated and experimental values for five polymers. There is agreement in many cases, especially for essentially nonpolar systems, but full understanding of the interrelationship has not yet been achieved. The lack of accounting for permanent dipole–permanent dipole and hydrogen bonding (electron interchange) in the “New Flory” theory leading to the coefficient “χ12” is thought to be largely responsible for this. It does appear, however, that the constant “4” (or 0.25) in the HSP correlations and the 0.25 in the leading term of the Prigogine theory have identical functions. They modify the specific interactions described by the Prigogine δ and also the polar and hydrogen bonding HSP (δP and δH). This could imply that the Prigogine ρ attempts to describe what the δD parameter describes, that is, the nondirectional (nonpolar) atomic interactions. Neither the Flory nor the Prigogine approaches can lead to the type of predictions possible with the HSP approach. The many correlations and other predictions contained in this book would not be possible with these theories, as they do not separate the polar and hydrogen bonding effects independently. The Prigogine theory must be used with the geometric mean to estimate the interaction between different species. The Hildebrand and HSP approaches inherently use the geometric mean. This implies that the geometric mean is capable of describing not only dispersion interactions but also those due to permanent dipoles and hydrogen bonding.

INTRODUCTION The Flory–Huggins “chi” parameter, χ, has been used for many years in connection with polymer solution behavior,1,2 but now the χ12 parameter derived from the New Flory theory is being currently accepted for general use instead of the older χ. It would be desirable to relate the widely used HSP3–10 more directly to χ12. This would allow estimates of χ12 for systems where the HSP are known, but χ12 is not. The reverse is not possible as a single χ12 parameter cannot be used to divide the cohesion energy into contributions from dispersion (nonpolar) forces, permanent 27

7248_C002.fm Page 28 Wednesday, May 9, 2007 8:25 AM

28

Hansen Solubility Parameters: A User’s Handbook

dipole–permanent dipole forces, and hydrogen bonding (electron interchange), which is the basis of the HSP. Reliable χ12 values for numerous solvents and the same polymer can be used to determine the HSP for the polymer, in the same manner as solvency or swelling data being used for a similar purpose. In principle, the weighting schemes (described in Chapter 5) to average the solvent parameters for obtaining the polymer HSP can also be used with the χ12 parameter, just as they are used with weight gain or intrinsic viscosity. Patterson15 and coworkers17 have shown how to predict the χ12 parameter using corresponding states theories (CST),2,11–17 as well as using the Hildebrand solubility parameter in (strictly) nonpolar systems. They use the symbol ν2 instead of χ12 for the same quantity. The Hildebrand solubility parameter is the square root of the cohesive energy density (ced).18,19 Also, it has been shown recently that HSP and the Prigogine/Patterson CST are mutually confirming and give similar predictions.20,21 This is discussed in more detail later. The customary equation to calculate χ12 from the Hildebrand solubility parameters for a nonpolar solvent and a nonpolar polymer is: χ12 = [V(δ1 – δ2)2]/RT + β

(2.1)

where V is the molar volume of the solvent, δ is the Hildebrand solubility parameter for the solvent (1) and polymer (2), R is the gas constant, and T is the absolute temperature. The empirical constant β has been discussed as being necessary for polymer systems,22 as a correction to the Flory combinatorial entropy. β, although combinatorial in origin, was attached to χ12 in order to preserve the Flory form of the chemical potential expression. β has a generally accepted average value near 0.34. Biros et al.17 state that this value of β presents difficulties as an explanation of an error in the Flory combinatorial entropy approximation. These authors state that β should be interpreted as aligning the χ12 values from the solubility parameters with those found from CST. The CST predict χ to be about 0.3 units larger than that found when using the Hildebrand solubility parameter. β is not required for essentially nonpolar systems when HSP are used in a relation similar to Equation 2.1, as shown later. The Hildebrand parameters are applicable to regular solutions, which, in the current context, implies strictly nonpolar systems. Hildebrand solubility parameters have been shown to reflect the noncombinatorial free energy directly via the first term in Equation 2.112,13 (see also Chapter 1, Equation 1.3 and Equation 1.4). It was previously thought that the heat of mixing was given by the Hildebrand theory as φ1φ2VM(δ1 – δ2)2, where VM is the volume of the mixture and the ϕs are volume fractions of the solvent and polymer, respectively. This is not true. The heat of mixing must be found by differentiating this relation as shown by Delmas and coworkers.12,13 The work of Delmas, Patterson, and coworkers has shown that predictions with the nonpolar Hildebrand solubility parameter and the Prigogine CST are in excellent agreement with each other with regard to heats of mixing in essentially nonpolar systems. Both positive and negative heats of mixing are allowed, predicted, and found. The argument that solubility parameters are inadequate, as they do not allow for negative heats of mixing, is not valid. These studies also show that subsequent increases in temperature lead to improved solvency when a solvent has higher solubility parameters than the polymer. When the solvent has lower solubility parameters than the polymer, an increase in temperature leads to poorer solvency. Precipitation can even occur with increasing temperature. This temperature is called the lower critical solution temperature. (See also the discussion in Chapter 1.)

HANSEN SOLUBILITY PARAMETERS (HSP) It has been shown that the total energy of vaporization can be divided into at least three parts.6 These parts come from the nonpolar/dispersion (atomic) forces, ED; the permanent dipole–permanent dipole

7248_C002.fm Page 29 Wednesday, May 9, 2007 8:25 AM

Theory

29

(molecular) forces, EP; and hydrogen bonding (molecular) forces, EH. The latter is more generally called the electron exchange energy. E = ED + EP + EH

(2.2)

E/V = ED/V + EP/V + EH/V

(2.3)

δ2 = δD2 + δP2 + δH2

(2.4)

δD, δP , and δH are the HSP for the dispersion, polar, and hydrogen bonding interactions, respectively. δ is the Hildebrand solubility parameter, (E/V)1/2. It might be noted that the value of a solubility parameter in MPa1/2 is 2.0455 times larger than in the often used (cal/cm3)1/2 units. As described in Chapter 1, a corresponding states calculation using hydrocarbons as reference is used to find the part of the cohesive energy of a liquid that is attributable to dispersion (nonpolar) forces. Subtracting this nonpolar contribution from the total cohesion energy then gives the sum of the permanent dipole–permanent dipole and hydrogen bonding (electron interchange) contributions to the total cohesion energy. These can then be separated by calculation and/or experiment into the polar and hydrogen bonding parameters. HSP also include volume effects, as they are based on cohesion energy density. Volume effects are also basic to the Prigogine CST. As described in Chapter 3, Panayiotou and coworkers have used a statistical thermodynamics approach to calculate all three parameters, thus giving support to the approach used in Equation 2.2 to Equation 2.4. HSPs have been applied to the study of polymer solubility and swelling, biological materials, barrier properties of polymers, surfaces,4,20,23–26 etc. and have been described in greater detail elsewhere7,8,10 (see also the following chapters). The three parameters described in Equation 2.4 are fundamental energy parameters that can be calculated from the mutual interactions of identical molecules in a pure liquid. The quantities required are E, V, the dipole moment (and perhaps the refractive index and the dielectric constant), and generalized corresponding states correlations for hydrocarbons, (ED). Group contribution methods and simpler calculational procedures have also been established.10 These procedures were described in Chapter 1. The calculated values for a large number of the liquids have been confirmed experimentally by solubility tests. The usual equation used in HSP correlations is: (Ra)2 = 4(δD2 – δD1)2 + (δP2 – δP1)2 + (δH2 – δH1)2

(2.5)

Ra, in this equation, is a modified difference between the HSP for a solvent (1) and polymer (2). Ra must not exceed Ro, the radius of interaction of a HSP solubility sphere, if solubility is to be maintained. Both Ra and Ro have the same units as solubility parameters. These correlations have been very convenient for practical use, for example, in solvent selection. The constant “4” has been found empirically useful to convert spheroidal plots of solubility into spherical ones using δD and either of the other parameters (see Chapter 5). It has been used with success in well over 1000 HSP correlations with a computer program that optimizes a solubility sphere according to Equation 2.5, where all the good solvents are within the sphere and the bad ones are outside. This program was described in Chapter 1. This experimental procedure is still thought to be the best way to arrive at the HSP for polymers; the polymer HSPs being given by the coordinates for the center of the sphere. The reliability of the spherical characterizations and the need to divide the total cohesion energy (E) into at least three parts has been confirmed by systematically locating nondissolving solvents that are synergistic and dissolve a given polymer when mixed.3 They only need to be located on opposite sides of the sphere of solubility for the given polymer.

7248_C002.fm Page 30 Wednesday, May 9, 2007 8:25 AM

30

Hansen Solubility Parameters: A User’s Handbook

For present purposes of comparison, Equation 2.5 must be normalized by 4RM2 to make its predictions consistent with the quantities commonly used in the literature, in connection with the CSTs and χ12: Η = (RA)2/RM2 = [(δD2 – δD1)2 + (δP2 – δP1)2/4 + (δH2 – δH1)2/4]/RM2

(2.6)

RA = Ra/2; RM = Ro/2

(2.7)

RM is the maximum solubility parameter difference that still allows the polymer to dissolve based on Equation 2.6. RM is the radius of a HSP sphere (spheroid) based on Equation 2.7. The HSP difference between solvent and polymer, RA, must be less than RM for solution to occur. It can be seen that the quantities RA/RM and H = (RA/RM)2 will be 1.0 on the boundary surface of a sphere describing polymer solubility. RA/RM is a ratio of solubility parameters, whereas H is a ratio of cohesion energy densities. H is zero when the solubility parameters for the solvent and polymer match and subsequently increases to larger values as the differences between solvent and polymer increase.

RESEMBLANCE BETWEEN PREDICTIONS OF HANSEN SOLUBILITY PARAMETERS AND CORRESPONDING STATES THEORIES Patterson and coworkers 15,17 have explained the Prigogine theory in concise form and simplified some of the most important aspects. A key parameter is the Prigogine δ. This describes normalized cohesive energy differences between polymer segments and solvents. The Prigogine δ parameter is defined as δ = (ε2 – ε1)/ε1

2.8)

where ε is the cohesive energy for a polymer segment (2) or for a solvent (1). For the present discussion, it is advantageous to define the Prigogine δ using cohesive energy densities as follows: δ = [(ced2)1/2 – (ced1)1/2]2/(ced1) = (δ1 – δ2)2/δ12

(2.9)

The numerator is the difference in cohesion energy densities between solvent and polymer, and this is normalized by the ced of the solvent. As indicated above, cohesion energies (HSP) for solvents can be calculated whereas those for polymers currently require experimental data on solubility or other relevant testing procedures. The Prigogine ρ accounts for differences in the size of the solvent and polymer segments. The segmental distance parameter is s. ρ is defined as ρ = (σ2 – σ1)/σ1

(2.10)

Another key parameter in the Prigogine/Patterson CST is ν2, which is in fact equal to χ12.15 “ν ” is approximated by 2

ν2 app = (δδ2/4 + 9ρ2) app = (δδ/2 – 3ρ/2)2

(2.11)

“ν2” includes effects from differences in segmental energy (the δ effect) and in segmental size (the ρ effect). The geometric mean rule [ε12 = (ε1ε2)1/2] was used to arrive at this result, just as it was used to arrive at the equations having differences in solubility parameters. Patterson states that the

7248_C002.fm Page 31 Wednesday, May 9, 2007 8:25 AM

Theory

31

coefficient in front of ρ is uncertain15 and, furthermore,27 that “ρ was a (misguided) attempt to take into account segment-size differences.” Patterson27 has recently helped the author to clarify some points about the relations among the theories. I feel a few quotes from these communications, in addition to the one just cited, are in order at this point: In my opinion, the Flory theory was a very usable and particularly successful case of the Prigogine theory (which was in fact difficult to use). An additional point that Flory made much of but was only touched by Prigogine, is that surface/volume fraction of the polymers are very different from those of solvents, i.e., the polymer is very bulky. These are interesting differences between the Flory theory and that of Prigogine. But, again in my opinion, Flory always presented his theory as something absolutely different from that of Prigogine’s, using different symbols, different names for concepts, etc. In reaction against this I have always liked to call the whole thing the Prigogine–Flory theory. However, since about 1970, I have done very little with polymer solutions and, hence, when using the term Prigogine–Flory theory, I have used it with respect to mixtures of small molecules and not polymer solutions. I think that the work by a lot of people has established the utility of the Prigogine–Flory theory, or if you like, the Flory theory, for small-molecule mixtures. Also: Very specifically, the Prigogine parameters delta and rho are now out of fashion, and they have been lumped together in the χ12 parameter. Particularly, the rho parameter does not have nearly as much importance as Prigogine thought, and Flory completely discarded it. I think, too, the origin of the parameter beta in the solubility parameter approach would, in the Prigogine–Flory approach, be ascribed to free-volume differences, which must inevitably exist between any polymer and any solvent and which give a contribution to the chi parameter ….

This demonstrates that there is not complete agreement among those who have concerned themselves with these theories. The following is an attempt to unify all of these thoughts. The ideas have not been fully tested as of yet, but the implications appear very clear to the author, at least. The discussion concentrates on the ν2 parameter, being loyal to the Patterson article15 (sometimes referred to as the Prigogine–Patterson theory) where this part of the book got its start. More specifically, ν2 accounts for segmental energy differences, and differences in size of solvent and polymer segments for breakage of solvent–solvent (1–1) bonds and polymer–polymer (2–2) bonds to allow formation of solvent–polymer (1–2) bonds. In nonpolar systems, the Prigogine δ is small (perhaps zero in this context), and the quantity ν2 depends on segmental-size differences only. The Prigogine δ parameter becomes important in systems with specific interactions, i.e., those with polar and hydrogen bonding. Differences in ced arising from these sources in such systems are modified by a factor of 0.25 according to Equation 2.11. If we now consider Equation 2.6, it can be seen that each of the three terms in this equation is in the form of a Prigogine δ as given by Equation 2.9. These terms describe normalized differences in the respective types of cohesive energy in corresponding states terminology. The cohesive energy differences in Equation 2.6 are normalized by RM2, the ced of the worst possible good solvent, i.e., a solvent located on the boundary of a Hansen solubility sphere rather than the ced of a given solvent under consideration. In strictly nonpolar systems, the polar and hydrogen bonding terms in Equation 2.6 are zero, and the interaction is described by the difference in δD. One is led to the conclusion that the first term in Equation 2.6 relates directly to the second term (the ρ effect) in Equation 2.11. In the future, this relationship could be explored in more detail with the hope of experimental verification of the coefficient in front of ρ.

7248_C002.fm Page 32 Wednesday, May 9, 2007 8:25 AM

32

Hansen Solubility Parameters: A User’s Handbook

If we now consider a system where δD1 is equal to δD2, the polymer–solvent interaction will be either polar or hydrogen bonding (or both) in nature, i.e., there will only be specific interactions. Such differences in δP and δH will be modified by 0.25 in Equation 2.6. It is noteworthy that the same factor, 0.25, modifies the Prigogine δ term in Equation 2.11, i.e., when there are specific interactions between solvent and polymer. The same 0.25 is present for a similar purpose both in Equation 2.6 (HSP) and in Equation 2.11 (CST) where the geometric mean was used. The geometric mean appears to be applicable to all types of energies discussed here.

THE χ12 PARAMETER AND HANSEN SOLUBILITY PARAMETERS Patterson and coworkers have shown that χ12 can be calculated, using ν2 as a symbol for the same quantity.17 Therefore, according to the previous discussion, it is expected that Equation 2.6 can be used in a similar way to predict χ12. There remains a general belief that χ12 can be calculated by Equation 2.1 using Hildebrand solubility parameters and a value of 0.34 for β. The change required to progress from the nonpolar Hildebrand solubility parameter to include polar and hydrogen bonding effects with the HSP, in calculating χ12, is to replace the Hildebrand solubility parameter difference of Equation 2.1 by a corresponding HSP term, i.e., A1,2 A1,2 = [(δD2 – δD1)2 + 0.25(δP2 – δP1)2 + 0.25(δH2 – δH1)2]

(2.12)

and χ12 is estimated from: χ12 = VA1,2/RT

(2.13)

The empirical factor β (0.34) in Equation 2.1 was found from studies on almost nonpolar systems using the Hildebrand solubility parameters. This is an average correction to these calculations because of the neglect of some relatively small but significant values of (δP1 – δP2) and/or (δH1 – δH2). β is not required in Equation 2.13. This same assumption was made by Zellers and coworkers in their approach to correlate the swelling and permeation of elastomers used in chemical protective clothing.28–31 An estimate for χ12 can also be found by noting that the total χ12 parameter in common solutions of polymers having high molecular weight is required to be close to 0.5 at the point of marginal solution or precipitation.1 This boundary value is called the critical chi parameter, χc. In HSP terminology, this is a boundary solvent with a placement directly on the sphere of solubility, and the quality is indicated by H in Equation 2.6 being equal to 1.0. This allows a simple estimate for χ12 for higher-molecular-weight polymers by the relation: χ12 = χc(RED)2 = Η/2

(2.14)

This last equation assumes an average V, just like the HSP correlations have done up to the point. It has been noted many times that liquids with lower V are often better solvents than liquids having essentially identical HSP but with larger V. This is seen with liquids like methanol (V = 40.7 cc/mol) and acetone (V = 74.0 cc/mol), which are sometimes good solvents, even when H is greater than 1.0 and for liquids like the phthalate and other plasticizers (V > 150 cc/mol), which are not good solvents in spite of H being less than 1.0. An explanation for this is found by comparing Equation 2.13 with Equation 2.14. Equation 2.15 can be derived from this comparison. The dependency of χc on polymer molecular size is also included in Equation 2.15, as this is a partial explanation for some of the results discussed later.1

7248_C002.fm Page 33 Wednesday, May 9, 2007 8:25 AM

Theory

33

RM2 = (Ro/2)2 = {0.5(1 + 1/r1/2)RT/V}

(2.15)

“r” is the ratio of the polymer size to that of the solvent and is usually considered as the approximate degree of polymerization, assuming the size of the solvent molecule as being close to that of the polymer segment size. For a solvent with V = 100 cc/mol, this added term leads to a correction of 1.1 for a polymer molecular weight of 10,000 and to a correction of 1.03 for a polymer molecular weight of 100,000. The correction term is expressed more generally by the relation in Equation 2.16. Correction = {0.5(r1–1/2 + r2–1/2)2}

(2.16)

Here, r1 and r2 are the number of statistical segments in the molecules in question. For mixtures of low molecular weight, where r1 is approximately equal to r2 and both are approximately equal to 1, the correction amounts to a factor of about 2; when one of the molecules is a high polymer, the correction amounts to 0.5 (as discussed earlier); when both the molecules have very high molecular weights, the correction approaches 0, meaning compatible mixtures of polymers are very difficult to achieve. A study of the kind discussed in the following for smaller molecules would seem appropriate to help clarify some of the questions raised. Table 2.1 gives the expected RM and Ro values based on Equation 2.15 for polymers of molecular weight high enough to neglect the effect of r. It is usually assumed that V for the average solvent is near 100 cc/mol. Table 2.1 indicates that all polymers of reasonably high molecular weight will be insoluble in solvents with V greater than about 100 cc/mol for a corresponding RA that is greater than an RM of 3.5 MPa1/2 (Ra greater than 7.0 MPa1/2). This is not generally the case as many values of RM have been reported that are much higher than this10 (see also Appendix, Table A.2); meaning they are more easily dissolved than Equation 2.15 indicates. Values for RM greater than 5 MPa1/2 are common, with some polymer RM values being considerably larger, although these are generally for lower-molecular-weight materials. This immediately points to potential problems in directly calculating χ12 from HSP data when RM is significantly larger than about 3.5 MPa1/2. Some improvement in the estimates of χ12 using Equation 2.14 is possible by including V in a correction term. Equation 2.14 has been retained for present purposes of comparison, however, because of its simplicity. The column for χ12, estimated from Equation 2.14 in Table 2.3 to Table 2.7, is placed adjacent to that of the solvent molar volume to allow an easy mental multiplication by V/100 if desired.

TABLE 2.1 Expected Solubility Parameter Differences for Marginal Solubility as a Function of the Molecular Volume of the Solvent V (cc/mol)

RM (MPa1/2)

Ro (MPa1/2)

50 100 200

5.0 3.5 2.5

10.0 7.0 5.0

7248_C002.fm Page 34 Wednesday, May 9, 2007 8:25 AM

34

Hansen Solubility Parameters: A User’s Handbook

TABLE 2.2 Hansen Solubility Parameter Data for Polymers Selected for the Comparisons Given in Table 2.2 to Table 2.6 Polymers

δD

δP

δH

Ro

Polybutadiene Polyisobutylene Polystyrene Polyvinylacetate Polyacrylonitrile

17.5 16.9 21.3 20.9 21.7

2.3 2.5 5.8 11.3 14.1

3.4 4.0 4.3 9.7 9.1

6.5 7.2 12.7 13.7 10.9

Note: Units are MPa1/2. Values in these units are 2.0455 times larger than in (cal/cm3)1/2.

TABLE 2.3 Comparison of Experimental, Indicative Chi Parameter Data, χlit, with Calculations Based on HSP for Polybutadiene, Buna Hüls CB 10 cis-Polybutadiene Raw Elastomer, Chemische Werke Hüls Solvent

V

Benzene Toluene Xylene Pentane n-Hexane n-Heptane n-Octane Chloroform Carbon tetrachloride Methanol Water

89.4 106.8 123.3 116.2 131.6 147.4 163.5 80.7 97.1 40.7 18.0

χ12 (Equation 2.14) 0.12 0.04 0.02 0.62 0.52 0.43 0.39 0.07 0.16 5.68 20.3

χ12 (Equation 2.13)

χlit

0.10 0.04 0.08 0.62 0.58 0.54 0.54 0.05 0.13 1.97 3.1

0.4 0.3 0.4 0.7 0.6 0.5 0.6 0.15 0.3 3.3 3.5

Source: Solubility data from Hansen, C.M., J. Paint Technol., 39(505), 104–117, 1967; Hansen, C.M., The Three Dimensional Solubility Parameter and Solvent Diffusion Coefficient, Their Importance in Surface Coating Formulation, Doctoral dissertation, Danish Technical Press, Copenhagen, 1967.

COMPARISON OF CALCULATED AND EXPERIMENTAL χ12 PARAMETERS The predictions of χ12 using Equation 2.13 and Equation 2.14 have been compared with χ12 parameter data in standard references.32,33 The polymers used for the comparisons are listed in Table 2.2 with their HSP data. The calculated and indicative experimental values for χ12 for the given solvent–polymer systems are reported in Table 2.3 to Table 2.7. The polymers were chosen because

7248_C002.fm Page 35 Wednesday, May 9, 2007 8:25 AM

Theory

35

TABLE 2.4 Comparison of Experimental, Indicative Chi Parameter Data, χlit, with Calculations Based on HSP for Polyisobutylene, Lutonal® I60, BASF Solvent

V

χ12 (Equation 2.14)

χ12 (Equation 2.13)

χlit

Benzene Toluene Decalin Cyclohexane Pentane n-Hexane n-Octane n-Nonane Chloroform Carbon tetrachloride Methylene dichloride

89.4 106.8 156.9 108.7 116.2 131.6 163.5 179.7 80.7 97.1 63.9

0.19 0.10 0.35 0.20 0.44 0.37 0.29 0.27 0.06 0.20 0.25

0.17 0.11 0.43 0.23 0.53 0.49 0.50 0.51 0.05 0.21 0.17

0.5 0.5 0.4 0.45 0.5 0.5 0.5 0.3+ 1.0 0.5 0.6

Source: Solubility parameter data from Hansen, C.M., Solubility parameters, in Paint Testing Manual, Manual 17, Koleske, J.V., Ed., American Society for Testing and Materials, Philadelphia, 1995, pp. 383–404.

of similarities/differences in the HSP data, as well as the availability of sufficient data on the χ12 parameter. This is not a complete evaluation of Equation 2.13 and Equation 2.14, but it points out clearly that there are factors which are not completely understood. Table 2.8 provides these solvents along with their HSP. A casual inspection of the measured and calculated χ12 values in Table 2.3 through Table 2.7 would give the impression that there are significant discrepancies between these values which require further explanation. The calculated and literature values for χ12 agree in some cases and differ significantly in others. Some possible reasons for this are discussed in the following sections.

POLYBUTADIENE The calculated and experimental χ12 values for polybutadiene are given in Table 2.3. The first three entries are for aromatic solvents. The solubility parameter predictions indicate that these are exceptionally good solvents, whereas the χ12 values indicate that they are moderately good. Before one adds on a constant value of about 0.3 to bring an agreement, it should be noted that the calculated and experimental χ12 values for the aliphatic solvents are in good agreement. The solubility parameters for the higher-molecular-weight homologs are closer to those of the polymer, but the size effect reduces solvent quality. Agreement for the aliphatic solvents is considered excellent. It should be noted that Ro is very near the ideal value for such calculations according to Table 2.1. Chloroform and carbon tetrachloride are predicted by HSP to be very good solvents, especially confirmed by the χlit for chloroform. HSP considerations indicate that chloroform and the aromatic solvents are near neighbors with similar HSP and might have similar qualities. This is not borne out by the χlit values for the aromatics, which are suspected as being too high for presently unknown reasons. The calculated and literature values for methanol and water are different enough to warrant a comment. HSP considerations indicate that the difference in behavior between these two liquids should be sizable, which the χlit values do not indicate. A problem of some significance in any

7248_C002.fm Page 36 Wednesday, May 9, 2007 8:25 AM

36

Hansen Solubility Parameters: A User’s Handbook

TABLE 2.5 Comparison of Experimental, Indicative Chi Parameter Data, χlit, with Calculations Based on HSP for Polystyrene, Polystyrene LG, BASF Solvent

V

χ12 (Equation 2.14)

χ12 (Equation 2.13)

χlit

Benzene Toluene Xylene Ethyl benzene Styrene Tetralin Decalin (cis) Cyclohexane Methyl cyclohexane n-Hexane n-Heptane n-Octane Acetone Methyl ethyl ketone Methyl isobutyl ketone Cyclohexanone Ethyl acetate n-Butyl acetate sec-Butyl acetate

89.4 106.8 123.3 123.1 115.6 136.0 156.9 108.7 128.3 131.6 147.1 163.5 74.0 90.1 125.8 104.0 98.5 132.5 133.6

0.23 0.21 0.25 0.26 0.16 0.09 0.24 0.41 0.49 0.67 0.61 0.58 0.51 0.38 0.45 0.15 0.40 0.40 0.54

0.66 0.73 0.99 1.05 0.61 0.38 1.22 1.44 2.03 2.87 2.92 3.08 1.22 1.12 1.83 0.52 1.29 1.72 2.35

0.40–0.44 0.40–0.44 0.4 0.45 0.35 0.4 0.5 0.50–1.0 0.5 0.8 0.8 0.9 0.6 0.49 0.5 0.5 0.5 0.5 0.4

Source: Solubility data from Hansen, C.M., J. Paint Technol., 39(505), 104–117, 1967; Hansen, C.M., The Three Dimensional Solubility Parameter and Solvent Diffusion Coefficient, Their Importance in Surface Coating Formulation, Doctoral dissertation, Danish Technical Press, Copenhagen, 1967.

study of solvents at low concentrations in polymers is that the smaller amounts of solvent relative to the polymer can lead to preferential association of the solvent with those local regions/segments/groups in the polymer which have similar energies (HSP). These local regions may not necessarily reflect the same affinities as the polymer as a whole, such as are reflected by the solubleor-not approach commonly used in HSP evaluations. These local association effects can influence results on swelling studies in both good and bad solvents, for example. Other types of studies carried out at low-solvent concentrations can also be influenced by these segregation/association phenomena. An extension of this type of situation can be cited in the tendencies of water to associate with itself, as well as with local regions within polymers. This has made simple predictions of its behavior impossible. A detailed discussion of this is beyond the scope of this chapter. It is suggested, however, that the potential differences observed here between HSP predictions and observed χlit may be derived from such phenomena.

POLYISOBUTYLENE The calculated and experimental χ12 values for polyisobutylene are given in Table 2.4. There are some similarities with polybutadiene both chemically and in the Ro value of 7.2 MPa1/2 being near the ideal for a polymer of very high molecular weight. The cyclic and aromatic solvents are again better as judged by HSP than the χlit values indicate, whereas the estimates for the aliphatic solvents are in excellent agreement with Equation 2.13, in particular. Again, HSP finds chloroform,

7248_C002.fm Page 37 Wednesday, May 9, 2007 8:25 AM

Theory

37

TABLE 2.6 Comparison of Experimental, Indicative Chi Parameter Data, χlit, with Calculations Based on HSP for Polyvinylacetate, Mowilith® 50, Farbwerke Hoechst Solvent

V

χ12 (Equation 2.14)

χ12 (Equation 2.13)

χlit

Benzene Toluene Decalin (cis) Tetralin Cyclohexane Methyl cyclohexane n-Nonane n-Decane Acetone Methyl ethyl ketone Methyl isobutyl ketone Ethyl acetate Dimethyl phthalate Dioxane Chloroform Chlorobenzene n-Propanol

89.4 106.8 156.9 136.0 108.7 128.3 179.7 195.9 74.0 90.1 125.8 98.5 163.0 85.7 80.7 102.1 75.2

0.56 0.51 0.64 0.37 0.76 0.49 0.88 0.88 0.33 0.33 0.45 0.39 0.12 0.29 0.32 0.33 0.47

1.91 2.05 3.80 1.92 3.13 2.03 6.00 6.54 0.92 1.11 1.83 1.46 0.58 0.95 0.99 1.27 1.34

0.3–0.5 0.5 2.7 1.3 2.4 0.5 3.3 3.4 0.3–0.46 0.4–0.44 0.5 0.4 0.4 0.4 0.4 0.5 1.2–1.6

Source: Solubility data from Hansen, C.M., J. Paint Technol., 39(505), 104–117, 1967; Hansen, C.M., The Three Dimensional Solubility Parameter and Solvent Diffusion Coefficient, Their Importance in Surface Coating Formulation, Doctoral dissertation, Danish Technical Press, Copenhagen, 1967.

TABLE 2.7 Comparison of Experimental, Indicative Chi Parameter Data, χlit, with Calculations Based on HSP for Polyacrylonitrile Solvent

V

χ12 (Equation 2.14)

χ12 (Equation 2.13)

χlit

Ethylene carbonate γ-Butyrolactone Ethanol Water N,N-Dimethyl formamide N,N-Dimethyl acetamide Dimethyl sulfoxide Tetramethylene sulfoxide

66.0 76.8 58.5 18.0 77.0 92.5 71.3 90.0

0.40 0.16 1.15 5.3 0.33 0.44 0.21 0.25

0.63 0.30 1.61 2.3 0.61 0.97 0.36 0.53

0.4 0.36–0.40 4.0 2.0 0.2–0.3 0.4 0.3–0.4 0.3

Source: Solubility data from Brandrup, J. and Immergut, E.H., Eds., Polymer Handbook, 3rd ed., Wiley-Interscience, New York, 1989. (a) Gundert, F. and Wolf, B.A., Polymersolvent interaction parameters, pp. VII/173–182. (b) Fuchs, O., Solvents and non-solvents for polymers, pp. VII/379–407.

7248_C002.fm Page 38 Wednesday, May 9, 2007 8:25 AM

38

Hansen Solubility Parameters: A User’s Handbook

TABLE 2.8 Hansen Solubility Parameters for the Solvent Included in Table 2.3 to Table 2.7 Solvent

δD

δP

δH

Solvent

δD

δP

δH

Benzene Toluene Xylene Ethyl benzene Styrene Decalin (cis) Tetralin Cyclohexane Methyl cyclohexane n-Pentane n-Hexane n-Heptane n-Octane n-Nonane Acetone Methyl ethyl ketone Methyl isobutyl ketone Cyclohexanone Ethyl acetate

18.4 18.0 17.6 17.8 18.6 18.0 19.6 16.8 16.0 15.6 14.9 15.3 15.5 15.7 15.5 16.0 15.3 17.8 15.8

0.0 1.4 1.0 0.6 1.0 0.0 2.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 10.4 9.0 6.1 6.3 5.3

2.0 2.0 3.1 1.4 4.1 0.0 2.9 0.2 1.0 0.0 0.0 0.0 0.0 0.0 7.0 5.1 4.1 5.1 7.2

n-Butyl acetate sec-Butyl acetate Dimethyl phthalate 1,4-Dioxane Chloroform Chlorobenzene Carbon tetrachloride Methylene dichloride Methanol Ethanol n-Propanol Ethylene carbonate γ-Butyrolactone N,N-dimethyl formamide N,N-dimethyl acetamide Dimethyl sulfoxide Tetramethylene sulfoxide Water

15.8 15.0 18.6 19.0 17.8 19.0 17.8 18.2 15.1 15.8 16.0 19.4 19.0 17.4 16.8 18.4 18.2 15.5

3.7 3.7 10.8 1.8 3.1 4.3 0.0a 6.3 12.3 8.8 6.8 21.7 16.6 13.7 11.5 16.4 11.0 16.0

6.3 7.6 4.9 7.4 5.7 2.0 0.6 6.1 22.3 19.4 17.4 5.1 7.4 11.3 10.2 10.2 9.1 42.3

Note: Units are MPa1/2. a

The value 0.0 is valid in nonpolar media and derives from a zero dipole moment; a progressively higher value in increasingly polar media is required because of induced dipoles10 (see also Chapter 1).

methylene dichloride, and carbon tetrachloride as being very good, in agreement with solubilityor-not experiments, whereas the χlit values indicate these are not good or at best, marginal in quality. The results of Equation 2.13 for the aliphatic hydrocarbons are particularly in good agreement with χlit.

POLYSTYRENE The calculated and experimental χ12 values for polystyrene are given in Table 2.5. The Ro value of 12.7 MPa1/2 is now much higher than the ideal value indicated in Table 2.1. The polymer molecular weight is thought to be reasonably high, but is unknown. As a consequence of the Ro value, practically all the χ12 values calculated by Equation 2.13 are too high. One is tempted to divide by a factor of 2 or 3, but there is no consistent pattern. Equation 2.14 includes the boundary value of χc equal to 0.5, so the results are more in agreement with χlit . HSP predicts that the aromatic and cyclic solvents are somewhat better than that expected from χlit. The agreement would be better if the χ12 values obtained from Equation 2.14 for these were increased by a factor 2. The values found by Equation 2.14 for the aliphatic hydrocarbons are also lower than χlit, but are qualitatively in agreement. The χ12 values for ketones and esters, using Equation 2.14, are in generally good agreement with the literature values. An exception of some note is the well-known good solvent cyclohexanone that is predicted as a much better solvent by HSP than the χlit value would indicate. There is considerably more differentiation in predictions of solvent quality found by Equation 2.14 than values indicated by χlit.

7248_C002.fm Page 39 Wednesday, May 9, 2007 8:25 AM

Theory

39

POLYVINYLACETATE The calculated and experimental χ12 values for polyvinylacetate are given in Table 2.6. The molecular weight for this polymer is reported as being 260,000. It can initially be noted that Ro is 13.7 MPa1/2, which again means χ12 values found from Equation 2.13 will be higher than those found in the literature. This difference varies considerably, but a factor of 2 to 4 is generally required to give reasonable agreement. Equation 2.13 is certainly not generally acceptable as an instrument to predict χlit. Equation 2.14 gives reasonably good approximations to χlit as long as the solvents are good enough to dissolve the polymer; however, there are some major disagreements. Tetralin dissolves the polymer, but has χlit equal to 1.3. n-Propanol is an error from the HSP prediction of it being a marginal solvent, whereas it is a nonsolvent. Alcohols of higher and lower molecular weight do have a significant effect on this polymer, however, and the azeotropic mixture of ethanol and water actually dissolves it.3,6 When dealing with nonsolvents, the HSP predictions of χ12 are generally lower than the data found in the literature. Once again, a factor of 3 to 4 is required to bring the values into agreement.

POLYACRYLONITRILE The calculated and experimental χ12 values for polyacrylonitrile are given in Table 2.7. This polymer has high polar and hydrogen-bonding parameters and Ro equal to 10.9 MPa1/2 which, once more, is somewhat above the ideal range. The agreement with Equation 2.14 is reasonably good for the good solvents. The nonsolvents are not in good agreement. Equation 2.13 agrees surprisingly well with the best solvents, γ-butyrolactone and dimethyl sulfoxide, but the agreement is not uniform when all the solvents are considered.

GENERAL DISCUSSION It should be noted in general that χ12 can either increase or decrease with concentration of the polymer. Barton32 presents data to examine the potential magnitude of this effect. The correlations given in Table 2.2 are based on whether or not the polymer dissolves at a concentration of 10%, with the exception of the data for polyacrylonitrile where no polymer concentration is indicated in the original solubility data.33 The HSP data for correlations of the type given in Table 2.2 can also be expected to change for higher polymer concentration and molecular weight. RM is expected to decrease only slightly for marginally higher polymer molecular weight, while considering a reasonably high molecular weight, and RM is expected to decrease somewhat for higher polymer concentration, although this can vary, especially for lower-molecular-weight species. An interesting fact to keep in mind is that a polymer with molecular weights (in millions) will only swell in even the best solvents. The present evaluations are at the same polymer concentration unless otherwise noted. No significant corrections of the type included in Equation 2.16 are required, as the polymer molecular weight is very high in all cases. Therefore, corrections of this type are not responsible for the differences in the calculated and observed χ12 parameters. A point of some concern is that negative values for χ12 are found in the literature, but these are not allowed in either the CST or HSP approaches. There is no obvious general explanation for this situation. A negative χ12 implies a solvent of a quality superior to anything that normal polymer–liquid interactions could provide. Normal here also includes the specific interactions attributable to permanent dipole–permanent dipole and hydrogen bonding interactions, as discussed earlier. A closer review of this situation is desired. No systems with negative χ12 are included in Table 2.2 to Table 2.7. An additional problem of some concern is that, in general, there is considerable scatter in the χ12 parameter data from different sources. Clearing up this situation is far beyond the scope of this book. However, one cannot help but wonder why, and the seeming discrepancies do not contribute

7248_C002.fm Page 40 Wednesday, May 9, 2007 8:25 AM

40

Hansen Solubility Parameters: A User’s Handbook

to blind confidence in any of the reported χ12 values. Indicative χ12 values are used here. One can also find variations in HSP values for polymers from different sources,32 so there are also problems in determining which values are best in this approach. The χ12 parameter does not specifically account for permanent dipole–permanent dipole or hydrogen bonding interactions, which must be considered a major source of potential differences. There has been some discussion as to whether the coefficient 4 in Equation 2.5 (corresponding to a coefficient of 0.25 in Equation 2.12) should be a different number. Barton cites a case where a coefficient of 0.2 (rather than 0.25) in Equation 2.12 was determined.34 Skaarup has mentioned a case of 5 (rather than 4) as a value for the coefficient in Equation 2.5, which, of course, gives 0.2 in Equation 2.12.35 The author has also explored situations involving water, where the DATA FIT was equal for either a constant 4 or 5 in Equation 2.5. Zellers and coworkers used this coefficient as an adjustable parameter for individual solvents in their studies.28–31 One significant factor in this discussion is that solvents with higher solubility parameters generally have lower molecular volumes. This means they will be better than expected by average comparisons of behavior. This fact tends to lead one to stretch the spheres a little more in the δP and δH directions to encompass these good solvents that would otherwise lie outside of the spheroids. This would lead to a number that is slightly higher than the 4 in Equation 2.5, and it is the author’s feeling that this could be the case whenever a complete understanding of the effect of solvent molecular volume, and other size effects is accomplished. The use of the coefficient 4 is also confirmed in Chapter 9. The Prigogine theory contains structural parameters that have not been explored in this context. There are also structural parameters in the New Flory theory. It is possible that the use of structural parameters will allow better understanding, enhance the possibility of improved calculations, and reduce the need for experimental studies. The experimentally-determined radius for the solubility spheres automatically takes these factors into account, but reliable calculation of the radius of interaction has not been possible as yet.

POSTSCRIPT The author has always experienced consistency in the quality of the predictions using the HSP. Care is required to generate the necessary data, and there should always be a reevaluation of experimental data based on an initial correlation. The solvent parameters have been used with success for many years in industrial practice to predict solvent quality using computer techniques by most, if not all, major solvent suppliers. Mixing rules have been established for even complicated solvent blends. These are usually based on summing up simple volume fraction times the solubility parameter values. (An evaluation of the quality of the χ12 values in the literature could be made with precipitation experiments for mixtures to see whether a mixing rule gives consistent results for these as well.) The solvent δD, δP, and δH values that were established with extensive calculations have been supported by tens of thousands of experimental data points based on solubility, permeation, surface wetting, etc.10 It has become quite clear that the HSP for the solvents are not precise enough for sophisticated calculations, but they certainly represent a good and satisfactory means for practical applications. The HSP for the solvents relative to each other are correct for the majority of the common solvents. The “nearest neighbors” to a good solvent are clearly expected to be of nearly comparable quality unless they are in a boundary region of the HSP solubility sphere. The solvent quality indicated by the ratio Ra/Ro (the RED number) has been particularly satisfactory. This ratio was defined years ago as a ratio of solubility parameters, as plotting and interpretation of data used solubility parameters. Use of the ratio of cohesion energy densities is also possible, of course, as this is indeed closer to an energy difference number and would agree more with the Flory approach as seen in Equation 2.6 and Equation 2.14, as H is really nothing other than (RED)2.

7248_C002.fm Page 41 Wednesday, May 9, 2007 8:25 AM

Theory

41

As a result of having reviewed all of this here, the author senses that the HSP approach is a practical extension in complete agreement with the Prigogine–Flory theory when the geometric mean is used, at least as far as the major factors discussed earlier are concerned. The comparisons presented previously confirm some relation, but the single χ12 parameter may have been oversimplified, such that the more complete HSP approach cannot be immediately recognized. The ability of HSP to describe molecular affinities among so many different materials listed in this book speaks for the general application of both the Prigogine and the HSP treatments. The Prigogine treatment is acknowledged as difficult to use in practice. This is not true of the HSP approach.

CONCLUSION The Prigogine/Patterson CST and the HSP approach (which also involves a corresponding states calculation) are shown to have very close resemblance. Both can be used to estimate the Flory χ12. Two equations involving HSP are given for this purpose. Reasonably good predictions are possible under favorable circumstances. Favorable circumstances involve a system with an essentially nonpolar polymer whose Ro value is not too different from 7.0 MPa1/2. χ12values for the better solvents are calculated by HSP at lower levels than those found in the literature. χ12 values for nonsolvents are also generally calculated by HSP as being significantly lower than the reported literature values. The most favorable circumstances are, of course, not always present, and some problems still exist and need to be solved before these calculations can be used with confidence to estimate χ12 values for any solvent–polymer system. The HSP values for the polymers used for the present comparisons are based on solubility-or-not type experiments which reflect the properties of the polymer as a whole. These may not completely correspond to the type of evaluation often used to find χ12, as less-than-dissolving amounts of solvent may be used, and the solvent may associate with given segments or groups in the polymer and not reflect the behavior of the polymer as a whole (see also the discussion in Chapter 5). An empirical factor, β, equal to about 0.34 appears in many sources in the literature in connection with calculation of χ12 using Hildebrand solubility parameters. β disappears when HSP are used for this purpose, but the resulting equation has not been studied enough to allow general use of HSP to calculate χ12 parameters. Studies on the effect of molecular size, segmental size, and polymer size are still required. It is suggested that the structural factors discussed by Prigogine be tried in this respect.11 Use of the geometric mean in conjunction with the Prigogine theory brings the HSP and Prigogine approaches into agreement. The massive amount of experimental data presented in this book strongly supports the use of the geometric mean. As a curiosity, it might be noted that the use of the geometric mean (Lorenz–Berthelot mixtures) generated an ellipsoidal miscibility plot essentially identical to those given in Chapter 5, Figure 5.1 and Figure 5.2.36 This approach was not continued because it was stated that “the boundary of this ellipse is of little practical importance as there are no known cases of immiscibility in mixtures known to conform to the Lorenz–Berthelot equations.” As stated in the Preface to this book, it has not been its purpose to recite the developments of polymer solution thermodynamics in a historical manner with full explanations of each theory or modifications thereof. The references cited in the Preface do this already. Chapter 3 and Chapter 4 have been added to this edition of this handbook to give broader coverage in this respect. This chapter has attempted to show relations between the classical theories of polymer solution thermodynamics and the HSP approach, which includes a quantitative accounting of both permanent dipole–permanent dipole and hydrogen bonding interactions as an integral part. The relation between the Prigogine–Patterson theory and HSP was the most obvious.

7248_C002.fm Page 42 Wednesday, May 9, 2007 8:25 AM

42

Hansen Solubility Parameters: A User’s Handbook

REFERENCES 1. Flory, P.J., Principles of Polymer Chemistry, Cornell University Press, New York, 1953. 2. Eichinger, B.E. and Flory, P.J., Thermodynamics of polymer solutions, Trans. Faraday Soc., 64(1), 2035–2052, 1968; Trans. Faraday Soc. (2), 2053–2060; Trans. Faraday Soc. (3), 2061–2065; Trans. Faraday Soc. (4), 2066–2072. 3. Hansen, C.M., The three dimensional solubility parameter — key to paint component affinities I. Solvents, plasticizers, polymers, and resins, J. Paint Technol., 39(505), 104–117, 1967. 4. Hansen, C.M., The three dimensional solubility parameter — key to paint component affinities II. Dyes, emulsifiers, mutual solubility and compatibility, and pigments, J. Paint Technol., 39(511), 505–510, 1967. 5. Hansen, C.M. and Skaarup, K., The three dimensional solubility parameter — key to paint component affinities III. Independent calculation of the parameter components, J. Paint Technol., 39(511), 511–514, 1967. 6. Hansen, C.M., The Three Dimensional Solubility Parameter and Solvent Diffusion Coefficient, Their Importance in Surface Coating Formulation, Doctoral dissertation, Danish Technical Press, Copenhagen, 1967. 7. Hansen, C.M., The universality of the solubility parameter, Ind. Eng. Chem. Prod. Res. Dev., 8(1), 2–11, 1969. 8. Hansen, C.M. and Beerbower, A., Solubility parameters, in Kirk-Othmer Encyclopedia of Chemical Technology, Suppl. Vol., 2nd ed., Standen, A., Ed., Interscience, New York, 1971, pp. 889–910. 9. Hansen, C.M., 25 Years with solubility parameters (25 År med Opløselighedsparametrene, in Danish), Dan. Kemi, 73(8), 18–22, 1992. 10. Hansen, C.M., Solubility parameters, in Paint Testing Manual, Manual 17, Koleske, J.V., Ed., American Society for Testing and Materials, Philadelphia, 1995, pp. 383–404. 11. Prigogine, I. (with the collaboration of Bellemans, A. and Mathot, A.), The Molecular Theory of Solutions, North-Holland, Amsterdam, 1957, chap. 16, 17. 12. Patterson, D. and Delmas, G., New aspects of polymer solution thermodynamics, Off. Dig. Fed. Soc. Paint Technol., 34(450), 677–692, 1962. 13. Delmas, D., Patterson, D., and Somcynsky, T., Thermodynamics of polyisobutylene-n-alkane systems, J. Polym. Sci., 57, 79–98, 1962. 14. Bhattacharyya, S.N., Patterson, D., and Somcynsky, T., The principle of corresponding states and the excess functions of n-alkane mixtures, Physica, 30, 1276–1292, 1964. 15. Patterson, D., Role of free volume changes in polymer solution thermodynamics, J. Polym. Sci. Part C, 16, 3379–3389, 1968. 16. Patterson, D.D., Introduction to thermodynamics of polymer solubility, J. Paint Technol., 41(536), 489–493, 1969. 17. Biros, J., Zeman, L., and Patterson, D., Prediction of the χ parameter by the solubility parameter and corresponding states theories, Macromolecules, 4(1), 30–35, 1971. 18. Hildebrand, J. and Scott, R.L., The Solubility of Nonelectrolytes, 3rd ed., Reinhold, New York, 1950. 19. Hildebrand, J. and Scott, R.L., Regular Solutions, Prentice-Hall, Englewood Cliffs, NJ, 1962. 20. Hansen, C.M., Cohesion parameters for surfaces, pigments, and fillers (Kohæsionsparametre for Overflader, Pigmenter, og Fyldstoffer, in Danish), Färg och Lack Scand., 43(1), 5–10, 1997. 21. Hansen, C.M., Polymer solubility — prigogine theory and Hansen Solubility parameter theory mutually confirmed (Polymeropløselighed — Prigogine Teori og Hansen Opløselighedsparameterteori Gensidigt Bekræftet, in Danish), Dan. Kemi, 78(9), 4–6, 1997. 22. Hildebrand, J. and Scott, R.L., The Solubility of Nonelectrolytes, 3rd ed., Reinhold, New York, 1950, chap. 20. 23. Hansen, C.M., Characterization of surfaces by spreading liquids, J. Paint Technol., 42(550), 660–664, 1970. 24. Hansen, C.M., Surface dewetting and coatings performance, J. Paint Technol., 44(570), 57–60, 1972. 25. Hansen, C.M. and Pierce, P.E., Surface effects in coatings processes, XII Federation d’Associations de Techniciens des Industries des Peintures, Vernis, Emaux et Encres d’Imprimerie de l’Europe Continentale, Congress Book, Verlag Chemie, Weinheim/Bergstrasse, 1974, 91–99; Ind. Eng. Chem., Prod. Res. Dev., 13(4), 218–225, 1974.

7248_C002.fm Page 43 Wednesday, May 9, 2007 8:25 AM

Theory

43

26. Hansen, C.M. and Wallström, E., On the use of cohesion parameters to characterize surfaces, J. Adhes., 15(3/4), 275–286, 1983. 27. Patterson, D., personal communication, 1997. 28. Zellers, E.T., Three-dimensional solubility parameters and chemical protective clothing permeation. I. Modeling the solubility of organic solvents in Viton® gloves, J. Appl. Polym. Sci., 50, 513–530, 1993. 29. Zellers, E.T. and Zhang G.-Z., Three-dimensional solubility parameters and chemical protective clothing permeation. II. Modeling diffusion coefficients, breakthrough times, and steady-state permeation rates of organic solvents in Viton® gloves, J. Appl. Polym. Sci., 50, 531–540, 1993. 30. Zellers, E.T., Anna, D.H., Sulewski, R., and Wei, X., Critical analysis of the graphical determination of Hansen’s solubility parameters for lightly crosslinked polymers, J. Appl. Polym. Sci., 62, 2069–2080, 1996. 31. Zellers, E.T., Anna, D.H., Sulewski, R., and Wei, X., Improved methods for the determination of Hansen’s solubility parameters and the estimation of solvent uptake for lightly crosslinked polymers, J. Appl. Polym. Sci., 62, 2081–2096, 1996. 32. Barton, A.F.M., Handbook of Polymer-Liquid Interaction Parameters and Solubility Parameters, CRC Press, Boca Raton, FL, 1990. 33. Brandrup, J. and Immergut, E.H., Eds., Polymer Handbook, 3rd ed., Wiley-Interscience, New York, 1989. (a) Gundert, F. and Wolf, B.A., Polymer-solvent interaction parameters, pp. VII/173–182. (b) Fuchs, O., Solvents and non-solvents for polymers, pp. VII/379–407. 34. Barton, A.F.M., Applications of solubility parameters and other cohesion energy parameters, Polym. Sci. Technol. Pure Appl. Chem., 57(7), 905–912, 1985. 35. Skaarup, K., private communication, 1997. 36. Rowlinson, J.S., Liquids and Liquid Mixtures, Butterworths Scientific Publications, London, 1959, chap. 9.

7248_C002.fm Page 44 Wednesday, May 9, 2007 8:25 AM

7248_book.fm Page 45 Tuesday, April 24, 2007 9:19 AM

Thermodynamic 3 Statistical Calculations of the Hydrogen Bonding, Dipolar, and Dispersion Solubility Parameters Costas Panayiotou KEY WORDS Statistical thermodynamics, cohesive energy density, Hansen solubility parameters, solvent screening.

ABSTRACT The main objective of this chapter is the presentation of an equation-of-state framework for the calculation of the hydrogen-bonding component of the solubility parameter as well as the other partial solubility parameters. A new statistical thermodynamic approach has been developed for the estimation of these partial components over a broad range of temperature and pressure. Key to this approach is the development of explicit expressions for the contribution of hydrogen bonding, dispersion, and dipolar interactions to the potential energy of the fluid. The approach is applicable to ordinary solvents, supercritical fluids, and high polymers. Information on various thermodynamic properties of fluids is used in order to estimate the three solubility parameter components. Extensive tables with the key parameters are presented. When information on hydrogen bonding interaction is available from other sources, the proposed method is essentially a predictive method for the hydrogen bonding component of the solubility parameter. On the other hand, available information on these separate components is exploited for extracting information on the thermodynamic behavior of the fluids over an extended range of external conditions.

INTRODUCTION The conceptual simplicity of the solubility parameter, δ, makes it most attractive in industry and in academia as well. Originally introduced by Hildebrand,1 it remains today one of the key parameters for selecting solvents in industry, characterizing surfaces, predicting solubility and degree of rubber swelling, polymer compatibility, chemical resistance, permeation rates, and for numerous other applications. There is also much interest in utilizing solubility parameters for rationally designing new processes, such as the supercritical fluid, the coating, and the drug delivery processes.2–8 Of course, the use of solubility parameter, or the closely related cohesive density is not always successful and this lack of total success stimulates continuing research. The central principle behind the use of δ is the historic alchemist maxim, similia similibus solvuntur (“like dissolves like”), 45

7248_book.fm Page 46 Tuesday, April 24, 2007 9:19 AM

46

Hansen Solubility Parameters: A User’s Handbook

probably the oldest rule of solubility. This rule can, indeed, be a good guide in the study of solubility, as long as it enables definition of the degree of likeness in the given system with sufficient precision. This need for precision in the definition of likeness led to the division of δ into its partial components or Hansen solubility parameters5 δd, δp, and δhb, for the dispersion, the polar, and the hydrogen bonding contributions, respectively. Thus, liquids with similar δd, δp, and δhb are very likely to be miscible. The bulk of the developments in solubility parameter reside on the principle of “similarity matching” of properties. As it is recognized, however, that a more appropriate principle would be the “complementary matching” of properties,9 the hydrogen bonding component, δhb, is further subdivided into an acidic component, δa, and a basic component, δb, in order to account for the Lewis acid and Lewis-base character of the substance.8–10 Over the years, the partial solubility parameters were determined for an enormous number of substances and led to critical compilations available in the open literature.3–5 These compilations are most valuable sources of information for the nature of the substances and their interactions with other substances. Starting from the original definition of cohesive energy density and solubility parameter, we have already proposed a systematic approach for estimating the solubility parameter over an extended range of temperature and pressure.11–12 In this work, it became clear that the hydrogen bonding contribution could be calculated rather accurately from the hydrogen bonding part of the potential energy E and the volume V of the system as obtained, for example, from the lattice-fluid hydrogen bonding (LFHB) equation-of-state model.13 The model, however, could not separate the dispersion and the polar components of the solubility parameters. The proposition was made to calculate δd from the solubility parameter of the corresponding homomorph hydrocarbon. Although this proposal could be valid for some classes of fluids, it could not be generalized. In a recent publication,14 we have proposed a group contribution method for the estimation of the total solubility parameter of a large variety of substances. The very same method could be used for the estimation of δd, as shown later. Knowledge of the hydrogen bonding and the dispersion components of the solubility parameter could lead to an estimation of the polar component as well. This chapter is, however, heavily based on a more recent publication15 in which the previous approach11–12 was extended in an effort to account for all three components of the solubility parameter. This was done by adopting the more recent and more accurate NRHB (nonrandom hydrogen-bonding) equation-of-state framework,16 which was modified in order to explicitly account for dipole–dipole interactions and, thus, explicitly calculate the polar component, δp.

THEORY THE EQUATION-OF-STATE FRAMEWORK Let us consider a system of N molecules of a fluid at temperature T, external pressure P, and of volume V, which are assumed to be arranged on a quasi-lattice that has a coordination number z, number of sites Nr , N0 and that denotes empty sites. Each molecule is assumed to be divided in r segments of segmental volumes v*, and to have zq = zrs external contacts, s being its surface-tovolume ratio, a geometric characteristic of the molecule. The total number (Nr) of lattice sites is given by: Nr = rN + N0

(3.1)

Following previous practice,15,16 one may write for the configurational partition function of the fluid in the N,P,T ensemble and in its maximum term approximation: Q ( N , T , P ) = QR QNR Qhb = Ω R Ω NR Ω hb exp

−E p − Ed − E hb − PV exp exp exp kT kT kT kT

(3.2)

7248_book.fm Page 47 Tuesday, April 24, 2007 9:19 AM

Statistical Thermodynamic Calculations

47

Ed, Ep, and Ehb in Equation 3.2 are the dispersion, polar, and hydrogen bonding components, respectively, of the potential energy of the system. The detailed rationale behind the form of the combinatorial term, ΩR, its correction factors for nonrandom distribution of free volume, ΩNR, and for the hydrogen bonding, Ωhb, can be found in the previous work.16 Here, the final equations are simply reproduced, namely: N r ! N rl N ⎛ N q ! ⎞ ΩR = ω N 0 ! N ! ⎜⎝ N r ! ⎟⎠

z 2

N

(3.3)

where l=

(

) (

)

z r −q − r −1 2

(3.4)

whereas the total number of intermolecular contacts in the system is given by: zNq = zqN + zN0

(3.5)

In Equation 3.3, ω is a characteristic quantity for each fluid that takes into account the flexibility and the symmetry of the molecule, and this quantity cancels out in all applications of interest here. In the following, we will need the site fractions f0 and f for the empty sites and the molecular segments, respectively. The relation is given by: f0 =

N 0 N r − rN = = 1− f Nr Nr

(3.6)

For the second factor, ΩR, we may use various expressions available in the open literature.16 The most classical is Guggenheim’s quasi-chemical expression17: 2

⎡⎛ N 0 ⎞ ⎤ N ! N ! ⎢ ⎜ r 0 ⎟ !⎥ ⎢⎣⎝ 2 ⎠ ⎥⎦ = 2 ⎡⎛ N r 0 ⎞ ⎤ N rr ! N 00 ! ⎢⎜ ⎟ !⎥ ⎢⎣⎝ 2 ⎠ ⎥⎦ 0 rr

Ω NR

0 00

(3.7)

where Nrr is the number of external contacts between the segments belonging to molecules; N00 is the number of contacts between the empty sites; and Nr0 is the number of contacts between a molecular segment and an empty site. The superscript 0 refers to the case of randomly distributed empty sites. In this random case, we have: N rr0 =

1 qN z zqN = qN θr 2 N 0 + qN 2

(3.8)

7248_book.fm Page 48 Tuesday, April 24, 2007 9:19 AM

48

Hansen Solubility Parameters: A User’s Handbook

0 N 00 =

1 N z N 0 z 0 = N 0 θ0 2 Nq 2

(3.9)

and N r00 = zqN

N0 qN = zN 0 = zqN θ0 = zN 0 θr Nq Nq

(3.10)

where θr = 1 − θ0 =

q r q r + v −1

(3.11)

and the reduced volume is defined as:

v=

V rNvv ∗ 1 = = ∗ ρ V rNv ∗

(3.12)

ρ˜ being the reduced density. The corresponding number of intersegmental contacts (Nij) in the nonrandom case are given by the following equations: N rr = N rr0 Γ rr =

z qN θr Γ rr 2

0 N 00 = N 00 Γ 00

(3.13)

N r 0 = N r00 Γ r 0 The nonrandom factors, Γ, in these equations are equal to unity in the random case. These numbers must satisfy the following material balance equations:16–18 2 N 00 + N 0 r = zN 0 2 N rr + N 0 r = zqN

(3.14)

By combining these equations, we obtain: θ0 Γ 00 + θr Γ r 0 = 1 θr Γ rr + θ0 Γ r 0 = 1 These two equations along with the quasi-chemical condition:16–18

(3.15)

7248_book.fm Page 49 Tuesday, April 24, 2007 9:19 AM

Statistical Thermodynamic Calculations

49

⎛ 2 * ε* / z ⎞ 4 N rr N 00 Γ rr Γ 00 = = exp ⎜⎝ RT ⎟⎠ = A N r20 Γ r20

(3.16)

form a system of three equations from which one may obtain the factor Γ. The reduced density needed in Equations 3.15, is obtained from the equation of state (cf. Equation 3.22 shown later). The intersegmental interaction energy, ε* = zε/2, in Equation 3.16 is related to the scaling temperature, T*, and scaling pressure, P*, of the fluid by: ε∗ = RT ∗ = P ∗ v ∗

(3.17)

whereas the reduced temperature and pressure are defined as: T=

T P , P= * * T P

(3.18)

Combining Equation 3.15 and Equation 3.16, one can obtain a quadratic equation for Γr0, with the physically meaningful solution: Γr0 =

2

(

)

1 + ⎡⎣1 − 4θ0 θr 1 − A ⎤⎦

1

(3.19) 2

Most general expressions for Ωhb may be found in the original work.13 In the case of a fluid with d proton donors and a proton acceptors forming NH hydrogen bonds, one has:

Ω hb =

⎡⎣ N !⎤⎦

(

2

NH ! ⎡ N − NH ⎣

⎛ ρ −S ⎞ exp H ⎟ 2 ⎜ R ⎠ ⎝ rN !⎤ ⎦

NH

)

(3.20)

where SH is the entropy change upon hydrogen bond formation and the NH is given by: B+d +a− N νH = H = rN

(B + d + a)

2

− 4ad

2r

(3.21)

With these definitions, the equation of state of the fluid is given by: ⎤ ⎡ ⎞ z ⎛ ⎛l q ⎞ z P + T ⎢ ln 1 − ρ − ρ ⎜ − ν H ⎟ − ln ⎜ 1 − ρ + ρ⎟ + ln Γ 00 ⎥ = 0 r ⎠ 2 ⎠ 2 ⎝ ⎝r ⎦ ⎣

(

)

(3.22)

and the chemical potential by: μ dp μ H μ = + RT RT RT

(3.23)

7248_book.fm Page 50 Tuesday, April 24, 2007 9:19 AM

50

Hansen Solubility Parameters: A User’s Handbook

where μ dp RT

= ln

⎡ q P z q ⎤ zq 1 ⎡⎣ ln Γ rr ⎤⎦ − + r − ρl + ln ρ − q ln ⎢1 − ρ + ρ⎥ + ωr r ⎦ 2 2 T T ⎣

(3.24)

is the chemical potential for the dispersion and polar interactions, and μH d a = r ν H − d ln − a ln RT d − νH a − νH

(3.25)

is the chemical potential for the hydrogen bonding interactions. The heat of vaporization is given by: ⎡⎛ ⎞ ⎤ ⎛ ⎞ q q HV = rN ε∗ ⎢⎜ Pv − θr Γ rr ⎟ − ⎜ Pv − θr Γ rr ⎟ ⎥ + E H ⎡⎢ N H ⎣ r r ⎠ liq ⎥ ⎠ vap ⎝ ⎢⎣⎝ ⎦

( )

liq

( )

− NH

vap

⎤ ⎥⎦

(3.26)

Equation 3.15, Equation 3.19, and Equation 3.22 are coupled equations and must be solved simultaneously for the reduced density and the nonrandom factors. In the above formalism, the contributions from dispersion and polar forces are lumped into one contribution. An attempt will be made in the next section to separate them.

THE CONTRIBUTION

FROM

DIPOLAR FORCES

In an initial attempt, the contribution of dipole–dipole interactions was approximated by the multipolar u-expansion of Twu and Gubbins19 by keeping the leading term of the point dipole–point dipole interaction and the Pade approximations,20–21 as well as by using the perturbation model of Nezbeda and Pavlicek.22–24 In an overwhelming majority of cases, this procedure led to underestimations of δp that often fall in the range of one to two orders of magnitude below the expected value. Thus, a drastically different approach was adopted that preserves the simplicity of the formalism of the previous paragraph. In the previous paragraph, as in NRHB16, it was assumed that only first neighbor segment–segment interaction contacts contribute to the potential energy (E) of the system and, thus, we may write for the non-hydrogen-bonding part (dispersion and polar): − E dp = N rr ε dp = Γ rr qN θr ε∗dp

(3.27)

Obviously, in the absence of dipolar interactions we should have εdp = ε and εdp* = ε*, that is, only the contribution from dispersion forces. We may then write quite generally:

(

)

ε∗dp = ε∗ ⎡1 + f m, T , ρ, r,... ⎤ ⎣ ⎦

(3.28)

The crux of the problem is the explicit form of the function f in Equation 3.28. This function should be zero in the absence of dipole–dipole interactions or when the dipole moment, m, of the fluid is zero. As, however, in Equation 3.27 we count segmental interactions, the function f might be approximated by:

7248_book.fm Page 51 Tuesday, April 24, 2007 9:19 AM

Statistical Thermodynamic Calculations

51

b

⎛m⎞ f m, T , ρ, r,... = ⎜ ⎟ g m, T , ρ, r,... ⎝r⎠

(

)

(

)

(3.29)

Hansen5, summarizing years of experience, has observed that δp is directly proportional to m, which implies that b in Equation 3.29 could be set equal to 2. His additional observation that δp is inversely proportional to V1/2 could be reconciled with Equation 3.27 and Equation 3.29 by writing: 2

⎛m⎞ c f m, T , ρ, r,... = ⎜ ⎟ ⎝ r ⎠ Γ rr θr

(

)

(3.30)

c being a constant. The second attempt for the estimation of δp was made by using Equation 3.30 along with Equation 3.27 and Equation 3.28. This approach with the constant c replaced by the expression: c = πs 4

(3.31)

could, indeed, lead to a simultaneously satisfactory estimation of δp and a satisfactory description of the thermodynamic behavior of the fluids. However, it turned out that the following simpler form of Equation 3.30: 2

⎛m⎞ f = ⎜ ⎟ πs 2 ⎝r⎠

(3.32)

led to better results and provided both a satisfactory description of the thermodynamic behavior of fluids over a broad range of external conditions and a satisfactory estimation of δp and the other partial solubility parameters for the overwhelming majority of fluids. One additional reason for adopting Equation 3.32 is that the only change that should be made in the formalism of the previous 2 ⎡ ⎤ ⎛m⎞ paragraph is to replace ε* by ε∗ ⎢1 + π ⎜ ⎟ s 2 ⎥ . ⎝r⎠ ⎢ ⎥ ⎣ ⎦ Thus, the final form of the potential energy that was adopted is:

2 ⎡ ⎛m⎞ 2⎤ ⎢ − E = Γ rr qN θr ε 1 + π ⎜ ⎟ s ⎥ − N H E H ⎝r⎠ ⎢ ⎥ ⎣ ⎦ ∗

(3.33)

On the basis of the above equation, the partial solubility parameters are given by:

δd =

Γ rr qN θr ε∗ V

(3.34)

7248_book.fm Page 52 Tuesday, April 24, 2007 9:19 AM

52

Hansen Solubility Parameters: A User’s Handbook

δp =

⎡ ⎛ m ⎞2 ⎤ Γ rr qN θr ε∗ ⎢ π ⎜ ⎟ s 2 ⎥ ⎢ ⎝r⎠ ⎥ ⎣ ⎦ V

δ hb =

− N H EH V

(3.35)

(3.36)

where, the total volume, V, of the system is given by: V = rNvv ∗ + N H VH

(3.37)

VH being the volume change upon hydrogen bond formation. An alternative expression for the factor f in Equation 3.32, which takes explicitly into consideration its dependence on temperature, is given in Appendix 3.II.

APPLICATIONS In this section, we will apply the model presented in the previous two sections in a multitude of cases. As a first step, we will describe the vapor pressure, the orthobaric densities, and the heat of vaporization of fluids by determining their three scaling constants through a least-squares fit.16,25–26 These constants are reported in Table 3.1a for a number of common fluids. The critical compilation (Design Institute for Physical Property Research [DIPPR])25 was used as a source for the thermodynamic data and the dipole moments of the studied fluids. As in NRHB,16 the geometric constant s of each fluid was obtained through the widely used group contribution calculation scheme of UNIFAC.27–28 Having determined the scaling constants through the previously mentioned procedure, this approach can estimate (essentially predict) the dispersion and the polar components of the solubility parameter over a broad range of temperature and pressure. In the case of hydrogen-bonded fluids, the energy, entropy, and volume change upon hydrogen bond formation are also needed. For simplicity, the volume change, VH, was set equal to zero for all fluids. In addition, for lack of reliable information pertinent to hydrogen bonding over an extended range of temperature and/or pressure, the entropy change, SH, was set equal to –26.5 JK1mol1, as for alkanols.13,16 This is a rather crude approximation but it permits a more direct comparison of the strength of the various types of hydrogen bonds through the mere comparison of the energy change, EH. Essentially, the energy change, EH, which is the only adjustable parameter for the description of hydrogen bonding in our approach, was adjusted through the experimental value5 of δhb. As this is only one datum, it does not suffice to reliably determine SH as well. The parameter EH is also reported in Table 3.1a. In the overwhelming majority of cases, the number of hydrogen bonds in the system is obtained through Equation 3.21. This equation, however, cannot be used in the case of carboxylic acids where the main mode of hydrogen bonding is dimerization. The case of carboxylic acids is treated separately, as shown in Appendix 3.I. In a similar manner, the scaling constants for high polymers were obtained by correlating the available extensive experimental pressure-volume-temperature (PVT) data29 with the equation of state, Equation 3.22, and are reported in Table 3.1b. Table 3.2a and Table 3.2b compare the experimental5 solubility parameters with those estimated/predicted by our approach for a number of common fluids. As observed and in view of the

7248_book.fm Page 53 Tuesday, April 24, 2007 9:19 AM

Statistical Thermodynamic Calculations

53

TABLE 3.1A Characteristic Constants of Pure Fluidsa Fluid

ε* = RT*/J.mol1

ν* = ε*P*–1/cm3 .mol–1

ν*sp = ρ*–1/ cm3 .g–1

m/ Debye

–EH/ J.mol–1

s = q/r s

Propane n-Butane n-Pentane n-Hexane n-Heptane n-Octane n-Nonane n-Decane n-Undecane n-Dodecane 3-Methyl pentane 2,4-Dimethyl hexane 2,2,4-Trimethyl pentane Cyclohexane

3319 3800 4295 4557 4734 4870 4999 5111 5212 5265 4514. 4821 4794 5171

Nonpolar Fluids 9.121 10.748 13.15 13.57 14.00 14.45 14.925 15.317 15.577 15.637 13.750 15.916 17.142 13.040

1.426 1.392 1.373 1.317 1.290 1.283 1.278 1.266 1.259 1.252 1.313 1.301 1.297 1.205

0 0 0 0 0 0 0 0 0 0 0 0 0 0

0 0 0 0 0 0 0 0 0 0 0 0 0 0

0.903 0.881 0.867 0.857 0.850 0.844 0.839 0.836 0.833 0.830 0.856 0.843 0.857 0.800

Benzene Toluene Styrene

4986 5247 5656 5639 5403 5845 5664 3197 5295 3888 4406 3992 4383 4226 3840 4460 1811 4827 4650 3859 5304 2632 3267 3787 3740 4130 4376 4683 4977 5294 6199 4824

0.370 0.360 0.130 0.330c

2720 3585 6421 6444 5100 5080 5120 8250 6274 11133 8705 6390 7384 6780 7547 6570 5860 8565 5116 4412 4275 25100 25100 24771b 25100 25100 25100 25100 25100 23000 22940 21130

0.753 0.757 0.760 0.760 0.759 0.720 0.720 0.908 0.769 0.896 0.869 0.894 0.903 0.818 0.888 1.030 0.909 0.840 0.881 0.800 0.746 0.941 0.903 0.881 0.881 0.867 0.857 0.850 0.839 0.833 0.757 0.806

o-Xylene Tetralin Acetone Acetophenone Ethyl acetate n-Butyl acetate Vinyl acetate Methyl methacrylate -Caprolactone Diethyl ether 1,4-Dioxane Carbon dioxide Chloroform Dichloromethane Vinyl chloride Chlorobenzene Methanol Ethanol 1-Propanol 1-Butanol 1-Pentanol 1-Hexanol 1-Octanol 1-Decanol Phenol Ethylene glycol

Polar/Hydrogen-Bonded Fluids 9.551 1.079 10.922 1.098 11.971 1.075 11.960 1.075 11.445 1.087 11.016 1.004 10.431 0.992 9.113 1.134 10.350 0.952 13.264 1.018 13.050 1.049 10.548 0.961 12.823 0.978 14.757 0.941 11.575 1.182 10.830 0.870 7.050 0.739 10.423 0.619 9.786 0.736 9.880 0.960 10.418 0.873 10.737 1.140 10.820 1.126 11.400 1.124 11.560 1.122 12.140 1.131 12.700 1.131 14.652 1.156 14.25 1.141 14.698 1.146 12.220 0.928 11.635 0.904

0.630 0.219 1.050c 2.887 3.028 1.780 1.841 1.790 1.670 4.437 1.151 0.400 2.320c 1.010 1.439 1.451 1.690 1.700 1.690 1.680 1.680 1.660 1.650 1.650 1.650 1.619 1.451 2.308

7248_book.fm Page 54 Tuesday, April 24, 2007 9:19 AM

54

Hansen Solubility Parameters: A User’s Handbook

TABLE 3.1A (CONTINUED) Characteristic Constants of Pure Fluidsa Fluid 1,2-Propylene glycol Glycerol Diethylamine n-Butylamine Tetrahydrofuran Formamide Dimethylformamide Acrylonitrile Dimethylsulfoxide Acetic acid Acrylic acid Propionic acid Butyric acid Methacrylic acid Octanoic acid Oleic acid Stearic acid Ammonia Water

a b c

ε* = RT*/J.mol1

ν* = ε*P*–1/cm3 .mol–1

2325 2751 4377 4417 4025 4193 4610 3634 3065 4175 4703 4840 4676 4787 5028 5525 5166 6400 5610 1234 2676 4222

13.650 16.61 12.411 12.430 12.474 11.432 5.959 12.146 8.110 9.160 6.75 8.95 7.963 9.088 9.752 10.052 10.014 19.413 12.322 6.000 8.709 10.527

ν*sp = ρ*–1/ cm3 .g–1

m/ Debye

–EH/ J.mol–1

s = q/r s

0.909 0.798 1.265 1.229 1.036 1.025 0.899 0.762 1.175 0.908 0.896 0.903 0.939 0.963 0.935 0.938 1.036 1.108 1.041 1.341 0.992 0.991

3.627 4.197 0.920 1.391 1.631 1.360c 3.717 3.807 3.867 3.957 1.739 1.460 1.751 1.649 1.649 0.650c

22450 22520 12268 12325 8700 8770 16238 16120 9152 11038 20882 31600 24700 28880 21638 21553 26369 21090 19200 12277 17493 18100

0.866 0.767 0.861 0.874 0.925 0.925 0.869 0.855 0.887 0.855 0.910 0.876 0.902 0.888 0.922 0.922 0.850 0.823 0.824 1.039 0.861 0.861

1.700 1.739 1.670 2.750 1.850 0.970c

SH was set equal to –26.5 JK–1mol–1 in all cases. Adjusted to fit δhb. Adjusted to fit δp.

TABLE 3.1B Characteristic Constants of Pure Fluids/Polymersa Fluid (Polymer)

ε* = RT*/J.mol1

ν* = ε*P*–1/cm3 mol1

ν*sp = ρ*–1/cm3 .g–1

Polyethylene-lin. Polypropylene Polystyrene Poly(vinyl chloride) Polyacrylonitrile Poly(methyl methacrylate) Polycarbonate (bisphenol A) Poly(ε-caprolactone) Poly(vinyl acetate) Nylon 66

5401 5993 6335 4876 5862 5294 5840 4817 5607 5029

13.169 13.348 11.692 8.472 9.614 11.569 11.915 10.976 13.121 9.169

1.130–0.000039P 1.147–0.000164P 0.710–0.000087P 0.676–0.000017P 0.839–0.000049P 0.821–0.000083P 0.806–0.000020P 0.870–0.000016P 0.810–0.000070P 0.898–0.000037P

a b

m/ Debye 0.080 0 1.298 1.451 1.810 2.600 4.300 1.750 0.450 10.00

Pressure, P, in column 4 is in MPa. Numbers in parentheses are proton-donors and proton-acceptors, respectively, per repeat unit.

–EH/ J.mol–1

s = q/r s

1350(4,2)b 400(6,2) 3600(8,1) 4400(3,1) 8700(3,1) 4800(8,2) 9550(6,3) 7800(10,2) 3700(6,2) 68000(2,2)

0.800 0.799 0.667 0.780 0.887 0.843 0.728 0.818 0.825 0.783

7248_book.fm Page 55 Tuesday, April 24, 2007 9:19 AM

Statistical Thermodynamic Calculations

55

TABLE 3.2A Total and Partial Solubility Parameters (in MPa1/2) of Pure Fluids δ Total

δ HB

5

5

Fluid

Exp

Propane n-Butane n-Pentane n-Hexane n-Heptane n-Octane n-Nonane n-Decane n-Undecane n-Dodecane 3-Methyl pentane 2,4-Dimethyl hexane 2,2,4-Trimethyl pentane Cyclohexane

13.10 14.10 14.40 14.90 15.20 15.40 15.60 15.70 15.80 15.90 14.67 14.65 14.08 16.76

Benzene Toluene Styrene o-Xylene Tetralin Acetone Acetophenone Ethyl acetate n-Butyl acetate Vinyl acetate Methyl methacrylate ε-Caprolactone Diethyl ether 1,4-Dioxane Carbon dioxide Chloroform Dichloromethane Vinyl chloride Chlorobenzene Methanol Ethanol 1-Propanol 1-Butanol 1-Pentanol 1-Octanol 1-Decanol Phenol Ethylene glycol 1,2-Propylene glycol

Calc

Exp

Nonpolar Fluids 12.85 0 14.10 0 14.34 0 14.58 0 15.12 0 15.21 0 15.25 0 15.30 0 15.40 0 15.52 0 14.65 0 14.39 0 13.90 0 16.30 0

δP Exp

0 0 0 0 0 0 0 0 0 0 0 0 0 0

Polar/Hydrogen-Bonded Fluids 18.41 18.27 2.05 2.05 18.32 17.78 2.00 2.00 19.07 18.29 4.10 4.04 19.07 18.30 4.10 4.05 18.20 18.01 3.10 3.10 19.80 18.80 2.90 2.90 19.80 19.07 2.90 2.90 19.95 20.04 6.95 7.00 21.73 20.48 3.68 3.68 18.48 18.31 9.20 9.33 17.59 17.43 6.30 6.30 18.58 18.33 5.90 5.90 17.92 17.77 5.40 5.40 21.41 21.28 7.40 7.40 15.66 15.73 5.11 5.12 20.47 20.08 7.36 7.36 14.56 11.64 4.10 4.09 18.94 19.18 5.73 5.74 20.79 19.92 4.09 4.09 17.77 16.00 2.40 2.40 19.61 18.91 2.05 2.05 29.61 29.89 22.30 24.08 26.13 26.08 19.43 19.98 24.45 24.19 17.40 17.41 24.45 24.17 17.40 17.58 23.35 22.90 15.80 15.80 21.65 21.95 13.91 14.52 20.87 20.27 11.86 11.94 20.32 19.37 10.00 10.03 24.63 24.69 14.90 14.95 33.70 33.64 25.77 25.74 29.52 29.19 23.32 23.73

5

Calc

0 0 0 0 0 0 0 0 0 0 0 0 0 0

0 0 0 0 0 0 0 0 0 0 0 0 0 0

1.02 1.40 1.00 1.00 1.00 2.00 2.00 10.43 8.59 5.85 3.70 7.20 6.50 15.0 2.86 1.84 6.90 3.07 7.36 6.50 4.30 12.27 8.80 6.80 6.80 5.70 4.50 3.27 2.60 5.90 11.05 9.41

1.01 0.92 0.33 1.00 1.48 0.43 2.00 10.14 7.04 6.07 4.70 5.90 5.60 13.22 3.48 1.91 6.86 3.79 5.63 5.07 4.33 11.34 8.24 6.79 6.79 5.72 4.99 3.75 3.15 5.16 12.23 12.56

7248_book.fm Page 56 Tuesday, April 24, 2007 9:19 AM

56

Hansen Solubility Parameters: A User’s Handbook

TABLE 3.2A (CONTINUED) Total and Partial Solubility Parameters (in MPa1/2) of Pure Fluids Total Fluid Glycerol Diethylamine n-Butylamine Tetrahydrofuran Formamide Dimethylformamide Acrylonitrile Dimethylsulfoxide Acetic acid Acrylic acid Propionic acid Butyric acid Methacrylic acid Octanoic acid Oleic acid Stearic acid Ammonia Water

HB

5

Calc

Exp

34.12 16.61 18.28 19.46 19.46 36.65 23.95 21.59 26.75 21.35 24.01 19.95 20.2 21.00 21.00 22.24 17.38 19.04 24.63 47.82

34.34 16.80 18.48 18.93 19.26 38.18 23.62 22.29 26.20 27.58 25.90 25.44 23.96 23.85 23.67 21.61 16.81 18.87 26.52 48.68

29.25 6.10 8.00 8.00 8.00 19.00 11.25 6.80 10.20 13.52 14.90 12.40 10.60 10.20 10.20 8.20 5.50 5.50 17.80 42.32

Exp

P

5

Exp 29.18 6.13 8.08 8.00 8.00 18.80 11.16 6.80 10.28 12.09 14.90 12.19 12.21 10.25 10.22 8.66 4.95 4.30 17.80 42.17

5

Calc

12.07 2.30 4.50 5.70 5.70 26.20 13.70 12.80 16.40 7.98 6.40 5.30 4.10 2.80 2.80 3.30 3.10 3.30 15.70 16.00

14.31 2.89 4.75 6.98 5.70 21.41 13.99 13.16 14.75 8.22 6.31 6.81 5.51 6.87 2.80 3.35 2.49 1.86 15.70 16.00

TABLE 3.2B Total and Partial Solubility Parameters (in MPa1/2) of Common Polymers Total

HB

Fluid

Exp

5

5

Calc

Exp

Polyethylene-lin. Polypropylene Polystyrene Poly(vinyl chloride) Polyacrylonitrile Poly(methyl methacrylate) Polycarbonate (bisphenol A) Poly(ε-caprolactone) Poly(vinyl acetate) Nylon 66

16.26 18.10 19.26 19.55 27.43 21.52 20.25 20.20 18.18 30.87

17.37 18.15 19.26 21.93 28.37 22.35 20.43 20.33 18.42 33.09

2.80 1.00 2.90 3.40 9.10 5.10 6.90 8.40 4.00 24.00

P 5

Calc

Exp

2.80 1.00 2.90 3.42 9.10 5.10 6.90 8.40 4.00 23.90

0.80 0.0 4.50 7.80 14.10 10.50 5.90 5.00 2.20 11.00

Calc 0.80 0.0 4.50 8.08 14.08 10.44 5.90 5.00 2.20 11.00

approximations made, the overall picture is rather satisfactory. We are not aware of any similar predictive approach in the literature in order to make the analogous comparison. There are a number of comments that should be made regarding Table 3.1a and Table 3.1b. First, the scaling constants for the nonpolar substances are identical to those reported previously.16 Thus, their calculated solubility parameters, reported in Table 3.2a, are essentially predictions of

7248_book.fm Page 57 Tuesday, April 24, 2007 9:19 AM

Statistical Thermodynamic Calculations

57

the model. Second, the series of 1-alkanols is another case where the calculated solubility parameters are, essentially, predictions of the model, as the hydrogen bonding parameters are the same as in the original NRHB model.16 Of course, the scaling constants have been changed as the interaction energy is now split into its dispersion and polar components, but the new parameter, the dipole moment, m, is not an adjustable parameter. Third, in all other cases of polar substances there is always an experimental hydrogen bonding contribution5 even in cases where there is no obvious proton-donor and proton-acceptor pair. There are cases where even the dipoles themselves are not obvious and, apparently, the polar component δp refers to quadrupole or higher-order interactions. A typical example is the case of carbon dioxide, where neither protons nor dipoles are present. In such cases, the donor–acceptor interaction was replaced by the acid–base or the electrophilic–nucleophilic (carbon–oxygen) interaction, and a fictitious value of m was adjusted on the basis of the corresponding experimental5 value for δp. In a similar manner, in aromatic hydrocarbons, all hydrogens were considered as equivalent proton donors and (the π-electrons of) the aromatic ring as the proton acceptor. Fourth, in the case of polymers, the picture is somewhat more complex. The repeating unit was considered the basis for the calculations, and the reported numbers in Table 3.1b of proton donors and acceptors refer to this basis. Thus, the total number of proton donors and acceptors (for the polymer) are the reported numbers in Table 3.1b multiplied by the degree of polymerization of each polymer. In this case, the dipole moment of the polymer was again adjusted on the basis of the experimental5 δp. Fifth, for some fluids there are two entries in Tables 3.1a and 3.1b. In these cases, the estimated δp on the basis of the literature value25 for m was largely deviating from the experimental one, and thus, in the second entry the value of m was adjusted on the basis of the experimental5 δp. Figure 3.1 shows the calculated components of the solubility parameter of water over an extended range of saturation temperatures. As was expected, the main contribution to δ of water, especially at low temperatures, is hydrogen bonding. This type of diagram is most useful for designing applications involving subcritical or supercritical water. An analogous diagram for ammonia is shown in Figure 3.III.1 of Appendix 3.III. In this case, the contribution of the polar component is as important as that of the hydrogen bonding component, and it appears to override the hydrogen bonding component at the supercritical region. In the same figure, one may compare the temperature dependence of these components as estimated by the alternative approach of Appendix 3.III. 50 δ 40

δ / MPa½

δhb 30

20

δd δp

10

0 250

300

350

400

450 T/K

FIGURE 3.1 Fractional solubility parameters for water.

500

550

600

650

7248_book.fm Page 58 Tuesday, April 24, 2007 9:19 AM

58

Hansen Solubility Parameters: A User’s Handbook

17.6

δd, alkanol

17.2

16.8

16.4

16.0

15.6 13.0

13.5

14.0

14.5

15.5 δalkane

15.5

16.0

FIGURE 3.2A Comparison of the experimental dispersion component of solubility parameters of 1-alkanols5 with the solubility parameters of the corresponding n-alkane homomorphs.

16.4

δd, alkanol

16.0

15.6

15.2

14.8

14.4 12.5

13.0

13.5

14.0

14.5

15.0

15.5

δalkane FIGURE 3.2B Comparison of the calculated dispersion component of solubility parameters of 1-alkanols with the solubility parameters of the corresponding n-alkane homomorphs. The correlation line is: δd,alkanol = 6.86 + 0.60 δalkane .

It is often proposed, as in our previous approach,11–12 to estimate the dispersion component, δd, in polar fluids from the total solubility parameter of the corresponding homomorph hydrocarbon. This approach, however, is not always successful as shown in Figure 3.2A, where the dispersion component, δd, of alkanols is compared with the total solubility parameter of the corresponding homomorph hydrocarbon. As shown, the experimental5 data not only fall away from the diagonal; they do not even fall on a straight line. As a consequence, the route for the estimation of the solubility parameter components through the homomorph concept is not always a safe way. In

7248_book.fm Page 59 Tuesday, April 24, 2007 9:19 AM

Statistical Thermodynamic Calculations

59

contrast, when we compare these two solubility parameters as calculated by the present approach, the data seem to fall at least on a straight line, as shown in Figure 3.2B. However, care must be exercised once again, as in this figure, the slope of the straight line is much lower than unity. The splitting of the potential energy into its dispersion, polar, and hydrogen bonding components in Equation 3.33, which led to the explicit forms of Equation 3.34 through Equation 3.36, is most useful for an additional reason: If the partial solubility parameters and the molar volume are known (e.g., compilations by Hansen,5 Barton,3 van Krevelen,30 etc.), and the normal boiling point (or the vapor pressure at some other temperature) are either known or can be estimated with reasonable accuracy, then one may use this information and the above formalism to estimate the scaling constants and the hydrogen bonding energy of the fluid through a least squares fit. Of course, the dipole moment is also needed and, if not known, it may be estimated by various rather widely available ab initio or semiempirical quantum mechanics calculations. We have applied the above procedure to acetophenone by using Hansen’s compilation5 for the partial solubility parameters and molar volume, and DIPPR25 for the normal boiling point and the dipole moment. The obtained scaling constants are: ε* = 5267 J/mol, v* = 10.177 cm3/mol, vsp* = 0.9495 cm3/g, and EH = 6293 J/mol, rather close to the corresponding parameters reported in Table 3.1A. These scaling constants can now be used for the estimation of the basic thermodynamic properties of the fluid at any temperature and pressure. The experimental25 and the calculated (essentially, predicted) vapor pressures and saturated liquid densities for acetophenone are compared in Figure 3.3A and Figure 3.3B, respectively. As observed, this procedure leads to a reasonably accurate estimation of the thermodynamic properties of fluids. A most useful concept that quantifies the similarity of two substances 1 and 2, especially the similarity of a polymer, 2, and a potential solvent, 1, for it, is the solubility parameter distance, Ra, defined by:5

(

Ra = ⎡ 4 δ d 2 − δ d 1 ⎣⎢

) + (δ 2

p2

− δ p1

) + (δ 2

hb 2

)

2 − δ hb1 ⎤ ⎥⎦

(3.38)

The idea is: the smaller the Ra, the better is the solvent for the polymer. A sphere with radius Ro encompasses the good solvents for this polymer. A refined discussion on Ra and the related quantities Ro and RED = Ra/Ro is provided by Hansen.5,31 The partial solubility parameters for (Bisphenol A) Polycarbonate as functions of temperature, as calculated using the scaling constants in Table 3.1B, are shown in Figure 3.4. It can be seen that all three components are nonnegligible and there is a cross-over in the polar and hydrogen bonding components for this polymer. The distances (Ra) of this polymer with three common solvents are compared in Figure 3.5. According to the calculations, chloroform is the best of the three solvents for this polymer, followed by tetrahydrofuran (THF). Heptane is the worst and, essentially, a nonsolvent for the polymer, and all these findings agree with experiment. The distances (Ra) for polypropylene with three solvents: tetrahydrofuran, chloroform, and tetralin are similarly compared in Figure 3.6. As shown, the best solvent for polypropylene appears to be tetralin, which is again corroborated by the experiment. This type of figure is most useful not only for the mere selection of the solvent, but also for the selection of the external conditions (especially, temperature) for the dissolution of the polymer or any other solute.

DISCUSSION AND CONCLUSIONS A new approach has been presented for the estimation of the partial solubility parameters of pure substances. The capacity of the approach appears satisfactory for both the estimation of the partial solubility parameters and the description of the thermodynamic behavior of fluids over a broad

7248_book.fm Page 60 Tuesday, April 24, 2007 9:19 AM

60

Hansen Solubility Parameters: A User’s Handbook

2.5

2.0

P / MPa

1.5

1.0

0.5

0.0 250

300

350

400

450

500 T/K

550

600

650

700

FIGURE 3.3A Experimental (symbols)25 and predicted (line) vapor pressures for acetophenone.

1.1

ρ / g.cm3

1.0

0.9

0.8

0.7

0.6 250

300

350

400

450

500 T/K

550

600

650

700

FIGURE 3.3B Experimental (symbols)25 and predicted (line) liquid densities for acetophenone.

range of temperature and pressure. The author and his coworkers are not aware of any similar integral approach in the literature in order to make comparison. The equation-of-state approach for the estimation of the partial solubility parameters, which is presented in this work, has a number of features. First, it is in principle, applicable to any fluid regardless of its size and shape. Second, it permits the estimation of the partial solubility parameters over an extended range of temperature and pressure. Third, it may utilize available information on

7248_book.fm Page 61 Tuesday, April 24, 2007 9:19 AM

Statistical Thermodynamic Calculations

61

20

16

d

δ / MPa½

12

8 p

4

hb

280

320

360

400 T/K

440

480

330

340

FIGURE 3.4 Partial solubility parameters for (bisphenol A) polycarbonate.

10

n-Heptane

Ra / MPa½

8

6

4

2

0 290

THF¹

THF²

Chloroform

300

310

320 T/K

FIGURE 3.5 The estimated solubility parameter distance, Ra, of (bisphenol A) polycarbonate (MW = 100000) with three common solvents, as a function of temperature. The lines marked with THF1 and THF2 were obtained by using the first and second set of scaling constants in Table 3.1A, respectively.

7248_book.fm Page 62 Tuesday, April 24, 2007 9:19 AM

62

Hansen Solubility Parameters: A User’s Handbook

10

THF

Ra / MPa½

8

Chloroform 6

4

Tetralin

2 290

300

310

320

330

340

T/K FIGURE 3.6 The estimated solubility parameter distance, Ra, of (linear) Polypropylene with three common solvents, as a function of temperature. The lines for tetralin and tetrahydrofuran (THF) were obtained by using the second set of scaling constants in Table 3.1A.

the partial solubility parameters for the estimation of the scaling constants of substances for which there are no available extensive experimental data on vapor pressures, heats of vaporization, orthobaric densities, etc. Fourth, it may act as useful guide for the selection of appropriate solvents and/or dissolution conditions. Of course, the statistical thermodynamic model, on which the above approach resides, can be used for a detailed description of the phase diagrams of pairs (mixtures) of fluids when the scaling constants and the binary interaction parameters are available. However, when a fast screening of potential solvents is needed, the approach of this work is sufficient and most valuable. A novel element in our approach is the way the potential energy is split into its dispersion, polar, and hydrogen bonding components. The calculation of the polar component, in particular, is rather oversimplified and there is much room for improvement if one wishes to use more involved expressions for the function f in Equation 3.30. The alternative approach in Appendix 3.II is one example. Significant progress could be made if experimental information on the partial solubility parameters as functions of temperature and pressure were available. One further possibility is the use of a group contribution method for the estimation of the dispersion component in much the same way as suggested in,14 as shown in Appendix 3.III. As shown, δd can be estimated with an average absolute error of 0.40, less than half the corresponding error for the estimation of total with the same method.14 Such a group contribution method fails dramatically, however, for the hydrogen bonding component, which enhances further the usefulness of the approach reported in the present work. Once the estimations of total δ, δhb, and δd are available, the polar component is obtained by a simple subtraction.

ACKNOWLEDGMENTS The contribution of E. Stefanis and I. Tsivintzelis in the preparation of tables and figures of this chapter is gratefully acknowledged. The contribution of C. M. Hansen through his valuable comments is also gratefully acknowledged.

7248_book.fm Page 63 Tuesday, April 24, 2007 9:19 AM

Statistical Thermodynamic Calculations

LIST OF SYMBOL SPECIAL TO THIS CHAPTER E G H K L N N Nr No Nij P R R m S s T Q(N, P, T) V* V X Y z zq

Potential energy Gibbs free energy Enthalpy Boltzmann’s constant Staverman’s parameter Number of experimental points Number of molecules Total number of lattice sites Number of empty lattice sites Number of contacts of type i-j Pressure Gas constant Number of segments per molecule Dipole moment Entropy Surface to volume fraction Temperature Configurational partition function of fluid in the N, P, T ensemble, see Equation 3.2, p. 46 Average segmental volume Volume ole fraction in liquid phase ole fraction in vapor phase Lattice coordination number Average number of external contacts per molecule

GREEK LETTERS ν* Γ Δ E Θ Θ M P φ Ω Ω

Segmental volume Non random factor Solubility parameter Interaction energy Surface (contact) fraction Hole-free surface (contact) fraction Chemical potential Density Segment fraction Geometric–flexibility parameter Combinatorial term (In Equation 3.7 and Equation 3.20)

63

7248_book.fm Page 64 Tuesday, April 24, 2007 9:19 AM

64

Hansen Solubility Parameters: A User’s Handbook

SUPERSCRIPT ~ * L V

Reduced quantity Scaling constant Liquid phase Vapor phase

SUBSCRIPT d, hb, p Dm H O R

Dispersion, hydrogen bonding, and polar component, respectively Quantity pertinent to dimer Hydrogen bonding quantity Property pertinent to holes Property pertinent to molecular segments

REFERENCES 1. Hildebrand, J. and Scott, R.L., Regular Solutions, Prentice-Hall, Englewood Cliffs, NJ, 1962. 2. Hansen, C.M., The three dimensional solubility parameter — key to paint component affinities I., J. Paint Technol., 39(505), 104–117, 1967. 3. Barton, A.F.M., Handbook of Solubility Parameters and Other Cohesion Parameters, CRC Press, Boca Raton, FL, 1983. 4. Barton, A.F.M., Applications of solubility parameters and other cohesion parameters, Polym. Sci. Technol. Pure Appl. Chem., 57(7), 905–912, 1985. 5. Hansen, C.M., Hansen Solubility Parameters: A User’s Handbook, CRC Press, Boca Raton, FL, 1999. 6. Hansen, C.M., Aspects of solubility, surfaces, and diffusion in polymers, Prog. Org. Coat., 51(1), 55–66, 2004. 7. Tehrani, J., Am. Lab., 40hh-40mm, February 1993. 8. Bustamante, P., Peña, M.A., and Barra, J., Int. J. Pharm., 174, 141–150, 1998. 9. Jensen, W.B., in Surface and Colloid Science in Computer Technology, Mittal, K.L., Ed., Plenum Press, New York, 1987, pp. 27–59. 10. Karger, B.L., Snyder, L.R., and Eon, C., J. Chromatogr., 125, 71–88, 1976. 11. Panayiotou, C., Fluid Phase Equilibria, 131, 21–35, 1997. 12. Panayiotou, C., Fluid Phase Equilibria, 236, 267, 2005. 13. Panayiotou, C. and Sanchez, I.C., J. Phys. Chem., 95, 10090–10097, 1991. 14. Stefanis, E., Constantinou, L., and Panayiotou, C., Ind. Eng. Chem. Res., 43, 6253–6360, 2004. 15. Stefanis, E., Tsivintzelis, I., and Panayiotou, C., Fluid Phase Equilibria, 240, 144–154, 2006. 16. Panayiotou, C., Pantoula, M., Stefanis, E., Tsivintzelis, I., and Economou, I., Ind. Eng. Chem. Res., 43, 6592–6606, 2004. 17. Guggenheim, E.A., Mixtures, Clarendon Press, Oxford, 1952. 18. Panayiotou, C. and Vera, J.H., Polym. J., 14, 681–694, 1982. 19. Twu, C.H. and Gubbins, K.E., Chem. Eng. Sci., 33, 863–878, 1978. 20. Kraska, T. and Gubbins, K.E., Ind. Eng. Chem. Res., 35, 4727–4737, 1996. 21. Stell, G., Rasaiah, J.C., and Narang, H., Mol. Phys., 27, 1393–1414, 1974. 22. Nezbeda, I. and Pavlíek, J., Fluid Phase Equilibria, 116, 530–536, 1996. 23. Nezbeda, I. and Weingerl, U., Mol. Phys., 99, 1595–1606, 2001. 24. Karakatsani, E., Spyriouni, T., and Economou, I., AIChE J., 51(2005), 2328–2342. 25. Daubert, T.E. and Danner, R.P., Eds., Data Compilation Tables of Properties of Pure Compounds, AIChE Symp. Ser. No. 203, American Institute of Chemical Engineers, New York, 1985. 26. Perry, R. and Green, D., Ed., Chemical Engineers’ Handbook, CD, McGraw Hill, New York, 1999. 27. Fredenslund, A., Jones, R.L., and Prausnitz, J.M., AIChE J., 21, 1086–1099, 1975.

7248_book.fm Page 65 Tuesday, April 24, 2007 9:19 AM

Statistical Thermodynamic Calculations

65

28. Fredenslund, A., in Models for Thermodynamic and Phase Equilibria Calculations, Sandler, S., Ed., Marcel Dekker, New York, 1994. 29. Zoller, P. and Walsh, D., PVT Data for Polymers, Technomic Publ. Co., Lancaster, Basel, 1995. 30. van Krevelen, D.W., Properties of Polymers, Elsevier, Amsterdam, 2nd ed., 1976. 31. Hansen, C.M., Fifty years with solubility parameters — past and future, Prog. Org. Coat., 51(1), 77–84, 2004.

APPENDIX 3.I: THE ACID DIMERIZATION Following the LFHB practice,13 one may derive the formalism for the acid dimerization in a rather straightforward manner. For simplicity, we will consider dimerization only, as dimers are the overwhelming majority of the association species in hydrogen-bonded acids. Let Ndm be the number of dimers in the system. One can select these dimerized molecules out of the N acid molecules in, N! ! N − 2Ndm !

(2 N ) ( dm

(3.I.1)

)

ways, and form the Ndm dimers in

(2 N ) ( dm

(

)(

)

N! N! 2 N dm − 1 2 N dm − 3 ...1 = ! N − 2 N dm ! N − 2 N dm ! N dm ! 2 N dm

)

(

)

(3.I.2)

ways. If Gdm = E dm + PVdm − TS dm

(3.I.3)

is the free energy change upon formation of one dimer, the hydrogen bonding factor in the partition function becomes: QH =

(

⎛ ρ ⎞ N! N dm ⎜ ⎝ rN ⎟⎠ N − 2 N dm ! N dm ! 2

N dm

)

⎛ N G ⎞ exp ⎜ − dm dm ⎟ RT ⎠ ⎝

(3.I.4)

The equilibrium number of dimers per mol of segments of acid, νdm, is obtained from the above equation through the usual free energy minimization condition, or 2+ ν dm =

1 − K dm

1 4 + 2 K dm K dm 4r

(3.I.5)

where, K dm =

⎛ −Gdm ⎞ ρ exp ⎜ r ⎝ RT ⎟⎠

(3.I.6)

In this case of dimerization, the hydrogen bonding contribution to the chemical potential is: μH 1 = r ν dm − ln RT 1 − 2 r ν dm

(3.I.7)

7248_book.fm Page 66 Tuesday, April 24, 2007 9:19 AM

66

Hansen Solubility Parameters: A User’s Handbook

APPENDIX 3.II: AN ALTERNATIVE FORM OF THE POLAR TERM The factor f in Equation 3.32 for the dipolar interactions implies that the dipolar forces depend on temperature and/or volume in the same manner as do the dispersion forces. This is a rather gross simplification as one would expect the dipolar forces to be inversely proportional to temperature or to a function of temperature.19–24 This has been explored by adopting the following alternative form for the factor f: 2

f =

2 ⎛ m ⎞ s2 π ⎜⎝ r ⎟⎠ T

(3.II.1)

With this expression, the formalism of the main text remains the same except for the equation for the potential energy, which now becomes: 2 ⎡ 4 ⎛ m ⎞ s2 ⎤ ⎥ − N H EH − E = Γ rr qN θr ε∗ ⎢1 + ⎜ ⎟ ⎢ π⎝ r ⎠ T ⎥ ⎣ ⎦

(3.II.2)

This change will change, of course, the scaling constants of the fluids. These new scaling constants are reported in Table 3.III.1 for some representative fluids. These constants describe the key thermodynamic properties of the fluids in a similar, almost identical manner to the one obtained by the corresponding scaling constants of the main text. In addition, the predicted partial solubility parameters by the two sets of the scaling constants are compared in Table 3.II.2. As observed, Equation 3.II.1 and Equation 3.II.2 do not lead to any clear improvement in this respect either. The essential difference is the dependence of δp on temperature, which is now given by:

δp =

⎡ 4 ⎛ m ⎞ 2 s2 ⎤ ⎥ Γ rr qN θr ε∗ ⎢ ⎜ ⎟ ⎢π ⎝ r ⎠ T ⎥ ⎣ ⎦ V

(3.II.3)

As shown in Figure 3.II.1, the two alternative approaches for the estimation of the polar component lead to differences not only in δp but also in δd and to the total . The hydrogen bonding component appears, however, intact. This is important, as δhb may be used in approaches like the one reported in Appendix 3.III.

APPENDIX 3.III: A GROUP-CONTRIBUTION METHOD FOR THE PREDICTION OF δ AND δD The details of the group contribution method may be found in the original work.14 Two kinds of functional groups are used: First-order (UNIFAC groups) and second-order groups that are based on the conjugation theory. The basic equation that gives the value of each property according to the molecular structure is: f(p) =

∑n F i

i

i

+

∑m j

j

Sj

(3.III.1)

7248_book.fm Page 67 Tuesday, April 24, 2007 9:19 AM

Statistical Thermodynamic Calculations

67

TABLE 3.II.1 Characteristic Constants of Pure Fluidsa Fluid

ε* = RT*/J.mol–1

Benzene Toluene Tetralin Acetone Acetophenone Ethyl acetate n-Butyl acetate Methyl methacrylate Diethyl ether 1,4-Dioxane Chloroform Dichloromethane Chlorobenzene Methanol Ethanol 1-Butanol 1-Octanol Phenol Ethylene glycol 1,2-Propylene glycol Glycerol Diethylamine n-Butylamine Tetrahydrofuran Acetic acid Butyric acid Ammonia Water

5041 5236 5858 4207 5875 4444 4741 5046 4040 4435 4859 5062 5517 3798 4042 4656 5280 6868 5681 4386 2595 4471 4927 4491 5110 4981 2506 2070

a b

ν* = ε*P*–1/cm3 .mol–1

ν*sp = ρ*1/cm3.g–1

9.526 10.684 11.034 10.709 10.899 14.170 13.484 17.060 12.293 11.216 10.026 10.704 10.420 13.710 14.040 14.031 15.250 14.805 20.048 12.800 30.000 12.232 14.685 11.261 7.019 7.776 7.500 23.862

1.078 1.093 1.000 1.171 0.958 1.033 1.054 1.019 1.192 0.865 0.617 0.747 0.873 1.178 1.155 1.145 1.145 0.940 0.961 0.942 0.798 1.262 1.257 1.029 0.902 0.962 1.399 0.997

–EH/J.mol–1 3724 3590 5055 7970 6192 10915 8674 5383 6630 6501 8610 4992 4269 25100 25100 25100 25100 23300 21775 22400 24600 12240 11980 8810 23735 23822 11940 18198

SH was set equal to 26.5 JK–1mol–1 in all cases. Adjusted to fit δhb.

where Fi is the contribution of the first-order group of type i that appears ni times in the compound and, Sj, is the contribution of the second-order group of type j that appears mj times in the compound. f(p) is a single equation of the property, p, and is selected after a thorough study of the physicochemical and thermodynamic behavior of the property. The determination of the contributions is done by a two-step regression analysis for the Fis and the Sjs, respectively. A least-square analysis has been carried out to estimate the first-order and second-order group contributions for the solubility parameters. In Table III.1, the first-order group contributions for total solubility parameter, δ, and the dispersion partial solubility parameter (Hansen), δd, at 25°C are presented. Table III.2 shows the second-order group contributions for the same properties. The selected equations for the estimation of each property are the following: Total solubility parameter, δ, at 25°C ((kJ/m3)(1/2)): δ=(

∑n F i

i

i

+

∑m j

j

S j + 75954.1)

0.383837

− 56.14

(3.III.2)

7248_book.fm Page 68 Tuesday, April 24, 2007 9:19 AM

68

Hansen Solubility Parameters: A User’s Handbook

TABLE 3.II.2 Total and Partial Solubility Parameters (in MPa1/2) of Pure Fluids Polar/Hydrogen-Bonded Fluids Total

HB

P

Fluid

Expa

Calc

Expa

Calc

Expa

Calc

Benzene Toluene Tetralin Acetone Acetophenone Ethyl acetate n-Butyl acetate Diethyl ether 1,4-Dioxane Chloroform Dichloromethane Methanol Ethanol 1-Butanol 1-Octanol Phenol Ethylene glycol 1,2-Propylene glycol Glycerol Diethylamine n-Butylamine Tetrahydrofuran Acetic acid Butyric acid Ammonia Water

18.41 18.32 19.80 19.95 21.73 18.48 17.59 15.66 20.47 18.94 20.79 29.61 26.50 23.35 20.87 24.63 33.70 29.52 34.12 16.61 18.31 19.46 21.35 20.20 27.40 47.82

18.41(18.27)a 17.95(17.78) 18.85(18.80) 21.29(20.04) 21.20(20.48) 18.66(18.31) 17.72(17.43) 15.71(15.73) 19.67(20.08) 19.46(19.18) 20.02(19.92) 30.86(29.89) 26.26(26.08) 22.92(22.90) 20.30(20.27) 24.59(24.69) 33.97(33.64) 31.90(29.19) 34.14(34.34) 16.97(16.80) 18.32(18.48) 20.05(18.93) 27.70(27.58) 24.74(23.96) 28.73(26.52) 47.80(48.68)

2.05 2.00 2.90 6.95 3.68 9.20 6.30 5.11 7.36 5.73 4.09 22.30 19.43 15.80 11.86 14.90 25.77 23.32 29.25 6.10 8.00 8.00 13.52 10.60 17.80 42.82

2.05 2.00 2.90 6.95 3.68 9.20 6.30 5.11 7.36 5.73 4.09 24.15(24.08) 20.08(19.98) 15.84(15.80) 11.97(11.94) 14.90 25.93 23.44 30.91 6.10 8.00 8.00 13.52 10.60(12.21) 17.80 43.15

1.02 1.40 2.0 10.43 8.59 5.32 3.70 2.86 1.84 3.07 7.36 12.27 8.80 5.70 3.27 5.90 11.05 9.41 12.07 2.30 4.50 5.70 7.98 4.10 15.70 16.00

0.93(1.01) 0.84(0.92) 1.16(0.41) 10.44(10.14) 7.46(7.04) 5.70(6.07) 4.39(4.70) 2.98(3.48) 1.84(1.91) 3.34(3.79) 5.56(5.63) 12.11(11.34) 8.49(8.24) 5.70(5.72) 3.27(3.75) 6.31(5.16) 16.01(12.2) 13.93(12.56) 12.10(14.31) 2.49(2.89) 4.86(4.75) 6.04(6.98) 7.98(8.22) 4.69(5.51) 16.10(15.70) 18.70(16.00)

a

Values in parenthesis from Table 3.2A.

Dispersion partial solubility parameter, δd, at 25°C ((kJ/m3)(1/2)): δd =

∑n F i

i

i

+

∑m

j

S j + 17.3231

(3.III.3)

j

The quantity mjSj is considered to be zero for compounds that do not have second-order groups. Table 3.III.3 illustrates the statistics concerning the overall improvement in the estimation of solubility parameters that has been achieved after the introduction of second-order groups in the regression. As observed, the method is rather quite satisfactory.

Standard Deviation =

∑ (X

est

− X exp )2

N

Average Absolute Error = AAE =

1 N

∑X

,

est

− X exp , and

7248_book.fm Page 69 Tuesday, April 24, 2007 9:19 AM

Statistical Thermodynamic Calculations

69

30

30 (A) 25

(B) 25

δT

20 δ / MPa½

δ / MPa½

20 δhb

15

10

δT

δp

δhb 15

δd

δp 10

δd

5 290 300 310 320 330 340 350 360 370 380 T/K

5 290 300 310 320 330 340 350 360 370 380 T/K

FIGURE 3.II.1 Fractional solubility parameters for ammonia. A: with the scaling constants from Table 3.1A. B: with the constants from Table 3.II.1.

Average Absolute Percent Error = AAPE =

1 N

∑

Xest − X expp X exp

100% ,

where N is the number of data points, Xest is the estimated value of the property, and Xexp the experimental value. It is worth pointing out that the solubility parameters of complex structures that occur in aromatic or multiring compounds of biochemical interest can now be predicted by only using their molecular structure and without any other data known. Syntactic isomers can be distinguished, whereas stereoisomers cannot. The estimation of one of the other Hansen solubility parameters, such as δhb, as described in the main text, could lead to the estimation of δp as well.

Example of Prediction of the Hansen Partial Solubility Parameter, δd, for 1-Butanol First-Order Groups

Occurrences, ni

Contributions, Fi

niFi

1 3 1 — —

–0.97135 –0.02686 –0.34621 — —

–0.97135 –0.08058 –0.34621 –1.39814 17.3231

–CH3 –CH2 –OH ΣniFi Universal constant, C

No second-order groups are involved. δd =

∑n F i

i

i

+ 17.3231 = 15.92496 (kJ/m 3 )(1/2)

7248_book.fm Page 70 Tuesday, April 24, 2007 9:19 AM

70

Hansen Solubility Parameters: A User’s Handbook

TABLE 3.III.1 First-Order Group Contributions to (Total) δ and δd at 25°C First-Order Groups -CH3 -CH2 -CH< >C< CH2 = CH-CH = CHCH2 = C< -CH = C< >C = C< CH2 = C = CHCHECCEC ACH AC ACCH3 ACCH2CH3CO CH2CO CHO COOH CH3COO CH2COO HCOO COO OH ACOH CH3O CH2O CHO CH2O (CYCLIC) CH2NH2 CHNH2 CH3NH CH2NH CHNH CH3N CH2N ACNH2 CONH2 CONHCH3 CON(CH3) 2 C5H4N C5H3N CH2SH CH2S I BR CH2CL CHCL

Contributions to δ –2308.6 –277.1 –355.5 –176.2 –2766.2 –381.9 –980.2 1887.1 1601.8 –3745.0 –975.5 2169.3 –6.4 684.3 –221.8 1023.4 3269.1 7274.2 5398.2 9477.8 1865.1 5194.2 1716.0 3671.8 12228.9 8456.1 –480.8 –206.7 1229.1 3733.9 3650.7 560.4 8616.2 4183.8 3381.8 2166.5 –2662.6 9228.4 14930.1 27386.9 12770.8 4686.3 6574.7 2191.2 3585.2 3183.8 2163.8 1923.3 426.3

Contributions to δd –0.97135 –0.02686 0.64501 1.26857 –1.05853 0.00476 –0.48289 0.53723 0.35922 –1.65178 0.23203 –0.20284 0.11050 0.84464 0.21737 0.69325 –0.35506 0.65267 –0.40303 –0.29100 –0.54006 0.29130 na 0.20386 –0.34621 0.52883 –0.58280 0.03098 0.88334 0.27531 –0.58277 0.01116 na 0.81162 na 0.87693 1.46810 1.69868 –0.06889 na 0.44822 na na 1.27971 1.05949 0.77971 0.57169 0.26226 0.44622

Sample Group Assignment (occurrences) Propane (2) Butane (2) Isobutane (1) Neopentane (1) Propylene (1) cis-2-Butene (1) Isobutene (1) 2-Methyl-2-butene (1) 2,3-Dimethyl-2-butene (1) 1,2-Butadiene (1) Propyne (1) 2-Butyne (1) Benzene (6) Naphthalene (2) Toluene (1) m-Ethyltoluene (1) Methyl ethyl ketone (1) Cyclopentanone (1) 1-Butanal (1) Vinyl acid (1) Ethyl acetate (1) Methyl propionate (1) n-Propyl formate (1) Ethyl acrylate (1) Isopropanol (1) Phenol (1) Methyl ethyl ether (1) Ethyl vinyl ether (1) Diisopropyl ether (1) 1,4-Dioxane (2) 1-Amino-2-propanol (1) Isopropylamine (1) n-Methylaniline (1) di-n-Propylamine (1) di-Isopropylamine (1) Trimethylamine (1) Triethylamine (1) Aniline (1) 2-Methacrylamide (1) n-Methylacetamide (1) N,N-Dimethylacetamide (1) 2-Methylpyridine (1) 2,6-Dimethylpyridine (1) n-Butyl mercaptan (1) Diethyl sulfide (1) Isopropyl iodide (1) 2-Bromopropane (1) n-Butyl chloride (1) Isopropyl chloride (1)

7248_book.fm Page 71 Tuesday, April 24, 2007 9:19 AM

Statistical Thermodynamic Calculations

71

TABLE 3.III.1 (CONTINUED) First-Order Group Contributions to (Total) δ and δd at 25°C First-Order Groups CCL CHCL2 CCL2 CCL3 ACCL ACF CL-(C=C) CF3 CH2NO2 CHNO2 ACNO2 CH2CN CF2 CF C4H3S F (except as above) CH2 = C = C< CH = C = CHCHCO O (except as above) Cl (except as above) NH2 (except as above) >C = N-CH = NNH (except as above) N = NCN (except as above) NO2 (except as above) O = C = NCHSH CSH SH (except as above) S (except as above) SO2 >C = S >P>C = 0 (except as above) N (except as above)

Contributions to δ –1415.6 1164.0 na –1208.7 1332.2 –701.5 –473.5 –5199.5 10030.7 12706.7 6303.5 9359.8 –3464.4 na 4722.7 –2965.3 –2326.1 –795.6 7805.8 2467.6 636.3 –841.5 3380.7 5026.4 3459.4 –7339.6 10253.0 1655.1 2694.6 1234.8 2230.2 na 4770.2 14215.0 26271.8 –1643.4 na na

Contributions to δd 2.75755 1.17971 0.36532 na 0.84750 0.11704 0.22893 –0.22931 na na 1.41953 –0.33919 –0.97290 0.17069 na –0.70693 –0.28043 na na 0.04716 0.22562 na –0.30737 0.96719 na na 0.08615 na –0.13065 na na 1.04271 1.48988 1.55021 0.77470 na –0.43429 1.54378

Sample Group Assignment (occurrences) t-Butyl chloride (1) 1,1-Dichloropropane (1) 2,2-dichloropropane (1) Benzotrichloride (1) m-Dichlorobenzene (2) Fluorobenzene (1) 2,3-Dichloropropene (1) Perfluorohexane (2) 1-Nitropropane (1) 2-Nitropropane (1) Nitrobenzene (1) n-Butyronitrile (1) Perfluoromethylcyclohexane (5) Perfluoromethylcyclohexane (1) 2-Methylthiophene (1) 2-Fluoropropane (1) 3-Methyl-1,2-butadiene (1) 2,3-Pentadiene (1) Diisopropyl ketone (1) Divinyl ether (1) Hexachlorocyclopentadiene (2) Melamine (3) 2,4,6-Trimethylpyridine (1) Isoquinoline (1) Dibenzopyrrole (1) p-Aminoazobenzene (1) cis-Crotonitrile (1) Nitroglycerine (3) n-Butyl isocyanate (1) Cyclohexyl mercaptan (1) tert-Butyl mercaptan (1) 2-Mercaptobenzothiazole (1) Thiophene (1) Sulfolene (1) N-Methylthiopyrrolidone (1) Triphenylphosphine (1) Anthraquinone (2) Triphenylamine (1)

Note: na = not available.

Thus, estimated δd =15.925 MPa1/2, experimental5 δd =16.00 MPa1/2 Percentage error = (16.00-15.925)/16.00 = 0.47% According to Table 3.II.2, this equation-of-state approach estimates = 22.92 MPa1/2 and δhb = 15.80 MPa1/2. These data combined with the group contribution result for δd give: δp = 4.70 MPa1/2. The experimental value5 is 5.72 MPa1/2.

7248_book.fm Page 72 Tuesday, April 24, 2007 9:19 AM

72

Hansen Solubility Parameters: A User’s Handbook

TABLE 3.III.2 Second-Order Group Contributions to (Total) δ and δd at 25°C Second-Order Groups

Contributions, δ

Contributions, δd

Sample Group Assignment (occurrences)

(CH3)2-CH(CH3)3-C-CH(CH3)-CH(CH3)-CH(CH3)-C(CH3)2-C(CH3)2-C(CH3)2ring of 5 carbons ring of 6 carbons -C = C - C = CCH3-C = -CH2-C = >C{H or C}-C = string in cyclic >CHCHO CH3(CO)CH2C(cyclic) = O ACCOOH >C{H or C}-COOH CH3(CO)OC{H or C}< (CO)O(CO) ACHO >CHOH >C < OH -C(OH)C(OH)-C(OH)C(N) C(in cyclic)-OH C-O-C = C AC-O-C >N{H or C}(in cyclic) -S-(in cyclic) ACBr ACI (C = C)-Br CH3(CO)CH< ring of 3 carbons ring of 4 carbons ring of 7 carbons ACCOO AC(ACHm)2AC(ACHn)2 Ocyclic-Ccyclic = O AC-O-AC CHn-O-OH CHm-O-O-CHn NcycHm-Ccyc = O Ocyc-CcycHm = Ncyc -O-CHm-O-CHnAC-NH-AC C(= O)-C-C(= O)

142.1 592.3 1581.2 2678.4 5677.6 –2637.7 –524.2 –426.8 11.9 –762.7 –1257.2 626.1 –1634.4 142.0 –3745.0 –3076.5 511.1 134.4 –2875.9 3315.0 –359.5 –23.4 5020.6 3306.4 4022.7 –228.5 2493.0 –492.7 2389.4 337.4 1267.1 na –437.1 –9764.5 –3673.4 –1486.4 –83.5 –69.8 9215.6 –4646.5 2002.5 –2029.1 11489.1 –8721.6 –620.3 2.8 –3668.9

0.04604 –0.07377 na na na –0.66808 0.38742 –0.13554 –0.07853 –0.32357 –0.27979 –0.19450 na –0.04509 –0.29806 –0.22930 na –0.52196 –0.27069 0.37724 0.11231 –0.06801 na –0.08088 –0.08764 0.20629 0.25679 0.22183 0.48916 0.12341 0.00000 –0.40589 na 0.02003 na na –0.18466 –0.37514 0.24676 –0.56461 na na 0.29563 na 0.08394 na –0.48615

Isobutane (1) Neopentane (1) 2,3-Dimethylbutane (1) 2,2,3-Trimethylbutane (1) 2,2,3,3-Tetramethylpentane (1) Cyclopentane (1) Cyclohexane (1) 1,3-Butadiene (1) Isobutene (2) 1-Butene (1) 3-Methyl-1-butene (1) Ethylcyclohexane (1) 2-Methylpropanal (1) Methyl ethyl ketone (1) Cyclopentanone (1) Benzoic acid (1) Isobutyric acid (1) Isopropyl acetate (1) Acetic anhydride (1) Benzaldehyde (1) 2-Propanol (1) Tert-Butanol (1) 1,2-Propanediol (1) 1-Amino-2-propanol (1) Cyclohexanol (1) Ethyl vinyl ether (1) Methyl phenyl ether (1) Cyclopentimine (1) Tetrahydrothiophene (1) Bromobenzene (1) Iodobenzene (1) 2-bromo-propene (1) Methyl isopropyl ketone (1) Cyclopropane (1) Cyclobutane (1) Cycloheptane (1) Methyl benzoate (1) Naphthalene (1) Diketene (1) Diphenyl ether (1) Ethylbenzene hydroperoxide (1) di-t-Butyl peroxide (1) 2-Pyrrolidone (1) Oxazole (1) Methylal (1) Dibenzopyrrole (1) 2,4-Pentanedione (1)

Note: na = not available.

7248_book.fm Page 73 Tuesday, April 24, 2007 9:19 AM

Statistical Thermodynamic Calculations

73

TABLE 3.III.3 Comparison of the First- and Second-Order Approximations Standard Standard Data Deviation Deviation AAE AAE AAPE (%) AAPE (%) Property Points First-Order Second-Order First-Order Second-Order First-Order Second-Order δ δd

1017 344

1.47 0.61

1.31 0.58

1.00 0.44

0.90 0.41

5.15 2.62

4.67 2.42

7248_book.fm Page 74 Tuesday, April 24, 2007 9:19 AM

7248_C004.fm Page 75 Thursday, May 10, 2007 12:51 PM

Hansen Solubility 4 The Parameters (HSP) in Thermodynamic Models for Polymer Solutions Georgios M. Kontogeorgis ABSTRACT Polymer thermodynamics plays an important role in a large number of processes and in the design of many different polymer-based products. Examples include: 1. The removal of unreacted monomers, colorants, by-products, toxic compounds, and other additives after polymerization 2. The selection of mixed solvents in the paints and coatings industry toward designing environmentally-friendly paints (water-based, fewer VOC) 3. The control of emissions from paints as well as the swelling of the film in the presence of water 4. The recycling of polymers based on physicochemical methods like selective dissolution 5. The compatibility of polymer blends including those with novel structures (star-like, dendrimers), permeabilities of gases in flexible polymeric pipes used in the North Sea and other major oil and gas producing areas for transporting of hydrocarbons on the seabed and from the seabed to the surface 6. Compatibility of plasticizers in PVC 7. In the biotechnology, aqueous two-phase systems based on polymers for separating proteins This is only a short list, and many more applications of polymer thermodynamics exist. In several of these cases it is not sufficient to employ only the Hansen solubility parameters (HSP), as much more detailed calculations may be needed, including solvent activities, for example, for solvent emission assessment or even full phase diagrams and at both low (e.g., biotechnology) and high pressures (e.g., polyolefin industry, gas permeabilities in polymers). Polymers form highly nonideal liquid solutions with low-molecular weight chemicals and liquid–liquid phase separation (LLE) is the rule rather than the exception in polymer-solvent mixtures. Moreover, such LLE may take various forms; UCST, LCST, closed loop, etc., and temperature, polymer molecular weight, and polydispersity have great effects. Free-volume effects, special structures and crystallinity cause additional complexities. It is rather tempting to combine the extensive use/tables available for the HSP with a rigorous thermodynamic approach and compare the performance of this method to more advanced approaches. This is the purpose of this chapter. As several of the literature approaches used for

75

7248_C004.fm Page 76 Thursday, May 10, 2007 12:51 PM

76

Hansen Solubility Parameters: A User’s Handbook

comparison purposes are based on the group-contribution principle, a short introduction to this principle is provided first.

GROUP CONTRIBUTION METHODS FOR ESTIMATING PROPERTIES OF POLYMERS THE GROUP-CONTRIBUTION PRINCIPLE SOLUBILITY PARAMETERS)

AND

SOME APPLICATIONS (DENSITY,

Many properties of pure polymers and polymer solutions can be estimated with group contributions (GC), e.g., density, the solubility parameter (Hildebrand and HSP), the melting and glass transition temperatures, activity coefficients, and the surface tension. The GC method is based on the assumption that the properties of molecules can be estimated using additive rules from the values of the corresponding groups they are composed of. For example, n-hexane (CH3-(CH2)4-CH3) can be considered to have two CH3 and four CH2 groups. Similarly, butanone has one CH3, one CH2, and one CH3CO group. If the group values are known for a specific property F, then the total value of the property for the whole molecule is often expressed by a general additive rule of the form: F=

∑n F i

i

(4.1)

i

or similar additive equations. In Equation 4.1, ni is the number of groups of type i and Fi is the corresponding group value. In some cases, Fi values are also functions of temperature for temperature-dependent properties such as the volume and the vapor pressure. For several properties, the general GC equation has a more complicated form than that indicated by Equation 4.1. The GC methodology has been applied to many properties and for both low molecular weight compounds and polymers. Activity coefficients have also been predicted with group contributions, e.g., the UNIFAC model by Fredenslund et al.1 Van Krevelen2 gives an overview of the application of group contribution methods to several properties of pure polymers, including also mechanical and environmentally-related properties. Van Krevelen provides extensive GC tables for the Hildebrand and Hansen solubility parameters as well. An alternative GC method for the polymer (and solvent) density has been developed by Elbro et al.3 (GCVOL) and was recently extended to cover several group families.4,5 A list of GCVOL parameters is provided elsewhere.6 GCVOL can predict satisfactorily the density of solvents, oligomers, and polymers, including copolymers, often within 2%.7 The great advantage of the group contribution method is its simplicity: Even though there may be thousands of different molecules (and mixtures), the corresponding number of groups is significantly smaller (no more than 100 or so). Thus, instead of knowing the parameter values of a specific property for thousands of molecules, it is sufficient to know the group parameters for a much smaller number of groups. Two limitations of the approach should be kept in mind: 1. The GC methodology is a very useful technique leading to good results in many cases. However, it is an approximation, based often on a somewhat unjustified division of the molecule into groups. For some properties, such as for density, GC methods perform much better than for others, e.g., melting point. Specific molecules are assigned as separate groups (e.g., methanol) because further division is not possible if good results are to be obtained. Problems can also be expected for multifunctional groups and where more than one polar groups are close to each other (e.g., in alcohols and acids with more

7248_C004.fm Page 77 Thursday, May 10, 2007 12:51 PM

The Hansen Solubility Parameters (HSP) in Thermodynamic Models for Polymer Solutions

77

than one OH and COOH groups; for hydroxyl-acids or for alkanolamines). However, despite these problems, the GC principle is extensively used for property calculations for specific molecules and also, in the reverse way, for selecting suitable compounds having a required set of properties. The latter technique is called computer aided product design. 2. The exact definition of groups may be different from method to method. In some cases, even two different methods for the same property may employ different definitions for the groups. For example, the division of butadiene rubber according to the van Krevelen and GCVOL methods for density is different. In other cases, and for the same GC method, a particular molecule can be divided into groups in two different ways that may yield different results. The latter problem will be further discussed later.

GC FREE-VOLUME-BASED MODELS

FOR

POLYMERS (ENTROPIC-FV, UNIFAC-FV)

The Free-Volume Concept The classical Flory–Huggins model (See Equation 4.9 later) provides a first approximation for polymer solutions. Both the combinatorial and the energetic terms need improvement via inclusion of free-volume effects and nonrandom local-composition terms such as those of the UNIQUAC, NRTL, and UNIFAC models. The concept of free-volume (FV) is rather loose but still very important. Elbro8 showed, using a simple definition for the FV (Equation 4.2), that the FV percentages of solvents (40–50%) are greater than those of polymers (30–40%), with the exception of water and polydimethyl siloxane (PDMS). Many mathematical expressions have been proposed for the FV. One of the simplest and most successful equations is: V f = V − V * = V − Vw

(4.2)

originally proposed by Bondi9 and later adopted by Elbro et al.10 and others11 in the Entropic-FV model. In Equation 4.2, free-volume is simply the “empty” volume available to the molecule when the molecules’ own (hard-core or closed-packed V*) volume is subtracted. But what is actually the hard-core volume? This also is rather difficult to determine. Elbro8,10 suggested using V* = Vw , i.e., equal to the van der Waals volume (Vw), which is obtained from the group increments of Bondi and is tabulated for almost all existing groups in the UNIFAC tables. Other investigators interpreted the hard-core volume somewhat differently; most agree today that V* > Vw due to the closed packed structure of molecules. For example, the closed-packed cubic structure suggests that V* = 1.35Vw, and Bondi mentions that for many organic molecules it can be expected that the ratio V*/Vw should be between 1.28 and 1.43. The UNIFAC-FV Model The original UNIFAC model does not account for the free-volume differences between solvents and polymers; consequently, it highly underestimates the solvent activities in polymer solutions. Empirical modified UNIFAC versions (developed in Lyngby and Dortmund) with exponential segment fractions are also inadequate for polymer solutions; they systematically overestimate the solvent activities. Various modifications — extensions of the classical UNIFAC approach to polymers — have been proposed. All of these approaches include the FV effects, which are neglected in the UNIFAC combinatorial term, and most of them employ the energetic (residual) term of UNIFAC. The most well-known is the UNIFAC-FV model by Oishi and Prausnitz12:

7248_C004.fm Page 78 Thursday, May 10, 2007 12:51 PM

78

Hansen Solubility Parameters: A User’s Handbook

fv ln γ i = ln γ comb + ln γ res i i + ln γ i

(4.3)

The combinatorial (comb) and residual (res) terms are taken from original UNIFAC.1 All the energetic parameters in the residual term are the same as in original UNIFAC, i.e., estimated based on vapor-liquid equilibria data for low molecular weight compounds. No parameter reestimation is performed. An additional term is added for the free-volume effects. The FV term used in UNIFAC-FV has a theoretical origin, and it is based on the Flory equation of state:

ln γ

fv i

( (

) ⎤⎥ − c ⎡⎢⎛ v ) ⎥⎥⎦ ⎢⎣⎜⎝ v

⎡ v1i / 3 − 1 ⎢ = 3ci ln ⎢ 1/ 3 ⎢ vm − 1 ⎣

i

i

m

−1 ⎞⎛ 1 ⎞ ⎤ − 1⎟ ⎜ 1 − 1/ 3 ⎟ ⎥ ⎠⎝ v1 ⎠ ⎥⎦

(4.4)

where the reduced volumes are defined as: vi =

Vi bVi,w

x1V1 + x2V2 vm = b x1V1,w + x2V2,w

(

(4.5)

)

In Equation 4.5, the volumes Vi and the van der Waals volumes Vi,W are all expressed in cm3/mol. In the UNIFAC-FV model as suggested by Oishi and Prausnitz12 the parameters ci (3ci is the number of external degrees of freedom) and b are set to constant values for all polymers and solvents (ci = 1.1 and b = 1.28). The performance of UNIFAC-FV is rather satisfactory, as shown by many investigators, for a large variety of polymer-solvent systems. Some researchers have suggested that, in some cases, better agreement is obtained when these parameters (ci and b) are fitted to experimental data.13 The UNIFAC-FV model was originally developed for solvent activities in polymers and does not give satisfactory results for polymer activities; thus, it has not been applied to polymer-solvent LLE. The Entropic Model A similar to UNIFAC-FV but somewhat simpler approach, which can be readily extended to multicomponent systems and liquid-liquid equilibria, is the so-called Entropic-FV model proposed by Elbro et al.10 and Kontogeorgis et al.11: ln γ i = ln γ icomb− fv + ln γ ires ln γ icomb− fv = ln ϕ ifv =

ϕ ifv ϕ fv +1− i xi xi

xiVi , fv

∑x V j

=

j . fv

j

ln γ ires → UNIFAC

xi (Vi − Vwi )

∑ x (V − V j

j

j

wj

(4.6)

)

7248_C004.fm Page 79 Thursday, May 10, 2007 12:51 PM

The Hansen Solubility Parameters (HSP) in Thermodynamic Models for Polymer Solutions

79

As can been seen from Equation 4.6, the free-volume definition given by Equation 4.2 is employed. The combinatorial term of Equation 4.6 is very similar to that of Flory–Huggins. However, instead of volume or segment fractions, free-volume fractions are used. In this way, combinatorial and free-volume effects are combined into a single expression. The combinatorial–FV expression of the Entropic-FV model is derived from statistical mechanics, using a suitable form of the generalized van der Waals partition function. The residual term of Entropic-FV is taken by the so-called new or linear UNIFAC model, which uses a linear temperature dependent parameter table:14

(

amn = amn,1 + amn,2 T − To

)

(4.7)

This parameter table has been developed using the combinatorial term of the original UNIFAC model (which, as mentioned, does not account for free-volume effects). As with UNIFAC-FV, no parameter reestimation has been performed. The same group parameters are used in linear-UNIFAC and Entropic-FV. Both UNIFAC-FV and Entropic-FV require as input the volumes of solvents and polymers (at the temperatures of interest). If not available, these can be estimated with the GC methods mentioned previously, e.g., GCVOL. Activity coefficient calculations with UNIFAC-FV and Entropic-FV, especially the former, are rather sensitive to the density values employed. As already mentioned, the Entropic-FV model has been derived from the van der Waals partition RT - – ----afunction. The similarity of the model with the van der Waals equation of state P = -----------V – b V2 becomes apparent if the latter is written (when the classical Van der Waals one fluid mixing and classical combining rules are used) as an activity coefficient model: ln γ i = ln γ icomb− fv + ln γ ires ⎛ ϕ fv 2 ϕ fv ⎞ ⎛ V ⎞ = ⎜ ln i + 1 − i ⎟ + ⎜ i ( δ i − δ j ) ϕ 2j ⎟ ⎠ xi ⎠ ⎝ RT ⎝ xi ϕ ifv =

(4.8)

xi (Vi − bi )

∑ x (V − b ) j

j

j

j

δi =

ai Vi

ϕi is the volume fraction as defined later in Equation 4.10. The first term in Equation 4.8 is the same as in Entropic-FV with Vw = b, whereas the latter term is a regular solution theory-type term. Various efforts in improving Entropic-FV have been published, focusing especially on its combinatorial-FV term; they are reviewed elsewhere.6,15 For example, Kouskoumvekaki et al.16 suggested using Equation 4.6 with V* = 1.2Vw, which yields better results for athermal polymer solutions, compared to the assumption V* = Vw adopted in the original Entropic-FV model. This is in agreement with what is stated previously, i.e., a covolume being higher than the van der Waals volume. Entropic-FV has been extensively applied to various types of phase equilibria (VLE, LLE, and SLE) of polymer–solvent and polymer–polymer (blends) systems as well as solutions including dendrimers, mixed solvents, copolymers, and paint-related polymers. It is considered one of the state-of-the-art models in the field.

7248_C004.fm Page 80 Thursday, May 10, 2007 12:51 PM

80

Hansen Solubility Parameters: A User’s Handbook

THE FLORY–HUGGINS MODEL

AND THE

REGULAR SOLUTION THEORY

The Flory–Huggins (FH) model is the most well-known approach for polymer solutions and can, for binary systems, be expressed under some assumptions as follows for the solvent activity coefficient: ln γ 1 = ln

ϕ1 ϕ + 1 − 1 + χ12 ϕ 22 x1 x1

(4.9)

ϕ ⎛ 1⎞ = ln 1 + ⎜ 1 − ⎟ ϕ 2 + χ12 ϕ 22 x1 ⎝ r⎠

The most important assumption in deriving Equation 4.9 is that the Flory–Huggins interaction parameter (χ12) is independent of composition (see Appendix 4.A.1). The parameter r is the ratio of the polymer volume to the solvent volume V2/V1 (approximately equal to the degree of polymerization), and the volume fraction is defined as: xiVi xiVi + x jV j

ϕi =

(4.10)

The first term of Equation 4.9 is due to combinatorial effects and is derived from the lattice theory, whereas the second rather empirical (van Laar-type) energetic term includes the only adjustable parameter of the model, the so-called FH interaction parameter χ12. The FH theory can be extended to multicomponent systems (see Appendix 4.A.1) but (at least) one χ12-value is required per binary. Moreover, unfortunately, the FH parameter is typically not a constant and should be estimated from experimental data. Usually it varies with both temperature and concentration, which renders the FH model useful basically for correlating experimental data. Accurate representation of miscibility curves with the FH model is possible using rather complex equations for the temperature and the concentration-dependence of the FH-parameters: ΔG = RT

∑ x ln ϕ + g i

i

12

ϕ1ϕ 2

i

⎞ ⎛ c g12 = B(ϕ)C (T ) = a + bϕ 2 ⎜ 1 + + dT + e ln T ⎟ T ⎝ ⎠

(

(4.11)

)

Although we could use equations like Equation 4.11, it should be mentioned that the adjustable parameters of such equations (a,b,c,d,e) have no apparent physical significance; they cannot be generalized and are specific for each polymer-solvent system. For practical applications, it often suffices to use Equation 4.9 with a composition-independent Flory–Huggins interaction parameter. Even in this way, the FH model cannot be used for predictions unless a predictive scheme for the FH parameter is available. Such a predictive scheme can be based on a solubility parameter, either the Hildebrand or the Hansen. Due to the similarity of the van Laar term with the regular solution theory (see Equation 4.8), we can relate the FH parameter with the solubility parameters. This is an approximate approach, but in some cases a reasonable value of the FH parameter can be obtained, using the following equation: χ12 = χs + χ h = 0.35 +

(

V1 δ1 − δ 2 RT

)

2

(4.12)

7248_C004.fm Page 81 Thursday, May 10, 2007 12:51 PM

The Hansen Solubility Parameters (HSP) in Thermodynamic Models for Polymer Solutions

81

Equation 4.12, without the empirical 0.35 term, is derived from the regular solution theory (compare Equation 4.8 and Equation 4.9). The constant 0.35 is added for correcting for the deficiencies of the FH combinatorial and residual terms. These deficiencies become evident when comparing experimental data for athermal polymer and other asymmetric solutions to the results obtained with the FH model. A consistent underestimation of the activity coefficient data is observed, which is often attributed to the inability of the FH model to account for the free-volume differences between polymers and solvents or between compounds differing significantly in size such as n-alkanes with very different chain lengths. The term, which contains the 0.35 factor, corrects in an empirical way for these free-volume effects. However, and although satisfactory results are obtained in some cases, we cannot generally recommend using Equation 4.12 for estimating the FH parameter. Moreover, for many nonpolar systems with compounds having similar solubility parameters, the empirical factor 0.35 should be dropped. Rules of Thumb and Solvent Selection Using the Flory–Huggins Model and Solubility Parameters The FH model and the solubility parameters offer various alternative approaches for solvent selection for polymers, and these rules of thumb are summarized here. A chemical (1) will be good solvent for a specific polymer (2), or in other words, the two compounds will be miscible if one (or more) of the following “rules of thumb” are valid: 1. Using Hildebrand solubility parameters. If the polymer and the solvent have “similar polar and hydrogen bonding degrees:” ⎛ cal ⎞ ≤ 1.8 ⎜ 3 ⎟ ⎝ cm ⎠

δ1 − δ 2

1/ 2

(4.13)

2. Using Hansen solubility parameters (HSP). If the polymer and the solvent have very different polar and hydrogen bonding degrees:

(

4 δd1 − δd 2

) + (δ 2

p1

− δ p2

) + (δ 2

h1

− δh2

)

2

≤R

(4.14)

where R is the Hansen solubility parameter sphere radius. 3. χ12 ≤ 0.5 (the lower the Flory–Huggins parameter value, the greater the miscibility or, in other words, the better a solvent is a specific chemical). Values much above 0.5 indicate nonsolvency. 4. Ω1∞ ≤ 6 (the lower the infinite dilution activity coefficient of the solvent, the greater the solvency of a chemical). Values of the infinite dilution activity coefficient above 8 indicate nonsolvency.17 In the intermediate region (between 6 and 8), it is difficult to conclude if the specific chemical is a solvent or a nonsolvent. This latter rule of thumb requires some further explanations. The weight-based activity coefficient at infinite dilution is defined as: γ i∞ = lim xi →0 γ i ⎛xγ ⎞ M Ωi∞ = lim wi→0 ⎜ i i ⎟ = γ 1∞ 2 M1 ⎝ wi ⎠

7248_C004.fm Page 82 Thursday, May 10, 2007 12:51 PM

82

Hansen Solubility Parameters: A User’s Handbook

(The latter part of the equation is valid for a binary solvent(1)-polymer(2) mixture.) Extensive collections of experimental Ω1∞ data are available;18 otherwise, these can be estimated by thermodynamic models like the ones mentioned above (UNIFAC-FV, Entropic-FV, and Flory–Huggins). Thermodynamic models often perform better for this type of calculation rather than for predicting full LLE phase diagrams. However, the results depend not only on the accuracy of the model but also on the reliability of the rule of thumb, which in turns depends on the assumptions of the Flory–Huggins approach. A thermodynamically more correct method is to calculate the activity–concentration diagram with a thermodynamic model like Entropic-FV or UNIFAC-FV; the maximum indicates phase split, whereas a monotonic increase of activity with concentration indicates a single liquid phase (homogeneous solution).

ACTIVITY COEFFICIENTS MODELS USING THE HSP FLORY–HUGGINS MODELS USING HILDEBRAND PARAMETERS (HSP)

AND

HANSEN SOLUBILITY

Several of the problems of the Flory–Huggins model are associated with the difficulties in predicting the FH interaction parameter and the fact that this parameter depends on both temperature and concentration. Recently, Lindvig et al.19 proposed an extension of the Flory–Huggins equation using the Hansen solubility parameters for estimating activity coefficients of complex polymer solutions. The expression for the solvent activity coefficient in a binary solvent-polymer solution is: ln γ 1 = ln

ϕ1 ϕ + 1 − 1 + χ12 ϕ 22 x1 x1

(4.15)

2 V 2 2 χ12 = α 1 ⎡( δ d1 − δ d 2 ) + 0.25 ( δ p1 − δ p 2 ) + 0.25 ( δ h1 − δ h 2 ) ⎤ ⎦ RT ⎣

This model is hereafter abbreviated as FH/Ha(nsen) or FH/HSP. Lindvig et al.19 have tested three different combinatorial expressions, i.e., different ways of expressing the composition fraction ϕi: 1. Based on volume fractions (Equation 4.10) 2. FV fractions (Equation 4.6) 3. Segment fractions, the latter being defined via Equation 4.10 but with van der Waals volumes used instead of volumes. The universal parameter α has been fitted in each case to a large number of polymer-solvent VLE data. In total, 358 data points have been considered for solutions containing acrylates and acetates (PBMA, PMMA, PEMA, PVAc). A minimum does exist for the different types of solutions, as can be seen in Figure 4.1. In particular, the minima for the nonpolar and hydrogen bonding solvents are very close, whereas there is little sensitivity to the parameter for polar solvents. This means that a universal α value can be established. These are shown for the various combinatorial terms in Table 4.1. In all cases, the results are better when the optimum value is used than when α is set equal to one or when the term with the Hansen solubility parameters is ignored (α = 0). The best results are obtained when the volume-based combinatorial term is used (Equation 4.10) together with α = 0.6 (see also Figure 4.1). Table 4.2 provides results for several polymer solutions with the FH/Hansen model (FHHa, Equation 4.15) using the volume-based combinatorial with both the optimum parameter and α = 1.

7248_C004.fm Page 83 Thursday, May 10, 2007 12:51 PM

The Hansen Solubility Parameters (HSP) in Thermodynamic Models for Polymer Solutions

Combinatorial part based on volume fractions

100

PBMA PEMA PMA PMMA PVAc Average

90 80 Average absolute percentage deviation

83

70 60 50 40 30 20 10 0

0

0.1

0.2

0.3

0.4 0.5 0.6 0.7 Value of the correction constant

0.8

0.9

1

FIGURE 4.1 Influence of the α-parameter on the performance of Equation 4.15 for all polymer solutions in the database when the Flory–Huggins part of the model is based on volume fractions. (From Lindvig, Th., Michelsen, M.L., and Kontogeorgis, G.M., Fluid Phase Equilibria, 203, 247, 2002. Reprinted with permission.)

Results are also shown with three different group-contribution models, the previously described Entropic-FV (EFV) and UNIFAC-FV (UFV) activity coefficients and an equation of state, the GC–Flory (GCFl) model by Bogdanic and Fredenslund.20 The presentation of the results is organized into three categories according to the nature of the solvents (nonpolar, polar, and hydrogen bonding). Finally, Table 4.3 provides an overall comparison of the FH/Hansen model using all three choices for the combinatorial term and the three GC models mentioned above. Results are shown for all systems considered in the database for the estimation of the α-parameter as well as two commercial epoxy resins for which activity coefficient data are available. It can be concluded that: 1. The α-parameter is higher when volume and especially segment fractions are used in the combinatorial term. This may be expected as entropic effects are not accounted for and compensation is required by a higher parameter value. It seems that the HSP account not only for energetic effects but also for some residual-free volume contributions. 2. For the FH/Hansen model: in all cases better results are obtained when the optimum αparameter is used compared to α = 1. Moreover, FH/Hansen performs in all cases better than FV/Hildebrand which is the best possible model implementing the Hildebrand parameters.21 The average deviations with FV/Hildebrand are: 36% (nonpolar), 24% (polar), and 48% (hydrogen bonding). The FV/Hildebrand model is given by Equation 4.8 with solubility parameters being the total Hildebrand parameters.

7248_C004.fm Page 84 Thursday, May 10, 2007 12:51 PM

84

Hansen Solubility Parameters: A User’s Handbook

TABLE 4.1 Optimum Values of the α-Parameter with the FH/HSP Model, Equation 4.1 (From Lindvig, Th., Michelsen, M.L., and Kontogeorgis, G.M., Fluid Phase Equilibria, 203, 247, 2002. Reprinted with permission.) Fraction

Non-Polar

Polar

H.B.

Total

Volume

αopt % AAD (αopt) % AAD (α = 0) % AAD (α = 1)

0.55 20 37 40

1.00 23 31 23

0.60 25 54 53

0.60 22 41 40

Segment

αopt % AAD (αopt) % AAD (α = 0) % AAD (α = 1)

0.85 19 47 22

1.00 34 48 34

0.75 28 63 40

0.80 25 51 29

Free volume

αopt % AAD (αopt) % AAD (α = 0) % AAD (α = 1)

0.25 28 31 76

0.05 20 20 33

0.40 22 40 87

0.30 26 31 71

Note: The average absolute deviations (AAD) are provided using these optimum values as well as when the α-parameter is equal to zero or one. Results are shown for all systems available in the database (denoted as “total”) and for the different types of solvents. H.B. indicates hydrogen bonding solvents.

3. The FH/Hansen model is as accurate as the other group-contribution models for the systems used in the database for estimating the α-parameter. It is particularly better than the GC models for hydrogen bonding solvents. 4. The FH/Hansen model is more accurate than the other models, especially the well-known UNIFAC-FV, for the two epoxy resins. 5. The deviations are within the reported experimental uncertainty for infinite dilution activity coefficients, which is typically between 10–20%. The FH/Hansen Model vs. the GC Methods There are similarities and differences between the FH/Hansen and the GC methods. The most important similarities are that they both need as input the density of polymer and solvents (though not GC–Flory, which is an equation of state) and that they are formulated as activity coefficient models, thus their application is limited to low pressures. An advantage of the FH/Hansen model compared to the GC methods is that the exact knowledge of the structure of the polymers is not needed. The only information required is the HSP and the densities, which are available for many polymers, solvents, and other chemicals, or can be readily estimated. Moreover, the FH/Hansen method does not suffer from the often problematic assignment of groups in the GC methods. As an example, we can mention that three different definitions have been proposed for the “acetate” group of, for example, PBMA: CCOO,21 COOCH2,12 and COO.22 Use of different main groups in models like EFV or UFV will have different results, and it is not always apparent beforehand which group should be chosen. A difficulty with the FH/Hansen model is that various values of HSP are sometimes reported for the same polymers in the literature.19

7248_C004.fm Page 85 Thursday, May 10, 2007 12:51 PM

The Hansen Solubility Parameters (HSP) in Thermodynamic Models for Polymer Solutions

85

TABLE 4.2 Average Absolute Percentage Deviations (% AAD) between Calculated and Experimental Activity Coefficients for Various Groups of Solvents at Infinite Dilution in Polymers often Used in Paints and Coatings Applications Non-Polar Solvents Polymer

Nsys

NDP

EFV

UFV

GCFl

α=1 FHHa

α = 0.6 FHHa

PBMA PEMA PMA PMMA PVAc Total

11 4 5 5 8 33

60 5 17 17 103 202

22 47 45 40 20 31

22 49 46 25 19 29

11 36 28 26 17 20

65 86 18 19 53 51

18 44 24 38 14 24

Polar Solvents Polymer

Nsys

NDP

EFV

UFV

GCFl

α=1 FHHa

α = 0.6 FHHa

PBMA PEMA PMA PMMA PVAc Total

4 2 2 3 4 15

12 6 6 10 30 64

9 10 35 25 36 23

13 11 36 25 25 21

11 11 25 23 24 19

28 22 37 25 17 25

35 27 22 35 17 25

Hydrogen Bonding Solvents Polymer

Nsys

NDP

EFV

UFV

GCFl

α=1 FHHa

α = 0.6 FHHa

PBMA PEMA PMA PMMA PVAc Total

6 3 2 3 2 16

40 10 4 21 17 92

65 46 16 108 15 57

87 47 34 117 63 26

24 21 9 12 15 18

65 86 18 19 53 53

18 44 24 38 14 25

Note: Nsys and NDP are, respectively, the number of systems and datapoints.

APPLICATIONS Solvent Selection for Paints (Activity Coefficients at Infinite Dilution) Lindvig et al.19,21 have performed an evaluation of various models for solvent selection for paints. These included Entropic-FV, UNIFAC-FV, GC-Flory as well as the the FH/Hansen model presented previously and the classical Hansen method (Equation 4.14). The results of these evaluations are summarized in Table 4.4a, whereas selected results are shown in Table 4.4b. For the three group-contribution models, the solvent selection is based on the rule of thumb: Ω1∞ ≤ 6 : good solvent Ω1∞ ≥ 8 : poor solvent (nonsolvent)

7248_C004.fm Page 86 Thursday, May 10, 2007 12:51 PM

86

Hansen Solubility Parameters: A User’s Handbook

TABLE 4.3 Average Absolute Percentage Deviations (% AAD) between Experimental and Calculated Activity Coefficients for PaintRelated Polymer Solutions Using the Flory-Huggins/Hansen Method and Three Group Contribution Models Model

%AAD (systems in database)

%AAD Araldit 488

%AAD Eponol-55

FH/Hansen volume fractions FH/Hansen segment fractions FH/Hansen FV fractions Entropic-FV UNIFAC-FV GC-Flory

22 25 26 35 39 18

31 — — 34 119 29

28 — — 30 62 37

Note: The second column represents the systems used for optimization of the universal parameter (solutions containing acrylates and acetates). The last two columns show predictions for two epoxy resins. The density of the epoxies is estimated using the GCVOL method. Source: Adapted from Lindvig, Th. et al., Fluid Phase Equilibria, 203, 247, 2002.

TABLE 4.4A Validity of the Solubility Answers Obtained from Five Methods for Solvent Screening in Various Polymer-Solvent Systems Model

Correct Answers

Incorrect Answers

No Answer

No Calculation

FH/HSP (Equation 4.16) Original Hansen (Equation 4.14) FV/Hildebrand (Equation 4.17) Entropic-FV UNIFAC-FV GC-Flory

102 99 86 91 78 72

20 23 23 19 21 26

— — 13 19 17 14

7 7 7 0 13 17

Source: Adapted from Lindvig, Th. et al., Fluid Phase Equilibria, 203, 247, 2002; Lindvig, Th, et al., Thermodynamics of paint-related systems with engineering models, AIChE J., 47(11), 2573, 2001.

No answer is obtained when the infinite dilution activity coefficient value is between 6 and 8. For the FH/Hansen and FV/Hildebrand models, the solvent selection is based on whether the value of the FH parameter is below or above 0.5 as explained previously. All the FH/Hansen results shown in this section are with the “best” combination, i.e., using the volume fraction-based combinatorial and α = 0.6. The FH/Hansen and FV/Hildebrand models are summarized as follows:

7248_C004.fm Page 87 Thursday, May 10, 2007 12:51 PM

The Hansen Solubility Parameters (HSP) in Thermodynamic Models for Polymer Solutions

87

TABLE 4.4B Prediction of the Solubility for Characteristic Polymer-Solvent Systems Using Various Rules of Thumb and Models for Solvent Selection System

Experiment

PBMA/nC10 PBMA/xylene PBMA/CHCl3 PBMA/acetone PBMA/ethyl acetate PBMA/ethanol PMMA/acetone PMMA/ethyl acetate PMMA/butanol PEMA/MEK PEMA/diethyl ether PEMA/nitropropane PVAc/hexane PVAc/methanol PVAc/ethanol PVAc/nitromethane PVAc/THF

NS S S S S NS S S NS S S NS NS S NS S S

EFV 6.5 2.3 1.9 0.2 6.7 29.2 10.0 6.6 26.8 8.1 5.8 4.5 38.7 18.9 15.2 3.9 8.4

UFV

(–) (S) (S) (NS) (–) (NS) (NS) (– ) (NS) (NS) (S) (S) (NS) (NS) (NS) (S) (NS)

6.1 3.6 9.1 14.1 6.7 31.3 16.5 8.4 14.4 11.7 7.6 1.4 38.6 19.4 38.9 3.8 5.6

(–) (S) (NS) (NS) (–) (NS) (NS) (NS) (NS) (NS) (–) (S) (NS) (NS) (NS) (S) (S)

FH/HSP 1.20 0.41 0.14 0.2 0.27 1.01 0.18 0.36 0.67 0.09 0.57 0.11 1.09 0.71 0.63 0.43 0.05

(NS) (S) (S) (S) (S) (NS) (S) (S) (NS) (S) (NS) (S) (NS) (NS) (NS) (S) (S)

Note: The values in the table for EFV and UFV indicate infinite dilution activity coefficients while those of FH/HSP are the Flory-Huggins parameters estimated based on Equation 4.16. S = good solvent, NS = non solvent, - = no answer according to the rule of thumb. Source: Adapted from Lindvig, Th. et al., Fluid Phase Equilibria, 203, 247, 2002.

FH–Hansen (FH/HSP) ln γ 1 = ln

ϕ1 ϕ + 1 − 1 + χ12 ϕ 22 x1 x1

χ12 = 0.6

2 V1 ⎡ (δ d1 − δ d 2 )2 + 0.25 (δ p1 − δ p 2 ) + 0.25 (δ h1 − δ h 2 )2 ⎤⎦ RT ⎣

ϕi =

(4.16)

xiVi xiVi + x jV j

FV–Hildebrand ln γ 1 = ln ϕ ifv =

xiVi , fv

∑ j

χ12 =

ϕ ifv ϕ fv + 1 − i + χ12 ϕ 22 xi xi

x jV j . fv

=

xi (Vi − Vwi )

∑

V1 ( δ1 − δ 2 ) 2 RT

j

x j (V j − Vwj )

(4.17)

7248_C004.fm Page 88 Thursday, May 10, 2007 12:51 PM

88

Hansen Solubility Parameters: A User’s Handbook

The basic conclusions are: 1. The FH/Hansen method is, on average, as good as the original method of Hansen and the three GC methods 2. The FH/Hansen method is better than the “best” combination based on the Hildebrand parameters, i.e., the FV/Hildebrand method 3. For most polymers (PBMA, PMMA, PEMA) the GC methods have problems for ketonecontaining solutions, where only the FH/Hansen method performs well 4. Most models have problems with nitrocompounds and PEMA 5. The behavior of the various models for PVAC is rather mixed and peculiar, and unlike the other polymers, each model has different strengths and weaknesses. Mixed Solvent–Polymer Phase Equilibria Lindvig et al.23 extended the applicability of various models to mixed solvent-polymer VLE and typical results are presented in Table 4.6 divided according to the experimental source. Only a few experimental data are available for such multicomponent systems, and the accuracy of the data may in some cases be doubtful. Two fundamentally different modelling approaches have been tested: 1. Purely predictive GC models for which calculations can be made without regressing parameters from binary data for the systems considered (EFV, UFV, FH/Ha, GC-Flory, and GCLF). These models are the fastest and simplest tools for thermodynamic calculations. All calculations are based on existing parameters (typically group-based ones). 2. Molecular models using binary molecular interaction parameters estimated from experimental data for the corresponding binary systems. This is a more time-consuming approach than the former one but is still a predictive one for ternary mixtures in the sense that no multicomponent data are used for the regression of the model parameters. All interaction parameters have been fitted to binary experimental data as explained by Lindvig et al.23 All correlative models contain a single binary parameter except EFV/UNIQUAC, which has two. In addition to this distinction (purely predictive and molecular models), the models tested for multicomponent systems have similarities and differences, and their various characteristics are summarized in Table 4.5.

TABLE 4.5 Presentation of the Characteristics of the Various Models Tested for Mixed Solvent–Polymer Phase Equilibria Model

Correlative

SAFT EFV/UQ FH Pa-Ve EFV/UN UFV GCFl GCLF FH/Ha

X X X X

Fully Predictive

Activity Coefficient Model

Equation of State

Group-Contribution

X X X

Partially X

X X X X X

X X X X

X X

X X X X Partially

7248_C004.fm Page 89 Thursday, May 10, 2007 12:51 PM

The Hansen Solubility Parameters (HSP) in Thermodynamic Models for Polymer Solutions

89

TABLE 4.6 Average Absolute Logarithmic Percentage Deviations between Experimental and Predicted Equilibrium Pressures and Average Absolute Deviation (X100) between Calculated and Experimental Vapor-Phase Compositions (Mole Fractions) for Various Ternary Polymer-Mixed Solvent Systems Sys. No.

Variable

SAFT

EFV/U

FH

Pa-Ve

EFV/U

UFV

GCFI

GCLF

FHHa

1

P y

13 28

16 30

15 29

16 29

16 32

36 32

16 28

15 27

17 21

2

P y

7 14

4 13

12 14

7 14

8 8

14 8

149 6

14 23

7 13

3

P y

– –

– –

– –

– –

30 17

37 18

16 8

22 5

50 5

4

P y

– –

– –

– –

– –

72 4

57 5

52 4

32 4

20 4

Note: 1. PMMA-butanone-toluene at 308 K. 2. PS-benzene-toluene at 308 K. 3. PMMA-butanone-acetone at 308 K. 4. PMMA-benzene-toluene at 308 K. Source of experimental data: Liu et al. (2002), Fluid Phase Equilibria, 2002, 194–197: 1067–1075]. Sys. No.

Variable

SAFT

EFV/U

FH

Pa-Ve

EFV/U

UFV

GCFI

GCLF

FHHa

1

P y

11 4

6 3

6 3

6 3

2 3

1 3

21 3

8 3

2 3

2

P y

4 5

2 2

14 5

8 1

2 2

2 2

– –

2 4

11 5

3

P y

– –

– –

– –

– –

3 3

3 3

12 2

5 4

18 4

4

P y

– –

– –

– –

– –

4 18

4 18

13 19

9 19

5 15

Note: 1. PS-toluene-ethylbenzene at 303 K. 2. PS-toluene-cyclohexane at 303 K. 3. PVAc-acetone-ethyl acetate at 303 K. 4. PVAc-acetone-methanol at 303 K. Source of experimental data: Katayama et al. (1971) [Kagaku Kogaku, 1971, 35: 1012]; Matsumara and Katayama (1974), Kagaku Kogaku, 1974, 38: 388. Sys. No. 1

Variable

SAFT

EFV/U

FH

Pa-Ve

EFV/U

UFV

GCFI

GCLF

FHHa

P y

14 17

16 17

11 16

4 11

17 17

19 13

93 2

5 18

52 14

Source of experimental data: Tanbonliong and Prausnitz (1997), Polymer, 38: 5775; PS-chloroform-carbon tetrachloride at 323.15 K. Adapted from Lindvig et al.23

Besides the activity coefficient models mentioned previously (Entropic-FV, UNIFAC-FV, and FH/Hansen) and the GC-Flory equation of state, three advanced equations of state are considered: the GCLF by Lee and Danner,24 SAFT by Chapman and coworkers,25 and Panayiotou-Vera.26 The basic conclusions are:

7248_C004.fm Page 90 Thursday, May 10, 2007 12:51 PM

90

Hansen Solubility Parameters: A User’s Handbook

1. The FH/Hansen is as successful as the GC methods and other more complex models, both the correlative and the predictive. 2. The models perform sometimes differently for isolated systems, but the overall differences are minor, and we believe that they fall within the experimental uncertainties of the data considered. In many cases the results with the various models are closer to each other than to the experimental data. 3. For 9 out of the 13 systems considered, good predictions of both the equilibrium pressures and the vapor phase compositions are obtained by all models. 4. The correlative models do not offer any improvements over the group-contribution models. This is a surprising result as it is expected that the molecular information in a model should lead to a better representation of phase equilibria for multicomponent mixtures. It may be that some experimental data for such multicomponent mixtures are of low accuracy. 5. This investigation does not point to a “clear winner” among the models, and more data are required for further investigations. However, this preliminary study provides confidence for use of the FH/Hansen approach.

CONCLUSIONS AND FUTURE CHALLENGES Many successful calculations in polymer thermodynamics can be carried out using simple groupcontribution methods based on UNIFAC, which contain corrections for the FV effects. Models like Entropic-FV and UNIFAC-FV can be used for such calculations and are shown to satisfactorily predict the solvent activities and vapor–liquid equilibria for binary and ternary polymer solutions. Such methods require accurate values of the densities and, moreover, are based on the availability of group parameters in the UNIFAC tables. An alternative equally successful approach is offered by the combination of the Flory–Huggins model with the Hansen solubility parameters (HSP). The FH/HSP (Equation 4.16) includes a single universal parameter that has been regressed to experimental data for many polymer solvent solutions. The combinatorial term that gives the best results is based on volume fractions. The FH/HSP model is shown to be as successful as the state of the art GC models (EntropicFV, UNIFAC-FV, and GC-Flory) in predicting infinite dilution activity coefficients including complex epoxy polymers, solvent selection for paints, and VLE for mixed solvent–polymer systems. Moreover, FH/HSP is as successful for mixed solvent–polymer phase equilibria as complex, theoretically-based equations of state like SAFT and the Group Contribution Lattice Fluid. This chapter has been limited to vapor–liquid equlibria (both at finite concentrations and infinite dilution), mixed solvents, and solvent selection. The methods can be, in principle, extended to polymer–solvent LLE, and this has been indeed done, for example, for Entropic-FV, GC-Flory, GCLF, and SAFT. The predictive group contribution methods are less successful for the prediction of liquid–liquid equilibria as their parameters are based on VLE. Some results are summarized in the recent literature.6,15 Better results are obtained when a molecular local composition model is used as, for example, in the Entropic-FV/UNIQUAC model as shown by Pappa et al.27 As such models include the liquid volume as input parameter, successful results have been reported even for high pressure LLE. However, solvent selection can be based on the use of infinite dilution activity coefficients for which these models are quite successful. So far most activity coefficient models, including FH/HSP, have been applied mainly to organic polymer solutions of rather simple structures and VLE/solvent selection studies, although some complex paint-related polymers have been considered as well. Thorlaksen et al.28 have recently combined the Entropic-FV term with Hildebrand's regular solution theory and developed a model for estimating gas solubilities in elastomers. A similar approach can be adopted for the FH/HSP model presented here. Future developments can include complex structures such as dendrimers (where EFV and UFV already have been applied29); star-like, hyperbranched polymers as well as various copolymers;

7248_C004.fm Page 91 Thursday, May 10, 2007 12:51 PM

The Hansen Solubility Parameters (HSP) in Thermodynamic Models for Polymer Solutions

91

liquid–liquid and solid–liquid equilibria, including the effect of crystallinity; and cross-linking, inorganic polymers and polyelectrolytes.

LIST OF ABBREVIATIONS % AAD Comb Comb-FV CST Exper./exp. EFV EFV/UN EFV/UQ FH FH/HSP FV GC GC-F(lory) GCLF GCVOL HB HSP LCST LLE MW NS Pa-Ve PBMA PDMS PEMA PMMA Pred. PVAC Res S SAFT SLE UCST U-FV UNIFAC UQ vdW vdW1f VLE VOC

Average percentage absolute deviation Combinatorial Combinatorial-free volume Critical solution temperature Experimental Entropic-FV (Same as EFV) Entropic-FV (using UNIFAC for the residual term) Entropic-FV using UNIQUAC as the residual term Flory–Huggins (model/equation/interaction parameter) The FH model using the Hansen solubility parameters, Equation 4.16 Free-volume Group contribution (method/principle) Group contribution Flory equation of state Group contribution lattice fluid Group contribution volume (method for estimating the density) Hydrogen bonding Hansen solubility parameters Lower critical solution temperature Liquid–liquid equilibria Molecular weight Nonsolvent/nonsoluble Panayiotou–Vera equation of state Polybutyl methacrylate Polydimethylsiloxane Polyethyl methacrylate Polymethyl methacrylate Predicted Polyvinyl acetate Residual Solvent/soluble Statistical associating fluid theory Solid–liquid equilibria Upper critical solution temperature UNIFAC-FV Universal functional activity coefficient UNIQUAC Van der Waals equation of state Van der Waals one fluid (mixing rules) Vapor–liquid equilibria Volatile organic content

7248_C004.fm Page 92 Thursday, May 10, 2007 12:51 PM

92

Hansen Solubility Parameters: A User’s Handbook

SYMBOLS IN THIS CHAPTER G R R T a amn b ni r x v a b c F Fi V Vf V* VW δ φ γ χ12 ΩI∞

Gibbs energy Radius of Hansen solubility sphere Gas constant (in connection with T) Temperature Constant in van der Waals equation of state Coefficient defined in Equation 4.7 Constant in van der Waals equation of state Number of given groups of type “i” in molecule Ratio of polymer volume to solvent volume Mole fraction Reduced volumes in Equation 4.4 and Equation 4.5 Constant in Equation 4.15 Constant in Equation 4.5 Degree of freedom in Equation 4.4 Property in Equation 4.1 Group value for given property in Equation 4.1 Total volume Free volume (Equation 4.2) Hard core or close packed volume in Equation 4.2 van der Waals volume Solubility parameter (as in rest of handbook) Volume fraction (see also Equation 4.10) Activity coefficient in Chapter 4 Flory–Huggins interaction parameter Infinite dilution activity coefficient

APPENDIX 4.I: AN EXPRESSION OF THE FLORY-HUGGINS MODEL FOR MULTICOMPONENT MIXTURES The Flory–Huggins model was originally developed as a model for the entropy of mixing for mixtures containing molecules of different size, but it was soon modified also to account for energetic interactions. The model can be formulated in terms of the excess Gibbs energy as follows (Lindvig et al.23): G E = G E,comb + G E,res G E,comb = RT G E,res = RT

N

∑ n ln φx

i

i

i

i =1 N

N

i =1

j =1

∑∑φ φ a i

χij = 2 aij νi

j ij

7248_C004.fm Page 93 Thursday, May 10, 2007 12:51 PM

The Hansen Solubility Parameters (HSP) in Thermodynamic Models for Polymer Solutions

93

Using basic thermodynamics, the following expression for the activity coefficient is obtained: ln γ i = ln γ comb + ln γ res i i where the combinatorial term is given by: ln γ icomb = ln

ϕi ϕ +1− i xi xi

and the residual term is: NC

= 2 vi ln γ res i

∑ j =1

NC

ϕ j aij − vi

NC

∑∑ϕ ϕ a j

k

jk

j =1 k =1

The above formulation of the FH model is slightly different from the conventionally used formulation using the Flory–Huggins interaction parameter (χ12), although there is an interrelationship based on the simple equation shown above. For a binary mixture, the multicomponent equation reduces to the traditional FH residual term: ln γ 1res = χ12 ϕ 22

REFERENCES 1. Fredenslund, Aa., Jones, R.L., and Prausnitz, J.M., Group contribution estimation of activity coefficients in nonideal liquid mixtures, AIChE J., 25(1), 1086–1098, 1975. 2. Van Krevelen, D.W., Properties of polymers, Their correlation with chemical structure; their numerical estimation and prediction from additive group contributions, Elsevier, 1990. 3. Elbro, H.S., Fredenslund, Aa., and Rasmussen, P., Group contribution method for the prediction of liquid densities as a function of temperature for solvents, oligomers, and polymers, Ind. Eng. Chem. Res., 30, 2576, 1991. 4. Tsibanogiannis, I.N., Kalospiros, N.S., and Tassios, D.P., Extension of the GCVOL method and application to some complex compounds, Ind. Eng. Chem. Res., 33, 1641, 1994. 5. Ihmels, E.C. and Gmehling, J., Extension and revision of the group contribution method GCVOL for the prediction of pure compound liquid densities, Ind. Eng. Chem. Res., 42(2), 408–412, 2003. 6. Kontogeorgis, G.M., Thermodynamics of polymer solutions, in Handbook of Surface and Colloid Chemistry, 2nd ed., Birdi, K.S., Ed., CRC Press, Boca Raton, FL, 2003, chap.16. 7. Bogdanic, G. and Fredenslund, Aa., Prediction of VLE for mixtures with co-polymers, Ind. Eng. Chem. Res., 34, 324, 1965. 8. Elbro, H.S., Phase Equilibria of Polymer Solutions — with Special Emphasis on Free Volumes, Ph.D thesis, Department of Chemical Engineering, Technical University of Denmark, 1992. 9. Bondi, A., Physical Properties of Molecular Crystals, Liquids and Glasses, John Wiley & Sons, New York, 1968. 10. Elbro, H.S., Fredenslund, Aa., and Rasmussen, P., A new simple equation for the prediction of solvent activities in polymer solutions, Macromolecules, 23, 4707, 1990. 11. Kontogeorgis, G.M., Fredenslund, Aa., and Tassios, D.P., Simple activity coefficient model for the prediction of solvent activities in polymer solutions, Ind. Eng. Chem. Res., 32, 362, 1993. 12. Oishi, T. and Prausnitz, M., Estimation of solvent activities in polymer solutions using a groupcontribution method, Ind. Eng. Chem. Process Des. Dev., 17(3), 333, 1978. 13. Fried, J.R., Jiang, J.S., and Yeh, E., Comput. Polym. Sci., 2, 95, 1992.

7248_C004.fm Page 94 Thursday, May 10, 2007 12:51 PM

94

Hansen Solubility Parameters: A User’s Handbook 14. Hansen, H.K., Coto, B., and Kuhlmann, B., UNIFAC with lineary temperature-dependent groupinteraction parameters, IVC-SEP Internal Report 9212, 1992. 15. Kontogeorgis, G.M., Models for polymer solutions, in Computer Aided Property Estimation for Process and Product Design, Kontogeorgis, G.M. and Gaani, R., Eds., Elsevier, 2004, chap. 7. 16. Kouskoumvekaki, I., Michelsen, M.L., and Kontogeorgis, G.M., Fluid Phase Equilibria, 202(2), 325, 2002. 17. Holten-Andersen, J. and Eng, K., Activity coefficients in polymer solutions, Progress in Organic Coatings, 16, 77, 1988. 18. High, M.S. and Danner, R.P., Polymer Solution Handbook; DIPPR 881 Project, Design Institute for Physical Property Data, 1992. 19. Lindvig, Th., Michelsen, M.L., and Kontogeorgis, G.M., Fluid Phase Equilibria, 203, 247, 2002. 20. Bogdanic, G. and Fredenslund, Aa., Ind. Eng. Chem. Res., 34, 324, 1995. 21. Lindvig, Th., Michelsen, M.L., and Kontogeorgis, G.M., Thermodynamics of paint-related systems with engineering models, AIChE J., 47(11), 2573, 2001. 22. Lee, B.C. and Danner, R.P., Prediction of infinite dilution activity coefficients in polymer solutions: comparison of prediction models, Fluid Phase Equilibria, 128, 97, 1997. 23. Lindvig, Th., Economou, I.G., Danner, R.P., Michelsen, M.L., and Kontogeorgis, G.M., Modeling of multicomponent vapor-liquid equilibria for polymer-solvent systems, Fluid Phase Equilibria, 220, 11–20, 2004. 24. Lee, B.C. and Danner, R.P., Prediction of polymer-solvent phase equilibria by a modified groupcontribution EoS, AIChE, 42, 837, 1996. 25. Chapman, W.G. et al., New reference equation of state for associating fluids, Ind. Eng. Chem. Res., 29, 1709–1721, 1990. 26. Panayiotou, C. and Vera, J.H., An improved lattice-fluid equation of state for pure component polymeric fluids, Polym. Eng. Sci., 22, 345, 1982. 27. Pappa, G.D., Voutsas, E.C., and Tassios, D.P., Liquid-liquid phase equilibrium in polymer-solvent systems: correlation and prediction of the polymer molecular weight and the pressure effect, Ind. Eng. Chem. Res., 40(21), 4654, 2001. 28. Thorlaksen, P., Abildskov, J., and Kontogeorgis, G.M., Prediction of gas solubilities in elastomeric polymers for the design of thermopane windows, Fluid Phase Equilibria, 211, 17, 2003. 29. Kouskoumvekaki, I., Giesen, R., Michelsen, M.L., and Kontogeorgis, G.M., Free-volume activity coefficient models for dendrimer solutions, Ind. Eng. Chem. Res., 41, 4848, 2002.

7248_C005.fm Page 95 Wednesday, May 23, 2007 10:51 AM

of Characterization 5 Methods — Polymers Charles M. Hansen ABSTRACT The simplest experimental method to determine the Hansen solubility parameters (HSP) for a polymer is to evaluate whether or not it dissolves in selected solvents. Those solvents dissolving the polymer will have HSP closer to those of the polymer than those solvents that do not. A computer program or graphical method can then be used to find the HSP for the polymer. Other types of evaluations can also lead to polymer HSP. These include swelling, melting point reduction, surface attack, chemical resistance, barrier properties, viscosity measurements, and any other measurement reflecting differences in polymer affinities among the solvents. Polymer HSP can be higher than the HSP of any of the test solvents. This means that some of the methods suggested in the literature to interpret data, i.e., those which use averages of solvent HSP to arrive at the polymer HSP, must be used with care.

INTRODUCTION Experience has shown that if it is at all possible, an experimental evaluation of the behavior of a polymer in contact with a series of selected liquids is the best way to arrive at its HSP. Experimental data can be generated and treated in various ways to arrive at the values of interest. Examples are included in the following. The author’s usual approach to generate data in solubility parameter studies is to contact a polymer of interest with 40 to 45 well-chosen liquids. One may then observe or measure a number of different phenomena including full solution at a given concentration, degree of swelling by visual observation or by measurement of weight change, volume change, clarity, surface attack, etc. The object of the studies is to determine differences in affinity of the polymer for the different solvents. These differences are then traditionally used to divide the solvents into two groups, one which is considered “good” and the other which is considered “bad.” Such data can be entered into the SPHERE program as discussed in Chapter 1. Whenever possible, the author uses a set of solvents as described below, often supplemented by selected solvents depending on the purpose of the investigation. Supplementary test solvents are usually in the boundary regions as it is these that determine the parameters of the sphere. Adding more good solvents well within the sphere or more bad solvents well outside of it will not change anything but the data fit. The goal of the experimental work is to arrive at a set of data showing differences in behavior among the test solvents. These data are then processed to arrive at the four parameters characteristic of HSP correlations, three describing the nonpolar, polar, and hydrogen-bonding interactions for the liquids and the fourth, Ro, a radius of interaction for the type of interaction described. The author has most often used computer techniques to evaluate the data to find the polymer HSP. In earlier work simple plots were used. A simple plot of δP vs. δH is also helpful in many practical situations to get guidance as discussed in Chapter 8. The approximate determination of polymer HSP can be done with three plots of experimental data using the HSP parameters pairwise. Figure 5.1 to Figure 5.3 demonstrate how this was attempted initially.1 The spheroids in the figures including the δD parameter gave problems. Hansen and Skaarup2 simply used a scaling factor of 2 (the coefficient “4” in Chapter 1, Equation 1.9) to produce spheres in all three plots. As Ro must 95

7248_C005.fm Page 96 Wednesday, May 23, 2007 10:51 AM

96

Hansen Solubility Parameters: A User’s Handbook

12

10

8 δp R

6 (P,H) 4

2

2

4

6

8 δh

10

12

14

FIGURE 5.1 Two-dimensional plot of δP vs. δH for the solubility of polymethyl methacrylate (Polymer B in Table 5.2). The circle is the projection of a sphere on the given coordinates. Units are (cal/cm3)1/2. (From Hansen, C.M., Färg och Lack, 17(4), 71, 1971. With permission.)

be the same in all of these plots, a single compass setting is tried for a set of δD, δP , and δH to see how well the separation into good and bad solvents is accomplished. Calculations for points in doubt can be made using Chapter 1, Equation 1.9. Plots with the modified δD axis are given for the solubility of polystyrene3 shown in Figure 5.4 to Figure 5.6. These are the original figures from this thesis, and the numbers refer to a table of solvents found there. An idea of the accuracy of the graphical approach can be found in Table 5.1, where comparisons are made between the “hand” method and results of the SPHERE program. Table 5.2 contains a listing of the polymers included in Table 5.1. Specific solubility data are given for these polymers in 88 solvents in Appendix A.3. Teas4 has developed a triangular plotting technique which helps visualization of three parameters on a plain sheet of paper. Examples are found in Reference 5 to Reference 7 and in Chapter 8. The triangular plotting technique uses parameters for the solvents, which, in fact, are modified HSP parameters. The individual Hansen parameters are normalized by the sum of the three parameters. This gives three fractional parameters defined by Equation 5.1 to Equation 5.3. fd = 100δD/(δD + δP + δH)

(5.1)

fP = 100δP/(δD + δP + δH)

(5.2)

fh = 100δH/(δD + δP + δH)

(5.3)

The sum of these three fractional parameters is 1.0. This allows the use of the special triangular technique. Some accuracy is lost, and there is no theoretical justification for this plotting technique,

7248_C005.fm Page 97 Wednesday, May 23, 2007 10:51 AM

Methods of Characterization — Polymers

97

12

10

8 δp 6

4

2

2

4

6

8

10

12

14

δh FIGURE 5.2 Two-dimensional plot of δH vs. δD for the solubility of polymethyl methacrylate (Polymer B in Table 5.2). Expansion of the δD scale by a factor of 2 would yield a circle (a sphere in projection). Units are (cal/cm3)1/2. (From Hansen, C.M., Färg och Lack, 17(4), 71, 1971. With permission.)

but one does get all three parameters onto a two-dimensional plot. This plotting technique is often used by those who conserve old paintings, because it was described in a standard reference book very shortly after it was developed.7 Figure 8.4 shows how such a plot can be used in finding a suitable solvent when dealing with such an older oil painting. HSP for the polymers and film formers discussed in the following examples are given in Table 5.3. These data are based on solubility determinations unless otherwise noted. Barton6,8 has also provided solubility parameters for many polymers. Values for a number of acrylic, epoxy, and other polymers potentially useful in self-stratifying coatings have been reported by Benjamin et al.9 (see Chapter 8). Rasmussen and Wahlström10 provide additional HSP data in relation to the use of replenishable natural products (oils) in connection with solvents. The data processing techniques and data accumulated by Zellers and coworkers11–14 on elastomers used in chemical protective clothing are also useful. Zellers et al. also point out many of the problems encountered with these characterizations. Such problems are also discussed below. There are other sources of HSP for polymers in the literature, but a full review of these and their uses is beyond the scope of this book.

CALCULATION OF POLYMER HSP Calculation of the HSP for polymers is also possible. The results are not yet fully satisfactory, but there is hope for the future. One of the more significant efforts in this has been made by Utracki and coworkers.15,16 They assumed the δD parameter for polymers did not differ too much between polymers and interpreted evaluations of polymer–polymer compatibility using calculated values for δP and δH. A word of caution is advisable here and that is that the constant “4” in Equation 1.9 is very often if not most often significant, and should not be replaced with a “1,” either. Group

7248_C005.fm Page 98 Wednesday, May 23, 2007 10:51 AM

98

Hansen Solubility Parameters: A User’s Handbook

12

10

8 δp 6

4

2

2

4

6

8 δd

10

12

14

FIGURE 5.3 Two-dimensional plot of δP vs. δD for the solubility of polymethyl methacrylate (Polymer B in Table 5.2). Expansion of the δD scale by a factor of 2 would yield a circle (a sphere in projection). Units are (cal/cm3)1/2. (From Hansen, C.M., Färg och Lack, 17(4), 72, 1971. With permission.)

calculations were used. This is probably the best calculation approach currently available, but improvements are thought possible. See Chapter 3. The group contributions given in Chapter 1 can be used for this purpose, although the estimated dispersion parameters are thought to be too low. It is suggested that HSP for polymers determined by these calculations not be mixed with experimentally determined HSP until confirmation of agreement is found. It can be presumed that the errors involved in either process will cancel internally, but these may not necessarily be the same for the calculated results as for the experimental ones. The author has never been particularly successful in calculating the same values as were found experimentally, although a serious effort to use weighting and similar factors, as discussed in the following, has never been tried.

SOLUBILITY — EXAMPLES The most direct method to determine the three HSP for polymers or other soluble materials is to evaluate their solubility or degree of swelling/uptake in a series of well-defined solvents. The solvents should have different HSP chosen for systematic exploration of the three parameters at all levels. As indicated earlier, a starting point could be the series of liquids used by the author for many years. These are essentially those included in Table 5.4. Sometimes boundaries are defined better by inclusion of additional test solvents. A computer analysis quickly gives a choice of many of these, as solvents with RED numbers (Chapter 1, Equation 1.10) near 1.0 are located near the sphere boundary. It is actually the boundary which is used to define the center point of the sphere using Chapter 1, Equation 1.9. Some changes are also possible to remove or replace solvents which are now considered too hazardous, although good laboratory practice should allow use of the ones

7248_C005.fm Page 99 Wednesday, May 23, 2007 10:51 AM

Methods of Characterization — Polymers

99

16

14

SOLUBLE NOT SOLUBLE STRONG INTERACTION

12

δp

10 28

8

45

6

43

38A

69

29 RA

4

2

32 23

0 δd 0

2

21 19 19A 15A 18A 18 42 15 67A 89 25

4

6

8 δh

10

12

14

16

FIGURE 5.4 Two-dimensional plot δP vs. δH of solubility data for polystyrene (Polymer G in Table 5.2). Units are (cal/cm3)1/2.

indicated. The HSP generally in use for liquids have all been evaluated/calculated at 25°C. These same values can also be used to correlate physical phenomena related to solubility at other test temperatures with some care, as noted in the following. Several examples of HSP correlations based on solubility are found in Table 5.3. The entry for polyethersulfone (PES) found in Table 5.3 was determined from data included in the computer output reported in Table 5.4. The solubility of PES was evaluated in 41 different solvents. It was found that five of them actually dissolved the polymer. The input data to the SPHERE program described in Chapter 1 are included in Table 5.4 in the SOLUB column. A “1” means a good solvent and a “0” means a bad solvent. A 1* means that a good solvent lies outside the sphere, where it should not, and a 0* means a bad solvent lies inside the sphere, which means it is an outlier. Each of these error situations reduces the data fit. D, P, H, and R for the solubility of PES are given at the top. In addition, there is an indication of the data fit, which is 0.999 here. A perfect fit is 1.000. A data fit slightly less than 1.0 is actually preferred, as the computer program has then optimized the data to a single set of values that are so close to being correct as they can be within experimental error. An unknown number of sets of the parameters can give a data fit of 1.0 whenever this result is found. Perfect fits are rather easily obtained with small sets of data, and the boundaries are rather poorly defined, which means the center is also poorly defined. One can continue testing with additional solvents located in the boundary regions of the established sphere as stated previously. These can be found easily by listing the solvents in order of their RED numbers and choosing

7248_C005.fm Page 100 Wednesday, May 23, 2007 10:51 AM

100

Hansen Solubility Parameters: A User’s Handbook

16

14 SOLUBLE NOT SOLUBLE STRONG INTERACTION

12

δh

10

8 15

9

6

67A21 19A 69 15B 35 25 29

4

45 42 47A 28

23

RA

2

56

32 45

0 δp

5

6

7

26A

8 63 48 9

10

11

12

δd

FIGURE 5.5 Two-dimensional plot δH vs. δD of solubility data for polystyrene (Polymer G in Table 5.2). Expansion of the δD scale by a factor of 2 has given a spherical representation according to Chapter 1, Equation 1.9. Units are (cal/cm3)1/2.

those with values not too different from 1.0. The RED number is given for each solvent in the RED column. A quality number, Q = 1 – RED, is also conceptually useful. Finally, there is a column in Table 5.4 indicating the molar volume, V, of the solvents in cc/mol. There was no need to analyze the influence of this parameter in the present case. A second example of this type of approach is given in Table 5.3. Data on good and bad solvents17 for polyacrylonitrile (PAN) have been used as input to the computer program. There are 13 solvents indicated as good, and 23 indicated as bad. These test solvents do not differ as widely from each other as the test series suggested earlier, but the data are still useful in finding the HSP for this polymer. These are reported in Table 5.3. The data fit of 0.931 is good for this kind of data. Having found the HSP for a polymer in this manner, one can then search a database for additional solvents for the polymer in question. This was done for the HSP database with over 800 solvent entries in Table A.1 of the first edition of this handbook. A significantly large number of the 123 additional solvents found to have RED numbers less than 1.0 can be expected to dissolve this polymer, but such an extensive experimental study was not undertaken to confirm the predictions. A special problem that can be encountered is when only a few solvents with very high solubility parameters dissolve a polymer. An example is polyvinyl alcohol with true solvents being 1-propanol and ethanol in a data set with 56 solvents.6 The entry in Table 5.3 places a big question mark over the solubility parameters, as well as with the radius 4.0 and the perfect fit of the data. The computer analysis quickly encompasses the two good solvents in the data set within a small sphere as they

7248_C005.fm Page 101 Wednesday, May 23, 2007 10:51 AM

Methods of Characterization — Polymers

101

16

14 SOLUBLE NOT SOLUBLE STRONG INTERACTION

12

δp

10 27

8

40 16 38A

44 26B

28

45 43

6

69 29

66

21 58

4

RA 8

67A

2

9

56

32 8

23 25

0 δh

5

6

7

26A

8

9

10

11

12

δd

FIGURE 5.6 Two-dimensional plot δP vs. δD of solubility data for polystyrene (Polymer G in Table 5.2). Expansion of the δD scale by a factor of 2 has given a spherical representation according to Chapter 1, Equation 1.9. Units are (cal/cm3)1/2.

have reasonably similar parameters. Based on reasonable similarity with other solubility correlations for water-soluble polymers, one anticipates spheres with a radius much larger than the distance between these solvents. This result is not good and should not be used. Another example of determining HSP for a polymer with very high solubility parameters is Dextran C (British Drug Houses). Only 5 out of 50 solvents were found to dissolve Dextran C.18 In this case, there was enough spread in the solubility parameters of the test solvents such that the spherical model correlation (Chapter 1, Equation 1.9) forced the program to find a radius of 17.4 MPa1/2. This appears to be a reasonable number for this situation. The problem can be made clearer by noting the dissolving solvents with their RED numbers in parentheses. These were dimethyl sulfoxide (1.000), ethanolamine (0.880), ethylene glycol (0.880), formamide (0.915), and glycerol (0.991). Some dissolving liquids had RED equal to 1.0 or higher and included diethylene glycol (1.000), propylene glycol (1.053), and 1,3-butanediol (1.054). These helped to define the boundary of the Hansen solubility sphere. Note that the HSP for the polymer are in a region higher than that defined by the values of test liquids. Any technique using an average of the HSP for the test solvents will inherently underestimate the solute HSP in such a situation. The solubility data for the polymer Dextran C led to the HSP data reported in Table 5.3 when the SPHERE program used a starting point based on averages of the HSP values for the good solvents. When the starting point was 25 MPa1/2 for D, P, H, and Ro, respectively, a perfect data fit was found for D, P, H, and Ro equal to 26, 26, 26, and 24, all in MPa1/2. When the starting point was for D, P, H, and Ro equal to 30, 30,

7248_C005.fm Page 102 Wednesday, May 23, 2007 10:51 AM

102

Hansen Solubility Parameters: A User’s Handbook

TABLE 5.1 Calculated vs. Trial-and-Error Solubility Parameter Data for Various Polymersa Computed Handtrials A B C D E F G H I J K L M N O P Q

(First Values) (Second Values) δD δP δH

Ro

FIT

8.60 9.2 9.11 9.2 9.95 8.5 9.98 8.5 9.79 9.4 9.09 8.5 10.40 8.6 10.23 9.3 10.17 9.5 7.53 7.0 9.90 9.3 9.08 9.5 11.37 9.0 9.65 9.4 10.62 8.9 8.58 8.5 9.87

5.20 5.3 4.20 4.0 6.20 4.7 6.70 5.3 5.70 5.0 5.20 4.8 6.20 3.5 6.70 4.9 6.20 4.7 5.60 5.5 5.20 4.2 3.70 4.5 9.70 6.4 6.20 5.5 7.70 4.5 5.20 5.0 5.70

0.960 0.923 0.945 0.923 0.853 0.829 0.974 0.957 0.930 0.929 0.948 0.871 0.955 0.915 0.891 0.855 0.924 0.909 0.933 0.918 0.949 0.933 0.921 0.896 0.978 0.923 0.897 0.867 1.000 0.952 0.942 0.940 0.942

4.72 5.3 5.14 5.0 5.88 5.5 1.68 2.5 2.84 3.2 2.13 4.3 2.81 3.0 5.51 5.0 4.05 4.0 7.20 7.0 3.09 3.7 6.22 6.0 3.20 4.0 5.68 5.3 0.46 3.0 4.58 4.7 6.43

1.94 2.1 3.67 4.2 5.61 5.5 2.23 3.0 5.34 5.1 6.37 5.5 2.10 2.0 4.72 4.0 7.31 6.4 4.32 4.3 2.64 2.1 5.38 6.0 4.08 5.5 7.13 7.4 4.17 3.8 7.00 6.5 6.39

Computed Handtrials

R S T U V X Y Z A B C D E F G L

(First Values) (Second Values) δD δP 9.04 9.2 10.53 8.8 8.58 8.7 9.10 9.3 8.10 8.5 7.10 7.8 8.57 8.8 8.52 8.2 9.60 8.7 9.95 9.5 8.05 8.5 10.34 9.2 8.58 8.5 8.91 9.4 9.49 8.8 9.86 10.8 9.3

4.50 4.5 7.30 7.0 1.64 1.8 4.29 4.5 0.69 1.5 1.23 1.0 1.10 2.5 –0.94 0.8 2.31 2.5 4.17 4.0 0.18 1.0 6.63 5.8 0.58 1.5 3.68 4.5 2.68 2.7 7.14 7.0 6.2

δH

Ro

FIT

2.40 2.6 6.00 5.5 1.32 1.8 2.04 2.0 –0.40 1.5 2.28 3.6 1.67 1.2 7.28 5.7 3.80 3.5 5.20 5.5 1.39 2.0 6.26 4.2 1.76 1.8 4.08 3.5 2.82 2.7 7.35 8.8 4.7

5.20 5.0 8.20 6.0 3.20 3.5 4.70 4.7 4.70 3.4 6.20 4.0 3.20 3.8 4.70 2.9 5.20 4.2 7.20 7.0 4.20 3.4 6.70 5.0 3.20 2.6 1.70 3.2 4.70 4.0 5.70 7.1 4.2

0.985 0.972 0.910 0.879 0.974 0.965 0.969 0.950 0.974 0.964 0.921 0.881 0.950 0.914 0.971 0.954 0.942 0.951 0.980 0.976 0.966 0.960 0.964 0.868 0.968 0.956 0.992 0.895 0.961 0.963 0.970 0.936 0.892

Note: Units are (cal/cm3)1/2. a

See Table 5.2 for polymer types.

Source: From Hansen, C.M., Färg och Lack, 17(4), 73, 1971. With permission.

30, and 30, all in MPa1/2, a perfect correlation was found to D, P, H, and Ro equal to 30, 28, 28, and 32, all in MPa1/2. These data show that extrapolations into regions where there are no data can be problematic. It is thought that the data given in Table 5.3 for Dectran C are the most representative, because of the data fit being slightly less than 1.0 giving a better definition of a boundary. The properties of good solvents alone cannot always lead to a good estimate of the solubility parameters for these polymers, and the radii of spheres using only a few solvents with high solubility parameters will be very uncertain. One can sometimes find better results by correlating degrees of swelling or uptake, rather than correlate on solubility or not. The work of Zellers and coworkers

7248_C005.fm Page 103 Wednesday, May 23, 2007 10:51 AM

Methods of Characterization — Polymers

103

TABLE 5.2 List of Polymers and Resins Studied A B C D E F G H I J K L M N O P Q R S T U V X Y Z A B C D E F G L

Lucite® 2042-poly (ethyl methacrylate), E. I. du Pont de Nemours & Co., Inc. Poly (methyl methacrylate), Rohm and Haas Co. Epikote® 1001-epoxy, Shell Chemical Co. Plexal P65-66% oil length alkyd, Polyplex. Pentalyn® 830-alcohol soluble rosin resin, Hercules Incorporated. Butvar® B76-poly (vinyl butyral), Shawinigan Resins Co. Polystyrene LG, Badische Anilin- und Soda Fabrik. Mowilith® 50-poly (vinyl acetate), Farbwerke Hoechst. Plastopal H-urea formaldehyde resin, Badische Anilin- und Soda Fabrik. H Sec. Nitrocellulose-H 23, A. Hagedorn and Co. Parlon® P10-chlorinated poly (propylene), Hercules Incorporated. Cellulose acetate, Cellidora A-Bayer AG. Super Beckacite® 1001-Pure Phenolic Resin, Reichhold Chemicals Co. Phenodur 373U-phenol-resol resin, Chemische Werke Albert. Cellolyn 102-modified pentaerythritol ester of rosin, Hercules Incorporated. Pentalyn 255-alcohol soluble resin, Hercules Incorporated. Suprasec F5100-blocked isocyanate (phenol), Imperial Chemical Ind. Ltd. Plexal C34-34% coconut oil-phthalic anhydride alkyd, Polyplex. Desmophen 850, Polyester-Farbenfabriken Bayer AG. Polysar 5630 — styrene-butadiene (SBR) raw elastomer, Polymer Corp. Hycar® 1052-acrylonitrile-butadiene raw elastomer, B. F. Goodrich Chemical Corp. Cariflex IR 305-isoprene raw elastomer, Shell Chemical Co. Lutanol IC/123-poly (isobutylene), Badische Anilin- und Soda Fabrik. Buna Huls CB 10-cis poly butadiene raw elastomer, Chemische Werke Huels. Versamid® 930-polyamide, General Mills, Inc. Ester gum BL, Hercules Incorporated. Cymel® 300-hexamethoxy melamine, American Cyanamid Co. Piccolyte® S100-terpene resin, Pennsylvania Industrial Chemical Corp. Durez® 14383-furfuryl alcohol resin, Hooker Chemical Co. Piccopale® 110-petroleum hydrocarbon resin, Pennsylvania Industrial Chemical Corp. Vipla KR-poly (vinyl chloride), K = 50, Montecatini. Piccoumarone 450L-cumarone-indene resin, Pennsylvania Industrial Chemical Corp. Milled wood lignin — special sample from Prof. A. Björkman.

TABLE 5.3 Hansen Solubility Parameter Correlations for Selected Materials Material

δD

δP

δH

Ro

FIT

G/T

PES solubility PAN solubility PP swelling Polyvinyl alcohol ? (see text) Hexamethylphosphoramide PVDC melting temperature 110°C PVDC melting temperature 130°C Dextran C solubility

19.6 21.7 18.0 17.0 18.5 17.6 20.4 24.3

10.8 14.1 3.0 9.0 8.6 9.1 10.0 19.9

9.2 9.1 3.0 18.0 11.3 7.8 10.2 22.5

6.2 10.9 8.0 4.0 — 3.9 7.6 17.4

0.999 0.931 1.00 1.00 — 0.992 0.826 0.999

5/41 13/36 13/21 2/56 — 6/24 13/24 5/50

Note: Units are (cal/cm3)1/2.

7248_C005.fm Page 104 Wednesday, May 23, 2007 10:51 AM

104

Hansen Solubility Parameters: A User’s Handbook

TABLE 5.4A Calculated Solubility SPHERE for PES Solubility D = 19.6 P = 10.8 H = 9.2 RAD = 6.2 FIT = 0.999 NO = 41 Solvent

D

P

H

SOLUB

RED

V

Acetone Acetophenone Benzene 1-Butanol Butyl acetate γ-Butyrolactone Carbon tetrachloride Chlorobenzene Chloroform Cyclohexanol Diacetone alcohol o-Dichlorobenzene Diethylene glycol Diethyl ether Dimethyl formamide Dimethyl sulfoxide 1,4-Dioxane Ethanol Ethanolamine Ethyl acetate Ethylene dichloride Ethylene glycol Ethylene glycol monobutyl ether Ethylene glycol monoethyl ether Ethylene glycol monomethyl ether Formamide Hexane Isophorone Methanol Methylene dichloride Methyl ethyl ketone Methyl isobutyl ketone Methyl-2-pyrrolidone Nitroethane Nitromethane 2-Nitropropane Propylene carbonate Propylene glycol Tetrahydrofuran Toluene Trichloroethylene

15.5 19.6 18.4 16.0 15.8 19.0 17.8 19.0 17.8 17.4 15.8 19.2 16.6 14.5 17.4 18.4 19.0 15.8 17.0 15.8 19.0 17.0 16.0 16.2 16.2 17.2 14.9 16.6 15.1 18.2 16.0 15.3 18.0 16.0 15.8 16.2 20.0 16.8 16.8 18.0 18.0

10.4 8.6 0.0 5.7 3.7 16.6 0.0 4.3 3.1 4.1 8.2 6.3 12.0 2.9 13.7 16.4 1.8 8.8 15.5 5.3 7.4 11.0 5.1 9.2 9.2 26.2 0.0 8.2 12.3 6.3 9.0 6.1 12.3 15.5 18.8 12.1 18.0 9.4 5.7 1.4 3.1

7.0 3.7 2.0 15.8 6.3 7.4 0.6 2.0 5.7 13.5 10.8 3.3 20.7 5.1 11.3 10.2 7.4 19.4 21.2 7.2 4.1 26.0 12.3 14.3 16.4 19.0 0.0 7.4 22.3 6.1 5.1 4.1 7.2 4.5 5.1 4.1 4.1 23.3 8.0 2.0 5.3

0 1 0 0 0 1 0 0 0 0 0 0 0 0 1 0*a 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 1 0 0 0 0 0 0 0 0

1.371 0.955 2.129 1.777 1.741 0.998 2.301 1.576 1.483 1.467 1.321 1.204 2.101 2.183 0.915 0.996 1.493 2.077 2.241 1.547 1.007 2.837 1.563 1.395 1.618 3.044 2.745 1.094 2.575 0.990 1.368 1.782 0.655 1.580 1.899 1.387 1.429 2.457 1.237 1.978 1.485

74.0 117.4 89.4 91.5 132.5 76.8 97.1 102.1 80.7 106.0 124.2 112.8 94.9 104.8 77.0 71.3 85.7 58.5 59.8 98.5 79.4 55.8 131.6 97.8 79.1 39.8 131.6 150.5 40.7 63.9 90.1 125.8 96.5 71.5 54.3 86.9 85.0 73.6 81.7 106.8 90.2

Note: Units are MPa1/2. a

Outlier (a bad solvent lying inside sphere).

7248_C005.fm Page 105 Wednesday, May 23, 2007 10:51 AM

Methods of Characterization — Polymers

105

ALTERNATE TABLE 5.4B Calculated Solubility SPHERE for PES Solubility (Listed in RED Order) D = 19.6 P = 10.8 H = 9.2 RAD = 6.2 FIT = 0.999 NO = 41 Solvent

D

P

H

SOLUB

RED

V

Methyl-2-pyrrolidone Dimethyl formamide Acetophenone Methylene dichloride Dimethyl sulfoxide γ-Butyrolactone Ethylene dichloride Isophorone o-Dichlorobenzene Tetrahydrofuran Diacetone alcohol Methyl ethyl ketone Acetone 2-Nitropropane Ethylene glycol monoethyl ether Propylene carbonate Cyclohexanol Chloroform Trichloroethylene 1,4-Dioxane Ethyl acetate Ethylene glycol monobutyl ether Chlorobenzene Nitroethane Ethylene glycol monomethyl ether Butyl acetate 1-Butanol Methyl isobutyl ketone Nitromethane Toluene Ethanol Diethylene glycol Benzene Diethyl ether Ethanolamine Carbon tetrachloride Propylene glycol Methanol Hexane Ethylene glycol Formamide

18.0 17.4 19.6 18.2 18.4 19.0 19.0 16.6 19.2 16.8 15.8 16.0 15.5 16.2 16.2 20.0 17.4 17.8 18.0 19.0 15.8 16.0 19.0 16.0 16.2 15.8 16.0 15.3 15.8 18.0 15.8 16.6 18.4 14.5 17.0 17.8 16.8 15.1 14.9 17.0 17.2

12.3 13.7 8.6 6.3 16.4 16.6 7.4 8.2 6.3 5.7 8.2 9.0 10.4 12.1 9.2 18.0 4.1 3.1 3.1 1.8 5.3 5.1 4.3 15.5 9.2 3.7 5.7 6.1 18.8 1.4 8.8 12.0 0.0 2.9 15.5 0.0 9.4 12.3 0.0 11.0 26.2

7.2 11.3 3.7 6.1 10.2 7.4 4.1 7.4 3.3 8.0 10.8 5.1 7.0 4.1 14.3 4.1 13.5 5.7 5.3 7.4 7.2 12.3 2.0 4.5 16.4 6.3 15.8 4.1 5.1 2.0 19.4 20.7 2.0 5.1 21.2 0.6 23.3 22.3 0.0 26.0 19.0

1 1 1 1 0*a 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

0.655 0.915 0.955 0.990 0.996 0.998 1.007 1.094 1.204 1.237 1.321 1.368 1.371 1.387 1.395 1.429 1.467 1.483 1.485 1.493 1.547 1.563 1.576 1.580 1.618 1.741 1.777 1.782 1.899 1.978 2.077 2.101 2.129 2.183 2.241 2.301 2.457 2.575 2.745 2.837 3.044

96.5 77.0 117.4 63.9 71.3 76.8 79.4 150.5 112.8 81.7 124.2 90.1 74.0 86.9 97.8 85.0 106.0 80.7 90.2 85.7 98.5 131.6 102.1 71.5 79.1 132.5 91.5 125.8 54.3 106.8 58.5 94.9 89.4 104.8 59.8 97.1 73.6 40.7 131.6 55.8 39.8

Note: Units are MPa1/2. a

Outlier (a bad solvent lying inside sphere).

7248_C005.fm Page 106 Wednesday, May 23, 2007 10:51 AM

106

Hansen Solubility Parameters: A User’s Handbook

reports extensive studies of this type.11–14 It should be noted, however, that the HSP-sphere parameters usually vary some from correlation to correlation based on the same data when different criteria are used for good and bad solvents. This is because the absorbed solvent tends to locate in regions with similar solubility parameters, and there are local variations in HSP within most, if not all, polymers. This is particularly true of polymers which are not homopolymers. This situation relates to self-assembly. Solvents or segments of molecules with similar HSP will tend to reside near each other. An example of this is water residing at local hydrophilic sites, such as alcohol groups, in polymers. Utilization of the HSP affinity between molecules or segments of molecules is a viable way to control self-assembly. See also Chapter 18.

SWELLING — EXAMPLES The correlation for swelling of polypropylene reported in Table 5.3 is based on solvent uptake data reported by Lieberman and Barbe at 22°C.19 The limit of 0.5% was arbitrarily set to differentiate good solvents from bad ones. As mentioned earlier, experience has shown that a different limit usually gives different parameters. It should be noted that swelling data reflect the properties of the regions in the polymer where the solvent has chosen to reside because of energetic similarity (self-assembly). The principle is not necessarily “like dissolves like,” but rather “like seeks like.” If the solvent is homogeneously distributed in the polymer, the solubility parameters found will reflect the properties of the whole polymer. Crystalline regions will not contain solvent. If the solvent collects locally in regions with chemical groups different from the bulk of the polymer, then the HSP so derived will reflect at least partially the physical nature of these chemical groups. The parameters reported in Table 5.3 seem appropriate for what is expected in terms of low polarity and low hydrogen-bonding properties for a polypropylene-type polymer. An example of a characterization using swelling data that did not result in a good correlation is that for Viton® (The Du Pont Company, Wilmington, DE). This problem has been discussed by Zellers and Zhang11,12 and is also discussed in Chapter 13. If one tries to force-fit data where there are several different comonomers into a single HSP sphere, the result is usually reflected in a poor correlation coefficient. Figure 13.3 shows that improvements can be made by using a separate sphere for each comonomer. One reason for the poor correlation of swelling behavior is that Viton is not a homopolymer, and also contains a cross-linking chemical. The different segments have different affinities. Indeed, there are several qualities of Viton, each of which has significantly differing chemical resistance. Swelling of Viton has also been treated by Evans and Hardy20 in connection with predictions related to chemical protective clothing, and by Nielsen and Hansen,21 who presented curves of swelling as a function of the RED number.

MELTING POINT DETERMINATIONS — EFFECT OF TEMPERATURE Partly crystalline polymers that are placed in different liquids will have melting points which are lowered to a degree depending on the solvent quality of the individual liquids. The melting points of polyvinylidine chloride (PVDC) have been measured in different solvents.22 These data have been analyzed by evaluating solubility parameter regions based on those solvents which dissolve the polymer at 110°C and above and also at 130°C and above. As expected, there are more solvents which dissolve the semicrystalline polymer at the higher temperature. The results for these correlations are included in Table 5.3. The main reasons for the somewhat lower data fit at 130°C include two nondissolving solvents within the solubility parameter sphere. These are dimethyl phthalate, where the large molecular size is a factor, and benzyl alcohol, where temperature effects can be larger than expected compared with the other solvents as discussed later and in Chapter 1. The solubility parameters for PVDC at this temperature, based on tabulated solvent values at 25°C, are not affected significantly by this type of situation. A single room temperature solvent for PVDC

7248_C005.fm Page 107 Wednesday, May 23, 2007 10:51 AM

Methods of Characterization — Polymers

107

is reported by Wessling.22 This is hexamethylphosphoramide and its solubility parameters are also reported in Table 5.3 for comparison. The change in the values of the individual solubility parameters with temperature is discussed in Chapter 1 (Equation 1.16 to Equation 1.18). Chapter 3 also treats the temperature dependence of the HSP. See also Chapter 10 where the HSP of specifically carbon dioxide are treated in depth as a function of temperature and pressure.

ENVIRONMENTAL STRESS CRACKING Environmental stress cracking (ESC) is unfortunately a very frequent mode of failure for plastics. For this reason a whole chapter is devoted to the topic (see Chapter 14). It has been possible to correlate HSP with ESC phenomena, and this can also provide an estimate of the HSP for the given polymers. Care is advised since the stress/strain level is important, as is the molecular size and shape of the chemicals involved. Several collections of ESC data in the older literature23–25 should not be forgotten in these days of “Internet and only Internet.” Such collections have particular value as it is considered impossible to get a commercially available polymer without some additives. These can also affect ESC behavior. These older data were the basis of the ESC correlations given in Chapter 14.

INTRINSIC VISCOSITY MEASUREMENTS One of the more promising methods to evaluate polymer HSP for limited data is that using the intrinsic viscosity. Van Dyk et al. found a correlation with the intrinsic viscosity of an acrylic polymer (polyethyl methacrylate) in various solvents and the polymer HSP26 (see the discussion on polymer compatibility in Chapter 8). Segarceanu and Leca27 have devised a method to calculate the polymer HSP from data on its intrinsic viscosity in different solvents. The intrinsic viscosities will be higher in the better solvents because of greater interaction and greater polymer chain extension. The intrinsic viscosity gives an indication of the solvent quality. It has been used earlier to calculate the Flory–Huggins chi parameter, for example.28 In the new technique, the intrinsic viscosities are normalized by the intrinsic viscosity of that solvent giving the highest value. These normalized data (numbers are 1.0 or less) are then used in a weighted averaging technique to arrive at the center of the Hansen sphere. δD2 = Σ(δDi × [η]i)/Σ[η]i

(5.4)

δP2 = Σ(δPi × [η]i)/Σ[η]i

(5.5)

δH2 = Σ(δHi × [η]i]/Σ[η]i

(5.6)

The subscript 2 is for the polymer, and the respective solvents are indicated by an “i.” The intrinsic viscosity in the i-th solvent is given by [η]i. Those solvents with the greatest weighting factor have higher intrinsic viscosities and are closest to the geometric center of the sphere. Those solvents which do not dissolve the polymer were assumed to have a zero weighting factor. The HSP for a polyesterimide were reported as an example. HSP values were assigned both by the “classical” evaluation and with this newer approach. These data are included as the first entries in Table 5.5. This is a very promising method of arriving at the polymer HSP with limited data. There are several aspects of this work which deserve comment. It was demonstrated earlier that many polymers have higher solubility parameters than any of the solvents which are or can be used to test them. The present method only allows for polymer HSP within the range attainable

7248_C005.fm Page 108 Wednesday, May 23, 2007 10:51 AM

108

Hansen Solubility Parameters: A User’s Handbook

TABLE 5.5 HSP Data for the Same Polyesterimide Polymer Based on Data Given in Reference 27 Correlation

δD

δP

δD

Ro

FIT

Classicala1 Newa1 HSP SPHEREa HSP SPHEREb Classicala Classicalb Newa Newb

17.4 18.0 20.0 19.0

12.3 11.1 11.0 11.0

8.6 8.8 10.0 9.0

4.1 8.6 8.3 7.0

— — 1.000 1.000 0.426 0.447 0.506 0.364

Note: Units are MPa1/2. a

Indicates use of the solubility parameters for the solvents given in Reference 27. b Indicates use of the solvent HSP data in the author’s files.

by the test solvents. The method will lead to values that are too low in some cases, including the example with the polyesterimide used as an example in Segarceanu and Leca.27 It is not surprising that the polymer HSP are often higher than solvent HSP, as they are in a physical state between that of a liquid and a solid. When the cohesion energy becomes too high, a material is a solid rather than a liquid. Low molecular weight solids frequently have HSP somewhat higher than the HSP of liquids. Many examples can be given, including urea, ethylene carbonate, etc. When the data (as soluble or not) for the 11 solvents were processed by the SPHERE computer program, the parameters found were those given by the third set of HSP in Table 5.5. The agreement with the “new” method is acceptable, even though none of the test solvents have δd as high as that of the polymer. Further inspection showed that the solubility parameters used in the study were not in agreement with those published in the latest reference to Hansen listed by Segarceanu and Leca.27 It also appears that the radius of the HSP sphere for the classical determination is in error, being far too low. To further clarify the situation, several runs with the SPHERE program were done with the parameters listed in this book, as well as with those listed in the article being discussed. In both cases the data fit is not good for the HSP reported by Segarceanu and Leca.27 In the classical case, the data fit is only 0.426 (1.0 is perfect), and four of the five good solvents are located outside of the sphere. Only N-methyl-2-pyrrolidone is inside. In the new case, the data fit is not much better, being 0.506. Here, four of the five bad solvents are inside the sphere with only one being outside. It has been possible to estimate the polymer parameters within acceptable variation, but the radius of the sphere has not been accounted for in a satisfactory manner. Further inspection of the data suggests that morpholine, the solvent with the highest [η] that was used to normalize the data, is not as good as might have been expected from the intrinsic viscosity data. This can be seen in Table 5.6. The reason for this is unknown, but experience has shown that amines often are seen to react with various materials in a manner which does not allow their inclusion in correlations of the type discussed here. To conclude this section, it is noted that a similar weighting technique was used by Zellers et al.13,14 where the weighted measurements were solvent uptake by elastomers customarily used to make chemical protective clothing. The same precautions must be taken in analyzing this type of

7248_C005.fm Page 109 Wednesday, May 23, 2007 10:51 AM

Methods of Characterization — Polymers

109

TABLE 5.6 Calculated Solubility SPHERE for Polyesterimide (Listed in RED Order) D = 19.0 P = 11.0 H = 9.0 RAD = 7.0 FIT = 1.000 NO = 11 Solvent

[η]aN

D

P

H

SOLUB

RED

V

Methyl-2-pyrrolidone Dimethyl formamide Dimethyl sulfoxide γ-Butyrolactone Morpholine Cyclohexanone Diacetone alcohol Acetone Diethylene glycol monomethyl ether Ethylene glycol monoethyl ether Ethylene glycol monoethyl ether acetate

0.970 0.947 0.182 0.689 1.000 0.718 — — — — —

18.0 17.4 18.4 19.0 18.8 17.8 15.8 15.5 16.2 16.2 15.9

12.3 13.7 16.4 16.6 4.9 6.3 8.2 10.4 7.8 9.2 4.7

7.2 11.3 10.2 7.4 9.2 5.1 10.8 7.0 12.6 14.3 10.6

1 1 1 1 1 1 0 0 0 0 0

0.427 0.682 0.809 0.832 0.874 0.937 1.031 1.044 1.055 1.131 1.283

96.5 77.0 71.3 76.8 87.1 104.0 124.2 74.0 118.0 97.8 136.1

Note: Units are MPa1/2. a

Normalized intrinsic viscosity data from Reference 27.

measurement, but as the polymers studied were reasonably nonpolar, some of the solvents had HSP which were higher than those of the polymers studied. Zellers et al.14 and Athey29 also describe multiple variable statistical analysis techniques to find the HSP of a given polymer. Barton’s work6 contains many literature sources of intrinsic viscosity studies using the solubility parameter for interpretation.

OTHER MEASUREMENT TECHNIQUES There are many other techniques to differentiate between the behavior of different solvents in contact with a polymer. Many of these are discussed in the following chapters and will not be treated here. These include permeation measurements, chemical resistance determinations of various kinds including ESC, and surface attack, etc. Some of the techniques can be very useful, depending on the polymer involved. Others will present problems because of the probable influence of other factors such as solvent molar volume and length of time before attainment of equilibrium. Several of these phenomena can be correlated with HSP, but the techniques used in the measurements will present problems in using the data for direct HSP characterization of polymers because other effects are also important.

CONCLUSION HSP for polymers can be evaluated experimentally by correlations of data where a suitably large number of well-chosen solvents are brought into contact with the polymer. The observed behavior which can be correlated includes true solubility, swelling, weight gain, dimensional change, degree of surface attack, reduction of melting point, permeation rate, breakthrough time, and tensile strength reduction. Correlations for simple evaluations of chemical resistance of the suitable-ornot type and ESC are also possible. In each case, the molecular size of the liquids used can affect the result and should be considered in some way. The use of water as a test liquid is not recommended for these purposes.

7248_C005.fm Page 110 Wednesday, May 23, 2007 10:51 AM

110

Hansen Solubility Parameters: A User’s Handbook

REFERENCES 1. Hansen, C.M., Solubility in the coatings industry, Färg och Lack, 17(4), 69–77, 1971. 2. Hansen, C.M. and Skaarup, K., The three dimensional solubility parameter — key to paint component affinities III. Independent calculation of the parameter components, J. Paint Technol., 39(511), 511–514, 1967. 3. Hansen, C.M., The Three Dimensional Solubility Parameter and Solvent Diffusion Coefficient, Their Importance in Surface Coating Formulation, Doctoral dissertation, Danish Technical Press, Copenhagen, 1967. 4. Teas, J.P., Graphic analysis of resin solubilities, J. Paint Technol., 40(516), 19–25, 1968. 5. Gardon, J.L. and Teas, J.P., Solubility parameters, in Treatise on Coatings, Vol. 2, Characterization of Coatings: Physical Techniques, Part II, Myers, R.R. and Long, J.S., Eds., Marcel Dekker, New York, 1976, chap. 8. 6. Barton, A.F.M., Handbook of Solubility Parameters and Other Cohesion Parameters, CRC Press, Boca Raton, FL, 1983; 2nd ed., 1991. 7. Torraca, G., Solubility and Solvents for Conservation Problems, 2nd ed., International Centre for the Study of the Preservation and the Restoration of Cultural Property (ICCROM), Rome, 1978. 8. Barton, A.F.M., Handbook of Polymer-Liquid Interaction Parameters and Solubility Parameters, CRC Press, Boca Raton, FL, 1990. 9. Benjamin, S., Carr, C., and Wallbridge, D.J., Self-stratifying coatings for metallic substrates, Prog. Org. Coat., 28(3), 197–207, 1996. 10. Rasmussen, D. and Wahlström, E., HSP — solubility parameters: a tool for development of new products — modelling of the solubility of binders in pure and used solvents, Surf. Coat. Int., 77(8), 323–333, 1994. 11. Zellers, E.T., Three-dimensional solubility parameters and chemical protective clothing permeation. I. Modeling the solubility of organic solvents in Viton® gloves, J. Appl. Polym. Sci., 50, 513–530, 1993. 12. Zellers, E.T. and Zhang, G.-Z., Three-dimensional solubility parameters and chemical protective clothing permeation. II. Modeling diffusion coefficients, breakthrough times, and steady-state permeation rates of organic solvents in Viton® gloves, J. Appl. Polym. Sci., 50, 531–540, 1993. 13. Zellers, E.T., Anna, D.H., Sulewski, R., and Wei, X., Critical analysis of the graphical determination of Hansen’s solubility parameters for lightly crosslinked polymers, J. Appl. Polym. Sci., 62, 2069–2080, 1996. 14. Zellers, E.T., Anna, D.H., Sulewski, R., and Wei, X., Improved methods for the determination of Hansen’s solubility parameters and the estimation of solvent uptake for lightly crosslinked polymers, J. Appl. Polym. Sci., 62, 2081–2096, 1996. 15. Luciani, A., Champagne, M.F., and Utracki, L.A., Interfacial tension in polymer blends. Part 1: Theory, Polym. Networks Blends, 6(1), 41–50, 1996. 16. Luciani, A., Champagne, M.F., and Utracki, L.A., Interfacial tension in polymer blends. Part 2: Measurements, Polym. Networks Blends, 6(2), 51–62, 1996. 17. Fuchs, O., Solvents and non-solvents for polymers, in Polymer Handbook, 3rd ed., Brandrup, J. and Immergut, E.H., Eds., Wiley-Interscience, New York, 1989, p. VII/385. 18. Hansen, C.M., The universality of the solubility parameter, Ind. Eng. Chem. Prod. Res. Dev., 8(1), 2–11, 1969. 19. Lieberman, R.B. and Barbe, P.C., Polypropylene polymers, in Encyclopedia of Polymer Science and Engineering, 2nd ed., Vol. 13, Mark, H.F., Bikales, N.M., Overberger, C.G., Menges, G., and Kroschwitz, J.I., Eds., Wiley-Interscience, New York, 1988, pp. 482–483. 20. Evans, K.M. and Hardy, J.K., Predicting solubility and permeation properties of organic solvents in Viton glove material using Hansen’s solubility parameters, J. Appl. Polym. Sci., 93, 2688–2698, 2004. 21. Nielsen, T.B. and Hansen, C.M., Elastomer swelling and Hansen solubility parameters, Polym. Testing, 24, 1054–1061, 2005. 22. Wessling, R.A., The solubility of poly(vinylidine chloride), J. Appl. Polym. Sci., 14, 1531–1545, 1970. 23. Wyzgoski, M.G., The role of solubility in stress cracking of nylon 6,6, in Macromolecular Solutions, Seymour, R.B. and Stahl, G.A., Eds., Pergamon Press, New York, 1982, pp. 41–60. 24. Mai, Y.-W., Environmental stress cracking of glassy polymers and solubility parameters, J. Mater. Sci., 21, 904–916, 1986.

7248_C005.fm Page 111 Wednesday, May 23, 2007 10:51 AM

Methods of Characterization — Polymers

111

25. White, S.A., Weissman, S.R., and Kambour, R.P., Resistance of a polyetherimide to environmental stress crazing and cracking, J. Appl. Polym. Sci., 27, 2675–2682, 1982. 26. Van Dyk, J.W., Frisch, H.L., and Wu, D.T., Solubility, solvency, and solubility parameters, Ind. Eng. Chem. Prod. Res. Dev., 24(3), 473–478, 1985. 27. Segarceanu, O. and Leca, M., Improved method to calculate Hansen solubility parameters of a polymer, Prog. Org. Coat., 31(4), 307–310, 1997. 28. Kok, C.M. and Rudin, A., Prediction of Flory-Huggins interaction parameters from intrinsic viscosities, J. Appl. Polym. Sci., 27, 353–362, 1982. 29. Athey, R.D., Testing coatings: 6. Solubility parameter determination, Eur. Coat. J., 5, 367–372, 1993.

7248_C005.fm Page 112 Wednesday, May 23, 2007 10:51 AM

7248_C006.fm Page 113 Wednesday, May 23, 2007 11:00 AM

of Characterization 6 Methods — Surfaces Charles M. Hansen ABSTRACT Relations between cohesion parameters and surface energy parameters and their practical significance are discussed. Cohesion parameters (solubility parameters) can be used with full theoretical justification to characterize many surfaces, including substrates, coatings, plastics, pigment and filler surfaces, etc., in addition to the binder or polymer used in a given product. Important molecular relations between a binder in a coating or adhesive and its surroundings then become obvious. Use of cohesion parameters, i.e., Hansen solubility parameters in a total characterization of surface energy, clearly shows how the single point concepts of the (Zisman) critical surface tension and the wetting tension fit into a larger energy concept. A complete match of surface energies of two surfaces requires that exactly the same liquids (in a larger number of well-chosen test liquids) spontaneously spread on both surfaces. The dewetting behavior (wetting tension test) of the liquids must also be the same, in that the same liquids should not retract when applied to the surfaces as films.

INTRODUCTION Interfacial free energy and adhesion properties result from intermolecular forces. It has been recognized for many years that molecules interact by (molecular) surface to (molecular) surface contacts to enable solutions to be formed.1 As molecular surface-to-surface contacts control both solution phenomena and surface phenomena, it is not surprising that various correlations of cohesion parameters and surface phenomena can be found. This idea has been well explored and dealt with elsewhere.2 The various treatments and correlations in the literature will not be explicitly dealt with here, other than those directly related to Hansen solubility parameters (HSP). In this chapter, solubility parameters are called cohesion (energy) parameters and refer more specifically to HSP. Solubility as such does not necessarily enter into the energetics of interfacial phenomena, but the energy characteristics of surfaces can still be correlated with HSP. This chapter will emphasize methods of surface characterization using HSP. The orientation of adsorbed molecules is a significant added effect that must also be considered in many cases. The “like dissolves like” concept is extended and applied as “like seeks like” (self-assembly).

HANSEN SOLUBILITY PARAMETER CORRELATIONS WITH SURFACE TENSION (SURFACE FREE ENERGY) Skaarup was the first to establish a correlation between liquid surface tension and HSP. This correlation with surface tension had been long lost in an internal report to members of the Danish Paint and Printing Ink Research Laboratory in 1967, as well as in an abstract for a presentation to the Nordic Chemical Congress in 1968.3,4 γ = 0.0688V1/3[δD2 + k(δP2 + δH2)]

(6.1)

113

7248_C006.fm Page 114 Wednesday, May 23, 2007 11:00 AM

114

Hansen Solubility Parameters: A User’s Handbook

γ is the surface tension, and k is a constant depending on the liquids involved. This k was reported as 0.8 for several homologous series, 0.265 for normal alcohols, and 10.3 for n-alkyl benzenes. Beerbower independently published essentially the same type of correlation in 1971.5,6 With the exception of aliphatic alcohols and alkali halides, Beerbower found γ = 0.0715V1/3[δD2 + 0.632(δP2 + δH2)]

(6.2)

where γ is the surface tension. The constant was actually found to be 0.7147 in the empirical correlation. The units for the cohesion parameters are (cal/cm3)1/2, and those of the surface tension are dyn/cm in both Equation 6.1 and Equation 6.2. However, values in dyn/cm are numerically equal to those in mN/m. The constant was separately derived as being equal to 0.7152 by a mathematical analysis in which the number of nearest neighbors lost in surface formation was considered, assuming that the molecules tend to occupy the corners of regular octahedra. The correlations presented by Koenhen and Smolders7 are also relevant to estimating surface tension from HSP. The author has never explored them in detail, however, so they are not discussed here. It is interesting to note that δP and δH have the same coefficient in the surface tension correlations. They also have the same coefficient when solubility is correlated (see Chapter 1, Equation 1.9 or Chapter 2, Equation 2.6). The reason for this is the molecular orientation in the specific interactions derived from permanent dipole–permanent dipole and hydrogen bonding (electron interchange) interactions. The dispersion or London forces arise because of electrons rotating around a positive atomic nucleus. This causes local dipoles and attraction among atoms. This is a completely different type of interaction and requires a different coefficient in the correlations. It is this difference between atomic and molecular interactions that is basic to the entire discussion of similarity between HSP and the Prigogine corresponding states theory in Chapter 2. The finding that the polar and hydrogen bonding (electron interchange) effects require the same coefficient for both bulk and surface correlations suggests that the net effects of the (often mentioned) unsymmetrical nature of hydrogen bonding are no different from the net effects occurring with permanent dipole–permanent dipole interactions. The lack of specific consideration that hydrogen bonding is an unsymmetrical interaction led Erbil8 to state that HSP has limited theoretical justification, for example. The previous discussion and the contents of Chapter 1 and Chapter 2 clearly indicate that the author is not in full agreement with this viewpoint. In fact, it appears that this book presents massive experimental evidence, related both to bulk and surface phenomena, which shows that the geometric mean is valid for estimating interactions between dissimilar liquids. This includes dispersion, permanent dipole–permanent dipole, and hydrogen bonding (electron interchange) interactions.

METHOD TO EVALUATE THE COHESION ENERGY PARAMETERS FOR SURFACES One can determine the cohesion parameters for surfaces by observing whether or not spontaneous spreading is found for a series of widely different liquids. The liquids used in standard solubility parameter determinations are suggested for this type of surface characterization. It is strongly suggested that none of the liquids be a mixture, as this introduces an additional factor into the evaluations. The liquids in the series often used by the author are indicated in Chapter 5, Table 5.4 or Chapter 7, Table 7.2. Droplets of each of the liquids are applied to the surface and one simply observes what happens. If a droplet remains as a droplet, there is an advancing contact angle and the cohesion energy/surface energy of the liquid is (significantly) higher than that of the surface. The contact angle need not necessarily be measured in this simplified procedure, however. Contact angles have generally been found to increase for greater differences in cohesion parameters between the surface and liquid 9 (see also Figure 6.5). If spontaneous spreading is found, there is presumed

7248_C006.fm Page 115 Wednesday, May 23, 2007 11:00 AM

Methods of Characterization — Surfaces

115

15 18 *

Wetting Tension – ADV = O

C

19 *

10 24 *

δP

B

20 *

25

A 26

5 27 *

12 *

21 * 14 15 *

22 23

7

17 *

11 *

13 *

3 *

4

2 *

1 *

42

6 5 *

8 16 9

5

10

δH

15

20

FIGURE 6.1 HSP surface characterization of an epoxy surface showing regions of spontaneous spreading of applied droplets (A), lack of dewetting of applied films (B), and dewetting of applied films (C). Note that this characterization may not be valid for all epoxy surfaces. Units are MPa1/2. (From Hansen, C.M. and Wallström, E., J. Adhes., 15[3/4], 281, 1983. With permission.)

to be some “similarity” in the energy properties of the liquid and the surface. The apparent similarity may be misleading. As discussed in greater detail later, the fact of spontaneous spreading for a given liquid does not mean that its HSP are identical with those of the surface being tested. If a given liquid does not spontaneously spread, it can be spread mechanically as a film and be observed to see whether it retracts. This can be done according to ASTM D 2578-84 or ISO 8296:1987 (E). This test determines whether or not there is a receding contact angle under the given conditions. Figure 6.1 shows a complete energy description for an epoxy polymer surface 10,11 based on the testing procedure described previously. The Hansen polar and hydrogen bonding parameters δP and δH are used to report the data. Further explanation of these parameters themselves can be found in Chapter 1. The circular lines can be considered as portraying portions of HSP spheres, but the third Hansen parameter, δD, has not been specifically accounted for in the two-dimensional figure. Figure 6.1 shows two curves that are concave toward the origin. The lower of these divides the test liquids into two groups based on spontaneous spreading or not. Below the line one finds that liquids applied as droplets will spontaneously spread. Liquids that are found in the region above the upper curve will retract when applied as films. A test method to determine this is found in the ASTM and ISO standards given previously, for example, except that one uses a large number of pure liquids instead of the liquid mixtures suggested in the standards. Receding contact angles will generally increase as one progresses to liquids with still higher HSP. Intermediate between the two curves in Figure 6.1 is a region where liquids applied as droplets will remain as droplets, whereas liquids applied as films will remain as films. This region deserves more attention in future research. The energy properties of these liquids are not as close to those of the surface as are the energy properties of the liquids that spontaneously spread. Spontaneous spreading is more related to adhesion since such liquids want to cover the surface spontaneously. The wetting tension test uses

7248_C006.fm Page 116 Wednesday, May 23, 2007 11:00 AM

116

Hansen Solubility Parameters: A User’s Handbook

an external force to spread the liquids, after which they may continue to remain as a film. The mobility of the surface layer(s) will play a role in the wetting tension test. Hydrophilic segments can (perhaps) rotate toward a water droplet at some rate, for example, and increase the hydrophilic nature of the surface accordingly. This is discussed more in Chapter 18. As mentioned earlier, there is still a problem in simplifying these results for easier use and improved understanding. Hexane, for example, does not dissolve an epoxy polymer, but in Figure 6.1 it is almost in the middle of the region describing spontaneous spreading of the liquids. Hexane will not contribute to a “bite” into an epoxy coating for improving intercoat adhesion with a subsequent coating. Hexane is within the region of spontaneous spreading because it has a lower surface energy (surface tension) than the epoxy surface. Nature reduces the free energy level of the surface by requiring hexane to cover the epoxy coating. The result of this is that the center of the normal HSP sphere for describing spontaneous spreading can be assigned sizable negative values.11 This is both impractical and impossible. A better method of handling this situation is still desired, and until it is found, one must presumably refer to simple plots or other simple comparisons rather than to refined computer techniques, which are more desirable in most cases. In the meantime, interest will still be focused onto the usual test method(s) for determining surface tensions based on the Zisman critical surface tension plots (lack of advancing contact angle) or by using the ASTM procedure for wetting tension (lack of receding contact angle). The following discussion relates these to the HSP-type characterizations discussed earlier. Additional surface characterization plots for spontaneous spreading and wetting tension using HSP are included in Figure 6.2 for a plasticized polyvinyl chloride (PVC) and in Figure 6.3 for a polyethylene (PE).

A CRITICAL VIEW OF THE CRITICAL SURFACE TENSIONS 12,13 The Zisman critical surface tension is determined by measuring the extent that affinity is lacking (contact angles) for a surface using pure liquids or liquid mixtures in a series. The surface tension of each of the liquids is known. One can then plot cosine of the contact angle vs. liquid surface tension and extrapolate to the limit where the contact angle is no longer present (see Figure 6.4). Liquids with higher surface tensions than this critical value allow measurement of a contact angle, whereas liquids with lower surface tensions than the critical value will spontaneously spread. The fact that the liquid with a surface tension just under the critical value spontaneously spreads is often taken as an indication of high affinity. This is difficult to understand and appears to be a misunderstanding. The limiting critical surface tension12,13 has very little to do with the “best” solvent for the surface. It is more appropriately compared with a very poor solvent which can only marginally dissolve a polymer, for example. This is similar to the condition for a RED number equal to 1.0 discussed in Chapter 1 and Chapter 2. Measuring the critical surface tension has been and still will be a useful technique to better understand surfaces, but it should be done with the following in mind. Who would determine the solubility parameter for a polymer by the following method? One makes up a series of liquids with different, known solubility parameters. The polymer dissolves in some of them, and the degree of swelling of the polymer in question is measured in those liquids which do not dissolve it fully. One subsequently determines the solubility parameter of the polymer by extrapolating the degree of swelling to infinity, which corresponds to total solution. This extrapolation can be done by plotting 1/(degree of swelling) vs. solvent composition (solubility parameter). One now focuses attention upon that liquid which (by extrapolation) just dissolves the polymer. One assumes that there is no better solvent than this one and, consequently, assigns the polymer solubility parameters corresponding to those of this boundary solvent. This is exactly what one does when the critical surface tension is measured. This method should clearly never be used to determine solubility parameters for polymers. At the same time, it sheds some light onto the true meaning of the critical surface tension.

7248_C006.fm Page 117 Wednesday, May 23, 2007 11:00 AM

Methods of Characterization — Surfaces

117

15

18 *

19 *

10 24 * δP

20 *

Wetting Tension

25 * 26 *

12 *

21 *

5

14 15 * *

22 * 27 *

17 *

11 *

23

13 *

3 *

4 *

1 *

C B

6 5 7 *

8

2 *

A –

ADV = O

16 9 10 5

10

δH

15

20

FIGURE 6.2 HSP surface characterization of spontaneous spreading of applied droplets and wetting tension for applied films for plasticized polyvinyl chloride (PVC). Note that these characterizations may not be valid for all PVC surfaces. Units are MPa1/2 (From Hansen, C.M. and Wallström, E., J. Adhes., 15[3/4], 280, 1983. With permission.)

If we now consider the region for spontaneous spreading in Figure 6.1 to Figure 6.3, it can be seen that the critical surface tension is a point on its boundary. In practice, one finds different critical surface tensions for the same surface depending on which liquids (or liquid mixtures) are used. This is explained by the fact that the cohesion parameter regions of the type shown in Figure 6.1 to Figure 6.3 are not symmetrical around the zero axis. The individual liquid series used to determine the critical surface tension will intersect the cohesion parameter spontaneous spreading boundary at different points. The corresponding total surface tension will vary from intersection to intersection as mentioned earlier. Hansen and Wallström11 compared the critical surface tension plotting technique with one where a difference in HSP was used instead of liquid surface tension. One arrives at the same general conclusions from both types of plotting techniques. This comparison is made in Figure 6.4 and Figure 6.5.

A CRITICAL VIEW OF THE WETTING TENSION A region larger than that for spontaneous spreading will be found on a δP vs. δH plot when one plots data for those liquids that remain as films (do not break up or contract) when they are applied as films. This type of experiment measures the wetting tension. Mixtures of formamide and ethylene glycol monoethyl ether are usually used in practice for these measurements according to ASTM

7248_C006.fm Page 118 Wednesday, May 23, 2007 11:00 AM

118

Hansen Solubility Parameters: A User’s Handbook

15

18 *

19 *

10 δP

24 *

20 *

25 * 26 *

B

5

11 *

C 21 * 14 15 * *

22 * 27

12 *

23

13 *

Wetting Tension

3 *

4

2 *

1 *

– ADV = O

6 5

A

7 *

8 16 9 10 5

10

δH

15

20

FIGURE 6.3 HSP surface characterization of spontaneous spreading of applied droplets and wetting tension for applied films for a polyethylene (PE) surface. Note that these characterizations may not be valid for all PE surfaces. Units are MPa1/2. (From Hansen, C.M. and Wallström, E., J. Adhes., 15[3/4], 279, 1983. With permission.)

D 2578-84 or ISO 8296:1987 (E). One can also use the same liquids suggested earlier for cohesion parameter determinations and make a plot like that in Figure 6.1. If two different surfaces are to have the same wetting tension behavior, their plots must be the same. The results of the ASTM test are usually stated in terms of the surface tension of the liquid or liquid mixture which just stays intact as a film for 2 sec. This simple single point determination corresponds to determining a single point on the boundary of the HSP plot describing wetting tension for all liquids. A single point determination may not always be sufficient information and certainly neglects the complete picture possible from HSP considerations. Comments identical in principle to those included in the earlier section, “A Critical View of the Critical Surface Tensions,” on measurement of the critical surface tension are also valid here. It is hoped the reader now better understands the total energy context of the simple ASTM wetting tension measurements.

ADDITIONAL HANSEN SOLUBILITY PARAMETER SURFACE CHARACTERIZATIONS AND COMPARISONS Beerbower14 has reported many other correlations of surface phenomena with HSP. Examples include the work of adhesion on mercury; frictional properties of untreated and treated polyethylene

7248_C006.fm Page 119 Wednesday, May 23, 2007 11:00 AM

Methods of Characterization — Surfaces

119

ADV REC

7

1.0

0.8

3

COS θ

2

0.6 1

0.4

0.2

20

30

40

γ mN.m-1

50

60

FIGURE 6.4 Zisman critical surface tension plot of cosine of the static advancing and receding contact angles vs. liquid surface tension for low density polyethylene. The same data are used in Figure 6.5. (From Hansen, C.M. and Wallström, E., J. Adhes., 15[3/4], 282, 1983. With permission.)

1.0

ADV. REC.

7

0.8

3

COS θ

2

0.6 1

0.4

0.2

16

18

20

22

RA

24

26

28

FIGURE 6.5 Critical HSP plot of cosine of the static advancing and receding contact angles vs. the HSP difference as defined by Chapter 1, Equation 1.9 for low density polyethylene. The same data are used in Figure 6.4. (From Hansen, C.M. and Wallström, E., J. Adhes., 15[3/4], 282, 1983. With permission.)

7248_C006.fm Page 120 Wednesday, May 23, 2007 11:00 AM

120

Hansen Solubility Parameters: A User’s Handbook

for 2 and 5 min, respectively, with H2S2O7; the Joffé effect — effect of liquid immersion on fracture strength of soda-lime glass; and the Rehbinder effect — crushing strength of Al2O3 granules under various liquids. Beerbower has also brought cohesion parameters into the discussion of wear and boundary lubrication.14 It appears that these factors should still have some consideration, even though recent progress and understanding in the area are much more advanced.15 Additional surface characterizations using HSP are reported in Chapter 7. These include characterizations of the surfaces of pigments, fillers, and fibers. Both organic and inorganic materials have been characterized. The test method used is to determine sedimentation rates for the materials of interest in the same large number of solvents traditionally used in HSP studies. Adsorption of given liquids onto the particle or fiber surface slows the sedimentation rate, and indeed some (fine) particles with rather high densities suspend for years in organic liquids with rather modest densities. A significant advantage in this testing method is that hexane, for example, is not able to retard sedimentation where it may spontaneously spread, as discussed above. Hexane is not an isolated example of this behavior. The characterizations using standard HSP procedures indicate it is truly high affinity for the surface, which is important in these characterizations and not just spontaneous spreading. The reason for this may be the extent (or depth) of the adsorption layer, as well as whether the adsorption occurs at specific sites, or both. Results may be affected when molecules in a surface can orient differently from their original state upon contact with a liquid, for example, with water (see the discussion in Chapter 18). An indirect correlation between HSP and the phenomena discussed above, spontaneous spreading and dewetting, has been established through measurements of environmental stress cracking (ESC).16 As discussed in Chapter 14, ESC correlates with the strain and the HSP and molecular size and shape of the cracking agent. The polymers polycarbonate (PC), cyclic olefinic copolymer (COC), and acrylonitrile/butadiene/styrene (ABS) terpolymer could be described in terms of the regions A, B, and C as shown in Figure 6.1 to Figure 6.3. A large number of test liquids in each category were used to evaluate the critical strain required for ESC. It was found that in every case tested, category A liquids gave ESC. All category B liquids also gave ESC, but the critical strains were somewhat higher on an average. Category C liquids could also give ESC in some cases. nhexane was a category C liquid for some of these polymers in spite of its low surface tension. The HSP differences outweighed the expected spreading based on surface tension differences. Although these observations should not replace testing, a simple test of applying a droplet of liquid and possibly spreading it, if it does not do so itself, is a rapid way to assess a potential problem. Before leaving this section, it is appropriate to mention that thinking of the type described above has led to a Nordtest Method, NT POLY 176, “Spreading Surface Tension by the Applied Droplet Method.” This method is based on visual observation of droplets of known surface tension after they are applied to a test surface. The test surface may be a polymer, metal, or other material. The spreading surface tension is found to within ±1 mN/m by locating two liquids in a series where one of them spreads spontaneously and the other with a slightly higher surface tension does not. The preferred set of liquids is made with ethanol and water with a difference of 2 mN/m between them. Surfaces of many different geometries (from 4 μm diameter wire to ships being painted), states of contamination (from clean for internal medical use to contamination with oil, silicones, pressure sensitive adhesive, etc.), and orientation (ceiling in a tunnel, inside pipes, etc.) have been tested with remarkable success using this simple test. The usual procedure is to assign a value to a clean(ed) surface and then compare test surfaces, wherever they may be, against this to determine the presence of contamination.

SELF-STRATIFYING COATINGS A newer development in the coatings industry is to apply a single coat of paint which separates by itself into a primer and topcoat. A special issue of the journal Progress in Organic Coatings was devoted to this type of coating.17 Misev has also discussed formulation of this type of product using

7248_C006.fm Page 121 Wednesday, May 23, 2007 11:00 AM

Methods of Characterization — Surfaces

121

HSP concepts.18 The separation of the binders into primer and topcoat must occur while the coating is still liquid enough to allow the necessary transport processes to occur. The solvent must just dissolve the binders such that they become incompatible when it begins to evaporate. The binder with the lowest energy (surface tension/cohesion parameters) will naturally migrate toward the low energy air interface and, therefore, this determines which of the binders makes up the topcoat. There are a number of other factors which are important for the process, including polymer molecular weight, rate of solvent evaporation, etc., but these will not be discussed here. This discussion is included because it once more demonstrates how cohesion parameters are coupled with surface energy and also to interfacial energy. The interface between the topcoat and primer is formed from an otherwise homogeneous system. The previous considerations lead to the expectation that the magnitude of the interfacial surface tension between two incompatible polymers is closely related to the difference in their cohesion parameters. Without going into greater detail, it is widely known among those who work with partially compatible polymers that this is indeed the case.19,20 See also Chapter 9 where partial compatibility in bitumen (asphalt) is discussed. Figure 6.6 shows the principles involved for selecting the solvent which can make these work. The polymer with HSP nearest the origin will be the topcoat, as it has the lower (surface or cohesion) energy of the two. A solvent is required which dissolves both polymers, so it will be located in the common region to the spheres portrayed. Mutual solubility of two polymers is promoted when the solvent favors the polymer which is most difficult to dissolve.21 This is usually the one with the higher molecular weight. It is clear that selection of the optimum solvent for this process of designed generation of an interface is aided by systematic use of HSP. This is a prime example of selfassembly where proper formulation can be aided by the concepts discussed above.

POLAR PARAMETER

PRIMER

PARAMETERS REQUIRED FOR COMMON SOLVENT TOP COAT (LOWEST ENERGY)

HYDROGEN BONDING SOLUBILITY PARAMETER

FIGURE 6.6 Sketch of HSP principles used to formulate a self-stratifying coating from an initially homogeneous solution (see discussion in text). (From Birdi, K.S., Ed., Handbook of Surface and Colloid Chemistry, CRC Press, Boca Raton, FL, 1997, p. 324.)

7248_C006.fm Page 122 Wednesday, May 23, 2007 11:00 AM

122

Hansen Solubility Parameters: A User’s Handbook

MAXIMIZING PHYSICAL ADHESION If one wishes to maximize physical adhesion, the physical similarity (same HSP) of the two interfaces being joined must be as close as possible. The previous discussion suggests that physical similarity can be obtained when two criteria are met. The first criterion is that exactly the same liquids spontaneously spread on each of the surfaces to be joined. The second criterion is that exactly the same liquids maintain films when spread (ASTM method for wetting tension) on each of the surfaces to be joined. Any differences in this spontaneous spreading or wetting tension behavior can be interpreted as being a difference in physical similarity. The differences in the behavior of liquid droplets or films that are observed may suggest which steps can be taken to minimize differences, if this is required. Should one add aliphatic segments to reduce the polar and hydrogen bonding contributions? Should alcohol and/or acid groups be incorporated to increase the hydrogen bonding in the system? This type of approach can be used to establish guidelines for action relative to each of the HSP parameters. Aromatic character and halogens other than fluorine characteristically increase δD; nitro and phosphate groups characteristically increase δP; and alcohol, acid, and primary amine groups characteristically increase δH. Reference can be made to the table of group contributions in Chapter 1 (Table 1.1) for more precise comparisons. The discussion of forming good anchors on pigments and other surfaces found in Chapter 8 is also relevant to the present discussion, as such anchors can also be used to enhance adhesion.

CONCLUSION Greater insight into the makeup of a product is possible when one not only knows the cohesion parameters, i.e., HSP, for polymers and solvents it contains, but also the HSP for the various surfaces which these encounter. The surfaces of substrates, pigments, fillers, plastics, fibers, and other materials can also be characterized by HSP (see Chapter 5 and Chapter 7). This allows mutual interactions to be inferred by comparisons of which materials are similar and which materials are different in terms of their HSP. Similar materials in this context have similar HSP regardless of differences in composition. The critical surface tension and wetting tension are single point determinations. Cohesion parameters allow a more complete characterization of surfaces than do these single point measurements and, at the same time, allow insight as to how the single point measurements fit into the overall energy picture for the product. Guidelines for systematically changing the affinities of surfaces can also be obtained from HSP concepts. Both the spontaneous spreading region and the wetting tension region on HSP plots for two different surfaces must be identical if they are to have identical overall surface characteristics.

REFERENCES 1. Flory, P.J., Principles of Polymer Chemistry, Cornell University Press, New York, 1953. 2. Barton, A.F.M., Applications of solubility parameters and other cohesion energy parameters, Polym. Sci. Technol. Pure Appl. Chem., 57(7), 905–912, 1985. 3. Skaarup, K. and Hansen, C.M., The Three-Dimensional Solubility Parameter and Its Use (Det Tredimensionale Opløselighedsparametersystem og dets Anvendelse), Rapport No. 54 (TM 2-67), Lakog Farveindustriens Forskningslaboratorium, København, 1967 (in Danish). 4. Skaarup, K., Surface Tension and 3-D Solubility Parameters (Overfladespænding og 3-D Opløselighedsparametre), Nordiske Kemikermøde, København, 1968 (in Danish). 5. Beerbower, A., Surface free energy: a new relationship to bulk energies, J. Colloid Interface Sci., 35, 126–132, 1971. 6. Hansen, C.M. and Beerbower, A., Solubility parameters, in Kirk-Othmer Encyclopedia of Chemical Technology, 2nd ed., Suppl. Vol., Standen, A., Ed., Interscience, New York, 1971, pp. 889–910.

7248_C006.fm Page 123 Wednesday, May 23, 2007 11:00 AM

Methods of Characterization — Surfaces

123

7. Koenhen, D.N. and Smolders, C.A., The determination of solubility parameters of solvents and polymers by means of correlation with other physical quantities, J. Appl. Polym. Sci., 19, 1163–1179, 1975. 8. Erbil, H.Y., Surface tension of polymers, in Handbook of Surface and Colloid Chemistry, Birdi, K.S., Ed., CRC Press, Boca Raton, FL, 1997, pp. 265–312. 9. Hansen, C.M., Characterization of liquids by spreading liquids, J. Paint Technol., 42(550), 660–664, 1970. 10. Hansen, C.M. and Pierce, P.E., Surface effects in coatings processes, XII Federation d’Associations de Techniciens des Industries des Peintures, Vernis, Emaux et Encres d’Imprimerie de l’Europe Continentale, Congress Book, Verlag Chemie, Weinheim/Bergstrasse, 91–99, 1974; Ind. Eng. Chem. Prod. Res. Dev., 13(4), 218–225, 1974. 11. Hansen, C.M. and Wallström, E., On the use of cohesion parameters to characterize surfaces, J. Adhes., 15(3/4), 275–286, 1983. 12. Zisman, W.A., Relation of the equilibrium contact angle to liquid and solid constitution, in Contact Angle, Wettability and Adhesion, Advances in Chemistry Series No. 43, Gould, R.F., Ed., American Chemical Society, Washington, D.C., 1964, chap. 1. 13. Zisman, W.A., Surface energetics of wetting, spreading, and adhesion, J. Paint Technol., 44(564), 41, 1972. 14. Beerbower, A., Boundary Lubrication — Scientific and Technical Applications Forecast, AD747336, Office of the Chief of Research and Development, Department of the Army, Washington, D.C., 1972. 15. Krim, J., Friction at the Atomic Scale, Scientific American, 275(4), October 1996, pp. 48–56. 16. Nielsen, T.B. and Hansen, C.M., Surface wetting and the prediction of environmental stress cracking (ESC) in polymers, Polym. Degradation Stability, 89, 513–516, 2005. 17. Special issue devoted to self-stratifying coatings, Prog. Org. Coat., 28(3), July 1996. 18. Misev, T.A., Thermodynamic analysis of phase separation in self-stratifying coatings — solubility parameters approach, J. Coat. Technol., 63(795), 23–28, 1991. 19. Luciani, A., Champagne, M.F., and Utracki, L.A., Interfacial tension in polymer blends. Part 1: Theory, Polym. Networks Blends, 6(1), 41–50, 1996. 20. Luciani, A., Champagne, M.F., and Utracki, L.A., Interfacial tension in polymer blends. Part 2: Measurements, Polym. Networks Blends, 6(2), 51–62, 1996. 21. Hansen, C.M., On application of the three dimensional solubility parameter to the prediction of mutual solubility and compatibility, Färg och Lack, 13(6), 132–138, 1967.

7248_C006.fm Page 124 Wednesday, May 23, 2007 11:00 AM

7248_C007.fm Page 125 Wednesday, April 4, 2007 1:00 PM

of Characterization 7 Methods for Pigments, Fillers, and Fibers Charles M. Hansen ABSTRACT Cohesion parameters for pigments, fillers, and fibers can often be evaluated by observation of the suspension and/or sedimentation behavior of particulate matter in different liquids. These characterizations are based on relatively stronger adsorption by some of the liquids compared with others. Those liquids with stronger interaction can suspend finer fractions of solids indefinitely or retard sedimentation, compared with the other liquids. Data should be interpreted by accounting for differences in the densities and viscosities of the test liquids, such that a relative sedimentation rate can be used for comparisons. The absolute sedimentation rates are generally not of primary interest. Data from such evaluations can be computer-processed to assign Hansen cohesion parameters (HSP) to the material in question. Cohesion parameter data are given for some newer pigments, fillers, and a carbon fiber to demonstrate the principles.

INTRODUCTION The possibilities offered by cohesion parameter characterization of pigments, fillers, and fibers have not been generally recognized, judging from the relatively small number of publications appearing on the topic. Pigments and a few fillers were characterized in some of the author’s first publications dealing with the solubility parameter.1,2 These were given δD, δP, and δH parameters (HSP) and a characteristic radius of interaction exactly analogous to the polymer characterizations discussed in Chapter 2 and Chapter 5. These data together with some more recent pigment characterizations are included in Table 7.1, Table 7.2A, and Table 7.2B. Shareef et al.3 have also characterized pigment surfaces, including metal oxides. Gardon and Teas4 clearly showed the differences between zinc oxides treated and untreated with organic phosphate using a cohesion parameter characterization. Inorganic fibers have also been characterized.5 All of these characterizations again confirm the universality possible with these parameters. They reflect molecule–molecule interactions whether at surfaces or in bulk. In the future, more systematic selection of dispersion aids should be possible, as these can also be described with the same energy parameters. Hansen and Beerbower have touched on this topic.6 Each segment of such molecules requires its own HSP. The discussion in Chapter 15 for the interactions within cell walls in wood demonstrates how this could be done. It has been shown by calculation that hemicelluloses act like surface-active agents, with some segments seeking lowerenergy lignin regions and some segments (those with alcohol groups) orienting toward the higherenergy cellulose.

125

7248_C007.fm Page 126 Wednesday, April 4, 2007 1:00 PM

126

Hansen Solubility Parameters: A User’s Handbook

TABLE 7.1 HSP Correlations for Older Inorganic Pigments1,2 and Metal Oxides3 Material

δD

δP

δH

Ro

Kronos® RN57 TiO2a Aluminum pulver lack 80a Red iron oxidea Synthetic red iron oxideb Synthetic yellow iron oxideb

24.1 19.0 20.7 16.1 17.3 16.1 16.9 16.2

14.9 6.1 12.3 8.6 6.0 8.6 7.8 10.8

19.4 7.2 14.3 15.0 14.5 15.0 10.6 12.7

17.2 4.9 11.5 11.3 12.5 11.3 13.2 9.8

Zinc oxide

Note: Units are MPa1/2. a

From Hansen, C.M., The Three Dimensional Solubility Parameter and Solvent Diffusion Coefficient, Doctoral dissertation, Danish Technical Press, Copenhagen, 1967. With permission. b From Shareef, K.M.A. et al., J. Coat. Technol., 58(733), 35–44, February 1986. With permission.

METHODS TO CHARACTERIZE PIGMENT, FILLER, AND FIBER SURFACES The cohesion parameter (HSP) approach to characterizing surfaces gained impetus by experiments where the suspension of fine particles in pigment powders was used to characterize 25 organic and inorganic pigment surfaces.1,2 Small amounts of the pigments are shaken in test tubes with a given volume of liquid (10 ml) of each of the test solvents, and then sedimentation or lack of the same is observed. When the solid has a lower density than the test liquid, it will float. Rates of floating have also been noted, but the term sedimentation will be retained here for both sedimentation and floating. The amounts of solid sample added to the liquids can vary depending on the sample in question, and some initial experimentation is usually advisable. If the pigment or filler particle size is large — say over 5 µm — the surface effects become less significant compared with a sample where the particle size is only 0.01 µm. Problems arise when the pigments are soluble enough to color the liquid such that sedimentation cannot be evaluated. The larger particle size pigments and fillers may sediment very rapidly. Sedimentation rates have still been used successfully in some of these cases. The sedimentation rate is most easily expressed as the time at which the amount of particles, at a given point in the test tubes, has fallen to some small amount, perhaps zero. Observations can be made visually, or perhaps instrumentally, in a direction perpendicular to the incidence of a laser light. A visual observation is required in any event, as some samples seem to coat out rapidly on glass surfaces. Some pigments have portions that suspend for years in spite of large-density differences and relatively low-solvent viscosity. Satisfactory results from this type of measurement require some experience regarding what to look for. This can vary from sample to sample. A characterization is less certain when there are only 4 or 5 good liquids out of the perhaps 40 to 45 tested, although this depends somewhat on which liquids are involved. “Good” in this context means suspension of particulates is prolonged significantly, compared with the other test solvents, after compensating for differences in density and viscosity. A corrected relative sedimentation time (RST) can be found by modifying the sedimentation time, ts

7248_C007.fm Page 127 Wednesday, April 4, 2007 1:00 PM

Methods of Characterization for Pigments, Fillers, and Fibers

127

TABLE 7.2A List of Pigments Studied. HSP Results are Given in Table 7.2B Pigment

Description

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14.

TiO2, Kronos RN 57, Titan Co. A/S., Frederikstad, Norway. Phthalocyanine Blue, B6, E. I. du Pont de Nemours and Co. (1949). Isolbonared Nr. 7522, C. I. Pigment Red 48 (C.I. 15865) (MnSalt), Køge Chemical Works, Køge, Denmark. Peerless Carbon Black Isol Fast Yellow IO GX 2505, C.I. Pigment Yellow 3, Køge Chemical Works, Køge, Denmark. Reflex Blau TBK Ext. (No C.I. Index-pigment mixture), Farbwerke Hoechst, Frankfurt (M), Germany. Isol Ruby BKS 7520, C.I. Pigment Red 57 (C.I. 15850) (Ca Salt), Køge Chemical Works, Køge, Denmark. Hansagelb 10 G, C.I. Pigment Yellow 3 (C.I. 11710), Farbwerke Hoechst, Frankfurt (M), Germany. Fanalrosa G Supra Pulver, Pigment Red 81 (C.I. 45160), BASF, Ludwigshafen, Germany. Heliogenblau B Pulver, C.I. Pigment Blue 15 (C.I. 74160), BASF, Ludwigshafen, Germany. Heliogengrün GN, C.I. Pigment Green 7, (C.I. 74260), BASF, Ludwigshafen, Germany. Permanentgelb H 10 G, C.I. Pigment Yellow 81, (No C.I. index), Farbwerke Hoechst, Frankfurt (M), Germany. Permanent Bordeaux FRR, C.I. Pigment Red 12 (C.I. 12385), Farbwerke Hoechst, Frankfurt (M), Germany. Permanent Violet RL Supra, C.I. Pigment Violet 23, (C.I. 12505), Farbwerke Hoechst, Frankfurt (M), Germany. Isol Benzidine Yellow G 2537, C.I. Pigment Yellow 12 (C.I. 21090), Køge Chemical Works, Køge, Denmark. Brillfast Sky Blue 3862, C.I. Pigment Blue 3 (C.I. 42140), J. W. and T. A. Smith Ltd., London. Permanent Orange G, C.I. Pigment Orange 13 (C.I. 21110), Farbwerke Hoechst, Frankfurt (M), Germany. Permanent Red, FGR Extra Pulver, C.I. Pigment Red 112, (C.I. 12370). Farbwerke Hoechst, Frankfurt (M), Germany. Isol Fast Red 2G 2516, C.I. Pigment Orange 5, (C.I. 12075), Køge Chemical Works, Køge, Denmark. Monolite Fast Blue 3 RS, Powder, C.I. Vat Blue 4 (C.I. 69801), Imperial Chemical Industries. Heliogenblau LG, Pulver, C.I. Pigment Blue 16 (C.I. 74100), BASF., Ludwigshafen, Germany. Red Iron Oxide. Carbon Black, Printex V (5519-1), Degussa, Frankfurt (M), Germany. Aluminum Pulver Lack 80, Eckart-Werke, 851 Fürth/Bayern, Germany. Isol Benzidene Yellow GA-PR, 9500, C.I. Pigment Yellow 12, Køge Chemical Works, Køge, Denmark.

15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25.

Source: From Hansen, C.M., The Three Dimensional Solubility Parameter and Solvent Diffusion Coefficient, Doctoral dissertation, Danish Technical Press, Copenhagen, 1967. With permission.

RST = ts(ρp – ρs)/η

(7.1)

ρp and ρs are densities of particle and test liquid, respectively, and η is the liquid viscosity. A prolonged RST implies greater adsorption of the given solvent onto the surface in question. Characterizations based on these techniques tend to place emphasis on the nature of the surfaces for the smaller-particle-size fractions. An example of a data sheet used for such studies is included in Table 7.3.

DISCUSSION — PIGMENTS, FILLERS, AND FIBERS It can be reasoned that a pigment, filler, or fiber is most beneficial when the pigment surface and the binder in question have the same cohesion parameters. There are apparently no publications indicating a systematic modification of pigment surfaces to achieve a given set of cohesion parameters. The characterizations for some organic pigments are given in Table 7.4. These data indicate that their respective surfaces are essentially identical. An exception is the first one in the table where the analysis is based on only three good solvents that were able to extend sedimentation significantly relative to the other solvents tested.

7248_C007.fm Page 128 Wednesday, April 4, 2007 1:00 PM

128

Hansen Solubility Parameters: A User’s Handbook

TABLE 7.2B Characteristic Parameters for Various Pigments Given in Table 7.2A Pigment

δt

δD

δP

δH

Ro

Comments

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25.

16.8 10.5 10.0 13.6 11.9 13.2 10.5 10.5 13.0 12.0 12.0 8.8 13.2 11.5 10.2 13.3 11.5 11.2 14.2 15.2 13.5 13.7 13.1 10.4 9.1

11.8 9.3 8.7 10.3 10.2 10.8 9.6 9.1 9.8 10.8 10.0 8.4 10.7 9.6 9.3 9.5 9.7 10.0 10.9 10.8 10.7 10.1 10.3 9.3 9.0

7.3 3.1 3.5 6.0 4.8 3.8 3.0 4.0 7.0 3.5 4.8 1.5 4.8 5.2 3.0 7.2 3.9 3.5 5.6 6.5 5.0 6.0 6.0 3.0 2.7

9.5 3.7 3.5 6.6 3.8 6.6 3.2 3.3 5.0 4.0 4.5 2.3 6.1 3.6 2.9 6.0 4.7 3.5 7.1 8.5 6.5 7.0 5.5 3.5 2.3

8.4 2.3 2.5 6.0 4.4 7.0 3.9 3.3 5.2 5.2 4.8 2.2 5.2 4.4 3.9 5.1 4.5 5.0 7.0 7.0 6.0 5.6 5.5 2.4 2.5

Suspension Few suspending solvents Few suspending solvents Suspension Color only Mixed color and suspension Suspension Color only Color only Suspension Primarily suspension Suspension Color only Mixed color and suspension Mixed color and suspension Suspension Color only Color only Primarily color Suspension Suspension Suspension Suspension Suspension Suspension

Note: Units are (cal/cm3)1/2. Source: From Hansen, C.M., The Three Dimensional Solubility Parameter and Solvent Diffusion Coefficient, Doctoral dissertation, Danish Technical Press, Copenhagen, 1967. With permission.

These results suggest that pigment manufacturers have essentially arrived at the same result — a surface energy compatible with a wide variety of currently used binders. The solvents most frequently appearing as good for adsorption onto these surfaces include several chlorinated solvents, toluene, and tetrahydrofuran. As these solvents dissolve the most commonly used binders, one can conclude that the common binders will adsorb readily onto these pigment surfaces. This will give a good result, provided the solvent is not so good for the binder that it can remove the binder from the pigment surface. Schröder7 (BASF) confirms that the optimum polymer adsorption will be found when the binder and pigment surface have the same HSP. He indicates that the solvent should be very poor for the pigment and located on the boundary region for the binder. He prefers the pigment to have HSP values placing it intermediate between the solvent and binder. This is suggested for conditions where the solvent has higher HSP than the pigment, as well as for conditions where the solvent has lower HSP than the pigment. This situation, with the solvent and binder on opposite sides of the pigment, means the composite vehicle has parameters very closely matching those of the pigment.

7248_C007.fm Page 129 Wednesday, April 4, 2007 1:00 PM

Methods of Characterization for Pigments, Fillers, and Fibers

129

TABLE 7.3 Sedimentation Study Date Sample: Density of sample (Dp):

Reference No.:

Solvent

D(Ds) 20°C

Acetone Acetophenone Benzene 1-Butanol Butyl acetate Butyrolactone Carbon tetrachloride Chlorobenzene Chloroform Cyclohexane Cyclohexanol Diacetone alcohol o-Dichlorobenzene Diethylene glycol Diethyl ether Dimethyl formamide DMSO 1,4-Dioxane Dipropylene glycol Ethanol Ethanolamine Ethyl acetate Ethylene dichloride Ethylene glycol Ethyelene glycol monobutyl ether Ethyelene glycol monoethyl ether Ethyelene glycol monomethyl ether Formamide Hexane Isophorone Methanol Methylene dichloride Methyl isobutyl ketone Methyl-2-pyrrolidone Nitrobenzene Nitroethane Nitromethane 2-Nitropropane Propylene carbonate Propylene glycol Tetrahydrofurane Toluene Trichloroethylene

0.79 1.03 0.88 0.81 0.87 1.29 1.59 1.10 1.48 0.78 0.95 0.94 1.31 1.12 0.72 0.95 1.10 1.04 1.03 0.82 0.91 0.89 1.25 1.12 0.90 0.93 0.96 1.13 0.66 0.92 0.79 1.33 0.96 1.03 1.21 1.05 1.13 0.99 1.20 1.04 0.89 0.87 1.47

Sedimentation Time (min)

Viscosity 20°C

No.

0.35 1.90 0.65 4.00 0.74 1.92 0.99 0.80 0.37 1.00 68.00 3.20 1.27 35.70 0.23 0.82 1.98 1.31 107.0 1.22 24.10 0.44 0.84 20.90 2.90 2.05 1.72 3.30 0.33 2.60 0.59 0.43 0.59 1.80 2.03 0.55 0.62 0.75 2.80 56.00 0.55 0.59 0.58

4 6 13 28 30 37 40 41 44 47 48 56 61 75 82 90 94 96 98 104 105 106 120 121 123 124 126 131 140 148 153 162 167 172 177 178 179 181 204 205 222 225 229

D p- D s

From

To

Relative Sedimentation Time (RST) From

To

7248_C007.fm Page 130 Wednesday, April 4, 2007 1:00 PM

130

Hansen Solubility Parameters: A User’s Handbook

TABLE 7.4 HSP Correlations for Selected Materials Material

δD

δP

δH

Ro

Fit

G/T

Organic Pigments Paliotol® Gelb L1820 BASF Heliogen® Blau 6930L BASF Socco Rosso L3855 BASF Perm Rubin F6B Hoechst Perm Gelb GRL02 Hoechst Perm Lackrot LC Hoechst

18.9 18.0 17.3 16.7 16.7 19.0

3.5 4.0 5.7 3.7 2.5 5.0

10.5 4.0 2.7 3.1 3.7 5.0

5.4 4.0 4.1 4.8 4.5 4.0

0.99 1.00 0.99 0.88 0.95 1.00

3/35 5/34 4/34 6/33 5/37 7/28

Inorganic Pigments, Fillers, etc. Cabot Hochdispersea 16.7 Cabot Hochdisperse 19.3 Zeta Potential Blanc Fixeb 26.5

9.3 9.5 19.1

11.5 10.3 14.5

11.7 12.7 20.4

— 0.79 0.95

23/23 23/31 5/19

Note: A perfect data fit of 1.0 means that there most probably are other sets of the same parameters that will have a data fit of 1.0 and also define a sphere that surrounds all the good solvents. A data fit of 0.99+ is preferred to define the optimum sphere for this reason. G/T is the number of good (G) liquids and the total (T) number of liquids in a correlation. Units are MPa1/2. a

Data analysis which only considers the good solvents to define the least sphere possible. See discussion of the SPHERE1 program in Chapter 1. b Data from Winkler, J., Eur. Coat. J., 1–2/97, 38–42, 1997. With permission.

There is also a relation between how clearly a pigment can be characterized by sedimentation measurements and its zeta potential. Low zeta potential means sedimentation is rapid in all solvents, and this type of characterization becomes difficult or perhaps impossible. The zeta potential reflects the intensity (percentage coverage and number of layers) of the surface energy characteristics. It does not clearly indicate specific affinity relations of a given binder for the pigment surface, as a result of a given surface treatment, for example. This is given by HSP. To obtain a complete picture of the energetics of the surface, one needs an intensity factor, i.e., the zeta potential, as well as a qualitative factor, i.e., the cohesion energy parameters. The latter are generally lacking. One can suspect that some pigments have such high-intensity zeta potential — at some cost — that even though the cohesion parameters match poorly with a given binder, a system can still function satisfactorily. An HSP correlation for the zeta potential of blanc fixe is given in Table 7.4 using data from Winkler.8 This is discussed further in the following section. Acid–base theories have been popular.9–11 The author has not found it necessary to resort to this type of approach in any activity, although many have clearly found them beneficial. More research is needed to fully understand the successes of the acid–base as well as the HSP approaches. It would seem that the HSP approach allows predictive ability that is not possible with an acid–base approach. However, the current problem is the lack of data. Organic pigments normally have a good organic substrate on which to base an organic surface modification. The characterizations may reflect both a surface treatment and the surface of the base particles, depending on how the test liquids interact with these. It should not be too difficult to modify an organic surface to an alternative organic surface with satisfactory properties, if desired. It is conceptually and, in practice, more difficult to modify an inorganic surface to make it compatible with organic systems. This requires a significant change in surface energy from high to much lower and, presumably, also requires a greater degree of coverage to mask the base inorganic surface. The producers of inorganic pigments and fillers must either give their products suitable

7248_C007.fm Page 131 Wednesday, April 4, 2007 1:00 PM

Methods of Characterization for Pigments, Fillers, and Fibers

131

surfaces, probably after much effort, or else one needs one or more supereffective additives to be able to achieve a good and stable dispersion. It helps to incorporate given (high-cohesion energy) groups in a grinding resin, such as acid, alcohol, amine, etc. The relatively high local cohesion parameters in the binder that are associated with these groups would indicate a high affinity for the high-cohesion-energy surface of the inorganic material. At the same time, these local regions of adsorbed polymer segments are not particularly soluble or are insoluble in the cheaper hydrocarbon solvents — for example, those that have much lower cohesion parameters. This provides a good, stable anchor on the pigment surface. The solvent will not dissolve that polymer or polymer segment away from the surface. Binders with high-acid numbers are frequently used, with success, in printing inks for the same reason. This is discussed in more detail in the following section. It is felt that those who understand the use of cohesion parameters are able to more systematically modify surfaces of inorganic materials to optimize or improve their compatibility with organic polymers and binders. This has been done for inorganic Rockwool® fibers that are to be incorporated into polypropylene.5 It must be presumed that this type of systematic procedure can guide surface treatment of other inorganic materials in a more directed way toward a desired goal. Data fits have been generally lower for characterizations of particulate solid surfaces, such as fillers, than for characterizations of polymers based on solubility. When testing is finished, the polymeric macromolecule is no longer a solid in the good solvents, whereas the particulate filler remains a solid. A macromolecule has various possibilities for contortions and the positioning of significant active groups in solution (or when swollen), giving a large number of possible (dynamic) structures that can be formed with the solvent. A rigid solid surface does not have this potential for adopting energetically desirable positions for its active groups. The adjustments for optimum local association must be made by the solvent molecules alone in the sedimentation testing. There are many solvents that do not retard sedimentation significantly, whereas the predictions based on the behavior of other solvents that do significantly retard sedimentation indicate that this should be the case. A contribution to the formation of the energetically desirable geometrical structures is not possible from the movement of rigid solid surfaces. Therefore, some solvents may not be able to retard sedimentation because they cannot adopt the geometrical positioning required to do this without the help of a mobile substrate. This lack of expected performance may be also partly due to solvent size, the location of the active groups, or combinations of these. These phenomena appear to be a significant area for future research.

HANSEN SOLUBILITY PARAMETER CORRELATION OF ZETA POTENTIAL FOR BLANC FIXE Winkler8 has reported zeta potentials measured for 1% v/v blanc fixe with 0.34% moisture content. There were 19 liquids included in this careful study. These liquids could easily be divided into two groups. There were 5 systems with zeta potentials greater than about 10 mV and 14 systems with zeta potentials less than about 5 mV. Table 7.3 includes the results of the correlation of these data with cohesion parameters. The only major “error” was for hexamethylphosphoramide, with a RED of 0.951 and a zeta potential of 1.9 mV. This correlation supports the contention that cohesion parameters are significant for characterization of pigment, filler, and fiber surfaces. This is a good correlation and supports the views presented earlier. According to Winkler, there was no correlation with the acceptor or donor numbers (acid–base).

CARBON FIBER SURFACE CHARACTERIZATION Hansen solubility parameters have been assigned to a carbon fiber surface, Panex 33 from Zoltek. After considerable refinement of the experimental technique, it was determined that two separate sets of HSP are required to describe the fiber surface. One of these sets has δD;δP;δH equal to

7248_C007.fm Page 132 Wednesday, April 4, 2007 1:00 PM

132

Hansen Solubility Parameters: A User’s Handbook

TABLE 7.5 HSP Correlations for Various “Carbon” Materials Material

δD

δP

δH

Ro

Carbon fiber (high region)b Carbon fiber (low region)b Carbon blacka Carbon black 1 Carbon black 1 HTb Carbon black 2 HTb Carbon black 3 HTb Carbon black 4 HTb Petroleum coke Coal tar pitch C60 fullerened

13.4 17.2 21.1 17.9 21.5 20.5 20.5 18.9 16.4 18.7 19.7

17.8 3.4 12.3 8.1 7.3 8.7 8.7 11.1 4.0 7.5 2.9

14.2 2.0 11.3 8.9 13.3 12.7 12.5 8.5 10.0 8.9 2.7

10.3 3.9 16.6 6.2 11.4 9.2 9.1 5.6 10.7 5.8 3.9

Comments Sedimentation Sedimentation Suspension Sedimentation Sedimentation Sedimentation Sedimentation Sedimentation Sedimentation Solubility Solubilityc

rate 3mm fibers rate 3mm fibers (slow) (rapid) (rapid) (rapid) (rapid) (slow)

Note: Units are MPa1/2. a b c d

Printex V (5519-1), Degussa HT indicates a special heat treatment was performed prior to testing. Log mole fraction solubility greater than –3. Data from Hansen, C.M. and Smith, A.L., Carbon, 42(8–9), 1591–1597, 2004. With permission.

13.4;17.8;14.2, all in MPa1/2. This corresponds to a highly polar surface with a significant hydrogen bonding component as well. The second set is characteristic of a hydrocarbon material with δD;δP;δH equal to 17.2;3.4;2.0, again in MPa1/2. The hydrocarbon-like surface can be the backbone of the polyacrylonitrile (PAN) precursor for the fiber. Two separate sets of HSP assignments are confirmed by x-ray photoelectron spectroscopy (XPS) analysis. Two separate regions are found to coexist on the carbon fiber surface. The hydrogen bonding and polar contributions arise from the both bound and unbound (sizing/finish agent) chemical functionalities mainly in the form of hydroxyl, ether, carbonyl, carboxyl, amide, and nitrile groups. The carbonaceous backbone of the carbon fiber primarily accounts for the nonpolar region. This work was done as a part of the Framework program Interface Design of Composite Materials with the support of the Danish Research Agency, Ministry of Science, Technology and Innovation (STVF). The HSP characterizations were done at FORCE Technology, Broendby, Denmark, whereas the analyses were done at the Risø National Laboratory, Roskilde, Denmark. Table 7.4 contains the HSP data used to construct Figure 7.1. These data confirm that “carbon” can be many things with widely different surface energies. The origins of the material, as well as the method in which it has been handled or treated, can completely dominate the nature of the surface of the given materials and their solubility, if this is possible.

CONTROLLED ADSORPTION (SELF-ASSEMBLY) Significant tasks for formulators are to control the surface and interfacial energies of products, especially if they are water reducible. This is required to allow substrate wetting, to maintain stable dispersions, and to provide/ensure adequate and durable adhesion to given substrates. Guidelines for courses of action are frequently available when cohesion energy parameters are referred to. Some guides are discussed in the following. It is a well-known fact that a small percentage of acid groups (or alcohol groups) on a polymer chain will promote adhesion and adsorption to many surfaces. The cohesion energy parameter of an isolated acid group is high. One can consider the cohesion energy properties of formic acid (δD;

7248_C007.fm Page 133 Wednesday, April 4, 2007 1:00 PM

Methods of Characterization for Pigments, Fillers, and Fibers

133

20

Carbon fiber, high

18

16

14

“Carbon Black”

12 δp 10

Carbon Black 1

8

Coal tar pitch 6

4

Carbon fiber, low

Petroleum Coke

Fullerene 2

0 0

2

4

6

8 δH

10

12

14

16

FIGURE 7.1 Characterization of a carbon fiber and comparisons of this with other carbon materials. Units are MPa1/2. The work on which this figure is based was supported by the Framework program Interface Design of Composite Materials (STVF fund No. 26-03-0160). Reproduced with permission.

δP; δH = 14.3; 11.9; 16.6) as an isolated part of a polymer chain. The polar cohesion energy parameter of an acid group is not so high. It would seem logical to systematically use acid groups for adsorption to high-energy surfaces and to make certain that the cohesion energy parameters for the solvent and bulk of the product are much lower, such that isolated acid groups would not be dissolved. This would provide an anchor that the product will not be able to remove. This type of adsorption may be called hydrophilic bonding. If, on the other hand, the solvent were too good for the anchor,

7248_C007.fm Page 134 Wednesday, April 4, 2007 1:00 PM

134

Hansen Solubility Parameters: A User’s Handbook

it could be presumed that even an acid group may be too readily dissolved off the surface or at least take part in a dynamic equilibrium of adsorption and desorption. Absorbed/adsorbed water can sometimes interfere with such anchors at high-energy surfaces. The reverse of this thinking is systematically used by those designing associative thickeners and also by nature, such as the hydrophobic bonding in proteins. Certain segments of given molecules have such low cohesive energy parameters that they are no longer soluble in the media, which is usually aqueous, and they either seek out their own kind (associate) or perhaps adsorb on or penetrate into a low-energy surface where cohesion energy parameters more suitably match. The positive effects of associative thickeners can be counteracted by the presence of solvents preferentially locating where the hydrophobic bonding is to occur. The hydrophobic bonds lose strength or may even dissolve away. A challenge to the creative mind is to derive new uses for high-energy groups that are not particularly water soluble or sensitive. The division of the cohesion energy into at least three parts allows these considerations to be made in a reasonably quantitative manner. One can choose nitro groups or perhaps groups containing phosphorus as examples of species characterized by highpolar-cohesion-energy parameters and low or moderate hydrogen bonding parameters. The total cohesion energy parameters for ethanol and nitromethane are very close: 26.1 and 25.1 MPa1/2, respectively. Ethanol is soluble in water, nitromethane is not. Ethanol has a relatively high-hydrogen bonding parameter (19.4 MPa1/2) compared with nitromethane (5.1 MPa1/2). This makes all the difference. Would not the nitro group be a suitable anchor analogous to the previous discussion concerning acid groups? Also, it would not be hydrophilic with the inherent problems of water sensitivity associated with high-hydrogen bonding parameters. Several of the pigments reported in Table 7.4 did indeed have moderate affinity for the nitroparaffins, for example, but they were included in the lesser interacting group by the arbitrary division into good and bad groups.

CONCLUSION Many pigments and fillers have now been characterized by Hansen cohesion parameters (HSP). Many examples are given. A method based on relative sedimentation time and/or suspension is described for doing this. This method has generally allowed useful characterizations, although some experience is helpful. For example, the data are often scattered and not nearly of the quality usually found when observing polymer solution behavior. This scatter of data for untreated surfaces in particular may cause some to disregard the method; hopefully, they can develop a better one. The obvious advantages of having solvents, plasticizers, polymers, pigments, fillers, fibers, etc., characterized with the same energy parameters should provide incentive for improving on the present state, both in terms of numbers of characterizations as well as improved methodology. One assumes that maximum physical adsorption is accompanied by closely matching HSP. Local adsorption by so-called active groups (alcohol, acid, amine) having the required match, can give anchors on a surface that may no longer be soluble in the continuous media, and therefore will remain in place as required.

REFERENCES 1. Hansen, C.M., The Three Dimensional Solubility Parameter and Solvent Diffusion Coefficient, Doctoral dissertation, Danish Technical Press, Copenhagen, 1967. 2. Hansen, C.M., The three dimensional solubility parameter — key to paint component affinities II. J. Paint Technol., 39(511), 505–510, 1967. 3. Shareef, K.M.A., Yaseen, M., Mahmood Ali, M., and Reddy, P.J., J. Coat. Technol., 58(733), 35–44, February 1986.

7248_C007.fm Page 135 Wednesday, April 4, 2007 1:00 PM

Methods of Characterization for Pigments, Fillers, and Fibers

135

4. Gardon, J.L. and Teas, J.P., Solubility parameters, in Treatise on Coatings, Vol. 2, Characterization of Coatings: Physical Techniques, Part II, Myers, R.R. and Long, J.S., Eds., Marcel Dekker, New York, 1976, chap. 8. 5. Hennissen, L., Systematic Modification of Filler/Fibre Surfaces to Achieve Maximum Compatibility with Matrix Polymers, Lecture for the Danish Society for Polymer Technology, Copenhagen, February 10, 1996. 6. Hansen, C.M. and Beerbower, A., Solubility parameters, in Kirk-Othmer Encyclopedia of Chemical Technology, Suppl. Vol., 2nd ed., Standen, A., Ed., Interscience, New York, 1971, pp. 889–910. 7. Schröder, J., Colloid chemistry aids to formulating inks and paints, Eur. Coat. J., 5/98, 334–340, 1998. 8. Winkler, J., Zeta potential of pigments and fillers, Eur. Coat. J., 1–2/97, 38–42, 1997. 9. Vinther, A., Application of the concepts solubility parameter and pigment charge, Chim. Peint. (England), 34(10), 363–372, 1971. 10. Soerensen, P., Application of the acid/base concept describing the interaction between pigments, binders, and solvents, J. Paint Technol., 47(602), 31–39, 1975. 11. Soerensen, P., Cohesion parameters used to formulate coatings (Kohaesionsparametre anvendt til formulering af farver og lak), Färg och Lack Scand., 34(4), 81–93, 1988 (in Danish). 12. Hansen, C.M. and Smith, A.L., Using Hansen solubility parameters to correlate solubility of C60 fullerene in organic solvents and in polymers, Carbon, 42(8–9), 1591–1597, 2004.

7248_C007.fm Page 136 Wednesday, April 4, 2007 1:00 PM

7248_C008.fm Page 137 Wednesday, May 23, 2007 11:05 AM

— Coatings and 8 Applications Other Filled Polymer Systems Charles M. Hansen ABSTRACT Hansen solubility parameters (HSP) are widely used in the coatings industry to help find optimum solvents and solvent combinations. They also aid in substitution to less hazardous formulations in various other types of products such as cleaners, printing inks, adhesives, etc. The discussion in this chapter includes the physical chemical reasons why solvents function as they do in many practical cases. The behavior of solvents in connection with surfaces of various kinds and the use of HSP to understand and control surface phenomena is especially emphasized. Products where HSP concepts can be used in a manner similar to coatings include other (filled) polymer systems of various types such as adhesives, printing inks, chewing gum, etc. There are many examples of controlled self-assembly.

INTRODUCTION There are many applications documented in the literature where HSP have aided in the selection of solvents, understanding and controlling processes, and, in general, offering guidance where affinities among materials are of prime importance1–5 (see also the following chapters and examples below). This chapter emphasizes coatings applications and discusses the practical application of HSP to solvent selection. Computer techniques are helpful, but not necessary. The same principles useful for understanding the behavior of coatings are useful in understanding behavior in a larger number of related products, including adhesives, printing inks, and chewing gum, to mention a few. These contain widely different materials, both liquid and solid, which can be characterized by HSP. This allows their relative affinities to be established. Previous chapters have discussed how to assign HSP to solvents, plasticizers, polymers, and resins, as well as to the surfaces of substrates, pigments, fillers, and fibers. Various additives such as resins, surfactants, flavors, aromas, scents, drugs, etc., can also be characterized by HSP to infer how they behave in seemingly complex systems.

SOLVENTS In order to find the optimum solvent for a polymer, one must have or estimate its HSP. Matching the HSP of an already existing solvent or combination of solvents can be done, but this procedure does not necessarily optimize the new situation. The optimum depends on what is desired of the system. A solvent with the highest possible affinity for the polymer is both expensive and probably not necessary and will rarely be optimum. In more recent years, optimization increasingly includes considerations of worker safety and the external environment. Volatile organic compounds (VOC) are to be reduced to the greatest extent possible. Chapter 11 is devoted to replacing ozone-depleting chemicals in cleaning operations.

137

7248_C008.fm Page 138 Wednesday, May 23, 2007 11:05 AM

138

Hansen Solubility Parameters: A User’s Handbook

Whereas hand calculations and plotting of data are still quite useful and at times more rapid than computer processing, it is becoming almost mandatory that computers be used. To this end, most solvent suppliers and many large users of solvents have computer programs to predict solution behavior as well as evaporation phenomena. In spite of these pressures to let the computer do the thinking, an experienced formulator can often arrive at a near-optimum result without recourse to paper or to computers. A major factor in this almost immediate overview is the decrease in the number of solvents useful in coatings. By putting this together with other necessary considerations such as flash point, proper evaporation rate, cost, odor, availability, etc., the experienced formulator who knows the HSP for the relatively few solvents possible in a given situation will be able to select a near-optimum combination by a process of exclusion and simple mental arithmetic. This does not mean the use of HSP is on the way out. The real benefit of this concept is in interpreting more complicated behavior, such as affinities of polymers with polymers and polymers with surfaces as described in the following. Much more work needs to be done in these areas, but the following gives an indication of what might be expected. As indicated previously, computer techniques can be very useful but are not always necessary, and simple two-dimensional plots using δP and δH can often be used by those with limited experience with these techniques to solve practical problems. The nonpolar cohesion parameter, δD, cannot be neglected in every case, but, for example, when comparing noncyclic solvents in practical situations, it has been found that their dispersion parameters will be rather close regardless of structure. Cyclic solvents, and those containing atoms significantly larger than carbon, such as chlorine, bromine, metals, etc., will have higher dispersion parameters. The total solubility parameter for aliphatic hydrocarbon solvents is identical with their dispersion parameter and increases only slightly with increased chain length. This same trend is expected for oligomers of a polymer as molecular weight increases. Regardless of the means of processing data, the following examples are intended to illustrate principles on which to base a systematic course of action. Most coatings applications involve solvents reasonably well within the solubility limit which is indicated by the boundary of a solubility plot such as that shown in Figure 8.1.1 A maximum of cheaper hydrocarbon solvent is also desired and can frequently be used to arrive at such a situation for common polymers used in coatings. Some safety margin in terms of extra solvency is advised because of temperature changes, potential variations in production, etc. These can lead to a situation where solvent quality changes in an adverse manner. Balance of solvent quality on evaporation of mixed solvents is also necessary. Here again, computer approaches are possible, and calculations of solvent quality can be made at all stages of evaporation. It is usually good practice to include a small or moderate amount of slowly evaporating solvent of good quality and low water sensitivity to take care of this situation. These have frequently been slowly evaporating ketones and esters. An oxygenated solvent which is frequently added to hydrocarbon solvents and has been cost effective in increasing the very important hydrogen-bonding parameter has been n-butanol (or sometimes 2-butanol). The mixture of equal parts xylene and n-butanol has been widely used in conjunction with many polymers such as epoxies, but a third solvent, such as a ketone, ester, or glycol ether, is often included in small amounts to increase the polar parameter/solvency of the mixture. Neither xylene nor n-butanol satisfactorily dissolves an epoxy of higher molecular weight by itself. These are located in boundary regions of the solubility region for epoxies, but on opposite sides of the characteristic Hansen spheres (see Figure 8.2).1 Glycol ethers also can be added to hydrocarbon solvents with advantage, and the polar and hydrogen-bonding parameters are higher than if n-butanol had been added to the same concentration. There are many possibilities, and a solubility parameter approach is particularly valuable in quickly limiting the number of candidates. The addition of glycol ethers or other water-soluble solvents can have adverse effects, such as increased water sensitivity and poorer corrosion resistance of the final film, as some solvent retention must be anticipated, and the least volatile solvent is enriched and left behind. Relative costs for improving solvency from a hydrocarbon base solvent can be estimated by the relation (δP2 + δH2)1/2/cost. This relation has generally pointed to the use of n-butanol, for

7248_C008.fm Page 139 Wednesday, May 23, 2007 11:05 AM

Applications — Coatings and Other Filled Polymer Systems

139

12

10

8

6 δp MEK •

4

• E GMEE •B • E GMEEA

2 ARH 0

ALH 0

2

4

6

8

10

12

δH FIGURE 8.1 Sketch showing location of typical solvents relative to the HSP of a binder. Aliphatic hydrocarbons (ALH) and aromatic hydrocarbons (ARH) do not always dissolve well enough so other solvents must be added to bring the mixed solvent composition into the region of solubility for the binder. Ketones (MEK, methyl ethyl ketone), alcohols (B, n-butanol), or other solvents such as glycol ethers and their acetates (here ethylene glycol monoethyl ether and ethylene glycol monoethyl ether acetate) can be used to do this. The expected solvent improvement at least cost is discussed in the text as the quantity (δP2 + δH2)1/2/cost. Units in the figure are in (cal/cm3)1/2. The choice of solvent today would involve glycol ethers based on propylene glycol as discussed in Chapter 18. (From Hansen, C.M., Färg och Lack, 17(4), 69–77, 1971. With permission.)

example, as a cost-efficient solvent to increase the hydrogen-bonding parameter in particular. Solvents can be ranked in this manner to arrive at the least cost solutions to given solvent selection problems. Coalescing solvents in water-reducible coatings are often (but not always) those with somewhat higher hydrogen-bonding parameters than the polymer, which also means they are water soluble or have considerable water solubility. The distribution between the water phase and the dispersed polymer phase depends on the relative affinities for water and the polymer. Solvents which are not particularly water soluble will preferentially be found in the polymer phase. Such coalescing solvents may be preferred for applications to porous substrates, making certain they are where they are needed when they are needed. Otherwise, water-soluble coalescing solvents would tend to follow the aqueous phase, penetrating the substrate faster than the polymer particles, which also get filtered out and they are not therefore available to do their job in the film when the water evaporates. When water evaporates, the solvent must dissolve to some extent in the polymer to promote coalescence. Of course, this affinity of the coalescence solvent for the polymer is a function of its HSP relative to those of the polymer.

7248_C008.fm Page 140 Wednesday, May 23, 2007 11:05 AM

140

Hansen Solubility Parameters: A User’s Handbook

12

10

8 δp 6 MEK •

4

MC •

0

•B THF •

2

2

4

6

8 δH

10

12

14

16

FIGURE 8.2 Sketch showing formulation principles using two relatively poor solvents in combination to arrive at a good solvent. Xylene (X) can be mixed with n-butanol (B) to arrive at a mixture which can be improved by additions of tetrahydrofuran (THF), methylene chloride (MC), or methyl ethyl ketone (MEK) among others. These three very volatile solvents have often been used in analytical work, paint removers, etc., because they dissolve all of the typical coatings binders shown in the figure. Labeling requirements have dictated other choices in more recent years. Units are (cal/cm3)1/2. (From Hansen, C.M., Färg och Lack, 17(4), 69–77, 1971. With permission.)

Amines are frequently added to water-reducible coatings to neutralize acid groups built into polymers, thus providing a water-solubilizing amine salt. Amine in excess of that required for total neutralization of the acid groups acts like a solvent. Such amine salts have been characterized separately to demonstrate that they have higher solubility parameters than either (acetic) acid or organic bases.6 These salts are hydrophilic and have very little affinity for the polymers used in coatings, which means they are to be found in a stabilizing role in the interface in the aqueous phase while still being attached to the polymer. Electrostatic repulsion contributes to stability as well, and the dispersed solubilized polymer can be visualized in terms of a porcupine with raised quills. Surface-active agents, whether nonionic or ionic, are also to be found where the affinities of the respective parts of their molecules dictate their placement; like seeks like. The hydrophilic end with a high hydrogen-bonding parameter will seek the aqueous phase, and the hydrophobic end will seek out an environment where energy differences are lowest (self-assembly). It might be noted here that some solvents have surfactant-like properties as well. Ethylene glycol monobutyl ether, in particular, has been shown to be a good coupling agent, as well as contributing to lowered surface tension.7 The hydrophobic end of such molecules may reside within the polymer if HSP relations dictate this. Otherwise, if the HSP differences are too great, the hydrophobic portion may be forced to remain in the interfacial region, not being accepted by the aqueous phase either. Increases in temperature especially lead to lower hydrogen-bonding parameters (see Chapter 1, Chapter 3, and Chapter 10). For this reason, solvents with high hydrogen-bonding parameters, such as glycols, glycol ethers, and alcohols, become better solvents for most polymers at higher

7248_C008.fm Page 141 Wednesday, May 23, 2007 11:05 AM

Applications — Coatings and Other Filled Polymer Systems

% H2O

141

T1 T2

T3

T1 > T2 > T3

Time FIGURE 8.3 Sketch of water uptake in a polymer as a function of temperature. Higher temperature leads to more rapid uptake and to higher equilibrium levels. Quenching to a lower temperature (arrow) leads to excess water in the film and possibly to water blisters and delamination (see text for further discussion). (Reprinted from Hansen, C.M., New developments in corrosion and blister formation in coatings, Prog. Org. Coat., 26, 113–120, 1995. With permission from Elsevier Science.)

temperatures. This can markedly affect hot-room stability in water-reducible coatings, for example, as more of the solvent will partition to the polymer phase, which swells, becomes more fluid, and has altered affinities for stabilizing surface-active agents. These may dissolve too readily in the swollen, dispersed polymer. When carefully controlled, these temperature effects are an advantage in water-reducible, oven-cured coatings, leading to higher film integrity, as poor solvents at room temperature become good solvents in the oven after the water has evaporated. A very special destructive effect of water is caused by the reduction of its hydrogen-bonding parameter with increases in temperature. The solubility of water in most polymers is higher at a higher temperature than it is at a lowered temperature because the HSP for the polymer and water match better at the higher temperature. It has been documented in many cases that a rapid quench from hot water to cold water can cause blisters in coatings.8 Previously dissolved water within the film now becomes in excess of that soluble in the film. This can be seen in Figure 8.3 where water uptake curves are shown for three temperatures. The amount and rate of uptake is higher for the higher temperatures. Rapid cooling to below the solubility limit at a lower temperature means the system is supersaturated. Excess water freed by this mechanism has been called SWEAT (soluble water exuded at lowered temperatures). If the SWEAT water cannot rapidly diffuse out of the coating, it will appear as a separate phase, perhaps first as clusters, but ultimately at hydrophilic sites or at a substrate. The coating fails by blistering or delamination. This special effect has been noted by the author in coatings (alkyd, polyester, and epoxy), in rigid plastics such as poly(phenylene sulfide) and poly(ether sulfone), and even in EPDM rubber. Examples of measurements of this type are shown in Chapter 12, Figure 12.3 and Figure 12.4 for an EPDM rubber gasket and for a poly(ether sulfone) tensile bar. This effect is not restricted to water; it has also been seen for an epoxy coating that was repeatedly removed from room temperature methanol to measure weight gain. The cooling due to the methanol evaporation was sufficient to produce methanol blisters near the air surface of the coating because of excess amounts of methanol over that soluble at the lower temperature resulting from the methanol evaporation. The use of supercritical gases as solvents has become possible in recent years. The solubility parameters for carbon dioxide have been reported 9 earlier and in the first edition of this handbook, based on the room temperature solubility of the gas in different liquids. These are now revised as discussed in Chapter 10 to δD, δP, and δH equal to 15.7, 6.3, and 5.7. HSP values are reported as a

7248_C008.fm Page 142 Wednesday, May 23, 2007 11:05 AM

142

Hansen Solubility Parameters: A User’s Handbook

function of temperature and pressure. This same type of analysis can be used to evaluate the temperature and pressure effects for other gases. See also Chapter 3. These parameters are found using the solvents that dissolve more than the theoretical amounts of carbon dioxide that are reported in Table 10.2. The use of such gases is considered an advantage for the environment, but their use has been limited to relatively smaller items because of the size of pressure equipment. Solvent technology has also been used in a wide variety of other products and processes as listed by Barton.2 One can mention the formulation of solvent cleaners based on vegetable oils as an additional example.10 Such “green” products have found increasing use, as have those solvents with low volatility, low VOC, and low labeling requirements.

TECHNIQUES FOR DATA TREATMENT As mentioned earlier, a simple approach to many practical problems is to make a two-dimensional plot of polar vs. hydrogen-bonding parameters with a circle (or estimated circle) for the polymer in question. The circle should encompass the good solvents. One can then plot points for potential solvents and quickly arrive at a starting composition for an experiment. Subsequently, this can be adjusted if necessary. A linear mixing rule based on the volume (or weight) fractions of the solvent components is usually satisfactory. Plasticizers should be included in the calculations. They will be very slow to dissolve rigid polymers, in particular, and are, of course, nonvolatile for all practical purposes. A special plotting technique for solvent selection developed by Teas11 is used frequently by those who restore old paintings. The art involved in this stage of the conservation process is to remove the old varnish without attacking the underlying original masterpiece. HSP principles have been used since the late 1960s for selecting solvents and solvent blends for this purpose.12 The triangular plotting technique uses parameters for the solvents, which, in fact, are modified HSP parameters. The individual Hansen parameters are normalized by the sum of the three parameters. This gives three fractional parameters defined by Equation 8.1 to Equation 8.3. fd = 100δD/(δD + δP + δH)

(8.1)

fP = 100δP/(δD + δP + δH)

(8.2)

fh = 100δH/(δD + δP + δH)

(8.3)

The sum of these three fractional parameters is 100 in the form the equations are written. This allows use of the special triangular technique. Some accuracy is lost, and there is no theoretical justification for this plotting technique, but one does get all three parameters onto a two-dimensional plot with enough accuracy that its use has survived for this type of application (at least). The Teas plot in Figure 8.4 includes an estimate of the solubility/strong attack of older, dried oil paint. A varnish which could be considered for use is Paraloid® B72, a copolymer of ethyl methacrylate and methyl methacrylate from Rohm and Haas. There is a region in the lower, right-hand part of this plot where the varnish is soluble and the dried oil is not. The varnish remover should be in this region. Mixtures of hydrocarbon solvent and ethanol are located in this region and could be considered. HSP correlations for materials of interest in restoration of older paintings are included in Table 8.1. A helpful simplifying relation to use in solvent selection calculations using solubility parameters is that the resultant values for mixtures can be estimated from volume fraction averages for each solubility parameter component. Solvent quality can be adjusted by the RED number concept, which is discussed in Chapter 1 (Equation 1.10), or graphically as described above.

7248_C008.fm Page 143 Wednesday, May 23, 2007 11:05 AM

Applications — Coatings and Other Filled Polymer Systems

0

143

100

fh

fp

DRIED OIL + PAR B72 E PAR B72

100 0

CH 0 100

fd

FIGURE 8.4 Teas plot for a typical painting conservation situation where a varnish is to be removed or applied without attacking the underlying original oil painting. Solvents indicated are cyclohexane (C), heptane (H), and ethanol (E) (see text for further discussion).

TABLE 8.1 HSP Correlations for Materials of Interest in the Conservation of Older Paintings MATERIAL

δD

δP

δH

Ro

FIT

G/T

Paraloid® 22 solubility Dammar gum dewaxed Dried oil (estimate)

17.6 18.4 16.0

7.4 4.2 6.0

5.6 7.8 7.0

9.4 8.3 5.0

1.000 0.915 1.000

17/26 30/56 9/22

A computer search with the SPHERE computer program (Chapter 1) for “nearest neighbors” for a given single solvent has been used many times to locate alternates for a wide variety of product types including coatings of various descriptions, cleaners, etc. A similar application is to predict which other solvents will probably be aggressive to a chemically resistant coating where very limited data have indicated a single solvent or two are somewhat aggressive. A nearest neighbor search involves calculation of the quantity Ra (Chapter 1, Equation 1.9) for a whole database, for example, and then arranging the printout in RED number order (Chapter 1, Equation 1.10). The potentially most aggressive liquids are at the top of the list. Solvents with RED less than 1.0 are “good” and therefore easily recognized. Sorting out these possibilities considering toxicity, evaporation rate, cost, etc. leads to the most promising candidates for the substitution.

7248_C008.fm Page 144 Wednesday, May 23, 2007 11:05 AM

144

Hansen Solubility Parameters: A User’s Handbook

SOLVENTS AND SURFACE PHENOMENA IN COATINGS (SELF-ASSEMBLY) Chapter 6 and Chapter 7 have been devoted to the characterization of surfaces for substrates, pigments, fillers, and the like. This means the interplay between solvent, polymer, and surfaces can be inferred by their relative affinities. These depend on their HSP relative to each other, and the RED number concept can be quite useful. As stated previously, the desired solvent quality in many coatings is just slightly better than that of a marginal solvent. This means RED numbers just under 1.0 relative to the polymer will be sought. One reason for the desired marginal solvent quality is that this will ensure that the polymer adsorbed onto pigment surfaces during pigment dispersion has little reason to dissolve away from that surface. The dispersion stabilizing polymer should remain on the pigment surface where it is desired. If this polymer is dissolved away, the result is most likely pigment flocculation, which leads to color change, undesired settling, and perhaps even rheological difficulties. The solvent in this case should have a RED number for the pigment surface greater than 1.0, or at least reasonably high, to aid in the planned affinity approach to pigment dispersion stability. Of course, the polymer, or some portion of it, and the pigment surface should have high affinity for each other. A sketch of the optimum relations in coatings is given in Figure 8.5 where the marginal solvent is number 1. Solvent 2 would probably be too expensive and, in addition, will probably dissolve the polymer too well. Schröder13 (BASF) confirms that the optimum polymer adsorption will be found when the binder and pigment surface have the same HSP. He indicates that the solvent should be very poor for the pigment and located in the boundary region for the binder. He prefers the pigment to have HSP values placing it intermediate between the solvent and binder. This is suggested for conditions where the solvent has higher HSP than the pigment, as well as for conditions where the solvent has lower HSP than the pigment. This situation, with the solvent and binder on opposite sides of the pigment, means the composite vehicle has parameters very closely matching those of the pigment. A very similar type of result was found by Skaarup,14 who especially emphasized that optimum color strength was found for solvents marginal in quality for the binder and poor for the pigment in question. Pigment

3

δp 2 1

Polymer δh FIGURE 8.5 Sketch showing influence of solvent quality on expected pigment dispersion stability (see text and Figure 1.1 for discussion). (Reprinted from Hansen, C.M., Paint and Coating Testing Manual, 14th ed. of the Gardner-Sward Handbook, 1995, p. 400. With permission. Copyright ASTM.)

7248_C008.fm Page 145 Wednesday, May 23, 2007 11:05 AM

Applications — Coatings and Other Filled Polymer Systems

145

In special applications, an extended polymer chain configuration is desirable, but a solid anchor to the pigment surface is also desired. This means a better-than-marginal solvent for the polymer is desired. A good anchor has high affinity for the pigment surface and marginal affinity for the solvent. Solvent 3 (Figure 8.5) would adsorb onto the pigment surface preferentially, and pigment dispersion stability would be poor. An extension of this thinking may be required for pigment pastes and other very highly filled products. In these cases, there is little dispersing vehicle relative to the pigment, and the solvent must be considered as being part of the dispersing vehicle. In such cases the solvent may have high affinity for the pigment surface as well as for dispersing polymer. An ideal situation here is where all the ingredients have the same HSP.

POLYMER COMPATIBILITY In some cases, closer-than-usual matches between solvent and polymer solubility parameters are required. This is true when two polymers are mixed and one of them precipitates. This is most likely the polymer with the larger molecular weight, and it must be dissolved even better. Lower RED numbers with respect to this polymer are desired, while still maintaining affinity for the other polymer. Miscible blends of two polymers have been systematically found using a solvent mixture composed exclusively of nonsolvents.15 This is demonstrated schematically in Figure 8.6, where it can be seen that different percentage blends of solvents 1 and 2 will have different relative affinities for the polymers. No other alternative theory of polymer solution thermodynamics can duplicate this predictive ability. Polymer miscibility is enhanced by larger overlapping solubility regions for the polymers as sketched in Figure 8.7. Polymers A and B should be compatible, whereas polymer C would not. Such a systematic analysis allows modification of a given polymer to provide more overlap and enhanced compatibility. The advantages of a copolymer containing the monomers of A or B and C should also be evident. Such a copolymer will essentially couple the system together.

1

B 1+2

δp A 2

δh FIGURE 8.6 Sketch showing how two otherwise immiscible polymers can be brought into a homogeneous solution by the use of mixed nonsolvents. (Reprinted from Hansen, C.M., Paint and Coating Testing Manual, 14th ed. of the Gardner-Sward Handbook, 1995, p. 400. With permission. Copyright ASTM.)

7248_C008.fm Page 146 Wednesday, May 23, 2007 11:05 AM

146

Hansen Solubility Parameters: A User’s Handbook

C δp

A B

δh FIGURE 8.7 Sketch describing expected polymer miscibility relations (see text for discussion). (Reprinted from Hansen, C.M., Paint and Coating Testing Manual, 14th ed. of the Gardner-Sward Handbook, 1995, p. 401. With permission. Copyright ASTM.)

Van Dyk et al.16 have correlated the inherent viscosity of polymer solutions with HSP. The inherent (intrinsic) viscosity used in this study, [η], is given by Equation 8.4. [η] = (ηs/η0)/c

(8.4)

ηs is the solution viscosity, η0 is the solvent viscosity, and c is the solution concentration. The concentration used was about 0.5 g/dl. This is an expression reflecting polymer chain extension in solution, with higher values reflecting greater chain entanglements because of greater polymer extension. This is interesting in that the solubility parameter is a thermodynamic consideration, whereas the viscosity is a kinetic phenomenon. Higher [η] were found for solvents with HSP nearest those of the polymer. As stated above additional uses of HSP (and the total solubility parameter) in solvent technology can be found in Barton,2 but these are too numerous to include here. However, a couple of examples relating to guided polymer compatibility are worthy of special mention. These are the formulation of asymmetric membranes for separations,17,18 where polymer solutions — having given HSP relations — and at least one solvent soluble in water are used. The solution is immersed in water, the solvent quality becomes bad, and a controlled porous membrane results. Another example of controlled phase relations during a dynamic process is found in the formulation of self-stratifying coatings. This is discussed in Chapter 6 in terms of the creation of interfaces and therefore interfacial surface tension. The HSP principles involved in this type of coating can be seen in Figure 8.8. The solvent must dissolve both the topcoat and primer and allow the lower surface tension topcoat to migrate to the surface during film formation. Formulation principles have been discussed in detail elsewhere.19,20 Before concluding this section, some of the recent work on miscible polymer blends should also be noted.21,22 This work used group contribution estimates of the δP and δH parameters only in an effort to correlate interfacial tension between polymers, assuming that the δD parameters would not be too different. Although this is a good starting point to prove the procedure has possibilities, further differentiation between the polymers and improved group contribution methods may offer even more improvement.

7248_C008.fm Page 147 Wednesday, May 23, 2007 11:05 AM

Applications — Coatings and Other Filled Polymer Systems

147

POLAR PARAMETER

PRIMER

PARAMETERS REQUIRED FOR COMMON SOLVENT TOP COAT (LOWEST ENERGY)

HYDROGEN BONDING SOLUBILITY PARAMETER

FIGURE 8.8 Sketch illustrating the principles of solvent selection for self-stratifying coatings. (From Birdi, K.S., Ed., Handbook of Surface and Colloid Chemistry, CRC Press, Boca Raton, FL, 1997, p. 324. With permission.)

HANSEN SOLUBILITY PARAMETER PRINCIPLES APPLIED TO UNDERSTANDING OTHER FILLED POLYMER SYSTEMS Recent characterizations of inorganic fillers and fibers23 have confirmed that HSP concepts can be applied to engineered fiber-filled systems such as those based on polypropylene. The behavior of chewing gum can also be analyzed in terms of solubility parameter principles.24 In addition to rheological behavior, appearance, and other performance considerations, a desired product characteristic is that the release of the taste components should be controlled. Greater differences in solubility parameters between flavoring agents and wax-free gum bases lead to enhanced flavor release. Similarity of HSP can lead to stopping the desired release too soon. Perhaps the most important practical work dealing with solubility parameters and the stability of pigment dispersions is that attributable to Stephen.25 He concludes that all the (solid) ingredients in a paint formulation should have the same energy characteristics. If they do not, there will be a driving force for this to occur. This can lead to problems. One can just as well make the formulation stable from the start, and then everything will remain stable just where it is because there are no driving forces for anything to move around. Although this sounds expensive, obvious, and perhaps too simple, the truth of the matter is well documented in very practical terms.

CONCLUSION Many practical uses of the solubility parameter concept have been described in detail, including optimizing solvent selection, improving polymer compatibility, and enhancing pigment dispersion. When all of the materials involved in a given product and application can be characterized with the same affinity (solubility/cohesion) parameters, the possibility exists to predict interactions

7248_C008.fm Page 148 Wednesday, May 23, 2007 11:05 AM

148

Hansen Solubility Parameters: A User’s Handbook

among them. This is true even in complicated situations, such as the formulation of various types of filled systems including coatings, printing inks, adhesives, and other filled polymer systems including chewing gum.

REFERENCES 1. Hansen, C.M., Solubility in the coatings industry, Färg och Lack, 17(4), 69–77, 1971. 2. Barton, A.F.M., Handbook of Solubility Parameters and Other Cohesion Parameters, CRC Press, Boca Raton, FL, 1983; 2nd ed., 1991. 3. Gardon, J.L. and Teas, J.P., Solubility parameters, in Treatise on Coatings, Vol. 2, Characterization of Coatings: Physical Techniques, Part II, Myers, R.R. and Long, J.S., Eds., Marcel Dekker, New York, 1976, chap. 8. 4. Beerbower, A., Boundary Lubrication — Scientific and Technical Applications Forecast, AD747336, Office of the Chief of Research and Development, Department of the Army, Washington, D.C., 1972. 5. Hansen, C.M. and Beerbower, A., Solubility parameters, in Kirk-Othmer Encyclopedia of Chemical Technology, Suppl. Vol., 2nd ed., Standen, A., Ed., Interscience, New York, 1971, pp. 889–910. 6. Hansen, C.M., Some aspects of acid/base interactions (Einige Aspekte der Säure/Base-Wechselwirkung, in German), Farbe und Lack, 83(7), 595–598, 1977. 7. Hansen, C.M., Solvents in water-borne coatings, Ind. Eng. Chem. Prod. Res. Dev., 16(3), 266–268, 1977. 8. Hansen, C.M., New developments in corrosion and blister formation in coatings, Prog. Org. Coat., 26, 113–120, 1995. 9. Hansen, C.M., 25 years with solubility parameters (25 År med Opløselighedsparametrene, in Danish), Dan. Kemi, 73(8), 18–22, 1992. 10. Rasmussen, D. and Wahlström, E., HSP-solubility parameters: a tool for development of new products — modelling of the solubility of binders in pure and used solvents, Surf. Coat. Int., 77(8), 323–333, 1994. 11. Teas, J.P., Graphic analysis of resin solubilities, J. Paint Technol., 40(516), 19–25, 1968. 12. Torraca, G., Solubility and Solvents for Conservation Problems, 2nd ed., International Centre for the Study of the Preservation and the Restoration of Cultural Property (ICCROM), Rome, 1978. (13, Via Di San Michelle, 00153 Rome) 13. Schröder, J., Colloid chemistry aids to formulating inks and paints, Eur. Coat. J., 5/98, 334–340, 1998. 14. Skaarup, K., The three dimensional solubility parameter and its use. II. Pigmented systems, skandinavisk tidskrift for Fårg och Lack, 14(2), 28–42, 1968; 14(3), 45–56, 1968. 15. Hansen, C.M., On application of the three dimensional solubility parameter to the prediction of mutual solubility and compatibility, Färg och Lack, 13(6), 132–138, 1967. 16. Van Dyk, J.W., Frisch, H.L., and Wu, D.T., Solubility, solvency, solubility parameters, Ind. Eng. Chem. Prod. Res. Dev., 24(3), 473–478, 1985. 17. Klein, E. and Smith, J.K., Assymetric membrane formation, Ind. Eng. Chem. Prod. Res. Dev., 11(2), 207–210, 1972. 18. Chawla, A.S. and Chang, T.M.S., Use of solubility parameters for the preparation of hemodialysis membranes, J. Appl. Polym. Sci., 19, 1723–1730, 1975. 19. Misev, T.A., Thermodynamic analysis of phase separation in self-stratifying coatings — solubility parameters approach, J. Coat. Technol., 63(795), 23–28, 1991. 20. Special issue devoted to self-stratifying coatings, Prog. Org. Coat., 28(3), July 1996. 21. Luciani, A., Champagne, M.F., and Utracki, L.A., Interfacial tension in polymer blends. Part 1: Theory, Polym. Networks Blends, 6(1), 41–50, 1996. 22. Luciani, A., Champagne, M.F., and Utracki, L.A., Interfacial tension in polymer blends. Part 2: Measurements, Polym. Networks Blends, 6(2), 51–62, 1996. 23. Hennissen, L., Systematic Modification of Filler/Fibre Surfaces to Achieve Maximum Compatibility with Matrix Polymers, Lecture for the Danish Society for Polymer Technology, Copenhagen, February 10, 1996.

7248_C008.fm Page 149 Wednesday, May 23, 2007 11:05 AM

Applications — Coatings and Other Filled Polymer Systems

149

24. Song, J.H. and Reed, M.A., Petroleum Wax-Free Chewing Gums Having Improved Flavor Release, U.S. Patent No. 5,286,501, February 15, 1994, assigned to Wm. Wrigley Jr. Company, Chicago, IL. 25. Stephen, H.G., Parameters controlling colour acceptance in latex paints, J. Oil Colour Chem. Assoc., 69(3), 53–61, 1986.

7248_C008.fm Page 150 Wednesday, May 23, 2007 11:05 AM

7248_book.fm Page 151 Tuesday, April 24, 2007 9:19 AM

Solubility Parameters 9 Hansen of Asphalt, Bitumen, and Crude Oils Per Redelius ABSTRACT Hansen solubility parameters (HSP) are shown to be a useful new tool for understanding compatibility relations among bitumens and crude oils. Bitumen and crude oils are complex mixtures of hydrocarbons which are kept in solution mainly by their mutual solubility. They are not colloidal dispersions as previously thought. Although the solubility of the hydrocarbons is mainly determined by the dispersive interactions, it is not possible to make correct estimates of their stability without also taking polar interactions and hydrogen-bonding interactions into consideration. HSP have proven their ability to give a good estimate of the stability of bitumen and/or crude oil having different origins in relation to solvents and polymers. Relations between the HSP of different materials is visualized using 3D plots showing the HSP as ellipsoids. A more precise determination of the extension of the ellipsoids can be found by turbidimetric titrations with three different titrants, each representing a direction in the HSP space, respectively. It is now possible with the help of simple laboratory experiments to predict the consequences of different courses of action, thus eliminating expensive trial and error testing.

SYMBOLS SPECIAL TO CHAPTER 9 C P FR pa po

Amount of bitumen/total amount of solvent and titrant Stability index given by Equation 9.3 Volume of solvent/total volume of solvent plus titrant Defined by Equation 9.1 Defined by Equation 9.2

INTRODUCTION Even if most of us are not familiar with bitumen, we all know the “black” roads on which we drive every day. The majority of road surfaces are black because the binding agent used to manufacture the surfacing is bitumen, which is mixed with crushed rock aggregate. Road surfaces can also be grey to white in color, in which case an alternative binder has been used: Portland cement concrete. Bitumen is a semisolid material that can be produced from certain crude oils by distillation. It can also be found in nature as “natural asphalt.” It consists of a mixture of hydrocarbons of different molecular sizes containing small amounts of heteroatoms such as sulfur, nitrogen, and oxygen, as well as traces of metals like vanadium and nickel. Bitumen behaves as a viscoelastic thermoplastic solid at ambient temperature and turns into a viscous liquid at high temperature. It presents unique adhesive and waterproofing properties, which make it ideal in the manufacture of asphalt for road 151

7248_book.fm Page 152 Tuesday, April 24, 2007 9:19 AM

152

Hansen Solubility Parameters: A User’s Handbook

construction and to use in a wide range of industrial application, from waterproofing in construction to sound dampening in the automotive industry. The term bitumen is not completely unambiguous as it has been given different meanings in different parts of our world. In Europe the term is defined as above, whereas in Canada, for example, it is used for heavy crude oils. In the U.S. the term asphalt is used instead of bitumen. Sometimes bitumen is confused with tar, which is a product of completely different origin. Tar is produced by dry distillation of coal or wood. The most common process for production of bitumen is by distillation under vacuum of properly selected crude oils. There are however just a limited number of crude oils which permit direct distillation to proper bitumen grades suitable for production of road asphalt. Although the reserves of such crude oils are very large worldwide, they are not primarily produced as they contain too small amounts of fuel, which is the most important and profitable product for refiners. The functional properties of bitumen are usually related to its use as binder in asphalt for roads. Thus, the most common properties are related to the rheology of bitumen. As the road construction area is very conservative, and bitumen has been used for about 100 years, most tests are empirical and have been used for a long time. Two of the most common tests are penetration at 25°C and softening point Ring&Ball. The penetration gives a measure of the stiffness of the bitumen at most common service temperatures of a road, whereas the Ring&Ball gives the stiffness close to the highest expected temperature in practice. In Europe bitumens are graded according to their penetration at 25°C — for example, 50/70, where the two numbers give the highest and lowest limit for the particular grade. It is also common, particularly in the U.S., to use viscosity gradation based on viscosity at 60°C. Bitumen is, however, a viscoelastic material with a complex rheology and can thus not be completely described by simple penetration testing and softening point. The development of modern and reliable rheometers — for example, the dynamic shear rheometer (DSR) — has made it possible to describe the full rheology of bitumen. During the last 20 years we have seen an increase in the use of polymer modified bitumen (PMB) with improved properties. The main reason for modification of bitumen is to improve the rheological properties, particularly to make the binder less sensitive to temperatures. It is desired to have a reasonable stiffness of the binder even at the highest surface temperatures a road can reach on a hot summer day, as well as being reasonably flexible at the lowest temperatures on a cold winter day. Another reason for modification with polymers is to increase durability. This will be improved if a proper polymer is selected. A large number of different polymers have been tested as modifiers for bitumen. In the end, just a few of them have reached larger commercial use. The main restriction for the choice of polymer is the expected improvement of the rheological properties in comparison with the price of the polymer. But even more important is the compatibility or the solubility of the polymer in the bitumen. Until now, there have been very few tools for prediction of compatibility between the polymer and bitumen, so the development of new PMB has to a large extent been done on a “trial and error” basis. The better understanding of the true nature and the solubility properties of bitumen provided by Hansen solubility parameters (HSP) has given a new tool for understanding of polymer compatibility with bitumen as discussed in the following.

MODELS OF BITUMEN Crude oils have been found in many places around the world. Although the true origin of crude oils is still under discussion, most scientists agree that they have been formed by degradation and transformation of ancient organisms. The properties of crude oil vary depending on age and conditions during formation. Some crude oils are liquids with low viscosity, whereas others are semisolid materials that have a viscosity making them impossible to handle at room temperature. The low viscosity crude oils contain large amounts of fuel but very little bitumen, if any, and the high viscosity crude oils contain very little fuel but large amounts of bitumen.

7248_book.fm Page 153 Tuesday, April 24, 2007 9:19 AM

Hansen Solubility Parameters of Asphalt, Bitumen, and Crude Oils

153

From a chemical point of view crude oil is an extremely complex mixture of hydrocarbons. Usually small amounts of heteroatom like nitrogen, oxygen, and sulfur, as well as trace amounts of metals like vanadium and nickel, are found, although the content varies depending on type and origin of the crude oil. The smallest molecules are the gaseous methane, ethane, and propane. These are dissolved in the liquid hydrocarbons. The heaviest molecules have molecular weights higher than 1000 and are thus hydrocarbons with 70 carbon atoms or more. The separation of crude oils into different fractions is done in refineries by distillation, with the different fractions being collected based on their boiling points. The low-boiling fractions consist of gasoline and gas oils. The constituents in these fractions have been characterized by modern analytical techniques until almost every single component has been identified. The heavier fractions (heavy gas oil), and particularly the residue after distillation, have escaped such detailed characterization. Most residual oils are further upgraded by different refining processes to fuels. Bitumen may be produced only after a proper distillation process of a selected crude oil using vacuum. Although the residual oil and bitumen have been extensively analyzed with modern equipment, most of the understanding is in terms of averages of different chemical functional groups or structures. From these data tentative structures of the molecules have been suggested.1 In fact hardly any one single molecule from the complex mixture has been chemically analyzed. There are several reason why this has been a superior challenge: • • • • •

The number of different molecules is very large. There is no major population of identical molecules. The material is black and viscous. The range of molecules of different polarities and sizes is continuous. The boiling point is higher than approximately 450°C, making the molecules fairly large.

The most common approach for chemical characterization of bitumen involves a separation into generic fractions based on chromatographic principles. The most common separation procedure is called SARA analysis (saturates, aromatics, resins, and asphaltenes). It consists of two principally different steps: first, creation and precipitation of a solid fraction by dilution of the bitumen with n-heptane, and then a separation of the soluble fraction with respect to polarity. The precipitated fraction is called asphaltenes and is defined as the fraction of bitumen that is insoluble in n-heptane. The n-heptane soluble fraction is named maltenes and is further separated by polarity into three more fractions. These fractions have been given names like “resins,” “aromatics,” and “saturates.” The most common and widespread hypothesis about the structure of bitumen, which is found in most books and papers on bitumen chemistry, teaches that bitumen is a colloidal dispersion of asphaltenes in maltenes. The dispersion is assumed to be stabilized by the resins. The first one to introduce this concept was Nelensteyn (1924).2 The model was later refined by Pfeiffer and Saal.3 Although the model might be attractive for mechanical engineering, it is more difficult to accept for an organic chemist, particularly since colloidal dispersions of hydrocarbons in other hydrocarbons are rare, except in the case of polymers. A number of questions are immediately raised: “Do the asphaltenes have enough different chemistry to permit dispersion rather than dissolution?” and “If it is a colloidal dispersion, what is the mechanism for its stabilization?” Other models that question the existence of micelles have also been proposed. Examples of models are the continuous thermodynamic model by Park and Mansoori4 and Buduszynski et al.,5 and the micro structural model as a result from the SHRP development program in the U.S.6 Recent research has shown that the asphaltenes do not form micelles but are soluble in the maltenes, and thus no micelles can exist in the bitumen.7,8 These models describe bitumen as a solution of organic material of different polarity and different molecular weight having a kind of mutual solubility in each other. When a solvent such as n-heptane is added to the system, the balance is disturbed. Part of the system precipitates. The precipitation behavior of asphaltenes is what could be predicted from regular solution theory and could be described as flocculation. In spite of the solubility model

7248_book.fm Page 154 Tuesday, April 24, 2007 9:19 AM

154

Hansen Solubility Parameters: A User’s Handbook

being a more precise description of the true nature of bitumen, it has received surprisingly low acceptance in the research on bitumen and crude oils.

ASPHALTENES During production, transport, and refining of certain crude oils there are sometimes problems with the formation of precipitates and deposits. The deposits have been claimed to be asphaltenes, and therefore there is considerable interest in them and money spent to save, if the formation of precipitates could be controlled. Thus, extensive research has been performed to investigate the chemistry of asphaltenes9,10 as well as mechanisms of formation of the precipitates. There is a clear definition of the term asphaltenes11,12 as the material that precipitates on dilution of bitumen or crude oil with n-heptane. Most of the characterization work has been conducted on precipitated asphaltenes, and very little attention has been given to asphaltenes in their natural environment in the bitumen. Much confusion has come from the misuse of the term asphaltenes to mean all kinds of precipitates from bitumen, suggesting that the insolubles in n-heptane could represent precipitates in general. This assumption might have been correct if the asphaltenes were a colloidal fraction in bitumen, but this it is not the case. As will be proven later in this chapter the cause of formation of precipitates is more related to general solubility rather than just solubility in n-heptane. The mechanism of precipitate formation is certainly not only an academic matter but is of major importance for the whole oil industry as precipitates may cause blocking and fouling of equipment used in crude oil production as well as in transport and refining. It is worth discussing some of the more common statements about the chemistry of asphaltenes and to compare them with experimental facts.

MOLECULAR WEIGHT A general statement about the molecular weight of asphaltenes would be that they are high molecular weight material. The true molecular weight of asphaltenes has been under discussion for many years. Investigations using vapor phase osmometry (VPO) on precipitated asphaltenes dissolved in different solvents have shown molecular weights from 1000 up to 10000, depending on the source of asphaltenes. The apparent molecular weight is strongly dependent on the solvent. This indicates that the asphaltenes associate in solution.1 Other attempts to determine molecular weight using field ionization mass spectrometry (FIMS) reveal an apparent molecular weight of 700–1000. These results also vary depending on crude oil source.5 It is obvious that the VPO overestimates the true molecular weight due to interactions between the molecules, and FIMS likely gives a more correct value, although there might be a risk that some degradation has taken place in the ion source. Recent studies with fluorescence depolarization techniques have confirmed the FIMS results.13 It may be speculated that large size molecules are less soluble in n-heptane, and thus asphaltenes should consist mainly of high molecular weight material. A high dependency of molecular weights on solubility is well known from polymers. There are, however, several hydrocarbons of lower molecular weight that are not soluble in n-heptane (for example, coronene or dibenz(a,h)anthracene, where the very high aromatic content leads to very high dispersion parameters compared with the relatively low dispersion parameter for n-heptane in the HSP concept), and similar molecules may be part of the asphaltenes fraction. It is thus reasonable to assume that the lowest molecular weight in the asphaltenes is equal to the smallest molecule with a boiling point at the cut-point of the bitumen. This varies with different crude oils but may be estimated as being 500°C. This roughly corresponds to hydrocarbons with 35 carbon atom, less for polycyclic aromatics and more for nalkanes. The conclusion is that the asphaltenes fraction likely consists of a range of molecules of different molecular weight, which might range from as low as 300 up to more than 1000.

7248_book.fm Page 155 Tuesday, April 24, 2007 9:19 AM

Hansen Solubility Parameters of Asphalt, Bitumen, and Crude Oils

155

POLARITY

Asphaltenes are claimed to be a “highly polar” fraction in bitumen, in contrast to the more nonpolar oils (maltenes). This statement is based on the fact that asphaltenes are insoluble in n-heptane, a nonpolar solvent. The asphaltenes are, however, easily soluble in relatively nonpolar solvents like benzene, toluene, and dichloromethane, whereas they are insoluble in polar solvents like water, glycerine, and methanol. It is thus more correct to state that the asphaltenes are not polar in a chemical sense, but they might be considered as more polar than the other hydrocarbons in bitumen and crude oil. As nitrogen and oxygen are the only atoms in asphaltenes that could contribute significantly to a permanent polarity, an estimation of the relative polarity can be made by considering the amounts of nitrogen and oxygen atoms compared to the amount of carbon atoms. Elemental analyses have revealed that the total amount of oxygen and nitrogen in the asphaltenes is usually lower than 4%.14 This is not more than about one to three nitrogen and oxygen atoms per asphaltene molecule assuming a molecular weight of about 1000. This is not enough to make them particularly polar. The apparent polarity might, however, be increased by the content of polyaromatic compounds in some asphaltenes. These are polarizable and thus may act as polar molecules in contact with other polar molecules. In spite of this, the asphaltenes remain mainly nonpolar, and the claims that they are highly polar have without any doubts been misleading in the attempts to understand the role of the asphaltenes in bitumen and crude oils.

SOLUBILITY PARAMETERS OF BITUMEN The first attempts to determine the solubility parameters of bitumens were made using the Hildebrand solubility parameter concept.15–20 The focus in these investigations was to study the onset of precipitation of asphaltenes and their solubility properties. In these investigations traditional systems using ratios between a good solvent and a poor solvent are used. The choice of good solvent was usually toluene and the poor solvent was usually n-heptane, but sometimes other n-alkanes were used. This approach gives reasonably good results, as long as it is in accordance with the definition of asphaltenes. As bitumen and crude oil mainly consist of hydrocarbons, the simple Hildebrand solubility parameters were believed to give a good prediction of solubility properties. When the solubility properties of bitumen are extended to more varied types of solvents than aromatic and aliphatic hydrocarbons, the good agreement with the Hildebrand solubility parameter is to some extent lost.21 The authors of Reference 22, for example, found that all good solvents for bitumen fall between = 15 MPa1/2 and = 23 MPa1/2, but not all solvents in this range were good solvents. This shows that the Hildebrand solubility parameters are not appropriate for bitumen, probably because there are other interactions between the molecules that are not taken into consideration. The authors of this paper and others23 found that using two-dimensional solubility parameters gives a better description of the solubility properties, but the best estimation was given by the Hansen three- dimensional solubility parameter.24,25 There are still some deviations. This indicates that the prediction could be slightly improved if more than three types of interactions are used, but this will make the model unnecessarily complicated. Determination of solubility parameters of bitumen and crude oil is rather complicated as these consist of a very complex mixture of hydrocarbons. In fact, it is not completely evident that solubility parameters should be applicable for such mixtures, and particularly not if the assumed colloidal model would be correct. Use of common methods based on physical and chemical parameters, which easily can estimate the solubility parameter of pure compounds, cannot be applied to such complicated mixtures as bitumen. The best approach is probably to make solubility tests of the material in a large number of solvents with known solubility parameters and then try to find the best average of the good solvents. Even this seemingly simple approach is rather complicated, however, when applied to bitumen. The first complication comes from the fact that bitumen is very black, and it is rather difficult to see with the eye whether the solution is clear or not. Another complication is that several solvents

7248_book.fm Page 156 Tuesday, April 24, 2007 9:19 AM

156

Hansen Solubility Parameters: A User’s Handbook

may partly dissolve the bitumen, leaving a small precipitate or residue. The third complication is that one has to take the mutual solubility between the bitumen molecules into consideration. The effect of the mutual solubility is that a higher concentration of bitumen results in better solubility, which is contradictory to normal solubility theory that teaches that a saturation level for the solute is reached. In this case solubility becomes better for higher concentrations.

TESTING OF BITUMEN SOLUBILITY Solubility testing may be used for calculation of the solubility parameters of bitumen by the method given in the following. In the testing of the solubility we find that most solvents give a kind of partial solubility with more or less residue. As bitumen is very black, it is sometimes difficult to notice small traces of precipitate. In uncertain cases a drop of the solution can be placed on a filter paper. If a black dot appears at the spot of the drop, the solution contains precipitate, but if the staining of the filter paper is a uniform darkish brown, it does not contain any precipitate. As it is so difficult to estimate true solubility, it is sometimes better to give a grading of the solubility in several steps, although the final calculation requires only “soluble” or “not soluble.” In an experiment using 15 different bitumens, the solubility was determined in 6 different grading levels, ranging from completely soluble to completely insoluble.26 Each level of solubility was designated as a solubility grade according to the following rules: 1. 2. 3. 4. 5. 6.

Totally dissolved: no residue by filter paper test. Almost totally dissolved: light residue was noticed by filter paper test. Partly dissolved: large residue was noticed in dark brown liquid. Slightly dissolved: large residue was noticed in red-brown liquid. Very slightly dissolved: mainly residue in brownish liquid. Not dissolved: colorless or almost colorless liquid.

The bitumens were selected to cover a wide variation of different properties. Some samples were taken from the market, and some were made experimentally for this purpose. It is known that the solubility of bitumen is concentration dependent. Thus, a fixed concentration was used in all experiments to get comparable data. In all experiments, 0.5 g bitumen was dissolved in 5 ml solvent. In most cases the samples were left to dissolve for at least 24 h and sometimes for up to 48 h.

HILDEBRAND SOLUBILITY PARAMETERS Solubility data for 15 different bitumens are given in Table 9.1. All solvents with no visible residue (grade 1) were considered as “good solvents,” and all others were considered “poor solvents.” A bar diagram of the solvents for bitumen No. 1 in relation to the Hildebrand solubility parameter is given in Figure 9.1. It is evident that the majority of the “good solvents” can be found in a range between = 17.8 MPa1/2 and = 25.8 MPa1/2, but it is also obvious that several “poor solvents” are found in the same range. The range of solubility parameters is slightly higher than claimed in Reference 21, which is probably due to a slightly different selection of solvents and bitumen types. The results confirm the earlier findings that the Hildebrand solubility parameter is of little or no value to predict solubility properties or compatibility between solvents or other materials with bitumen. One may speculate that the reason could be the presence of other kinds of interactions in bitumen such as, for example, polar interactions, hydrogen bonding, or π-interactions between the molecules. If these interactions are of significant importance, it explains the poor correlation with the Hildebrand solubility parameter, and also indicates that a better correlation may be achieved when more interactions are taken into consideration.

7248_book.fm Page 157 Tuesday, April 24, 2007 9:19 AM

Hansen Solubility Parameters of Asphalt, Bitumen, and Crude Oils

157

TABLE 9.1 Solubility Test of 15 Different Bitumens in 42 Different Solvents Solvent HSP No. and Name

7 11 46 52 92 102 115 122 148 182 209 234 255 263 297 303 306 325 326 328 367 368 375 376 380 397 417 438 456 481 491 521 524 532 534 536 584 585 617 637 649 698

– Acetone – Acetophenone – Aniline – Benzene – 1-Butanol – n-Butyrolactone – y-Butyrolactone – Carbon tetrachioride – Chloro benzene – Cyciohexanol – Diacetone alcohol – Dichlorobenzene – Diethylether – Diethylene glycol – Dimethylformamide – Dimethylsulfoxide – 1,4-Dioxan – Ethanol – Ethanolamine – Ethyl acetate – 1,2-Dichloroethane – Ethylene glycol – Ethylene glycol butyl ether – Ethylene glycol ethyl ether – Ethylene glycol methyl ether – Formamide – n-Hexane – Isophorone – Methanol – Methylethyl ketone – Methylisobutyl ketone – N-Methyl-2-pyrrolidone – Methylene chloride – Nitroethane – Nitromethane – 2-Nitropropane – Propylene carbonate – Propylene glycol – Tetrahydrofuran – Toluene – Trichloroethylene – Xylene

1

2

3

4

5

4 1 4 1 5 2 5 1 1 3 5 1 3 6 5 5 3 5 6 4 1 6 3 5 5 6 3 1 6 3 2 3 1 5 5 4 6 6 1 1 1 1

4 2 4 1 5 2 5 1 1 4 5 1 3 6 4 5 3 6 6 4 1 6 3 5 5 6 3 1 6 3 1 3 1 4 5 4 5 6 1 1 1 1

4 2 4 1 5 3 5 1 1 5 5 1 3 5 4 5 3 5 5 4 1 6 4 5 5 6 3 1 6 4 3 3 1 4 5 4 6 6 1 1 1 1

4 3 4 1 5 2 5 1 1 3 5 1 3 6 4 5 3 5 5 4 1 6 3 5 5 6 3 1 6 3 2 3 1 4 5 4 5 6 1 1 1 1

4 3 4 1 5 2 5 1 1 3 5 1 3 6 5 5 3 6 6 4 1 6 3 5 5 6 3 1 6 3 2 3 1 5 5 4 6 6 1 1 1 1

Bitumen Sample – Code No. 6 7 8 9 10 11

12

13

14

15

Solubility Grade 4 4 4 4 1 1 2 2 4 5 4 4 1 1 1 1 5 5 5 5 2 3 3 2 5 5 5 5 1 1 1 1 1 1 1 1 3 4 3 3 5 5 5 5 1 1 1 1 2 3 3 2 6 6 6 6 4 5 4 4 5 5 5 5 2 3 3 3 6 6 6 6 6 6 6 6 3 3 4 4 1 1 1 1 6 6 6 6 3 3 3 3 5 5 5 5 5 5 5 5 6 6 6 6 3 3 3 3 1 1 1 1 6 6 6 6 3 3 3 3 2 3 2 2 3 3 4 3 1 1 1 1 4 5 4 5 5 5 5 5 4 4 4 4 6 6 6 6 6 6 6 6 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

4 2 4 1 5 2 5 1 1 3 5 1 2 6 4 5 3 5 5 4 1 6 3 4 5 6 3 1 6 3 2 3 1 4 4 3 5 6 1 1 1 1

4 2 4 1 5 2 5 1 1 3 5 1 2 6 4 5 3 6 6 3 1 6 3 5 5 6 3 1 6 3 2 3 1 4 5 4 5 6 1 1 1 1

4 1 5 1 5 2 5 1 1 5 5 1 3 6 4 5 3 6 6 4 1 6 3 5 6 6 3 1 6 3 2 3 1 4 5 4 6 6 1 1 1 1

4 2 4 1 5 2 5 1 1 5 5 1 3 6 4 5 3 6 6 4 1 6 3 5 5 6 3 1 6 3 3 3 1 5 5 4 6 6 1 1 1 1

4 1 4 1 5 2 6 1 1 3 5 1 2 6 4 5 3 6 6 3 1 6 3 5 5 6 3 1 6 3 2 3 1 5 5 5 6 6 1 1 1 1

Note: The solubility is graded from 1 (completely soluble) to 6 (completely insoluble).

4 2 4 1 5 2 5 1 1 3 5 1 2 6 4 5 3 6 6 4 1 6 3 5 5 6 3 1 6 3 2 3 1 4 5 4 6 6 1 1 1 1

7248_book.fm Page 158 Tuesday, April 24, 2007 9:19 AM

158

Hansen Solubility Parameters: A User’s Handbook

FORMAMIDE ETHYLENGLYCOL ETHANOLAMINE PROPYLENGLYCOL METHANOL DIETHYLENGLYCOL PROPYLENCARBONAT DIMETHYLSULFOXIDE ETHANOL γ-BUTYROLACTONE DICHLORBENZENE NITROMETHANE DIMETHYLFORMAMIDE GLYCOLMETHYLETHER GLYCOLETHYLETHER METHYLPYRROLIDONE NITROETHANE ANILINE CYCLOHEXANOL 1-BUTANOL CHLORBENZEN ACETOPHENONE GLYCOLBUTYLETHER 1,2-DICHLOROETHANE 2-NITROPROPANE 1,4-DIOXAN METHYLENECHLORIDE ISOPHORONE ACETONE TETRAHYDROFURAN METHYLETHYLKETONE TRICHLORETHYLENE BENZENE TOLUENE ETHYL ACETATE o-XYLENE C-TETRACHLORIDE n-BUTYLACETATE DIACETONALCOHOL METHYLBUTYLKETONE DIETHYLETHER n-HEXANE

0.0

5.0

10.0

15.0

20.0

25.0

Hildebrand solubility parameter

30.0

35.0 MPa0.5

FIGURE 9.1 Solubility of bitumen No 1 (Table 9.1) in different solvents of known Hildebrand solubility parameter. White bars = poor solvents, gray bars = good solvents.

HANSEN SOLUBILITY PARAMETERS (HSP) The data set for bitumen No.1 in Table 9.1 was used for testing whether HSP gives a better model for bitumen solubility than Hildebrand solubility parameters. HSP consists of three components, each giving a quantitative value for the dispersion (D), polar (P), and hydrogen bonding (H)

7248_book.fm Page 159 Tuesday, April 24, 2007 9:19 AM

Hansen Solubility Parameters of Asphalt, Bitumen, and Crude Oils

159

insoluble ▼ soluble

Bitumen 25 ctions polar intera

20 15 10 5

di sp

er

▼▼ ▼ 0 ▼ ▼ ▼ ▼ 15 ▼

ei

siv

er nt

20

ac ns

tio

25 0

20 15 ng 10 nd i o b 5 en rog hyd

25

FIGURE 9.2 Plot of the solvents in Table 9.1, bitumen No.1, in a 3D, x-y-z plot, where each axis is one of the Hansen solubility parameters.

interactions, respectively. The suitability of HSP may be illustrated by using a three-dimensional (3D) diagram where each axis constitutes one of the interactions. All solvents with a solubility grade 1 were considered as “good solvents” and all other grades as “poor solvents.” The result is illustrated in Figure 9.2 where all “good solvents” are falling within a certain region separated from the “poor solvents.” This confirms that the solubility properties of bitumen can be reasonably well predicted by HSP. Although the “good solvents” are found in a region of relatively high dispersion interaction and relatively low polar and hydrogen bonding interaction, it seems like the latter two types of interactions are still of fundamental importance for understanding the properties of bitumen. Even if we cannot completely rule out the possibility that there exist other types of interactions, we may, however, conclude that the HSP estimate is good enough, and particularly for understanding the true nature of bitumen. It can be assumed that the same situation is valid also for crude oils, which indicates that the use of HSP would be a valuable tool, also, for crude oil production, transport, and processing.

THE SOLUBILITY SPHERE Chapter 1 includes a discussion of a computer program called SPHERE for calculation of the best estimated HSP as well as the radius of the best fitted pseudo sphere, which includes the “good solvents” and excludes the “poor solvents,” based on a set of solubility data. The program was applied on the data in Table 9.1 for calculation of the best estimate for HSP for 15 bitumens. The program permits only two levels of solubility, “good solvents” and “poor solvents,” however, whereas the solubility in Table 9.1 was determined in 6 grades. For comparison, the HSP were calculated using two different criteria for “good solvents.” In the first calculation only the best solvents (grade 1) were selected as “good solvents” and then in a second calculation the two best grades (1 and 2) were taken as “good solvents.” All other solvents were considered as “poor solvents.” The results are listed in Table 9.2. It is obvious that the calculated HSP for the different bitumens become slightly different, depending on the choice of solubility grade for the “good solvents.” Although the different bitumens are selected to represent a range of products produced

7248_book.fm Page 160 Tuesday, April 24, 2007 9:19 AM

160

Hansen Solubility Parameters: A User’s Handbook

TABLE 9.2 Calculated HSP for 15 Bitumens Using Two Levels of Solubility as the “Good Solvents” Grade 1 = “Good Solvents” Bitumen

D

P

H

RAD

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Average

18.0 17.6 17.9 17.9 17.9 18.0 18.0 18.0 17.9 17.9 17.9 17.9 17.9 18.0 17.9 17.91

4.8 5.0 4.5 4.5 4.5 4.8 4.8 4.8 4.5 4.5 4.5 4.5 4.5 4.8 4.5 4.63

3.2 2.8 3.3 3.3 3.3 3.2 3.2 3.2 3.3 3.3 3.3 3.3 3.3 3.2 3.3 3.23

5.5 5.5 5.3 5.3 5.3 5.5 5.5 5.5 5.5 5.5 5.5 5.5 5.5 5.5 5.5 5.46

Grades 1 and 2 = “Good Solvents” D 17.9 17.9 18.0 17.5 17.5 17.4 17.9 18.0 18.1 17.4 17.4 17.4 17.4 17.9 18.1 17.72

P

H

5.1 5.1 4.8 4.7 4.7 4.0 3.3 4.8 5.5 4.0 4.0 4.0 4.0 5.1 5.3 4.56

3.1 3.1 3.2 2.7 2.7 2.0 2.5 3.2 2.9 2.0 2.0 2.0 2.0 3.1 3.1 2.64

RAD 5.8 5.8 5.5 5.7 5.7 6.6 7.3 5.5 6.0 6.6 6.6 6.6 6.6 5.8 5.9 6.13

from different crude oils as well as different process conditions, the difference in HSP is surprisingly small. The average HSP for bitumen based on calculations using “grade 1” as “good solvents” are D = 17.9 MPa0.5, P = 4.6 MPa0.5, and H = 3.2 MPa0.5. The “sphere” radius (RAD) is 5.5 in the same units. The small variation in HSP between the different bitumens is a result of the small variation in solubility as seen in the data from Table 9.1, with only a few solvents giving different solubility for different types of bitumen. If solvents giving a small residue (solubility grade 2) are accepted as “good solvents,” one still gets a very similar average HSP, but the variation between the different binders becomes more evident. The main general trend is a small shift toward lower hydrogenbonding interactions and a larger radius of the solubility sphere. The larger radius is an expected consequence when more solvents are accepted as “good solvents.” The decrease in hydrogen bonding is more difficult to explain, but it might indicate that the “sphere” is not completely symmetrical. In applications where bitumen is used — for example road building and water proofing — it is well known that bitumen produced by different methods and from different crude oils have different performance. Although the 15 bitumens listed in Table 9.1 are primarily intended for use in the water-proofing industry, they are selected and manufactured to cover a wide variety of crude sources as well as different types of manufacturing processes. Laboratory experiments, and field experience for some of the samples, show that there is a large variation in performance of the bitumens. One example is the compatibility with polymers, such as styrene/butadiene/styrene (SBS), which varies to a large extent. It is expected that some of these differences should be reflected in the different chemical compositions of the bitumens and that these same differences should also be reflected in the HSP. The results given in Table 9.2 show, however, that there are only very small differences, particularly when calculated with only the best solvents as “good solvents.” If “grade 2” is also accepted as “good solvent,” the variation between the binders becomes more evident, but a comparison with known composition and performance still does not allow a simple correlation.

7248_book.fm Page 161 Tuesday, April 24, 2007 9:19 AM

Hansen Solubility Parameters of Asphalt, Bitumen, and Crude Oils

161

This lack of correlation is without any doubt disappointing. We may however speculate that it is mainly due to a lack of precision. The solvents in Table 9.1 are selected to cover a large area in the 3D solubility space, whereas most bitumens are mixtures of hydrocarbons where the differences in chemical properties are relatively small. Obviously it is necessary to have much better precision than the solubility testing as shown in Table 9.2. The better precision may be achieved in two ways. The first improvement is to use a better selection of solvents for the solubility testing. Solvents that have HSP close to the border of solubility are preferred to better define the border. Another approach is to perform turbidimetric titrations to estimate the exact HSP at the precipitation point calculated from the ratio of a “good solvent” and a “poor solvent” at precipitation. This approach is further discussed as BISOM titrations below. An improved selection of solvents should focus on solvents with RED values around 1, as these are close to the boundary. The RED (relative energy difference) concept is discussed in Chapter 1. As much variation as possible with respect to the dispersion, polar, and hydrogen bonding interactions is desired. This requires, of course, that an approximate HSP of the material is already available. And finally, nontoxic and inexpensive solvents are preferred. A suggested set of solvents, optimized for determination of HSP of bitumens and similar materials is presented in Table 9.3. These solvents have RED between 0 and 2 related to the estimated HSP of bitumen as presented above. When using this set of solvents for a Venezuelan binder, the HSP is D = 18.6 MPa0.5, P = 3.0 MPa0.5, H = 3.4 MPa0.5, and the radius of the sphere is 6.3 in the same units. This set of numbers is different from the previously estimated values in Table 9.2. A comparison can be made with binder No. 9 (D = 17.9 MPa0.5, P = 4.5 MPa0.5, H = 3.3 MPa0.5, and a radius of 5.5 MPa0.5), which is similar to the binder used to obtain the data reported in Table 9.3. If the HSP of other types of materials than bitumen are going to be measured, also in the petroleum area, it is suggested that other sets of solvents may be needed to get the best precision. Examples are light crude oils, distillates, base oils, petroleum waxes, etc.

COMPUTER PROGRAM FOR CALCULATION AND PLOTTING OF THE HANSEN 3D PSEUDOSPHERE The SPHERE program described in Chapter 1 has given very good approximations of the HSP as well as the diameter of the (solubility) sphere for a large number of materials. In the SPHERE program, a factor 4 is used as a multiplier for the difference in the dispersion interactions of the species concerned. This means that the “sphere,” with the three different types of interactions as coordinates, is in fact an ellipsoid (spheroid). A disadvantage with the SPHERE program is the lack of a tool for plotting the ellipsoid in a diagram that would be beneficial for illustration purposes. Thus, an improved program which permits 3D plotting of the ellipsoid was developed. During the development it was discussed that although the factor 4 has been proven to be a good approximation for most materials there might be complex mixtures which could give a better fit with experimental data if other factors were used. The new program has the following features: • • • • •

Permits plotting of the HSP solubility ellipsoid in a 3D diagram. Permits plotting of up to three ellipsoids representing different materials in the same 3D diagram. The input data should be based on “poor solvents,” “good solvents,” and “borderline solvents.” There should be an option to make other types of fitting than the SPHERE program to the available data. Negative values of HSP interaction coefficients are not taken into consideration.

7248_book.fm Page 162 Tuesday, April 24, 2007 9:19 AM

162

Hansen Solubility Parameters: A User’s Handbook

TABLE 9.3 Solvents Used for Determination of the Solubility of Bitumen with Their HSP in MPa0.5 HSP No. 56 93 717 1060 118 955 156 182 183 184 188 194 1019 791 269 1084 889 328 333 353 345 758 412 419 440 1063 450 464 472 481 500 502 1029 524 531 546 1051 704 617 618 885 637 953

Solvent

D

P

H

Solubility

Benzophenone 2-Butanol 2-Butyl octanol Butyraldehyde Caprolactone (epsilon) 1-Chloro pentane Chloroform Cyclohexanol Cyclohexanone Cyclohexylamine Cyclopentanone cis-Decahydronaphthalene 1.4-Dichlorobutane 1.1-Diethoxy ethanol (acetal) Ethylene glycol monoethyl ether acetate Diisopropylamine 1.2-Dimethoxybenzene Ethyl acetate Ethyl benzene Ethyl lactate 2-Ethyl-hexanol Ethylene glycol dibutyl ether Hexadecane Hexyl acetate Isopropyl acetate Laurylalcohol Mesityl oxide Methyl acetate Methyl benzoate Methyl ethyl ketone 1-Methyl naphthalene Methyl oleate 3-Methyl-2-butanol Methylene dichloride Nitrobenzene Oleyl alcohol Pyrrolidine Salicylaldehyde Tetrahydrofuran Tetrahydronaphthalene 1.2.3.5-Tetramethylbenzene Toluene 2-Toluidine

19.6 15.8 16.1 15.6 19.7 16.0 17.8 17.4 17.8 17.2 17.9 18.8 18.3 15.2 16.2 14.8 19.2 15.8 17.8 16.0 15.9 15.7 16.3 15.8 14.9 17.2 16.4 15.5 17.0 16.0 20.6 14.5 15.6 18.2 20.0 14.3 17.9 19.4 16.8 19.6 18.6 18.0 19.4

8.6 5.7 3.6 10.1 15.0 6.9 3.1 4.1 6.3 3.1 11.9 0 7.7 5.4 5.1 1.7 4.4 5.3 0.6 7.6 3.3 4.5 0 2.9 4.5 3.8 6.1 7.2 8.2 9.0 0.8 3.9 5.2 6.3 8.6 2.6 6.5 10.7 5.7 2.0 0.5 1.4 5.8

5.7 14.5 9.3 6.2 7.4 1.9 5.7 13.5 5.1 6.5 5.2 0 2.8 5.3 9.2 3.5 9.4 7.2 1.4 12.5 11.8 4.2 0 5.9 8.2 9.3 6.1 7.6 4.7 5.1 4.7 3.7 13.4 6.1 4,1 8.0 7.4 14.7 8.0 2.9 0.5 2.0 9.4

1 0 0 0 0 1 1 0 1 1 0 1 1 0 0 0 0 0 1 0 0 0 0 1 0 0 0 0 1 0 1 0 0 1 0 0 1 0 1 1 1 1 0

7248_book.fm Page 163 Tuesday, April 24, 2007 9:19 AM

Hansen Solubility Parameters of Asphalt, Bitumen, and Crude Oils

163

TABLE 9.3 (CONTINUED) Solvents Used for Determination of the Solubility of Bitumen with Their HSP in MPa0.5 HSP No. 648 653 667 670 698

Solvent 1.1.2-Trichloroethane Tricresyl phosphate 1.2.4-Trimethylbenzene 2.2.4-Trimethylpentane o-Xylene

D

P

H

Solubility

18.2 19.0 18.0 14.1 17.8

5.3 12.3 1.0 0 1.0

6.8 4.5 1.0 0 3.1

1 0 1 0 1

Note: Good solvents are indicated with a “1” and poor solvents are indicated with a “0.” This set of solvents better defines the boundary region as discussed in the text.

The computer program hsp3D was developed on a MATLAB platform.27 The program permits 6 different kinds of fit to create a 3D body, based on a large set of solubility data. In each case all good solvents are included and all poor solvents are excluded. 1. Convex hull fit which could be described as the points for the good solvents being wrapped with a flexible membrane. This fit makes use of only the good solvents. 2. The Hansen fit is the same type of fit as in the SPHERE program using Equation 1.9. The search algorithm is however slightly different, so the results compared to the SPHERE program might be slightly different. 3. Axis-aligned ellipsoid fit, which is similar to the Hansen fit above, but with variable coefficients for the three axes (the three types of interactions). In the normal Hansen fit a factor 4 is used for transformation of the dispersion interactions, in the axis-aligned fit this factor as well as the factors for the other two axes are adjusted to optimize the fit. 4. Rotated ellipsoid fit, which is similar to the Axis-aligned ellipsoid above but allows the body to rotate and tilt to obtain a better fit. In all of the fits above it is assumed that the axis of the ellipsoid is aligned along the three axes. In the rotated ellipsoid the program can tilt the axes to improve the fitting, and at the same time also optimize the transformation factors for the axes. 5. Rotated ellipsoid with convex hull center and volume. The body for this fit has the same center coordinates and volume as the convex hull but attempts to align with the “good solvents” to minimize distance to its surface. 6. Minimum enclosing ellipsoid is the body with the smallest volume that encloses all the “good solvents.” The features of the improved computer program hsp3D were further examined using the solubility data from Table 9.3. The results from the different available fits were compared in 3D diagrams with three different fits in each (Figure 9.3 and Figure 9.4). From Figure 9.3 it is evident that there is a very small difference between the resulting ellipsoids using different fitting algorithms. Transformation or tilting of the axis did not give any major improvement compared to the SPHERE data. This indicates that the factor 4 in Equation 1.9 seems also to be valid for such complicated mixtures as bitumen. In Figure 9.4 we see a comparison between the convex hulls, which probably is the best figure to describe the solubility properties, as it is the truest body constructed without approximations. This might be the first choice if different materials are going to be compared. Another way of comparing the quality of the fit using the different algorithms is to compare some indicators like volume, number of outliers, and fitting coefficient.

7248_book.fm Page 164 Tuesday, April 24, 2007 9:19 AM

164

Hansen Solubility Parameters: A User’s Handbook

Venezuelan bitumen soluble Venezuelan bitumen insoluble

14

δH H-bonding

12 10 8 6 4 2 0 15 10

20 18

5 16 δP Polar

0

δD Dispersive

FIGURE 9.3 3D solubility body of bitumen using computer program hsp3D. The ellipsoids according to Hansen fit, axis-aligned ellipsoid, and rotated ellipsoid are compared.

It can be seen in Table 9.4 that the HSP for the particular Venezuelan bitumen, and most likely also for other bitumens, is more or less independent of the fitting method. This shows that the approximation with an ellipsoid is rather robust. The best solubility body is the one having the smallest volume, the least number of outliers, and the highest fitting coefficient. The Hansen sphere and the axis aligned ellipsoid give almost the same results. The rotated ellipsoid gives a smaller volume but at the expense of more outliers and less good fitting. The most extreme case is the ellipsoid with the same center point (HSP) and the same volume as the convex hull, which gives the smallest volume, most outliers, and less good fitting. This is, of course, a result of the algorithm. If a body with multiple corners is transferred to an ellipsoid with the same volume, most of the corners mathematically will fall outside the ellipsoid. The fact that the coordinates are different indicates that the convex hull is skewing for this material compared to the Hansen Sphere. This might, however, also be due to an uneven selection of solvents rather than properties in the material.

COMPONENTS OF BITUMEN Bitumen is a very complex mixture of different hydrocarbons but yet with very similar properties. It is almost impossible to isolate chemically uniform fractions; instead, bitumen is usually divided into fractions that are defined by the selection of the separation method. Perhaps the most common separation of bitumen is the precipitation of asphaltenes from the maltenes. As stated above, the definition of asphaltenes is the material that precipitates upon dilution of bitumen (or oil) with nheptane.11,12 The fractionation could also be considered as an extraction of n-heptane soluble molecules from the bitumen, leaving a residue named “asphaltenes.” The asphaltene-free fraction from bitumen is called “maltenes.” In almost all of the literature about bitumen and asphalt it is

7248_book.fm Page 165 Tuesday, April 24, 2007 9:19 AM

Hansen Solubility Parameters of Asphalt, Bitumen, and Crude Oils

165

Venezuelan bitumen soluble Venezuelan bitumen insoluble

14 12 δH H-bonding

10 8 6 4 2 0 15 10 20 5

δP Polar

18 16

0

δD Dispersive

FIGURE 9.4 Plots of fit of bitumen solubility data using hsp3D showing the Convex hull model, the ellipsoid with the same center and volume, and also the minimum enclosing ellipsoid.

TABLE 9.4 Precision Indicators for Fitting the Data in Table 9.3 to Ellipsoids Type of Fitting

D

P

H

Volume

Outliers

Fit Coeficient

Hansen Sphere Axis aligned ellipsoid Rotated ellipsoid Ellipsoid: convex hull c and v Minimum enclosing ellipsoid

18.4 18.3 18.4 18.0 18.4

3.9 3.9 4.1 4.4 4.1

3.6 3.5 3.6 4.1 3.7

399 399 242 99 371

3 3 5 10 6

0.980 0.987 0.939 0.798 0.983

Note: Outliers = number of “good solvents” with RED > 1 + number of “poor solvents” with RED < 1.

erroneously claimed that the asphaltenes are dispersed in the maltenes as a colloidal dispersion. That this is not correct can easily be proven by solubility testing and plotting of the solubility ellipsoids using the hsp3D program. Asphaltenes isolated by the standard method ASTM D656012 have been tested for solubility in the set of solvents listed in Table 9.3. The isolated maltene fraction is also tested for solubility in the same set of solvents. The solubility ellipsoids for the two materials are plotted using the hsp3D program (Figure 9.5). Figure 9.5 confirms that there is no overlap of the HSP for n-heptane and the ellipsoid for asphaltene, and it can be considered that they are so far apart that the asphaltenes are not soluble in n-heptane. This agrees with the definition of asphaltenes. It is also evident that the HSP of the

7248_book.fm Page 166 Tuesday, April 24, 2007 9:19 AM

166

Hansen Solubility Parameters: A User’s Handbook

14 12

δH H-bonding

10

Asfaltenes

8 6

Maltenes 2 0 15 20

10 18

n-heptane δP Polar

16 0 δD Dispersive

FIGURE 9.5 Solubility ellipsoids for asphaltenes and maltenes compared with n-heptane.

maltenes is different from the HSP of n-heptane. Thus, there is no reason to believe that the asphaltenes will appear in the same state in maltenes as in n-heptane. The fact that they are insoluble in n-heptane is no evidence that they are insoluble in the maltenes. In fact, there is such a large overlap between the solubility ellipsoids of the maltenes and the asphaltenes that they are quite likely to be soluble in each other. This strongly suggests that the asphaltenes are not dispersed in the maltenes as a colloidal dispersion but are more likely dissolved. It might be argued that some of the asphaltenes molecules with extreme HSP might not be soluble in the maltenes, and thus could still be dispersed rather than dissolved. This is, however, less likely as long as the continuum in the asphaltenes and the maltenes is kept intact. In some experiments the asphaltenes have been further fractionated into “soluble” asphaltenes and “insoluble” asphaltenes.28 If a fraction of the “insoluble” asphaltenes is mixed with the maltenes they might be insoluble. The reason is that the continuum has been broken and would probably not reflect the conditions in the original sample. In fact, removal of fractions from either the maltenes or the asphaltenes will create a risk for phase separation. This is also the reason why one should be very careful in making any claims or predictions of bitumen properties based on the properties of fractions.

BITUMEN AND POLYMERS It is a very common practice to improve bitumen properties by adding different additives. The reason is to improve the low temperature properties by making the bitumen softer at very low temperatures (<20°C) and at the same time make the bitumen more stiff at higher temperatures (+60°C). The temperatures are representative of the highest and lowest temperatures on the surface of an asphalt road during winter and summer, although in reality the maximum and minimum temperatures vary considerably with geographical location. The most common and well known modification is the addition of different types of polymers to bitumen. Probably all possible types

7248_book.fm Page 167 Tuesday, April 24, 2007 9:19 AM

Hansen Solubility Parameters of Asphalt, Bitumen, and Crude Oils

167

of polymers have been tested in bitumen — for example, plastomers, elastomers, two component curing systems, and even recycled rubber and plastics. The requirements on such products are, however, very strict, so in practice very few polymers have found a wider use as modifiers for bitumen. One of the most important requirements is the “compatibility” between the bitumen and the polymer. In this case, the meaning of “compatibility” is the stability against phase separation. Another important factor is the cost efficiency, which means that a good improvement of the bitumen properties is achieved with fairly small levels of polymeric additives. In the road building industry where the volumes are large and the price constraints are strong, the maximum level of modification is typically below 5%. In the roofing industry higher levels are generally accepted as product quality is more important than price. The Hansen solubility parameter concept provides a good tool for selection of suitable polymers, based on predictions of compatibility between different polymers and bitumen. If the HSP of a particular polymer is not known, it can easily be determined with a simple solubility test as described above. To illustrate the usefulness we may compare two types of polymers with known HSP with the HSP of bitumen. To make it simple we selected two polymers, polyethersulfone (PES) and polyethylensulfide, for which solubility data are presented in Chapter 5 and Chapter 18, respectively. Neither of these polymers is a common modifier for bitumen. The solubility ellipsoids of the two polymers compared to the HSP sphere of Venezuelan bitumen are illustrated in Figure 9.6. It is evident that PES is not soluble in bitumen, as the solubility ellipsoid is almost completely outside the ellipsoid of bitumen. In case PES is mixed with bitumen it will be dispersed rather than dissolved.

20

δH H-bonding

15 Polyether sulphone

10

5

Bitumen

0 Polyethylene sulphide

25 20

20

15 10

18 5

16 0

δP Polar

δD Dispersive

FIGURE 9.6 Solubility ellipsoid for bitumen compared with solubility ellipsoid for polyether sulfone and polyethylene sulfide.

7248_book.fm Page 168 Tuesday, April 24, 2007 9:19 AM

168

Hansen Solubility Parameters: A User’s Handbook

H-bonding

10 SBS 5 Bitumen

0 15

10 Polar 20

5 18 16 0

14

Dispersion

FIGURE 9.7 Solubility ellipsoid of bitumen compared to a solubility ellipsoid of SBS.

The effect will be an increase in stiffness at temperatures where PES will act as a solid filler. In case of polyethylenesulphide the solubility ellipsoid is inside the ellipsoid of bitumen, and thus polyethylenesulphide is expected to be completely soluble (compatible). As the polymer is completely soluble we do expect the effect to be related to the concentration of the modifier. No problem with storage stability is foreseen. None of the polymers discussed above have been frequently used for modification of bitumen. The polymer most commonly used for bitumen is styrene butadiene block copolymer (SBS) or similar polymers based on styrene and butadiene. This polymer gives a good modification effect at fairly low concentration (3–5%). The major advantages are increased stiffness at fairly high temperatures (60°C) and improved flexibility at low temperature. The higher stiffness will decrease the risk for rutting (permanent deformation). This risk is highest on hot, sunny summer days. The HSP of SBS was determined with a solubility test as above and the solubility ellipsoid was plotted together with bitumen in Figure 9.7. It is evident that there is a considerable overlap between the SBS and the bitumen. This implies partial solubility. In reality the situation is even more complicated as SBS consists of two different kinds of polymer segments based on butadiene and styrene, respectively. Each of these segments has different HSP. SBS belongs to the group of thermoplastic elastomers. These become plastic-like and can be processed at higher temperatures, at the same time having rubber-like properties at room temperature because of physical crosslinking caused by polystyrene and polybutadiene being mutually insoluble. The polystyrene blocks have a glass transition temperature of approximately 100°C, and therefore SBS is fairly workable at temperatures above 100°C but is still rubber-like at lower temperatures. It has been proposed that the same mechanism is also giving a good effect in bitumen, with the polystyrene being soluble/compatible in bitumen at the mixing temperature (180°C) but becoming less soluble or insoluble at lower temperatures. The effect is the same physical crosslinking as in pure SBS. Figure 9.7 supports this

7248_book.fm Page 169 Tuesday, April 24, 2007 9:19 AM

Hansen Solubility Parameters of Asphalt, Bitumen, and Crude Oils

169

δH Hydrogen bonding

10 8 6 4 2 0 15 10 δP Polar

5 0

12

14

16

18

20

22

24

δD Dispersive

FIGURE 9.8 Comparison between a heavy Venezuelan crude oil (Laguna) and a light crude oil (Leadon) from the North Sea.

picture as the part of the SBS ellipsoid located outside bitumen presumably represent the polystyrene, although this has not been verified by experiments.

CRUDE OIL Crude oil is found almost all over the world with large reserves in the Middle East, Russia, China, North America, Venezuela, and the North Sea, just to mention a few examples. It is produced by drilling wells in the ground or under the sea. Crude oil is pumped up to the surface where it is transported by pipeline or ships to refineries for further processing into desired products. The crude oils are very different, depending on origin. Some crude oils are very light and contain a large percentage of the most desired products, gasoline and diesel fuel, whereas other crude oils are heavy and bitumen-rich. The heaviest of the crude oils have such a high viscosity that they can not be pumped at normal ambient temperature but always have to be handled at higher temperature. Only a few selected crude oils can be used for production of high quality bitumen suitable for making asphalt for roads. Under certain conditions of storage and transport of crude oils there are sometimes problems with the formation of precipitates and/or deposits. These might decrease the capacity of pipelines by formation of solid contaminants in the crude oil. These deposits are sometimes blamed on asphaltenes and sometimes on waxes. The exact nature of these precipitates and the mechanism of their formation are not fully understood and is thus the subject for intense research. There are large economic benefits to be gained if the problem with deposits could be decreased. The use of HSP to study the precipitates in comparison with the solubility parameters of the crude oils is a good tool for better understanding of the precipitation mechanism. To have the complete picture it is also necessary to understand how temperature and pressure influence the HSP of different molecules in the crude oil. The difference between two crude oils, heavy Venezuelan Laguna and medium Leadon from the North Sea, may be illustrated by 3D plots of the solubility ellipsoids of each crude oil calculated with the hsp3D program (Figure 9.8). It is shown that Leadon is covering a larger space than Laguna and is thus expected to have better solubility properties. This is probably an effect of Leadon being a lighter crude which contains more low viscosity oils. These are better solvents than the

7248_book.fm Page 170 Tuesday, April 24, 2007 9:19 AM

170

Hansen Solubility Parameters: A User’s Handbook

higher molecular weight components of the Laguna. It is also seen that the solubility ellipsoid for the Laguna is located completely inside the solubility ellipsoid for the Leadon. This means that the Laguna crude should be completely soluble in the Leadon crude, and no problems with the formation of precipitates are to be expected by dilution of Laguna crude oil with Leadon crude oil.

TURBIDIMETRIC TITRATIONS Although the determination and visualization of solubility parameters for bitumen and other mineraloil-derived materials have proven to be very illustrative, there is still a desire for better precision. This has partly been met, as discussed above, by a better selection of test solvents and by better methods to optimize the ellipsoid. There are very obvious differences when it comes to practical applications, and particularly with modification with SBS polymers, among bitumens having very similar solubility ellipsoids (Table 9.2). Bitumens that are seemingly very similar with respect to solubility give still very different properties after mixing with SBS. One of the most important properties is the separation stability. Most mixtures of bitumen and SBS show a tendency for separation if they are stored at high temperature for a long time. A typical separation test is made at 180°C for 3 d. The separation is usually seen as the polymer floating to the surface, but sometimes, particularly at concentrations of SBS between 10 and 15%, a phase separation can take place, also in the bitumen. This is seen as a hard precipitate at the bottom of the bitumen tank. The separation tendency can be overcome by a proper selection of bitumen, alternatively by selection of a suitable polymer. The selection of components is mainly done on a “trial and error” basis, although there are some empirical rules. HSP may be an excellent tool for selection of suitable combinations of bitumen and polymer, but better precision is required than can be obtained with simple solubility testing with pure liquids. Improved estimation of the best solubility ellipsoid is required for optimal use of the HSP concept.

BISOM TEST The procedure discussed in the following has been developed at Nynas Bitumen based on turbidimetric titrations to precisely determine the boundary of the surface of solubility. The procedure is called BISOM, an acronym for BItumen Solubility Model. The principle of the test method is visualized in Figure 9.9. The HSP ellipsoid of the bitumen is constructed using the “poor solvents” illustrated as solid triangles in the figure and the “good solvents” being illustrated with open triangles. Three nonsolvents have been selected. These have HSP placing them just outside the surface of the solubility ellipsoid. They may be seen as the black triangles in the center of the circles in Figure 9.9. In the case of BISOM titration the selected “poor solvents” are 2,2,4-trimethyl pentane (isooctane), 2-butanone (methyl ethyl ketone), and 2-ethyl hexanol. As bitumen is a high viscosity liquid or a semisolid material, it has to be diluted to decrease the viscosity to permit proper stirring during the titrations. For this purpose, a solvent with a solubility parameter as close to the center of the ellipsoid as possible has to be selected. For the BISOM titration we have selected toluene (or in some cases xylene) as the good solvent. The titration is illustrated as arrows, going from the HSP of the good solvent toluene to any of the three poor solvents. The titration can be considered as a dilution of the bitumen with a mixture of a good solvent and a poor solvent. The HSP of the mixture is proportional to the concentration of each solvent. A precipitate will appear when the HSP of the mixture has a value placing it on the surface of the solubility ellipsoid. An important effect to take into consideration in the turbidimetric titration of bitumen is the concentration effect. This comes from the fact that bitumen is kept homogeneous by the mutual solubility of all its different molecules. The effect is seen as a higher concentration of bitumen giving better solubility. This situation is contradictory to what is usually known for solubility of pure substances. To understand this phenomenon we must consider that the first sign of turbidity

7248_book.fm Page 171 Tuesday, April 24, 2007 9:19 AM

Hansen Solubility Parameters of Asphalt, Bitumen, and Crude Oils

171

Laguna B180 soluble Laguna B180 insoluble

14

δH H-bonding

12

2-ethyl-1-hexanol

10 8 6 4

MEK

2 0 15

Toluene 20

10

Iso-octane

18

5

16 0

δP Polar

δD Dispersive

FIGURE 9.9 Summary of the BISOM titration with the “good solvent” and the three “poor solvents” titrants.

is interpreted as “insolubility,” but more precisely, it is the insolubility of the molecule(s) that are least soluble in the particular titrant/solvent mixture that has been used. The concentration effect was first described by Hithaus.29 He developed a kind of turbidimetric test for what he called “peptization of asphaltenes.” In this titration only one good solvent, toluene, and one poor solvent, n-heptane, were used. To account for the concentration effect, Heithaus performed the titration at several different concentrations of bitumen. The details of the calculations can be found in Reference 29, but Figure 9.10 gives an illustration of the principle. Heithaus Titration 1 FR = Vs/(VS + VT)

0.8 0.6 VT = volume titrant

0.4

VS = volume solvent 0.2

-0.1

0

-0.2

0

0.1

0.2

0.3

0.4

0.5

C(g/mL) = Wa/(VS + VT)

FIGURE 9.10 Illustration of the Heithaus titration of a Venezuelan bitumen.

7248_book.fm Page 172 Tuesday, April 24, 2007 9:19 AM

172

Hansen Solubility Parameters: A User’s Handbook

In Figure 9.10 the dilution ratio FR = volume solvent/total volume solvent and titrant is plotted against C = amount of bitumen/total amount of solvent and titrant. At the start of the titration FR = 1 as no titrant (VT = 0) has been added, whereas at the same time C is equal to the concentration of bitumen in the solvent. During the addition of titrant both FR and C become smaller and smaller. At infinite dilution FR = 0 and C = 0, but before this a precipitate has been formed, provided that the titrant has been properly selected. The point where the first sign of precipitate is noticed is marked with a black dot in the diagram. For each experiment the titration is repeated several times using different concentrations of bitumen. In Figure 9.10 the titration has been repeated four times, illustrated by the four titration arrows showing the decrease of the FR and the C value during the titration, and four black dots indicating the first sign of precipitation. A straight line is fitted through the four points using the least R-squared method. The equation for the line is used for extrapolation to find the intercepts for FR when C = 0 and for C when FR = 0. The meaning of the FR value at C = 0 is the ratio of solvent to titrant where there is solubility for all concentrations of bitumen. The solvent to titrant ratio can be used to calculate the HSP at the precipitation point for infinite dilution of the bitumen. The fact that a higher concentration of bitumen requires more titrant to reach the precipitation point confirms that the solubility of bitumen increases as the concentration increases. This effect is also visible with very dilute solutions. The meaning of the intercept for FR = 0 is the lowest concentration of bitumen that is needed to give full stability in pure titrant. It could also be expressed as the maximum titrant which can be added to bitumen without causing precipitation. In practice the precipitation point at the BISOM titration can be determined by different methods. There are at least two commercial instruments which can be used for BISOM titrations although they are both originally developed for automatic Hethaus titration. The testing procedures used by the instruments are described in two ASTM standards. One of the instruments measures the transmission of light through a cuvette with a short beam length30 and the other instrument uses variation in the intensity of a reflected beam of light (attenuated total reflectance principle [ATR]).31 Both instruments can be equipped with more than one titrant for BISOM titration. There is a modified version of the ATR instrument31 which is suitable for BISOM titrations. The computer program hsp3D also has the possibility to handle HSP of solvents or solvent mixtures considered as being exactly on the surface of the ellipsoid. This is the case with HSP calculated from the ratio of good and poor solvents at the precipitation point. Thus, the BISOM data may be combined with solubility data for better precision. The BISOM data may easily be recalculated from FRmax to the FR value for the same concentration used in the solubility testing by using the equation for the line in Figure 9.10. The results from a BISOM titration may be reported in different ways. The most simple is to give the FRmax for the intercept C = 0 and Cmin for the intercept FR = 0. Heithaus29 proposed further calculations of factors: pa = 1 – FRmax

(9.1)

⎡⎛ 1 ⎞ ⎤ p 0 = FR max ⎢⎜ ⎟ + 1⎥ ⎣⎝ C min ⎠ ⎦

(9.2)

P=

p0 1 − pa

(9.3)

where pa is considered to be related to the solubility of the molecules in bitumen with the lowest solubility, p0 is related to the solubility power of the bitumen, and finally, P is a balanced stability index describing the internal stability of the bitumen. A higher number indicates a higher stability.

7248_book.fm Page 173 Tuesday, April 24, 2007 9:19 AM

Hansen Solubility Parameters of Asphalt, Bitumen, and Crude Oils

173

A high internal stability can be seen as a high allowance for blending with additives, polymers, solvents, or other types of bitumen or crude oil. The interpretation of the BISOM titration is a determination of the internal stability of the bitumen or crude oil, rather than determination of HSP, although it is based on the principles of HSP. The BISOM could also be seen as an identification of those molecules that are the least soluble in the bitumen or crude oil, and how close to insolubility they are. To have a complete picture it is necessary to have several titrants with different HSP as the traditional determinations of internal stability using only n-heptane as precipitant will usually give an incomplete picture. It is not expected that the material precipitated by addition of n-heptane would be the same as that precipitated by the addition of 2-ethyl-1-hexanol or 2-butanone. The use of three titrants permits calculation of the HSP at three precipitation points, each of which could be considered to be on the surface of the solubility ellipsoid. It is, however, not possible to estimate the solubility ellipsoid from only three points on the surface, particularly since the exact center point is not known. If the HSP calculated from turbidimetric titrations are going to be compared to the HSP ellipsoid of the same material, the concentration at the solubility testing has to be taken into consideration. The precipitation point at a certain concentration can easily be estimated from data such as are reported in Figure 9.10.

CONCLUSION •

• •

•

• • • • • •

• •

The Hansen solubility parameters of a complex material such as bitumen or crude oil can be estimated by determination of its solubility in a large number of solvents with known HSP. To have the best precision, the test solvents should be selected with respect to the material to be tested. It is preferred to have solvents near or at the borderline of solubility. The factor 4 in Equation 1.9 as a multiplier for the difference in the dispersion parameters for the materials being considered seems to be valid, also for complex mixtures like bitumen. Comparison of HSP for bitumen, asphaltenes, and maltenes confirms that asphaltenes are soluble in maltenes, and bitumen should thus not be considered as a colloidal dispersion as is frequently claimed. The best estimated HSP values for Venezuelan bitumen are D = 18.6 MPa0.5, P = 3.0 MPa0,5, H = 3.4 MPa0.5. A computer program hsp3D can estimate the best fit to the solubility data and from this calculate the HSP. The program can also estimate the best coefficients for the ellipsoid model to illustrate the extension of solubility regions in a 3D diagram. Up to three different materials can be compared in the same 3D plot. The program is not limited to bitumen and crude oils, but could equally well be used for other types of materials. A procedure for turbidimetric titrations has been developed to further improve the precision to determine the surface of the HSP solubility ellipsoid. This procedure is called a BISOM titration. BISOM titration is well suited for measurement of the internal stability of complex hydrocarbon mixtures like bitumen or crude oils. BISOM titration is also a determination of the least soluble molecules in bitumen or crude oils.

7248_book.fm Page 174 Tuesday, April 24, 2007 9:19 AM

174

Hansen Solubility Parameters: A User’s Handbook

REFERENCES 1. Speight, J.G., The Chemistry and Technology of Petroleum, 2nd ed., Marcel Dekker, New York, 1991. 2. Nellensteyn, F.J., Relation of the micelle to the medium in asphalt, Inst. Pet. Technol., 14, 134–138, 1928. 3. Pfeiffer, J.P. and Saal, R.N.J., Asphaltic bitumen as colloidal system, Phys. Chem., 44 139–149, 1940. 4. Park, S.J. and Mansoori, G.A., Aggregation and deposition of heavy organics in petroleum crudes, Energy Sources, 10, 109–125, 1988. 5. Boduszynski, M.M., McKay, J.F., and Lathham, D.R., Asphaltenes where are you? Proc. Assoc of Asphalt Paving Technologists, 49, 1980, pp. 124–143. 6. Petersen, J.C., Robertson, R.E., Branthaver, J.F., Harnsberger, P.M., Duvall, J.J., Kim, S.S., Anderson, D.A., Christiansen, D.W., and Bahia, H.U., Binder Characterization and Evaluation, Vol. 1: Physical Characterization, Strategic Highway Research Program SHRP-A-367, 1994. 7. Redelius, P.G., Bitumen solubility model using Hansen solubility parameter, Energy Fuels, 18, 1087–1092, 2004. 8. Sirota, E.B., Understanding the Physical Structure of Asphaltenes to Optimize Bitumen Manufacture, 3rd Euroasphalt and Eurobitume Congress Vienna, Paper 097, 930–939, 2004. 9. Yen, T.F. and Chilingarian, G.V., Asphaltenes and Asphalts, 1 in Developments in Petroleum Science 40 A, Elsevier Science BV, Amsterdam 1994. 10. Sheu, E.Y. and Mullins, O.C., Asphaltenes, Fundamentals and Applications, Plenum Press, New York, 1995. 11. Test Method for n-heptane Insolubles. Annual book of ASTM Standards Section 04.03, ASTM D327997, 2001. 12. Test Method for Determination of Asphaltenes (Heptane insolubles) in Crude Petroleum and Petroleum Products, Annual book of ASTM Standards Section 05.04, ASTM D6560-00. 13. Buch, L., Groenzin, H., Buenrostro-Gonzales, E., Andersen, S.I., Lira-Galena, C., Mullins, O.C., Molecular size of asphaltene fractions obtained from residuum hydrotreatment, Fuel, 82, 1075–1084, 2003. 14. Boduszynski, M.M., Asphaltenes in petroleum asphalts: composition and formation, Am. Chem. Soc. Meet., Div. Pet. Chem., Washington, D.C., September 9–14, 1979. 15. Mitchell, D.L. and Speight, J.G., The solubility of asphaltenes in hydrocarbon solvents, Fuel, 52, 149–152, 1973. 16. Hirschberg, A., deJong, L.N.J., Schipper, B.A., Meijer, J.G., Influence of temperature and pressure on asphaltene flocculation, Soc. Pet. Eng. J., 283–293, June 1984. 17. Hildebrand, J.H. and Scott, R.L., The Solubility of Nonelectrolytes, 3rd ed., Dover, New York, 1964. 18. Laux, H. and Rahimian, I., Colloid-disperse Crude Oil Systems; Phase Behaviour and Stability, Erdöl und Kohle — Erdgas — Petrochemie vereinigt mit Brennstoff-Chemie, 47(11), 430–435, 1994. 19. Hirschberg, A. and Hermans, L., Asphaltene phase behaviour: a molecular thermodynamic model, Proc. Characterisation of Heavy Crude Oils and Petroleum Residues, Lyon, June 25–27, 1984. 20. Burke, N.E., Hobbs, R.E., Kashou, S.F., Measurement and modelling of asphaltene precipitating, J. Pet. Technol., November, 1440–1446, 1990. 21. Rahimian, I. and Zenke, G., Zum Verhalten organicher Lösemittel gegenüber Bitumen, Bitumen 1, 2–8, 1986 (in German). 22. Neumann, H.J., Rahimian, I., Zenke, G., Einfluss der Löslichkeitseigenschaften von Asphaltene auf die Rückstandsverarbeitung. Erdöl und Kohle –Erdgas, 42(7/8), 278–286, 1989 (in German). 23. Hagen, A.H., Jones, R., Hofner, R.M., Randolf, B.B., Johnson, M.P., Characterisation of asphalt by solubility profiles, J. Assoc. Asphalt Paving Technol., 53, 119–137, 1984. 24. Redelius, P.G., Solubility parameters and bitumen, Fuel, 79, 27–35, 2000. 25. Hansen, C.M., Skaarup, K., The three dimensional parameter — key to paint component affinities. III. Independent calculation of the parameter components, J. Paint Technol., 39(511), 511–514, 1967. 26. Warnez, M. and Redelius, P.G., unpublished results from Icopal – Nynas joint research project: Correlation between Bitumen Characteristics and Fire Properties of Roofing Membranes, 2001–2003. 27. Turner, F. and Redelius, P., The program hsp3D is available as share ware in Europe from: Nynas Petroleum, S-149 82 Nynäshamn, Sweden, www.nynas.com or in USA from Western Research Institute, 265 North 9th Street, Laramie, Wyoming 82070–3380, www.westernresearch.org.

7248_book.fm Page 175 Tuesday, April 24, 2007 9:19 AM

Hansen Solubility Parameters of Asphalt, Bitumen, and Crude Oils

175

28. Östlund, J.-A., Wattana, P., Nydén, M., and Fogler, H.S., Characterisation of fractionated asphaltenes by UV-VIS and NMR self-diffusion spectroscopy, J. Colloid Interface Sci., 271(2), 372–380, 2004. 29. Heithaus J.J., Measurement and significance of asphatene pepetization, J. Inst. Pet., 48(458), 45–53, 1962. 30. Standard Test Method for Automated Heithaus Titrimetry, Annual book of ASTM Standards Section 04.03, ASTM D6703-01. 31. Standard Test Method for Determining Stability and Compatibility of Heavy Fuel Oils and Crude Oils by Heavy Fuel Oil Stability Analyzer (Optical Detection), Annual book of ASTM Standards Section 05.04, ASTM D7112-05a

7248_book.fm Page 176 Tuesday, April 24, 2007 9:19 AM

7248_book.fm Page 177 Tuesday, April 24, 2007 9:19 AM

of Hansen 10 Determination Solubility Parameter Values for Carbon Dioxide Laurie L. Williams ABSTRACT Reference values of the Hansen solubility parameters (HSP) for carbon dioxide (CO2) have been determined. The methodology adopted for this is based on room temperature solubility of the gas in different liquids of known HSP values. CO2 gas solubility data, at 25°C and a CO2 partial pressure of 1 atmosphere, in 101 liquid solvents have been gathered and evaluated, yielding the values: δd = 15.6 MPa1/2, δp = 5.2 MPa1/2, and δh = 5.8 MPa1/2. A methodology for extending this reference set of HSP values to any temperature and pressure has been developed utilizing an equation of state (EOS) of the form P = f(ρ, T).

INTRODUCTION The HSP concept is widely used for selecting suitable solvents for organic compounds. In addition, HSPs have been applied to biological materials, barrier properties of polymers, as well as the characterization of surfaces, pigments, fillers, and fibers.1 The basis of the HSP approach is the assumption that the total cohesive energy (E) of a pure compound is made up of the additive contributions from nonpolar (dispersion) interactions (Ed), polar (dipole–dipole and dipole–induced dipole) interactions (Ep), and hydrogen bonding or other specific association interactions including Lewis acid–base interactions (Eh): E = Ed + EP + Eh

(10.1)

Dividing each contribution by the molar volume, E Ed E p Eh = + + V V V V

(10.2)

gives the square of the total solubility parameter as the sum of the squares of the Hansen dispersion (δd), polar (δp), and hydrogen bonding (δh) solubility parameters, so that δ t2 = δ 2d + δ 2p + δ 2h

(10.3)

where

177

7248_book.fm Page 178 Tuesday, April 24, 2007 9:19 AM

178

Hansen Solubility Parameters: A User’s Handbook

δ 2d =

Ed ; V

δ 2p =

Ep V

;

and δ 2h =

Eh V

(10.4)

The determination of HSPs for compounds that are gases at ambient conditions is usually based on room temperature solubility of the gas in a range of different liquids of known δd, δp, and δh. Those liquids that show the highest solubility for the gas are assumed to have HSPs closer to those of the gas than those liquids that have lower solubilities for the gas. In the following section, published data of CO2 gas solubility at 25°C and a CO2 partial pressure of 1 atmosphere in a large number of liquid solvents are evaluated. From this data, a set of HSP values at a single reference temperature and pressure is determined.

METHODOLOGY Literature values of CO2 solubility in various liquid solvents at 25°C have been collected and are summarized in Table 10.1. All the solubility data shown in Table 10.1 was either experimentally determined at a CO2 partial pressure of 0.1 MPa, or has been corrected to pCO2 MPa using Henry’s law,2 xCO2 =

pCO2 KH

(10.5)

where KH is the Henry’s law coefficient, pCO2 is the partial pressure of CO2, and xCO2 is the mole fraction of dissolved CO2. In addition to correcting CO2 partial pressure to 0.1 MPa, it has been necessary to correct the reported values of CO2 gas solubility to a common set of units. Generally, solubilities have been reported as mole fraction of dissolved CO2, xCO2, or as one of the two dimensionless quantities; the Bunsen coefficient or the Ostwald coefficient. The Bunsen coefficient, Ω, is defined as the volume of gas, reduced to 0°C and 0.1 MPa, dissolved per unit volume of solvent at a system temperature, T, under a gas pressure of 1 atmosphere. The Ostwald coefficient, L, is defined as the ratio of the volume of gas absorbed to the volume of the absorbing liquid, all measured at the same temperature.3 If the solubility is small and the gas phase is ideal, the Ostwald coefficient is independent of total pressure and these two coefficients are simply related by4: L=

T Ω 273

(10.6)

The mole fraction of dissolved CO2 can then be calculated using2

xCO2

⎡⎛ RT ⎞ ⎤ = ⎢⎜ ⎟ + 1⎥ ⎢⎣⎝ LpCO2 V10 ⎠ ⎥⎦

−1

(10.7)

where R is the gas constant, T is the absolute temperature, and V10 the molar volume of the pure solvent. The Hansen dispersion, polar, and hydrogen bonding parameters in Table 10.1 are from Hansen’s solublity parameter handbook,5 and the total solubility parameter value is calculated from Equation 10.3, noted earlier. Using the collected values of xCO2 at 25°C and pCO2 = .1 MPa, HSP values were calculated based on a simple weighted average using the data set in Table 10.1, hereafter called data set #1. Where multiple CO2 gas solubility data were available for an individual solvent, an average value for that solvent was used.

7248_book.fm Page 179 Tuesday, April 24, 2007 9:19 AM

Determination of Hansen Solubility Parameter Values for Carbon Dioxide

179

TABLE 10.1 CO2 Solubility in Various Solvents at T = 25ºC and a CO2 Partial Pressure of 1 Atmosphere

Formula C 2 H 4O 2 C 2 H 4O 2 C 4 H 6O 3 C 3 H 6O C 3 H 6O C 3 H 6O C 3 H 6O C 3 H 6O C 3 H 6O C7H14O2 C7H14O2 C7H14O2 C5H11Br C5H11Cl C 6 H 7N C 6 H 7N C 6 H 7N C 6 H 7N C 7 H 6O C 7 H 6O C 6H 6 C 6H 6 C7H7Cl C6H5Br C4H10O C4H10O C4H10O C4H10O C4H10O C22H42O2 C4H8O2 CS2 CS2 CCl4 CCl4 CCl4 C6H5Cl C6H5Cl CHCl3 CHCl3 CHCl3 C9H12 C7H14 C7H12O C6H12 C6H12 C6H12 C6H12O

Solvent Acetic acid Acetic acid Acetic anhydride Acetone Acetone Acetone Acetone Acetone Acetone Amyl acetate Amyl acetate Amyl acetate Amyl bromide Amyl chloride Aniline Aniline Aniline Aniline Benzaldehyde Benzaldehyde Benzene Benzene Benzyl chloride Bromobenzene Butanol Butanol Butanol 2-Butanol t-Butanol Butyl oleate Butyric acid Carbon disulfide Carbon disulfide Carbon tetrachloride Carbon tetrachloride Carbon tetrachloride Chlorobenzene Chlorobenzene Chloroform Chloroform Chloroform Methyl ethyl benzene Cycloheptane Cycloheptanone Cyclohexane Cyclohexane Cyclohexane Cyclohexanol

Mole Fraction CO2

Actual/ Ideal

δd (MPa)1/2

0.01089 0.01122 0.01977 0.01876 0.01903 0.01896 0.01896 0.02259 0.02092 0.02800 0.02452 0.02612 0.01235 0.01424 0.00493 0.00486 0.00487 0.00487 0.01140 0.01171 0.00962 0.00880 0.00912 0.00789 0.00887 0.00734 0.00718 0.00660 0.00725 0.02790 0.01297 0.00215 0.00328 0.01100 0.01059 0.00904 0.00938 0.00981 0.01121 0.01277 0.01126 0.01008 0.00721 0.01588 0.00771 0.00760 0.00774 0.00471

0.4757 0.4899 0.8635 0.8193 0.8309 0.8279 0.8279 0.9865 0.9136 1.2227 1.0708 1.1407 0.5394 0.6220 0.2151 0.2121 0.2129 0.2129 0.4979 0.5114 0.4201 0.3843 0.3983 0.3444 0.3872 0.3207 0.3135 0.2882 0.3166 1.2183 0.5665 0.0940 0.1432 0.4803 0.4624 0.3948 0.4095 0.4283 0.4897 0.5576 0.4918 0.4401 0.3148 0.6934 0.3367 0.3319 0.3380 0.2057

14.5 14.5 16.0 15.5 15.5 15.5 15.5 15.5 15.5 15.8 15.8 15.8 20.3 15.5 19.4 19.4 19.4 19.4 19.4 19.4 18.4 18.4 18.8 20.5 16.0 16.0 16.0 15.8 15.2 14.7 14.9 20.5 20.5 17.8 17.8 17.8 19.0 19.0 17.8 17.8 17.8 16.1 17.2 17.2 16.8 16.8 16.8 17.4

δp (MPa)1/2 8.0 8.0 11.7 10.4 10.4 10.4 10.4 10.4 10.4 3.3 3.3 3.3 4.8 5.0 5.1 5.1 5.1 5.1 7.4 7.4 0.0 0.0 7.1 5.5 5.7 5.7 5.7 5.7 5.1 3.4 4.1 0.0 0.0 0.0 0.0 0.0 4.3 4.3 3.1 3.1 3.1 7.0 0.0 10.6 0.0 0.0 0.0 4.1

δh (MPa)1/2 13.5 13.5 10.2 7.0 7.0 7.0 7.0 7.0 7.0 6.1 6.1 6.1 2.8 1.3 10.2 10.2 10.2 10.2 5.3 5.3 2.0 2.0 2.6 4.1 15.8 15.8 15.8 14.5 14.7 3.4 10.6 0.6 0.6 0.6 0.6 0.6 2.0 2.0 5.7 5.7 5.7 0.0 0.0 4.8 0.2 0.2 0.2 13.5

δt (MPa)1/2

Ref.

21.37 21.37 22.29 19.94 19.94 19.94 19.94 19.94 19.94 17.26 17.26 17.26 21.05 16.34 22.50 22.50 22.50 22.50 21.43 21.43 18.51 18.51 20.26 21.62 23.20 23.20 23.20 22.19 21.75 15.47 18.74 20.51 20.51 17.81 17.81 17.81 19.58 19.58 18.95 18.95 18.95 17.56 17.20 20.77 16.80 16.80 16.80 22.40

29 30 29 29 30 31 32 33 34 35 29 30 29 29 29 30 33 3 30 29 36 29 29 29 32 37 38 39 40 38 29 29 41 35 36 29 29 34 29 34 30 29 42 43 44 2 42 45

7248_book.fm Page 180 Tuesday, April 24, 2007 9:19 AM

180

Hansen Solubility Parameters: A User’s Handbook

TABLE 10.1 (CONTINUED) CO2 Solubility in Various Solvents at T = 25ºC and a CO2 Partial Pressure of 1 Atmosphere

Formula

Solvent

Mole Fraction CO2

C6H10O C8H16 C8H16 C5H10 C5H8O C10H22 C10H22 C10H22O C2H4Br2 C2H4Br2 C2H4Br2 C2H4Br2 C3H6Cl2O CH2Cl2 C8H14O C2H6OS C2H6OS C2H6OS C3H7NO C4H8O2 C12H26 C12H26 C12H26 C12H26O C2H6O C2H6O C2H6O C4H8O2 C8H10 C2H4Cl2 C2H6O2 C10H12O2 C5H9NO2 C3H8O3 C7H16 C7H16 C7H16 C7H16 C7H16O C16H34 C16H34 C6H14 C6H14 C6H14O C5H5I C5H12O C4H10O C4H10O

Cyclohexanone Cyclooctane Cyclooctane Cyclopentane Cyclopentanone Decane Decane Decanol 1,2-Dibromoethane 1,2-Dibromoethane 1,2-Dibromoethane 1,2-Dibromoethane 1,3-dichloro-2-propanol Dichloromethane Dimethyl cyclohexanone Dimethyl sulfoxide Dimethyl sulfoxide Dimethyl sulfoxide Dimethyl formamide 1,4-Dioxane Dodecane Dodecane Dodecane Dodecanol Ethanol Ethanol Ethanol Ethyl acetate Ethyl benzene Ethylene chloride Ethylene glycol Eugenol N-formyl morpholine Glycerol (glycerin) Heptane Heptane Heptane Heptane Heptanol Hexadecane Hexadecane Hexane Hexane Hexanol Iodobenzene Isoamyl alcohol Isobutanol Isobutanol

0.01600 0.00688 0.00686 0.00491 0.01641 0.01204 0.01250 0.00973 0.00812 0.00804 0.00797 0.00838 0.00746 0.01250 0.01680 0.00908 0.00907 0.00945 0.01610 0.02272 0.01428 0.01290 0.01191 0.01811 0.00642 0.00642 0.00725 0.02300 0.01006 0.01133 0.00220 0.01023 0.01475 0.00009 0.01328 0.01190 0.01194 0.01202 0.01258 0.01161 0.01420 0.01318 0.01190 0.01174 0.00592 0.00809 0.00696 0.00690

Actual/ Ideal

δd (MPa)1/2

δp (MPa)1/2

0.6987 0.3004 0.2996 0.2146 0.7166 0.5256 0.5459 0.4249 0.3546 0.3509 0.3481 0.3659 0.3258 0.5459 0.7336 0.3965 0.3962 0.4125 0.7031 0.9921 0.6235 0.5633 0.5202 0.7908 0.2804 0.2805 0.3166 1.0044 0.4393 0.4946 0.0961 0.4469 0.6441 0.0039 0.5801 0.5197 0.5212 0.5248 0.5495 0.5069 0.6201 0.5754 0.5197 0.5124 0.2585 0.3530 0.3041 0.3014

17.8 17.5 17.5 16.4 17.9 15.7 15.7 17.5 17.8 17.8 17.8 17.8 17.5 18.2 15.2 18.4 18.4 18.4 17.4 19.0 16.0 16.0 16.0 15.5 15.8 15.8 15.8 15.8 17.8 19.0 17.0 15.1 16.6 17.4 15.3 15.3 15.3 15.3 15.1 16.3 16.3 14.9 14.9 14.1 19.5 15.8 15.1 15.1

6.3 0.0 0.0 0.0 11.9 0.0 0.0 2.6 6.4 6.40 6.4 6.4 9.9 6.3 8.8 16.4 16.4 16.4 13.7 1.8 0.0 0.0 0.0 6.5 8.8 8.8 8.8 5.3 0.6 7.4 11.0 8.8 11.7 12.1 0.0 0.0 0.0 0.0 8.0 0.0 0.0 0.0 0.0 8.6 6.0 5.2 5.7 5.7

δh (MPa)1/2 5.1 0.0 0.0 1.8 5.2 0.0 0.0 10.0 7.0 7.0 7.0 7.0 14.6 6.1 3.3 10.2 10.2 10.2 11.3 7.4 0.0 0.0 0.0 10.8 19.4 19.4 19.4 7.2 1.4 4.1 26.2 9.8 10.0 29.3 0.0 0.0 0.0 0.0 13.0 0.0 0.0 0.0 0.0 12.7 6.1 13.3 15.9 15.9

δt (MPa)1/2

Ref.

19.56 17.50 17.50 16.50 22.11 15.70 15.70 20.32 20.17 20.17 20.17 20.17 24.85 20.20 17.87 26.68 26.68 26.68 24.86 20.47 16.00 16.00 16.00 19.98 26.52 26.52 26.52 18.15 17.87 20.80 33.11 20.04 22.64 36.16 15.30 15.30 15.30 15.30 21.47 16.30 16.30 14.90 14.90 20.83 21.29 21.30 22.66 22.66

46 42 47 42 48 32 49 50 34 29 30 45 29 38 51 44 52 33 32 38 53 49 32 32 30 31 38 38 38 29 38 29 54 29 53 49 36 52 32 32 49 32 49 32 29 30 2 29

7248_book.fm Page 181 Tuesday, April 24, 2007 9:19 AM

Determination of Hansen Solubility Parameter Values for Carbon Dioxide

181

TABLE 10.1 (CONTINUED) CO2 Solubility in Various Solvents at T = 25ºC and a CO2 Partial Pressure of 1 Atmosphere

Formula

Solvent

Mole Fraction CO2

C6H12O2 C4H9Cl C3H8O CH4O CH4O CH4O CH4O C3H6O2 C3H6O2 C7H14 C7H12O C4H8O C11H10 C19H36O2 C6H5NO2 C6H5NO2 C5H9NO C9H20 C9H20O C8H18 C8H18 C8H18O C18H34O2 C15H32 C5H12 C5H12O C7F16 C8H7N C3H8O C3H8O C3H8O C3H6O2 C3H5N C5H10O2 C3H6Br2 C4H6O3 C4H6O3 C4H6O3 C5H5N C5H5N C5H5N C9H7N C4H8O2S C14H30 C14H30 C4H8O C10H12 C7H9N

Isobutyl acetate Isobutyl chloride Isopropanol Methanol Methanol Methanol Methanol Methyl acetate Methyl acetate Methyl cyclo hexane 2-Methylcyclohexanone Methyl ethyl ketone 1-Methyl naphthalene Methyl oleate Nitrobenzene Nitrobenzene N-methyl-2-pyrrolidone Nonane Nonanol Octane Octane Octanol Oleic acid Pentadecane Pentane Pentanol Perfluroheptane Phenyl acetonitrile Propanol Propanol Propanol Propionic acid Propionitrile Propyl acetate Propylene bromide Propylene carbonate Propylene carbonate Propylene carbonate Pyridine Pyridine Pyridine Quinoline Sulfolane Tetradecane Tetradecane Tetrahydrofuran Tetrahydronaphthalene m-Toluidine

0.02500 0.01410 0.00654 0.00641 0.00564 0.00563 0.00635 0.02089 0.02253 0.00927 0.01660 0.02444 0.00674 0.02690 0.01020 0.00997 0.01590 0.01231 0.01481 0.01254 0.01210 0.00930 0.01570 0.01167 0.01385 0.00806 0.02088 0.00957 0.00782 0.00680 0.00759 0.01234 0.01677 0.02429 0.00977 0.01074 0.01162 0.01210 0.01193 0.01169 0.01198 0.00912 0.00799 0.01171 0.01360 0.02700 0.00752 0.00635

Actual/ Ideal

δd (MPa)1/2

1.0919 0.6158 0.2856 0.2797 0.2463 0.2460 0.2774 0.9123 0.9837 0.4046 0.7249 1.0672 0.2943 1.1747 0.4455 0.4355 0.6943 0.5376 0.6465 0.5474 0.5284 0.4062 0.6856 0.5094 0.6048 0.3519 0.9118 0.4180 0.3414 0.2969 0.3316 0.5390 0.7323 1.0607 0.4266 0.4688 0.5073 0.5284 0.5211 0.5104 0.5231 0.3983 0.3489 0.5115 0.5939 1.1790 0.3285 0.2771

15.1 14.7 15.8 15.1 15.1 15.1 15.1 15.5 15.5 16.0 17.6 16.0 20.6 14.5 20.0 20.0 18.0 15.7 15.3 15.5 15.5 17.0 16.2 16.8 14.5 15.9 12.0 19.5 16.0 16.0 16.0 14.7 15.3 15.3 17.4 20.0 20.0 20.0 19.0 19.0 19.0 19.4 18.4 16.2 16.2 16.8 19.6 19.3

δp (MPa)1/2 3.7 5.3 6.1 12.3 12.3 12.3 12.3 7.2 7.2 0.0 6.3 9.0 0.8 3.9 8.6 8.6 12.3 0.0 7.3 0.0 0.0 3.3 3.1 0.0 0.0 4.5 0.0 12.3 6.8 6.8 6.8 5.3 14.3 4.3 7.5 18.0 18.0 18.0 8.8 8.8 8.8 7.0 16.6 0.0 0.0 5.7 2.0 3.8

δh (MPa)1/2 6.3 0.9 16.4 22.3 22.3 22.3 22.3 7.6 7.6 1.0 4.7 5.1 4.7 3.7 4.1 4.1 7.2 0.0 12.0 0.0 0.0 11.9 5.5 0.0 0.0 13.9 0.0 3.8 17.4 17.4 17.4 12.4 5.5 7.6 2.9 4.1 4.1 4.1 5.9 5.9 5.9 7.6 7.4 0.0 0.0 8.0 2.9 10.1

δt (MPa)1/2

Ref.

16.77 15.65 23.58 29.61 29.61 29.61 29.61 18.70 18.70 16.03 19.28 19.05 21.14 15.46 22.15 22.15 22.96 15.70 20.77 15.50 15.50 21.01 17.39 16.80 14.50 21.59 12.00 23.37 24.60 24.60 24.60 19.95 21.65 17.62 19.17 27.22 27.22 27.22 21.75 21.75 21.75 21.98 25.86 16.20 16.20 19.46 19.91 22.11

29 29 38 32 30 31 29 29 34 55 56 33 38 38 29 34 38 32 32 32 49 50 38 32 32 29 41 34 57 58 29 29 34 34 29 32 33 38 29 30 34 38 32 32 49 38 38 29

7248_book.fm Page 182 Tuesday, April 24, 2007 9:19 AM

182

Hansen Solubility Parameters: A User’s Handbook

TABLE 10.1 (CONTINUED) CO2 Solubility in Various Solvents at T = 25ºC and a CO2 Partial Pressure of 1 Atmosphere

Formula

Solvent

Mole Fraction CO2

C7H9N C 7H 8 C 7H 8 C 7H 8 C 7H 8 C12H27O4P C2Cl3F3 C13H28 C6H15N C8H18 C11H24 C11H24O H2O H2O H2O H2O C8H10 C8H10 C8H10 C8H10

o-Toluidine Toluene Toluene Toluene Toluene Tributyl phosphate Trichlorotrifluoroethane Tridecane Triethylamine 2,2,4-Trimethylpentane Undecane Undecanol Water Water Water Water o-Xylene m-Xylene m-Xylene p-Xylene

0.00605 0.00994 0.01010 0.01042 0.01039 0.03550 0.01823 0.01175 0.02321 0.01387 0.01148 0.01708 0.00070 0.00059 0.00060 0.00061 0.00994 0.01042 0.01063 0.01087

Note:

ideal xCO 2

Actual/ Ideal

δd (MPa)1/2

0.2641 0.4340 0.4411 0.4550 0.4536 1.5502 0.7961 0.5132 1.0134 0.6057 0.5014 0.7457 0.0306 0.0258 0.0261 0.0267 0.4339 0.4552 0.4642 0.4744

19.4 18.0 18.0 18.0 18.0 16.3 14.7 16.4 17.8 14.1 16.0 15.4 15.5 15.5 15.5 15.5 17.8 17.4 17.4 17.4

δp (MPa)1/2 4.2 1.4 1.4 1.4 1.4 6.3 1.6 0.0 0.4 0.0 0.0 6.7 16.0 16.0 16.0 16.0 1.0 1.0 1.0 1.0

δh (MPa)1/2 10.7 2.0 2.0 2.0 2.0 4.3 0.0 0.0 1.0 0.0 0.0 11.2 42.3 42.3 42.3 42.3 3.1 3.1 3.1 3.1

δt (MPa)1/2

Ref.

22.55 18.16 18.16 18.16 18.16 18.00 14.79 16.40 17.83 14.10 16.00 20.19 47.81 47.81 47.81 47.81 18.10 17.70 17.70 17.70

29 29 55 52 34 38 59 32 33 59 32 32 35 29 30 31 60 29 60 60

= 0.0229.

n

∑xδ

i di

δ

CO 2 d

=

i =1 n

∑x

’

(10.8)

’

(10.9)

.

(10.10)

i

i =1

n

∑xδ i

δ

CO 2 p

=

pi

i =1 n

∑x

i

i =1

n

∑xδ

i hi

δ

CO 2 h

=

i =1 n

∑x i =1

i

7248_book.fm Page 183 Tuesday, April 24, 2007 9:19 AM

Determination of Hansen Solubility Parameter Values for Carbon Dioxide

183

TABLE 10.2 Solvents Showing Greater Than Ideal Solubility for CO2 at 25°C Solvent Tributyl phosphate (C12H27O4P) Amyl acetate (C7H14O2) Butyl oleate (C22H42O2) Tetrahydrofuran (C4H8O) Methyl oleate (C19H36O2) Isobutyl acetate (C6H12O2) Methyl ethyl ketone (C4H8O) Propyl acetate (C5H10O2) Ethyl acetate (C4H8O2) Methyl acetate (C3H6O2) Note: PCO2 = 1,

ideal xCO 2

Exptl xCO 2

δd (MPa)1/2

δp (MPa)1/2

δh (MPa)1/2

0.03550

16.3

6.3

4.3

0.02800

15.8

3.3

6.1

0.02790

14.7

3.4

3.4

0.02700

16.8

5.7

8.0

0.02690

14.5

3.9

3.7

0.02500

15.1

3.7

6.3

0.02444

16.0

9.0

5.1

0.02429

15.3

4.3

7.6

0.02300

15.8

5.3

7.2

0.02253

15.5

7.2

7.6

= 0.0229.

A second HSP evaluation was undertaken using a subset of data set # 1. The subset was chosen to consist of solvents where the measured CO2 solubility was greater than the ideal solubility at ideal 25°C and pCO2 = .1 MPa, xCO 2 = 0.0229. (The calculation of this ideal solubility value is given in the Appendix 10.A.1). This data subset, hereafter called data set #2, is comprised of the 10 solvents shown in Table 10.2. These two evaluations resulted in the following HSP values for CO2 at 25°C: Data set #1: δd = 16.4 MPa1/2 δp = 5.5 MPa1/2 δh = 5.8 MPa1/2 Data set #2: δd = 15.6 MPa1/2 δp = 5.2 MPa1/2 δh = 5.8 MPa1/2

7248_book.fm Page 184 Tuesday, April 24, 2007 9:19 AM

184

Hansen Solubility Parameters: A User’s Handbook

20 15 10

δh (M Pa)½

25

5

5

10

δd (M Pa)½

15

20

25 0

5

10

15

20

0

δp (M Pa)½

FIGURE 10.1 Interaction radius, where Ro incorporates all good solvents and excludes all bad solvents.

The HSP values derived from the two data sets gave identical results in terms of the hydrogen bonding parameter, δh, and similar values for δp and δd. To assist in the determination of a final set of HSP values, a second approach, known as the solubility sphere,5–7 was also used to evaluate the published solubility data and resulting HSP values for data set #1 and data set #2. The solubility sphere approach is essentially a trial and error method, whereby all the good solvents are included within a sphere in δd, δp, and δh space, whereas simultaneously excluding all the bad solvents. The criterion of good versus bad is arbitrary and is defined based on the particular interaction being evaluated, such as degree of polymer swelling, dissolution, barrier breakthrough time, permeation coefficients higher than a given value, suspension time of a pigment, etc. In this evaluation, we are concerned with optimizing solvents for their CO2 solubility. Based upon the selected criteria, two-dimensional plots are produced for δd vs. δp, δd vs. δh, and δp vs. δh, and the three circle radii are adjusted until a single optimized radius for all three plots is found. This solubility sphere approach is essentially a trial and error method whereby all the good solvents are included within a sphere in δd, δp, and δh space while simultaneously excluding the bad ones. The resulting radius for the three plots of δd vs. δp, δd vs. δh, and δp vs. δh, is then used to plot a sphere in a three-dimensional plot of δd vs. δp vs. δh. The radius of this sphere is known as the interaction radius, Ro, and is considered a fourth parameter in HSP value determinations. Figure 10.1 is a schematic representation of the solubility sphere approach. The advantage of the solubility sphere approach is that once an interaction radius has been determined, solvents that have not yet been experimentally tested for the desired interaction can be quickly screened and, therefore, should be considered for further study. This solubility sphere evaluation is aided by an equation developed by Skaarup for determining the straight-line distance, Ra, in a plot of δd vs. δp vs. δh between two materials based on their respective HSP values,5

( Ra )

2

(

= 4 δd 2 − δd1

) + (δ 2

p2

− δ p1

) + (δ 2

h2

− δ h1

)

2

(10.11)

where δd2, δp2, and δh2 are associated with a given solvent and δd1, δp1, and δh1 with the center of the optimized solubility sphere. This equation was developed from plots of experimental data where the leading constant of 4 in the leading right-hand term was found to correctly represent the solubility data as a sphere encompassing the good solvents. An extended discussion of the validity of this coefficient is found in Chapter 2. Further confirmation is found in Chapter 9. For the present evaluation, a favorable interaction is defined as CO2 solubility greater than ideal at 25°C and pCO2 = 0.1 MPa, whereas an unfavorable interaction is one where the CO2 solubility is less than ideal at these same conditions. It is clear that for cases where CO2 solubility is greater than ideal, and therefore where the (attractive) CO2–solvent interactions are greater than

7248_book.fm Page 185 Tuesday, April 24, 2007 9:19 AM

Determination of Hansen Solubility Parameter Values for Carbon Dioxide

185

solvent–solvent interactions, Ra should be less than Ro. A convenient index for relative goodness of a solvent is the ratio Ra/Ro, which has been called the relative energy difference (RED) number5 RED = Ra Ro

(10.12)

For an individual solvent, a value of RED < 1 indicates favorable CO2–solvent interactions, whereas a RED ≈ 1 represents a boundary condition between good and bad. Progressively higher values of RED indicate progressively more unfavorable interactions. Computing Ra from Equation 10.11 and the RED from Equation 10.12 allows for easy scanning of large data sets, such as the 101 solvents listed in data set #1. The solubility spheres optimized for data set #1 and data set #2 with the two HSP center points are shown in Figure 10.2a, Figure 10.2b, and Figure 10.2c. As can be seen from these plots, an interaction radius Ro = 4.0 for data set #2 incorporates the good solvents, whereas for data set #1, an interaction radius Ro = 4.7 is necessary to incorporate the good solvents. In addition, the solubility sphere analysis for data set #1 results in the inclusion of 7 bad solvents (2-methylcyclohexanone, cyclohexanone, oleic acid, dichloromethane, trichloromethane, propylene bromide, and 1,2-dibromoethane), whereas the sphere analysis generated for data set #2 results in the inclusion of only 1 bad solvent (oleic acid). In terms of the solubility sphere technique, occurrences of good solvents falling outside of the sphere radius, and bad solvents falling inside the sphere radius can be viewed as an indication of the goodness of the fit.5

CO2 Solubility in Liquid Solvents, PCO2 = 1 atm., T=25°C 1 Data Set #1

2 Data Set #2 20 18 16

10 8 6 4

δP (MPa)

12

½

14

Alcohols Alkanes/Nonpolar compounds Halogenated compounds Simple acids/Carboxylic acids Esters Ketones Amides Cyclic Alkenes Nitro compounds Aldehyde Amines Nitriles Anhydrides Sulfoxide compounds

2 0 12

13

14

15

16 17 18 ½ δd (MPa) (a)

19

20

21

22

FIGURE 10.2 Two-dimensional plots of CO2 in organic solvents. (a) δp vs. δd, (b) δh vs. δp, and (c) δh vs. δd.

7248_book.fm Page 186 Tuesday, April 24, 2007 9:19 AM

186

Hansen Solubility Parameters: A User’s Handbook

CO2 Solubility in Liquid Solvents, PCO2 = 1 atm., T=25°C 1 Data Set #1

2 Data Set #2

22 Alcohols Alkanes/Nonpolar compounds

20 18

Halogenated compounds Simple acids/Carboxylic acids Esters Ketones

16

δh (MPa)½

14

Amides

12

Cyclic Alkenes Nitro compounds Aldehyde Amines

10 8 6

Nitriles Anhydrides Sulfoxide compounds

4 2 0

2

4

6

8

10 δp (MPa)

12

14

16

18

20

½

(b) FIGURE 10.2 (Continued)

From the refinement of the two sets of CO2 HSP values, using the solubility sphere methodology, the HSP values from data set #2 were selected as the optimum reference values for CO2 at T=25°C; δ d = 15.6 MPa1/ 2 δ p = 5.2 MPa1/ 2 δ h = 5.8 MPa1/ 2 This determination is further supported based on problems noted by Hansen,5 who observed that the approach of using all solvents to establish the center of a solubility sphere can result in this sphere boundary (and center) being determined by the poor solvents or nonsolvents, rather than the best solvents in the middle. A comparison of these CO2 values can be made with the large database of HSP values found in Hansen Solubility Parameters: A User’s Handbook.5 Similar values are reported for liquid solvents such as dipropyl ketone: δd = 15.8 MPa1/2, δp = 5.7 MPa1/2, and δh = 4.9 MPa1/2; 1,3dimethoxybutane: δd = 15.6 MPa1/2, δp = 5.5 MPa1/2, and δh = 5.2 MPa1/2; and ethyl acetate: δd = 15.8 MPa1/2, δp = 5.3 MPa1/2, and δh = 7.2 MPa1/2. It should be noted, however, that the CO2 optimum HSP reference values correspond to a reference temperature of 25°C and a reference pressure of 90.5 MPa8, that is, a higher operating pressure than found in common industrial applications. A methodology for extending this reference set of HSP values to any temperature and pressure has been developed and is discussed in the following sections.

7248_book.fm Page 187 Tuesday, April 24, 2007 9:19 AM

Determination of Hansen Solubility Parameter Values for Carbon Dioxide

187

CO2 Solubility in Liquid Solvents, PCO2 = 1 atm., T=25°C 1 Data Set #1

2 Data Set #2

20 18

Alcohols Alkanes/Nonpolar compounds Halogenated compounds

16

Simple acids/Carboxylic acids Esters Ketones Amides

δh (MPa)½

14 12 10 8

Cyclic Alkenes Nitro compounds Aldehyde

6

Amines

4

Nitriles Anhydrides Sulfoxide compounds

2 0

13

14

15

16

17

δd (MPa)

18

19

20

21

22

½

(c) FIGURE 10.2 (Continued)

ONE-COMPONENT HILDEBRAND PARAMETER AS A FUNCTION OF TEMPERATURE AND PRESSURE Hildebrand’s solubility parameter theory was derived from an approximation of the internal pressure of a fluid. This was later termed the cohesive energy density (ced), based on work conducted in 1928,9 1929,10 1932,11 and 195012 where the two terms, internal pressure and cohesive energy density were found to be related by the quantity, n, as shown in Equation 10.13: ⎛ ∂E ⎞ n ΔE ⎜⎝ ∂V ⎟⎠ = V T

(10.13)

where (ΔE/V) is defined by Hildebrand as ced, and (∂E/∂V)T is the internal pressure. Hildebrand and coworkers found that for nonpolar/nonassociating liquids, where intermolecular interactions are weak, n is not far from unity9–11 and the equality of ced and internal pressure is a good approximation. This same work also demonstrates that n is near unity for nonpolar liquids and also for polar liquids where the dipole moment is less than 2 D, and where specific interactions (particularly hydrogen bonding) are largely absent. 1 D is equal to 1.0 × 10-18 (ESO) or 3.336 × 10–30 Cm. Whereas no direct evaluation of the value of n has been found in the literature for carbon dioxide (CO2), a comparison of the values found by Hildebrand and others13–16 strongly suggests that the value of n for CO2 is expected to be near unity. As a result, the internal pressure and ced are approximately equal.

7248_book.fm Page 188 Tuesday, April 24, 2007 9:19 AM

188

Hansen Solubility Parameters: A User’s Handbook

Accordingly, Hildebrand's solubility parameter, defined as the square root of the ced,12 can also be approximated by the square root of the internal pressure for nonpolar/nonassociating fluids. 1/ 2

1/ 2

⎛ ΔE ⎞ ⎛ ∂E ⎞ δ=⎜ ≈⎜ ⎝ V ⎟⎠ T ⎝ ∂V ⎟⎠ T

(10.14)

And the internal pressure can be calculated from the thermodynamic equation of state, Equation 10.15. ⎛ ∂E ⎞ ⎛ ∂P ⎞ ⎜⎝ ∂V ⎟⎠ = T ⎜⎝ ∂T ⎟⎠ − P T V

(10.15)

⎛ ∂E ⎞ ⎛ ∂P ⎞ δ2 ≈ ⎜ =T⎜ −P ⎟ ⎝ ∂V ⎠ T ⎝ ∂T ⎟⎠ V

(10.16)

so that

Total (one-component) solubility parameters can therefore be calculated using an EOS of the form, P = f (ρ,T). This approach is used in this work to calculate the total solubility parameter for pure CO2, using the empirical EOS of Huang et al.17 ⎡ ⎢1 + b ρ' + b ρ' 2 + b ρ' 3 + b ρ' 4 + b ρ' 5 + b ρ' 2 exp p −c21ρ' 2 + b8 ρ' 4 exp −c21ρ' 2 2 3 4 5 6 7 ⎢ ⎢ 2 2 2 Δρ P = ρ R T ⎢⎢ +c22 ρ' 2 exp ⎡ −c27 ΔT ⎤ + c23 ' exp ⎡ −c25 Δρ − c27 ΔT ⎤ ⎢⎣ ⎦⎥ ⎣⎢ ⎦⎥ ρ ⎢ ⎢ 2 2 ⎢ +c24 Δρ exp ⎡ −c26 Δρ − c27 ΔT ⎤ ' ⎢ ⎥⎦ ⎣ ρ ⎣⎢

(

( )

( )

)

( )

(

( )

( )

⎤

)⎥⎥

⎥ ⎥ (10.17) ⎥ ⎥ ⎥ ⎥ ⎥⎦

where T ′ = T Tc ;

ΔT = 1 − T ′;

ρ′ = ρ ρc ;

Δρ = 1 − 1 / ρ′

(10.18)

and, ⎛ c c c c c ⎞ b2 = ⎜ c1 + 2' + 3' 2 + 4' 3 + 5' 4 + 6' 5 ⎟ ; ⎝ T T T T T ⎠

⎛c ⎞ b6 = ⎜ 14' ⎟ ⎝T ⎠

⎛ c c ⎞ b3 = ⎜ c 7 + 8' + 9' 2 ⎟ ; ⎝ T T ⎠

⎛c c c ⎞ b7 = ⎜ 15' 3 + 16' 4 + 17' 5 ⎟ ⎝T T T ⎠

⎛ c ⎞ b4 = ⎜ c10 + 11' ⎟ ; ⎝ T ⎠

⎛c c c ⎞ b8 = ⎜ 18' 3 + 19' 4 + 20' 5 ⎟ ⎝T T T ⎠

⎛ c ⎞ b5 = ⎜ c12 + 13' ⎟ ⎝ T ⎠ The state constants (ci) are as defined in the Huang reference.

(10.19)

7248_book.fm Page 189 Tuesday, April 24, 2007 9:19 AM

Determination of Hansen Solubility Parameter Values for Carbon Dioxide

189

25 25 δ (MPa)½

15

15 10

SUPERCRITICAL FLUID

10

5

5

δ (MPa)½

20 LIQUID

20

0

0 GAS

0

25

0

30 T (K

)

0 35

0

40

0 0

10

20

30

40

50

)

Pa

M P(

45

FIGURE 10.3 Total (one-component) solubility parameter of pure CO2 calculated using Equation 10.16 and Equation 10.17.

It should be noted that there are a wide range of available EOSs for carbon dioxide and a comparison of Huang’s and others can be found in a review by Span and Wagner.18 These equations and the appropriate derivatives are then used to calculate CO2 solubility parameters over the temperature and pressure range for which the EOS is stated to be valid (220 K ≤ T ≤ 420 K, and 0.1 MPa ≤ P ≤ 60 MPa). Figure 10.3 is a plot of the resulting one-component solubility parameters. Other notable works include Allada’s19 proposed generalized solubility parameter (that uses analytical equations of state (EOSs), such as Lee-Kelser or modified Redlich-Kwong for evaluation), the modified solubility parameter proposed by Ikushima et al.20 (where the solubility parameter is expressed in terms of reduced parameters), and the EOS model proposed by Panayiotou.21 This later work utilizes the lattice fluid theory and a lattice fluid hydrogen bonding model to evaluate solubility parameters and two separate components: physical (or van der Waals) and chemical (or specific, e.g., hydrogen bonding).

THREE-COMPONENT (HANSEN) SOLUBILITY PARAMETERS — PURE CO2 Extending the HSP methodology to supercritical fluids would significantly enhance the understanding of their solvent properties; however, no such studies appear to have been done. The pressure volume temperature (PVT) EOS that calculates the total (Hildebrand) CO2 solubility parameter value (Equation 10.16) was used to determine the combination of pressure and molar volume ⎛ ⎛ ∂P ⎞ ⎞ corresponding to T = 25°C and δt = ⎜ T ⎜ − P⎟ ⎟ ⎝ ⎝ ∂T ⎠ V ⎠

1/ 2

= 17.4 MPa1/2 that gave:

P = 91.7 MPa VCO2 = 39.13 cm 3 mole

(10.20)

7248_book.fm Page 190 Tuesday, April 24, 2007 9:19 AM

190

Hansen Solubility Parameters: A User’s Handbook

HSP values at other pressures and temperatures will be based on this set of HSP values, using pressure and temperature integral functions, which will be derived subsequently. Both temperature and pressure will influence total solubility parameters. However, other than Giddings’ extension of the one-component (Hildebrand) solubility parameter model to supercritical fluids,22 there appears to be no published reports on methods to calculate total solubility parameters as a function of pressure and only limited reports on the calculation of solubility parameters as a function of temperature.5,7,23 Generally, an increase in pressure at constant temperature will increase the total solubility parameter through an increase in the solvent density. Similarly, an increase in temperature at constant pressure will decrease the total solubility parameter. Both of these trends can be seen in Figure 10.3, where the total CO2 solubility parameter was calculated using Equation 10.16 and Equation 10.17. The temperature and pressure dependence of individual HSPs as a function of temperature and pressure has apparently not been evaluated for any liquid, gas, or supercritical fluid. A suggested approach for this calculation is outlined as follows where the temperature derivatives, originally derived by Hansen and Beerbower,24 are verified. Pressure derivatives, not found in any literature search, are derived in a manner parallel to the temperature derivatives. In addition, integral forms are developed.

TEMPERATURE AND PRESSURE EFFECTS ON HSPS: δd Hildebrand12 in his 1950 work considered the effect of temperature on solubility parameters by recalling the expression for the dependence of E on the volume: E=−

k Vn

(10.21)

where k is a constant dependent upon the nature of the particular liquid, and n is about 1.5 for normal (nonassociating or van der Waals) liquids. Substituting Equation 10.21 into Hansen’s definition for the dispersion solubility parameter, k1/2 δ d = − ( n+1) 2 V

(10.22)

allows one to calculate the change in δd produced by a change in volume by differentiating Equation 10.22. ⎛ ∂δ d ⎞ ⎛ n + 1⎞ = k 1/ 2 ⎜ ⎜⎝ ∂V ⎟⎠ ⎝ 2 ⎟⎠ T ,P

⎡ − ( n + 1) ⎤ ⎢V 2 ⎥ ⎡V −1 ⎤ ⎦ ⎥⎣ ⎢ ⎣ ⎦

(10.23)

⎛ n + 1⎞ ⎛ 1 ⎞ = −δ d ⎜ ⎝ 2 ⎟⎠ ⎜⎝ V ⎟⎠ and, ∂δ d ∂V = −1.25 δd V

(10.24)

7248_book.fm Page 191 Tuesday, April 24, 2007 9:19 AM

Determination of Hansen Solubility Parameter Values for Carbon Dioxide

191

Equation 10.24 can now be differentiated for either a change in temperature or pressure, or integrated. Results of both derivations are shown in Table 10.4 and Table 10.5.

TEMPERATURE AND PRESSURE EFFECTS ON HSPS: δp The first values of δp were assigned by Hansen and Skaarup using the Böttcher equation, shown here as Equation 10.25,

δ 2P =

⎡ cal ⎤ 12108 ε − 1 nD2 + 2 μ 2 ⎢ 3 ⎥ V 2 2 ε + nD2 ⎣ cm ⎦

(

)

(10.25)

A simplified equation was later developed by Hansen and Beerbower,24 δp =

37.4μ ⎡ MPa1/ 2 ⎤⎦ V 1/ 2 ⎣

(10.26)

where μ is the dipole moment (in Debyes). This equation is utilized for determining the change in δp with respect to either temperature at constant pressure or with respect to pressure at constant temperature. ⎛ ∂δ p ⎞ ⎛ 1 ⎞ −3 2 ( 37.4μ ) ⎜⎝ ∂V ⎟⎠ = − ⎜⎝ 2 ⎟⎠ V T ,P δp 1 ⎛ 37.4 μ ⎞ =− =− ⎜ 1/ 2 ⎟ ⎠ 2V ⎝ V 2V

(10.27)

and ∂δ p δp

=−

∂V . 2V

(10.28)

Equation 10.28 can now be differentiated for either a change in temperature or pressure, or integrated. Results of both derivations are shown in Table 10.4 and Table 10.5.

TEMPERATURE AND PRESSURE EFFECTS ON HSPs: δh In Hansen’s early work, the hydrogen bonding parameter was almost always found by subtracting the polar and dispersion energies of vaporization from the total energy of vaporization. This is still widely used where the required data are available and reliable. Hansen,5 however, noting that there is no rigorous way of arriving at values of the temperature dependence of the hydrogen bonding solubility parameter, developed an empirical approach for the determination of the temperature dependence of δh, which involves experimental heats of vaporization data for hydrogen-bonded substances, which, in turn, are taken from Bondi.25 From Equation 10.4, the hydrogen bonding solubility parameter, δh, is defined as: δ 2h =

Eh V

(10.29)

7248_book.fm Page 192 Tuesday, April 24, 2007 9:19 AM

192

Hansen Solubility Parameters: A User’s Handbook

so that E h = V δ 2h

(10.30)

where Eh is the hydrogen bonding contribution to the total cohesive energy. Differentiating Equation 10.30 with respect to temperature at constant pressure, ⎛ ∂E h ⎞ ⎛ ∂δ h ⎞ 2 ⎛ ∂V ⎞ ⎜⎝ ∂T ⎟⎠ = V 2 δ h ⎜⎝ ∂T ⎟⎠ + δ h ⎜⎝ ∂T ⎟⎠ P P P

( )

⎛ ∂δ ⎞ ⎛ ∂E ⎞ ⎛ ∂V ⎞ 2V δ h ⎜ h ⎟ = ⎜ h ⎟ − δ 2h ⎜ ⎝ ∂T ⎠ P ⎝ ∂T ⎠ P ⎝ ∂T ⎟⎠ P

⎛ ∂δ h ⎞ ⎜⎝ ∂T ⎟⎠ = P

(10.31)

⎛ ∂E h ⎞ 2 ⎛ ∂V ⎞ ⎜⎝ ∂T ⎟⎠ − δ h ⎜⎝ ∂T ⎟⎠ P P 2V δ h

Simplifying, rearranging terms, and substituting in the isobaric coefficient of thermal expansion, α,

⎛ ⎛ ∂E h ⎞ ⎞ ⎜ ⎜⎝ ∂T ⎟⎠ ⎟ ⎛ ∂δ h ⎞ α⎟ P ⎜ = − δ h ⎜⎝ ∂T ⎟⎠ ⎜ 2 Eh 2⎟ P ⎟ ⎜ ⎝ ⎠

(10.32)

Bondi,25 through exploratory calculations, has shown that the difference between the heat of vaporization of a hydroxylic compound (a compound displaying strong hydrogen bonding) and that of its hydrocarbon (or other nonpolar) homomorph constitutes a good measure of hydrogen bond strength. This work also discusses the decrease in the heat of formation of the hydrogen bond with increasing temperature. Reference curves of (dEh/dT) were constructed23 for various functional groups and are shown in Table 10.3 along with experimentally derived values of Eh.25 Averaging the rate of change of the hydrogen bond heat of vaporization with temperature (dEh/dT), and dividing by the average excess heats of vaporization (heat of vaporization of the hydrogen bonding compound minus the heat of vaporization of its nonpolar homomorph) results in the following form of Equation 10.32, ⎛ 2.64 × 10 −3 α ⎞ ⎛ ∂δ h ⎞ δ = − + ⎟ h ⎜ ⎜⎝ ∂T ⎟⎠ 2 2⎠ ⎝ P ⎛ α⎞ = −δ h ⎜ 1.32 × 10 −3 + ⎟ 2⎠ ⎝

(10.33)

7248_book.fm Page 193 Tuesday, April 24, 2007 9:19 AM

Determination of Hansen Solubility Parameter Values for Carbon Dioxide

193

TABLE 10.3 Experimentally Determined Values of Eh and (dEh /dT) Functional Group

Hydrogen-Bond Parameter, Eh (cal/mole)

dEh/dT (cal/mole·K)

± ± ± ±

–10 –4.5 –7.0 –2.9

–OH (aliphatic) –NH2 (aliphatic) –CN (aliphatic) –COOH (aliphatic)

4650 1350 500 2750

400 200 200 250

The change in the δh with respect to pressure at constant temperature is obtained by utilizing the relationship: ∂δ h ∂δ h ∂T = ⋅ ∂P ∂T ∂P

(10.34)

∂T β =− ∂P α

(10.35)

⎛ 1.32 × 10 −3 β β ⎞ ⎛ ∂δ h ⎞ δ = + ⎟ h ⎜ ⎜⎝ ∂P ⎟⎠ 2⎠ α ⎝ T

(10.36)

where,

so that

Equation 10.36 can be rearranged to a form that can also be easily integrated, Table 10.5. The derivative forms are summarized in Table 10.4 and the integrated forms in Table 10.5. The total solubility parameter, incremented for small changes in temperature and pressure, can be calculated from equations (derivative form) in Table 10.4,

TABLE 10.4 Equations (Derivative Form) for the Temperature and Pressure Effects on HSP Temperature Increment

δd

δp

δh

Pressure Increment

⎛ ∂δ d ⎞ ⎜ ∂T ⎟ = −1.25δ d α ⎝ ⎠P

⎛ ∂δ d ⎞ ⎜ ∂P ⎟ = 1.25δ d β ⎝ ⎠T

⎛ ∂δ p ⎞ ⎛ α⎞ ⎜ ⎟ = −δ p ⎜ ⎟ ⎝ 2⎠ ⎝ ∂T ⎠ P

⎛ ∂δ p ⎞ ⎛ β⎞ ⎜ ⎟ = δp ⎜ ⎟ ⎝ 2⎠ ⎝ ∂P ⎠ T

⎛ ∂δ h ⎞ ⎛ α⎞ −3 ⎜⎝ ∂T ⎟⎠ = −δ h ⎜⎝ 1.32 × 10 + 2 ⎟⎠ P

⎛ 1.32 × 10 −3 β β ⎞ ⎛ ∂δ h ⎞ δ = + ⎟ h ⎜ ⎜⎝ ∂P ⎟⎠ α 2⎠ ⎝ T

7248_book.fm Page 194 Tuesday, April 24, 2007 9:19 AM

194

Hansen Solubility Parameters: A User’s Handbook

TABLE 10.5 Equations (Integrated Form) for the Temperature and Pressure Effects on HSP δd

δ dref δd

δp

δ pref δp

δh

δ href δh

⎛ Vref ⎞ =⎜ ⎟ ⎝ V ⎠

−1.25

⎛ Vref ⎞ =⎜ ⎟ ⎝ V ⎠

−0.5

.5 ⎡ ⎛ Vref ⎞ ⎤ −3 ⎢ ⎥ = exp −1.32 × 10 Tref − T − ln ⎜ ⎟ ⎢ ⎝ V ⎠ ⎥ ⎣ ⎦

(

)

2 ⎡ ⎤ ⎡ ⎤ ⎛ ∂δ p ⎞ ⎛ ∂δ p ⎞ ⎛ ∂δ d ⎞ ⎛ ∂δ d ⎞ 2 ⎢δ p + ⎥ + δ = ⎢δ d + ⎜ + + Δ T Δ P Δ T Δ P ⎥ ⎟ ⎜ ⎟ ⎜ ⎜⎝ ∂P ⎟⎠ ⎢ ⎥ ∂ ∂ T P ⎝ ∂T ⎟⎠ P ⎠ ⎝ ⎠ ⎝ ⎢⎣ ⎥ T ⎦ P T ⎣ ⎦

⎡ ⎛ ∂δ ⎞ ⎛ ∂δ ⎞ + ⎢δ h + ⎜ h ⎟ ΔT + ⎜ h ⎟ ΔP ∂ T ⎠P ⎝ ⎝ ∂P ⎠ T ⎢ ⎣

⎤ ⎥ ⎥ ⎦

2

(10.37)

2

or from the equations (integrated form) in Table 10.5 2 2 ⎡ ⎤ ⎡ ⎤ ⎡ ⎤ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ δ pref ⎥ ⎢ δ dref ⎥ δ href 2 ⎢ ⎥ + +⎢ δ =⎢ −1.25 ⎥ −0.5 ⎥ 0.5 ⎞ ⎥ ⎢ ⎛ ⎢ ⎛ Vref ⎞ ⎥ ⎢ ⎛ Vref ⎞ ⎥ ⎢ exp ⎜ −1.32 × 10 −3 T − T − ln ⎛ Vref ⎞ ⎟ ⎥ ⎢⎜ ⎥ ⎢⎜ ⎥ ⎟ ⎟ ref ⎜ V ⎟ ⎢ V V ⎠ ⎠ ⎜⎝ ⎝ ⎠ ⎟⎠ ⎥ ⎢⎣ ⎝ ⎥⎦ ⎢⎣ ⎝ ⎥⎦ ⎢⎣ ⎥⎦

(

2

)

(10.38) where the reference values are as determined earlier; δdref = 15.6 MPa1/2, δpref = 5.2 MPa1/2, δhref = 5.8 MPa1/2, Vref = 39.13 cm3/mole, and Tref =298.15 K. CO2 HSP values calculated with the equations in Table 10.4, as a function of temperature and pressure, are shown in the CO2 HSP surface diagrams illustrated in Figure 10.4. From this work though, CO2 HSP values at a temperature of 25°C and a pressure of 200 bar, δd = 12.2 MPa1/2, δp = 4.7 MPa1/2, and δh = 5.2 MPa1/2 can also be compared to liquid solvent HSP values. Comparible liquid solvents include chlorodifluoromethane: δd = 12.3 MPa1/2, δp = 6.3 MPa1/2, and δh = 5.7 MPa1/2; isopropyl ether: δd = 13.7 MPa1/2, δp = 3.9 MPa1/2, and δh = 2.3 MPa1/2; and vinyl trifluoroacetate: δd = 13.9 MPa1/2, δp = 4.3 MPa1/2, and δh = 7.6 MPa1/2. It is also interesting to note that the total solubility parameter value of hexane, a solvent that CO2 is often closely compared to26–28 (δt = 14.9 MPa1/2) is very near the total solubility parameter of CO2 (at 25°C and 200 bar), δt = 14.0 MPa1/2. Yet, when the two are compared in terms of HSP values, little similarity is noted; hexane: δd = 14.9 MPa1/2, δp = 0.0 MPa1/2, and δh = 0.0 MPa1/2.

7248_book.fm Page 195 Tuesday, April 24, 2007 9:19 AM

20 17.5 15 12.5 10 7.5 5 2.5 0 2

50

30

0

0 35

T (K

)

40

0 45

10

00

20

30

40

20 17.5 15 12.5 10 7.5 5 2.5 0

50

195

δd (MPa)½

δd (MPa)½

Determination of Hansen Solubility Parameter Values for Carbon Dioxide

)

Pa

M P(

(a) CO2 dispersion parameter as a function of temperature and pressure.

5

5

4

4

3

3

2

2

1

1

δp (MPa)½

δp (MPa)½

6 6

0

0 25

0 3

00 3

T (K

50

40

)

0

0 45

0

10

20

30

40

50 )

Pa

M P(

7 6 5 4 3 2 1 0

7 6 5 4 3 2 1 0 2

50

30

T (K

0

)

3

50

40

0

0 45

0

10

20

30

40

δh (MPa)½

δh (MPa)½

(b) CO2 polar parameter as a function of temperature and pressure 45K)

50 )

Pa

M P(

(c) CO2 hydrogen bonding parameter as a function of temperature and pressure

FIGURE 10.4 HSP values for CO2 as a function of T and P. (a) CO2 dispersion parameter as a function of temperature and pressure. (b) CO2 polar parameter as a function of temperature and pressure. (c) CO2 hydrogen bonding parameter as a function of temperature and pressure.

7248_book.fm Page 196 Tuesday, April 24, 2007 9:19 AM

196

Hansen Solubility Parameters: A User’s Handbook

CONCLUSION A set of Hansen solubility parameters at T = 25°C have been determined for CO2, based on the room temperature solubility in different liquid solvents of known HSP values: δd = 15.6 MPa1/2, δp = 5.2 MPa1/2, and δh = 5.8 MPa1/2. Further, this set of HSPs were refined using the RED affinity number and Hansen solubility plots to correlate the solubility of carbon dioxide in the solvents identified in Table 10.1. It is important to note that the solubility parameter, or rather the difference in solubility parameters, for a given solvent–solute combination has been foremost in determining the mutual solubility of the system.5 An analogy to “like dissolves like” is appropriate. Therefore, the accuracy to which the solubility parameters for a binary pair can be known will be valuable in predicting the system’s behavior. This work introduces a theoretical methodology for generating solubility parameter values, both one-component Hildebrand and three-component Hansen parameters, for a pure supercritical fluid, using CO2 as an example. The ability to express molecular interactions, in terms of HSPs, for a pure fluid solvent in a way that unites the liquid, gas, and supercritical phases, represents an advancement in the understanding of the role of solvents in both existing and new applications.

ACKNOWLEDGMENTS Special thanks to James Rubin of Los Alamos National Laboratory for his excellent assistance and collaboration in the development of this work.

CHAPTER 10 ADDENDUM Research has been the key to many progressive developments. Significant steps toward improved understanding have been made with limited resources, and these result in still other improvements in the same direction when other resources are applied. The HSP for carbon dioxide reported in the chapter (δd, δp, and δh equal to 15.6 MPa1/2, 5.2 MPa1/2, and 5.8 MPa1/2) were found by a plotting technique with a radius for the solubility sphere of 4.0 MPa1/2. The data used were collected from a wide variety of sources, and the criterion for a good solvent was solubility in excess of the theoretical. A computer analysis of the same data has now shown that it is possible to describe the solubility of carbon dioxide with a slightly different solubility sphere. The δd, δp, and δh values 15.7 MPa1/2, 6.3 MPa1/2, and 5.7 MPa1/2 were found to give a perfect data fit of 1.000 (versus 0.981) with a radius of only 3.3 MPa1/2. Both correlations emphatically show the ability of this procedure, whether by hand or by computer, to correlate gas/liquid solubility data for a wide variety of chemically different solvents. It has not been possible, nor has it been deemed necessary, to revise the contents of this chapter using these slightly different numbers. This note is only to indicate why a slightly different set of HSP is reported elsewhere in this handbook (Table A-1 and Chapter 13, Table 13.4).

7248_book.fm Page 197 Tuesday, April 24, 2007 9:19 AM

Determination of Hansen Solubility Parameter Values for Carbon Dioxide

197

SYMBOLS SPECIAL TO THIS CHAPTER KH L P ai fi fio k pi pis Ω β ρ

Henry’s law constant (Equation 10.5) Ostwald coefficient in Equation 10.7 Pressure Activity coefficient of the “i”th component in Appendix 10.A.1, Equation 10.A.1 Fugacity of the “i”th component in Appendix 10.A.1, Equation 10.A.1 Fugacity at standard state in Appendix 10.A.1, Equation 10.A.1 Constant in Equation 10.21–Equation 10.23 Partial pressure of the “i”th component in Equation 10.A.2 Saturation pressure of the “i”th component in Equation 10.A.2 Bunsen coefficient in Equation 10.6 Compressibility Density

REFERENCES 1. Hansen, C.M., The three dimensional solubility parameter — key to paint component affinities. I. Solvents, plasticizers, polymers, and resins, J. Paint Technol., Vol. 39, No. 505, 104–117, 1967. 2. Wilhelm, E. and Battino, R., Thermodynamic functions of the solubilities of gases in liquids at 25°C, Chem. Rev., Vol. 73, No. 1, 1–9, 1973. 3. Battino, R. and Clever, H.L., The solubility of gases in liquids, Solutions and Solubilities, Dack, M.R.J., Ed., John Wiley & Sons, New York, 1965, p. 379, chap. 7. 4. Reid, R.C., Prausnitz, J.M., and Poling, B.E., The Properties of Gases and Liquids, 4th ed., McGraw Hill, New York, 1987, p. 334. 5. Hansen, C.M., Hansen Solubility Parameters: A User’s Handbook, CRC Press, Boca Raton, FL, 1999. 6. Hansen, C.M., The Three Dimensional Solubility Parameter and Solvent Diffusion Coefficient, Copenhagen Danish Technical Press, Denmark, 1967, pp. 33–38. 7. Barton, A.F.M., CRC Handbook of Solubility Parameters and Other Cohesion Parameters, 2nd ed., CRC Press, Boca Raton, FL, 1991. 8. Williams, L.L., Rubin, J.B., and Edwards, H.W., Calculation of Hansen solubility parameter values for a range of pressure and temperature conditions, including the supercritical fluid region, Ind. Eng. Chem. Res., Vol. 43, 4967–4972, 2004. 9. Westwater, W., Frantz, H.W., and Hildebrand, J.H., The internal pressure of pure and mixed liquids, Phys. Rev., Vol. 31, 135–144, 1928. 10. Hildebrand, J.H., Intermolecular forces in liquids, Phys. Rev., Vol. 34, 984–993, 1929. 11. Hildebrand, J.H. and Carter, J.M., A study of van der Waals forces between tetrahalide molecules, J. Am. Chem. Soc., Vol. 54, 3592–3603, 1932. 12. Hildebrand, J.H. and Scott, R.L., The Solubility of Nonelectrolytes, 3rd ed., Reinhold, New York, 1950. 13. Dack, M.R.J., Solvent structure. The use of internal pressure and cohesive energy density to examine contributions to solvent-solvent interactions, Aust. J. Chem., Vol. 28, 1643–1648, 1975. 14. Renuncio, J.A.R., Breedveld, G.J.F., and Prausnitz, J.M., Internal pressures and solubility parameters for carbon disulfide, benzene, and cyclohexane, J. Phys. Chem., Vol. 81, No. 4, 324–327, 1977. 15. Allen, G., Gee, G., and Wilson, G.J., Intermolecular forces and chain flexibilities in polymers: I. Internal presssures and cohesive energy densities of simple liquids, Polymer, Vol. 1, No. 4, 456–476, 1960. 16. MacDonald, D.D. and Hyne, J.B., The thermal pressure and energy-volume coefficients of the methyl alcohol-water and t-butyl alcohol-water systems, Can. J. Chem., Vol. 49, 2636–2642, 1971. 17. Huang, F., Li, M., Lee, L., and Starling, K., An accurate equation of state for carbon dioxide, J. Chem. Eng. Jpn., Vol. 18, No. 6, 490–496, 1985.

7248_book.fm Page 198 Tuesday, April 24, 2007 9:19 AM

198

Hansen Solubility Parameters: A User’s Handbook

18. Span, R. and Wagner, W., A new equation of state for carbon dioxide covering the fluid region from the triple-point temperature to 1100 K at pressures up to 800 MPa, J. Phys. Chem. Ref. Data, Vol. 25 No. 6, 1509, 1996. 19. Allada, S.R., Solubility parameters of supercritical fluids, Ind. Eng. Chem. Process Des. Dev., Vol. 23, 344, 1984. 20. Ikushima, Y., Goto, T., and Arai, M., Modified solubility parameter as an index to correlate the solubility in supercritical fluids, Bull. Chem. Soc. Jpn., Vol. 60, 4145, 1987. 21. Panayiotou, C., Solubility parameter revisited: an equation-of-state approach for its estimation, Fluid Phase Equilibria, Vol. 131, 21, 1997. 22. Giddings, C.J., Myers, M.N., McLaren, L., and Keller, R.A., High Pressure Gas Chromatography of Nonvolatile Species, Science, Vol. 162, 1968, pp. 67–73. 23. Bondi, A., Physical Properties of Molecular Crystals, Liquids, and Glasses, John Wiley & Sons, New York, 1968. 24. Hansen, C. and Beerbower, A., Solubility parameters, Kirk-Othmer Encyclopedia of Chemical Technology, Suppl. Vol., Interscience, New York, 1971. 25. Bondi, A. and Simkin, D.J., Heats of vaporization of hydrogen bonded substances, AIChE J., Vol. 3 No. 4, 473, 1957. 26. O’Neill, M.L., Cao, Q., Fang, M., Johnston, K.P., Wilkinson, S.P., Smith, C.D., Kerschner, J.L., and Jureller, S.H., Solubility of homopolymers and copolymers in carbon dioxide, Ind. Eng. Chem. Res., Vol. 37, 3067–3079, 1998. 27 Fedotov, A.N., Sinevich, E.A., and Simonov, A.P., Intermolecular interactions of supercritical carbon dioxide with polymers of different types, Russ. J. Phys. Chem., Vol. 71, No. 11, 1900–1903, 1997. 28. Hyatt, J.A., Liquid and supercritical carbon dioxide as organic solvents, J. Org. Chem., Vol. 49, 5097–5101, 1984. 29. Just, G., Z. Phys. Chem., Vol. 37, p. 342, 1901. 30. Kunerth, W., Solubility of CO2 and N2O in certain solvents, Phys. Rev., Vol. 19, 512–524, 1922. 31. De Ligny, C.L. and van der Veen, N.G., On the applicability of Pierotti’s theory for the solubiliy of gases in liquids, J. Solution Chem., Vol. 4, No. 10, 841–851, 1975. 32. Pirig, Y.N., Polyuzhin, I.V., and Makitra, R.G., Carbon dioxide solubility, Russ. J. Appl. Chem., Vol. 66, No. 4, Part 2, 691–695, 1993. 33. Podvigaylova, I.G., Zaynalov, B.K., Kruglikov, A.A., Radzhabov, D.T., Shagidanov, E.N., Shestakova, T.G., Korbutova, Z.V., and Mel’nikova, L.I., Solubility of CO2 in Organic Solvents, The Soviet Chemical Industry, No. 5, pp. 19–21, 1970. 34. Gjaldbaek, J.C. and Anderson, E.K., The solubility of carbon dioxide, oxygen, carbon monoxide, and nitrogen in polar solvents, Acta Chem. Scand., Vol. 8, 1398–1413, 954. 35. Gerrard, W., Solubility of Gases and Liquids: A Graphic Approach, Plenum Press, New York, 1976, p. 73. 36. Gjaldbaek, J.C., The solubility of carbon dioxide in perfluoro-n-heptane, normal heptane, cyclohexane, carbon tetrachloride, benzene, carbon disulphide, and aqueous solution in aerosol, Acta Chem. Scand., Vol. 7, 537–544, 1953. 37. Pardo, J., Lopez, M.C., Santafe, J., Royo, F.M., and Urieta, J.S., Solubility of gases in butanols. I., Fluid Phase Equilibria, Vol. 109, 29–37, 1995. 38. Fogg, P.G.T., Ed., Carbon Dioxide in Nonaqueous Solvents, Solubility Data Series, Vol. 50 1992 International Union of Pure and Applied Chemistry (IUPAC), Research Triangle Park, NC. 39. Pardo, J., Lopez, M.C., Mayoral, J.A., Royo, F.M., and Urieta, J.S., Solubility of gases in butanols. III., Fluid Phase Equilibria, Vol. 134, 133–140, 1997. 40. Pardo, J., Mainar, A.M., Lopez, M.C., Royo, F.M., and Urieta, J.S., Solubility of gases in butanols. IV., Fluid Phase Equilibria, Vol. 155, 127–137, 1999. 41. Kobatake, Y. and Hildebrand, J.H., Solubility and entropy of solution of He, N2, A, O2, CH4, C2H6, CO2 and SF6 in various solvents; regularity of gas solubilities, J. Phys. Chem., Vol. 65, 331–334, 1961. 42. Gironi, F. and Lavecchia, R., A simple method for determining the solubility of gases in liquids: application to CO2-cycloparaffin systems, Fluid Phase Equilibria, Vol. 87, 153–161, 1993. 43. Gallardo, M.A., Lopez, M.C., Urieta, J.S., and Losa, C.G., Solubility of 15 non-polar gases in cycloheptanone, Fluid Phase Equilibria, Vol. 58, 159–172, 1990.

7248_book.fm Page 199 Tuesday, April 24, 2007 9:19 AM

Determination of Hansen Solubility Parameter Values for Carbon Dioxide

199

44. Dymond, J.H., The solubility of a series of gases in cyclohexane and dimethylsulfoxide, J. Phys. Chem., Vol. 71, 1829–1831, 1967. 45. Begley, J.W., Maget, H.J.R., and Williams, B., Solubility of carbon dioxide in cyclohexanol, 1,2dibromoethane, a mixture of 1-chloro-2-bromopropane and 2-chloro-1-bromopropane, and mineral oil, J. Chem. Eng. Data, Vol. 10, No. 1, 4–8, 1965. 46. Gallardo, M.A., Melendo, J.M., Urieta, J.S., and Losa, C.G., Solubility of non-polar gases in cyclohexanone between 273.15 and 303.15 K at 101.32 kPa partial pressure of gas, Can. J. Chem., Vol. 65, 2198–2202, 1987. 47. Wilcock, R.J., Battino, R., and Wilhelm, E., The solubility of gases in liquids 10., J. Chem. Thermodynamics, Vol. 9, 111–115, 1977. 48. Gallardo, M.A., Melendo, J.M., Urieta, J.S., and Losa, C.G., Solubility of He, Ne, Ar, Kr, Xe, H2, D2, N2, O2, CH4, C2H4, C2H6, CF4, SF6, and CO2 in cyclopentanone from 273.15 and 303.15 K at 101.32 kPa partial pressure of gas, Fluid Phase Equilibria, Vol. 50, 223–233, 1989. 49. King, M.B. and Al-Najjar, H., The solubilities of carbon dioxide, hydrogen sulphide and propane in some normal alkane solvents I, Chem. Eng. Sci., Vol. 32, 241–1246, 1977. 50. Wilcock, R.J., Battino, R., Danforth, W.F., and Wilhelm, E., J. Chem. Thermodynamics, Vol. 10, 817–822, 1978. 51. Gallardo, M.A., Melendo, J.M., Urieta, J.S., and Losa, C.G., Solubility of nonpolar gases in 2,6dimethylcyclohexanone, Can. J. Chem., Vol. 68, 435–439, 1990. 52. De Ligny, C.L. and van der Veen, N.G., A test of Pierotti’s theory for the solubility of gases in liquids, by means of literature data on solubility and entropy of solution, Chem. Eng. Sci., Vol. 27, 391–401, 1972. 53. Hayduk, W., Walter, E.B., and Simpson, P., Solubility of propane and carbon dioxide in heptane, dodecane, and hexadecane, J. Chem. Eng. Data, Vol. 17, No. 1, 59–61, 1972. 54. Jou, F.-Y., Deshmukh, R.D., Otto, F.D., and Mather, A.E., Solubility of H2S, CO2 and CH4 in NFormyl Morpholine, J. Chem. Soc., Faraday Trans. I, Vol. 85, No. 9, 2675–2682, 1989. 55. Field, L.R., Wilhelm, E., and Battino, R., The solubility of gases in liquids. 6., J. Chem. Thermodynamics, Vol. 6, 237–243, 1974. 56. Gallardo, M.A., Lopez, M.C., Urieta, J.S., and Losa, C.G., Solubility of nonpolar gases in 2-methylcyclohexanone between 273.15 and 303.15 K at 101.32 kPa partial pressure of gas, Can. J. Chem., Vol. 67, 809–811, 1989. 57. Makranczy, J., Rusz, L., and Balog-Megyery, K., Hung. J. Ind. Chem., Vol. 7, No. 1, 41–46, 1979. 58. Tokunaga, J., J. Chem. Eng. Data, Vol. 20, 41–46, 1975. 59. Hiraoka, H. and Hildebrand, J.H., The solubility and entropy solution of certain gases in (C4F9)3N, CCl2F CClF2, and 2,24-(CH3)3C5H9, J. Phys. Chem., Vol. 68, No. 1, 213–214, 1964. 60. Byrne, J.E., Battino, R., and Wilhelm, The solubility of gases in liquids. 8., J. Chem. Thermodynamics, Vol. 7, 515–522, 1975.

APPENDIX 10.A.1: IDEAL SOLUBILITY OF GASES IN LIQUIDS AND PUBLISHED CO2 SOLUBILITY DATA. Published CO2 gas solubility data in 103 liquid solvents were gathered from the available literature and is presented in Table 10.1. Two of these solvents, triethylamine and 1,4-dioxane were subsequently deleted from the data set based on their known tendency to chemically react with CO2.1

IDEAL SOLUBILITY OF GASES IN LIQUIDS Solutions that come close to approximating ideal solutions are those which are very dilute, or those where the molecular species are so nearly alike that a given molecule is subject to the same intermolecular forces (both attractive and repulsive) in the mixture as in its own pure phase. (In very dilute solutions, the intermolecular forces on a solute molecule may be quite different than in the pure solute phase, but the solute molecules are far enough apart that solute–solute interactions do not manifest themselves.) The concept of an ideal solution is often an appropriate approximation

7248_book.fm Page 200 Tuesday, April 24, 2007 9:19 AM

200

Hansen Solubility Parameters: A User’s Handbook

TABLE 10.A.1 Ideal Carbon Dioxide Solubility Calculated Using Raoult’s Law T (°C)

s PCO2 (atm)

0 15 20 25 30

34.40 50.19 56.60 63.50 71.12

ideal =1/P s xCO2 CO2

0.0291 0.0199 0.0177 0.0157 0.0141

Note: pCO2 = 1 atm.

for gases dissolved in liquids, as at modest pressures; most gases are only sparingly soluble in typical liquids.2 Thermodynamically, an ideal solution is defined as one in which the activity, a, equals the mole fraction, xi, over the entire composition range and over a nonzero range of temperature and pressure.3 xi = ai =

fi fi o

(10.A.1)

The activity of a substance gives an indication of how active a substance is relative to its standard state, as it provides a measure of the difference in chemical potential at the state of interest and that at the standard state.4 The term fugacity, f, was introduced by Lewis5 as a measure of thermodynamic escaping tendency and is equal to the effective gas pressure corrected for deviations from ideality. In Equation 10.A.1, fi is the fugacity of component i at partial pressure pi, and fio is the fugacity at the saturation pressure of i, Pis, at the solution temperature. Equation 10.A.1 is an empirical rule suggested by Lewis and Randall5 that assumes imperfect gas mixtures to behave as ideal mixtures. When deviations from the ideal gas law are small, generally at low pressures, the effect of pressure on the fugacity of component i is negligible, and the fugacity terms in Equation 10.A.1 approach the partial pressure and saturation pressure of i, respectively. In this situation, therefore, the ratio of the partial pressure and saturation pressure can now be used to express the mole fraction, xi. xi =

pi Pi s

(10.A.2)

Equation 10.A.2 is known as Raoult’s law, and the mole fraction, as calculated from Raoult’s law, is referred to as the ideal gas solubility. The ideal solubility calculated from Equation 10.A.2 usually gives correct order of magnitude results provided that Pis is not large and the solution temperature is well below the critical temperature of the solvent and not excessively above the critical temperature of the gaseous solute.4 Table 10.A.1 evaluates the ideal solubility of CO2, calculated using Equation 10.A.2, for the temperature range 0°C to 30°C. ideal From Table 10.A.1, Raoult’s law predicts an ideal CO2 solubility of xCO 2 = 0.0157 at T = 25°C, and this value has been used in several of the published CO2 solubility studies.2,4,60

7248_book.fm Page 201 Tuesday, April 24, 2007 9:19 AM

Determination of Hansen Solubility Parameter Values for Carbon Dioxide

201

TABLE 10.A.2 Ideal Carbon Dioxide Solubility Calculated Using Raoult’s Law and Fugacities T (°C)

s PCO2 (atm)

fCO2 PCO2 = 1 atm

0 15 20 25 30

34.4 50.19 56.6 63.5 71.12

.9928 .9941 .9944 .9947 .9950

o fCO2 at s PCO2

26.51 36.04 39.6 43.3 47.17

ideal xCO 2

ideal xCO 2

s = 1/PCO2

o = fCO2/fCO2

0.0291 0.0199 0.0177 0.0157 0.0141

0.0375 0.0276 0.0251 0.0229 0.0211

% Diff. 28 39 42 46 49

Note: pCO2 = 1 atmosphere.

It has been noted by Prausnitz et al.4 that the simplest way to reduce Equation 10.A.1 to a more useful form is to rewrite it in the manner suggested by Raoult’s law, Equation 10.A.2. In doing so, however, they caution that several assumptions are made, and errors in the use of this estimation technique for the solubility of gases in liquids can be significant, especially, when the saturation pressure of the gas is high. Therefore, in cases where the saturation pressure of the gas is above 1 atmosphere, it is necessary to consider the error in using pi /pis instead of fi /fi.0 Table 10.A.2 gives the saturation pressures of CO2 for the temperature range 0°C to 30°C, as well as the fugacities and calculated ideal solubilities using both methodologies. From Table 10.A.2 it appears that the assumption of Raoult’s law for the determination of ideal CO2 solubility in liquids results in significant error. The ideal CO2 solubility at 25°C and 1 ideal atmosphere partial pressure, as calculated from CO2 fugacities,7 is xCO 2 = 0.0229 compared with ideal a Raoult’s law prediction of xCO 2 = 0.0157. This value was also used by Gjaldbaek et al.8,9 in their work comparing experimental and calculated CO2 gas solubilities.

REFERENCES 1. Charles M. Hansen, communication. 2. Reid, R.C., Prausnitz, J.M., and Poling, B.E., The Properties of Gases and Liquids, 4th ed., McGraw Hill, New York, 1987. 3. Hildebrand, J.H., Prausnitz, J.M., and Scott, R.L., Regular and Related Solutions, Van Nostrand Reinhold Company, New York, 1970. 4. Prausnitz, J.M., Lichtenthaler, R.N., de Azevedo, E.G., Molecular Thermodynamics of Fluid-Phase Equilibria, 2nd ed., Prentice-Hall, Englewood Cliffs, New Jersey, 1986. 5. Lewis, G.N. and Randall, M., Thermodynamics and the Free Energy of Chemical Substances, McGrawHill, New York, 1923. 6. Pirig, Y.N., Polyuzhin, I.V., and Makitra, R.G., Carbon dioxide solubility, Russ. J. Appl. Chem., Vol. 66, No. 4, Part 2, 691–695, 1993. 7. King, M.B., Phase Equilibrium in Mixtures, Pergamon Press Ltd., Oxford, U.K., 1969. 8. Gjaldbaek, J.C., The solubility of carbon dioxide in perfluoro-n-heptane, normal heptane, cyclohexane, carbon tetrachloride, benzene, carbon disulphide and aqueous solution of aerosol, Acta Chem. Scand., Vol. 7, 537–544, 1953. 9. Gjaldbaek, J.C. and Anderson, E.K., The solubility of carbon dioxide, oxygen, carbon monoxide and nitrogen in polar solvents, Acta Chem. Scand., Vol. 8 1398–1413, 1954.

7248_book.fm Page 202 Tuesday, April 24, 2007 9:19 AM

7248_C011.fm Page 203 Thursday, May 10, 2007 8:00 AM

of Hansen Solubility 11 Use Parameters to Identify Cleaning Applications for “Designer” Solvents John Durkee ABSTRACT Choosing the most effective solvents for cleaning requires characterization of both solvents and soils. Hansen solubility parameters (HSP) can be used whether or not single or multiple components are present in either the soil composite or the solvent blend. These HSP characterizations of solvents and soils can be compared as values in HSP space just as similar values of solvents and polymers are compared to determine if a solvent can dissolve a polymer. A previously developed value of RA for cleaning operations allows prediction of useful cleaning performance by solvent on soil. In this chapter, six representative soils and 13 base solvents, so-called “designer” solvents, were selected as feedstocks for formulation of binary azeotropes, using a literature database. The selected solvent azeotropes were predicted to clean five or six representative soils. In this chapter, it will be shown that the base azeotropic feedstock must display significant polar and hydrogen bonding intermolecular forces if the azeotrope is to clean a soil composite that also contains significant polar and hydrogen bonding intermolecular forces.

INTRODUCTION Since the early 1990s, more than a dozen new solvents have been commercialized for cleaning1, such as vapor degreasing, and other applications. The sobriquet “designer” has been applied to them as their structure was designed to provide certain benefits. Chiefly, these benefits are a relatively nonhazardous exposure to humans and compliance with nearly all environmental regulations (including the Montreal Protocol) of most nations. As designer goods vs. commodities, they are high priced, which limits applications. Another limitation is their character as solvents. Given the relative inertness of these designer solvents to both humans and the environment, it is not surprising to find them to be inert to many soils. The purpose of this chapter is to illustrate how to use HSPs to aid in identifying valuable applications for these and other solvents in cleaning applications. The approach taken is the traditional one used in cleaning work: matching the solvent to the soil.2 HSP values of the solvents will be those matched to soil materials.3

203

7248_C011.fm Page 204 Thursday, May 10, 2007 8:00 AM

204

Hansen Solubility Parameters: A User’s Handbook

A VARIETY OF SOLVENTS There are almost too many types of cleaning solvents. It can be hard to discriminate among them. Simple chemicals such as hexane, more complex chemicals such as N-methyl-2-pyrrolidone, flammable chemicals such as acetone, carcinogenic chemicals such as benzene, ozone-depleting chemicals such as carbon tetrachloride, and low-cost common chemicals such as water have all been used as cleaning solvents. One of the strengths of solvent-cleaning technology is the variety of solvents that can be used in various solvent cleaning processes. Yet, no cleaning solvent is perfect. All have flaws, both general and specific. Solvents with general flaws4 may pose environmental, safety, or health hazards. They may be physically unsuited to the application because of mismatched physical properties such as surface tension, density, or viscosity; have excessive or limited volatility; or simply be too expensive. However, the specific flaw that is usually fatal to an application is the solvent’s not being compatible with the soil materials. Prevention of a mismatch of intermolecular forces between a cleaning solvent and a soil (an incompatibility) is a role uniquely fulfilled by HSP.5 A soil material is probably soluble in a solvent (or solvent blend) if the Hansen parameters for the solvent lie within the solubility sphere for the soil.6

PATHOLOGY OF SOILS Those managing cleaning work must know at least as much about soils as they do about cleaning agents, if their cleaning work is to meet their requirements. After all, the soil is the enemy to be conquered. Soils are multicomponent mixtures. They are composed of occasionally unknown ingredients prepared with unknown and variable proportions, which are not homogenized. They may include “tramp” materials or contain unexpected components that are not always present or recognized at best, often inadvertently produced by users. Soil can also be the outcome of incomplete cleaning work. Cleaning of multiple soil components usually produces multiple cleaning results. The better matched cleaning agents and processes are to a soil, the better the result, while poor matches will still be apparent. A good example is hydrocarbon-based wax, which must normally be melted in order to be removed from surfaces. Microcrystalline wax, used as a binder in polishing compounds, has a sharp melting point because the range of molecular weights of its component paraffins is narrow. Wax used in box-coating operations has a mixture of molecular weights and a broad melting point. If the boiling point of the cleaning solvent only matches the average melting point of both wax mixtures, the higher melting fractions of the coating wax probably will not be well-cleaned because they are not heated to a high enough temperature. In other words, the soil least matched to the cleaning solvent is likely to be the soil least cleaned. (See “More Realistic View about Evaluating HSP of Composite Soils.”)

HSP OF MULTIPLE-COMPONENT SOILS How then is a user to match a single (or azeotropic) cleaning solvent with a multicomponent soil? The answer combines observation along with conventional solubility theory and is seen in the following list: •

First, one observes7 if the soil is a single- or multiple-phase mixture. If the soil material is a single phase (presumably a solution and not an emulsion), the components must have similar intermolecular forces, and a useful solvent can probably be found from their study. If the mixture contains multiple phases, it is likely that a cleaning process must

7248_C011.fm Page 205 Thursday, May 10, 2007 8:00 AM

Polar HSP (soil or solvent)

Use of Hansen Solubility Parameters

205

#2 Soil Component #1 HSP Parameters for Common Soil or Solvent

H2 Bonding HSP (soil or solvent) FIGURE 11.1

•

•

be chosen that is not based on solvency but on application of mechanical force (aqueous cleaning, for example) or the chosen cleaning process must have multiple steps. Second, one estimates the HSP values for a composite of soil materials. Information about the individual soil components is used here. This usually involves some compromise. For example, if the soil contains multiple components, HSP values for the two that represent the greatest volume fraction should be used. This is illustrated in Figure 11.1.8 If that means other components are not represented in estimation of HSP values, then it is assumed that the observation of the soil form as a single phase would mean that the unrepresented components must be similar to those represented. Whereas it is perfectly possible (see next two sections) to compute the composite HSP using the entire composition, the recipe for that composition is only seldom known.9 Third, one chooses a solvent whose HSP values are similar. This, too, involves some compromise. The choice of a solvent (or blend) similar to a soil component (or composite) is accomplished in a similar manner as solvent for a polymer10 is chosen (see section on “Method for Choice of Suitable Solvents”). In other words, the required compromise solvent should be no further than the outer boundary of the largest circle (see discussion following).

METHOD FOR CALCULATING HSP OF COMPOSITES (SOILS OR SOLVENTS) Solubility parameters11 of mixtures are linear. That is, each of the three HSPs (disperse, polar, and hydrogen bonding) of a solvent mixture is a linear function of composition. In this case, the composition value to be used in calculating solubility parameters for solvent mixtures is the volume fraction (φ) for each component.12 For a binary (two-solvent) mixture, the equation for all four13 solubility parameters is Equation 11.1.14 This equation is correct for more than two components where the HSP14 values are known.15,16 δ blend ≡ ⎡⎣ϕ comp1 × σ com1 ⎤⎦ + ⎡⎣ϕ comp 2 × σ com 2 ⎤⎦

(11.1)

Traditionally, without specific data, it is normally assumed that there is no volume change upon mixing of solvents. That is:

7248_C011.fm Page 206 Thursday, May 10, 2007 8:00 AM

206

Hansen Solubility Parameters: A User’s Handbook

TABLE 11.1 Calculation of HSP for Mixtures via Equation 11.1 Component 1 2

Solvent

Mol Density δDispersion δPolar δHydrogen-Bonding Wt. g/cc MPa1/2 MPa1/2 MPa1/2

Cyclohexane 84.2 Isopropanol 60.1

0.779 0.789

16.8 15.8

0.0 6.1

0.2 16.4

Wt. Vol. δHydrogen% % ϕ δDispersion δPolar Bonding 67.0 67.2 33.0 133.8

16.5

⎛ Wt . Fraction ⎞ ⎜ ⎟ ⎝ Density ⎠1 (Vol. Fraction)1 = ⎛ Wt . Fraction ⎞ ⎛ Wt . Fraction ⎞ ⎟ ⎜ ⎟ +⎜ ⎝ Density ⎠1 ⎝ Density ⎠2

2.0

5.5

(11.2)

An azeotropic mixture of cyclohexane and isopropanol is used as an example of this calculation. The data needed and results are given in Table 11.1. Note, for this example only, that the volume and weight17 concentrations are essentially the same. This is because the individual solvent density values are essentially the same. Also note that the value of adding isopropanol (2-propanol) to cyclohexane is to add polar and, especially, hydrogen bonding intermolecular forces to the blend. Cyclohexane has essentially none. In this way, the solvency character of the azeotropic blend is very different from that of the neat cyclohexane. Equation 11.1 is useful for both solvents and soils. It is also useful for azeotropes, nearazeotropes, or nonazeotropic blends.

MORE REALISTIC VIEW ABOUT EVALUATING HSP OF COMPOSITE SOILS Judgment is also a factor required for evaluation of the HSP of composite soils. In some cases, the volumetric-proportional HSP values (disperse, polar, and hydrogen bonding) of the soil may not represent the actual nature of the cleaning problem. Calculated values of HSP for a composite soil (as are shown in Figure 11.1) may not lead to the right choice of cleaning solvents. This happens when a single component in the composite soil limits cleaning performance vs. the well-mixed composite being the limit of cleaning performance.18 Unless the soil is a single component, one component of the soil composite will always be less soluble with the cleaning solvent in comparison with the composite of soil components as a whole. And other components of the composites will also have different rates of solubility. The least compatible component may well be the one that should define the solvency of the cleaning solvent. This is illustrated, with component number 3, in Figure 11.2. In other words, cleaning (where all soil components must be equally well removed) is a situation where the volumetric average may not represent the true situation.19

METHOD FOR CHOICE OF SUITABLE SOLVENTS RA is the distance in HSP space between the soil and the solvent. That distance should be as small as possible (RA ≈ 0).20,21 The equation to be used for solvent selection,22 which defines RA, is:

7248_C011.fm Page 207 Thursday, May 10, 2007 8:00 AM

Polar HSP (soil or solvent)

Use of Hansen Solubility Parameters

207

#2 Soil Component #1

#3

HSP Parameters for Common Soil or Solvent

H2 Bonding HSP (soil or solvent) FIGURE 11.2

RA2 Solution = ( δ Polar for Solvent – δ Polar for Soil )

2

+ ( δ H 2 – Bonding for Solvent – δ H 2 – Bonding for Soil )

2

(11.3)

2 + ⎡ 4 × ( δ Disperse for Solvent – δ Disperse for Soil ) ⎤ ⎣ ⎦

The preceding equation comprises the following data: •

• •

The basic data for Equation 11.3 is the three previously estimated HSP values for either a soil composite (via Equation 11.1) or HSP values for a component expected to be limiting. The independent variable in the equation is the choice of solvent. The dependent variable in the equation shows how well the soil is dissolved, that is, how closely RA approaches zero.

Polar HSP (soil or solvent)

The individual values for HSPs for the soil are either those computed from Equation 11.1 or that for the single soil component judged to limit cleaning performance as shown in Figure 11.2. HSP data and a spreadsheet can aid in making this complex choice because they allow evaluation of solution performance for many solvents (or blends) against all components of the soil. Graphically, the matching of solvent (or blend of solvents) to soil (of one of more components) is described in Reference 3 as Figure 4.6 (reproduced here as Figure 11.323). If the distance between the loci

Soil or Solvent #1 HSP Parameters for Common Soil or Solvent Soil or Solvent #2

H2 Bonding HSP (soil or solvent) FIGURE 11.3

7248_C011.fm Page 208 Thursday, May 10, 2007 8:00 AM

208

Hansen Solubility Parameters: A User’s Handbook

of HSP values for solvent and soil is too great (RA is too large), there will be little graphical overlap between the two spheres, and the proposed solvent cleaning process would be expected to be an unsatisfactory choice. Where there is HSP data for soil components and solvents, an optimum choice of solvents (or solvent blends) can be made (see section on “Identification of ‘Designer’ Solvents.”). No parts need be found, wet, cleaned, weighed, or examined.

REFERENCE SOILS FOR COMPARISON To evaluate the suitability and limitations of designer solvents and their azeotropes in solvent cleaning operations, one must be able to position them relative to common soils. This situation is common to that found in marketing. The question is: Against which soils should these solvents (and blends) be positioned (evaluated)? The approach taken here is to choose six different single-component organic soils for reference. They are identified and described in Table 11.2. Each represents a soil type frequently encountered in industrial operations, and their collective HSP values cover a broad range of HSP space. The HSP values of each soil are plotted as in Figure 11.4. These reference soils serve both a specific and a general purpose. Specific azeotropes of designer solvents will be proposed to clean each soil. But in general, the method of developing successful solvent cleaning application is illustrated by matching solvents to soil components.

IDENTIFICATION OF DESIGNER SOLVENTS Choice of the cleaning solvents of the designer type is somewhat arbitrary. Four families of solvents are considered: Hydrofluoroethers (HFEs): HFE-7100, HFE-7200, and HFE-820024 are molecules uniquely containing both fluorine and oxygen, as well as three or five hydrogen atoms. Hydrocarbon ether solvents partially based on silicon (not on carbon): OS-10, OS-20, and OS-30 are molecules also containing ether linkages, but they are based on a silicon to oxygen bond rather than a carbon to oxygen bond. The hydrocarbon methyl group is plentiful. Solvents more useful as refrigerants which form azeotropes: These are HFC-245fa (based on partially fluorinated ethane), HFC-365mfc (based on partially fluorinated propane), and (HFC-4310mee), based on partially fluorinated propane, which does find applications as a nonazeotropic cleaning solvent. In addition, the fully fluorinated solvent based on butane (PFC-506025) is also included for comparison. Solvents with harmful environmental impacts but useable in developing countries: HCFC141b, HCFC-225 ca/cb, and CFC-113. All three have ozone depletion potentials (ODP) of significance, which has led to a ban on their production in industrialized countries per the Montreal Protocol. However, their production and use in undeveloped countries is permitted for several years beyond the publication date for this book. Their structural formulas, physical, and solubility properties are given in Table 11.3. The two-dimensional graphs26 of the latter are illustrated in Figure 11.5. Numerically, they total a “baker’s dozen.”

AN OPEN QUESTION — ANSWERED It is interesting to note the overlap of Figure 11.4 and Figure 11.5. This overlap would represent the similarity (see Endnote 8) of these designer solvents for use in cleaning typical industrial soils. There is observed to be little overlap.

7248_C011.fm Page 209 Thursday, May 10, 2007 8:00 AM

Use of Hansen Solubility Parameters

209

Table 11.2 Reference Soil Materials Soil

#

ASTM Fuel “A” 1

Butyl Stearate

27

Castor Oil

Ethyl Cinnamate

2

3

4

5

δ Dispersion δ Polar δ Hydrogen-Bonding Molecular Model Image MPa½ MPa½ MPa½

14.3

12.6

13.6

16.0

13.5

0

6.3

6.0

10.8

3.5

0

6

15.9

13.9

Industrial Use

CH3

Isooctane (2,2,4-trimethylpentane, or isobutyltrimethylpentane) is a component of gasoline (Octane Rating 100) and representative of the general class of hydrocarbon chemicals.

6.1

CH3, Ester

10.5

CH3, Acid Double Bond Hydroxyl

7.5

Aromatic, Ester, Double Bond

3.7

CH3, Esters (3), Double Bonds (6)

Linseed Oil

Tricresyl Phosphate

Contained Functional Groups

13.5

Phosphate, Aromatic

Representative of a class of soils as fatty acid esters, which are used as used as raw material of emulsifiers or oiling agents for foods; spin finishes and textiles; lubricants for plastics, paint and ink additives; surfactants and base materials for perfumery; solvents or cosolvents; and oil carriers in agricultural operations. Castor oil is a vegetable oil. It and its derivatives have applications in the manufacturing of soaps, lubricants, hydraulic and brake fluids, paints, dyes, coatings, inks, plastics, waxes and polishes, pharmaceuticals and perfumes. (Ethyl Phenyl Acrylate) is a natural chemical and a common fragrance and flavoring component. It is the ester primarily responsible for the smell of cinnamon.

(Triglyceride Ester of mainly Linoleic Acid28), a natural vegetable oil, is the most important drying oil of the oil painting industry.

This material is found in common metalworking fluids: anti-wear agents or lubricants. It is also used as a plasticizer (PVC and alkyd resins), as a detergent, and as flame retardant.

Other than cleaning hydrocarbon-based soils that do not have substantial polarity or hydrogen bonding character, these designer solvents are no more functionally useful than a simple solvent such as heptane (which costs perhaps one twentieth as much per kg). Questions asked during the last decade by those who practice cleaning science are: What should these designer solvents be used for?, why should manufacturers charge exorbitant prices?, and “how can we gain the value of the United States volatile organic compound (USVOC) exemption and negligible ODP promised by their manufacturers, when these designer solvents appear useless to remove typical industrial soils?”

7248_C011.fm Page 210 Thursday, May 10, 2007 8:00 AM

210

Hansen Solubility Parameters: A User’s Handbook

HSP OF REPRESENTATIVE SOILS

HSP OF SINGLE COMPONENT SOLVENTS

Chosen to Cover a Broad Range of Value

13 ‘Designer’ Solvents 14 Hydrogen Bonding HSP, MPaˆ (½)

Hydrogen Bonding HSP, MPaˆ (½)

14 12 10 8 6 4 2 0

12 10 8 6 4 2 0

0

2

FIGURE 11.4

4 6 8 10 Polar HSP, MPaˆ (½)

12

14

0

2

4 6 8 Polar HSP, MPaˆ (½)

10

12

14

FIGURE 11.5

The approach to answer these questions without commercial bias will be two fold: 1. To extend the power of designer solvents by using them not in neat form (as a comparison of Figure 11.4 and Figure 11.5 shows that approach to be fruitless) but in combination with secondary components in binary azeotropes identified by the manufacturers of these solvents. In other words, the type of cleaning29 solvent evaluated in this study is a binary azeotrope with one component being a designer solvent and the other being a more common (and low-priced) chemical. Naturally, this formulation of solvent blends (reduces) purchase price and blends (dilutes) environmental benefits. 2. To use existing solvent data of physical and chemical properties as well as HSP values to match one of the six common industrial soils to a binary azeotrope one of whose components as a designer solvent. In all cases, the cleaning process contemplated is the conventional vapor degreasing.

LIMITING RA VALUE FOR EXPECTED GOOD CLEANING PERFORMANCE A good cleaning performance that is attained by limiting RA requires the following: • • •

A method for calculation of HSP values of blends that are identified (See Equation 11.1); Target soils to be identified (See Figure 11.4 for HSP values); and A library of azeotropic formulations identified (See Reference 16), including their blend HSP values calculated through literature component values from Reference 15 and Equation 11.1. The questions remain as to what standard of cleaning performance speaks to how well the soil is dissolved and how closely does RA approach zero?

The answer is in a study also reported in Reference 16. It is that RA is related to the amount of soil not removed (uncleaned) from the parts being cleaned. Cleaning trials30 have been and are being conducted by the Surface Cleaning Laboratory within the Toxic Use Reduction Institute (TURI), located at the University of Massachusetts–Lowell. The work is aimed at providing practical, useful, and environmentally sound solutions to businesses within the New England region of the U.S.

7248_C011.fm Page 211 Thursday, May 10, 2007 8:00 AM

Use of Hansen Solubility Parameters

211

Table 11.3 Properties of “Designer” Solvents Solvent

Structure

Specific Mol. Gravity, Boiling δ Dispersion δ Polar δ Hydrogen-Bonding δ Overall Wt. g/cc Point, °C Values in Mp½

Cleaning Solvents Designed for Minimal Environmental Impact

HFE-7100

250.1 1.520

60.0

13.7

2.2

1.0

13.9

HFE-7200

264.1 1.430

76.1

13.3

2.0

1.0

13.5

HFE-8200

250.1 1.43

76

13.0

4.0

1.0

13.6

HFC-43-10 mee

252.1 1.580

54.4

11.6

0.0

0.0

11.6

OS-10 (OS-10 Hexamethyldisiloxane)

162.3 0.764

100.6

12.4

0.0

0.0

12.4

OS-20 (OS-20 Octamethyldisiloxane)

236.5 0.820

152.8

11.7

0.0

0.0

11.7

OS-30 (Decamethyltrisiloxane)

310.7 0.854

193.9

12.2

0.0

0.0

12.2

Solvents More Useful as Refrigerants, But Which Form Azeotropes

HFC-245 fa

134.0 1.320

15.3

15.7

0.0

0.0

15.7

HFC-365 mfc

148.1 1.270

40.2

16.4

0.0

0.0

16.4

PFC-5060

288.0 1.680

55

12.1

0.0

0.0

12.1

Solvents with Harmful Environmental Impacts But Usable in Developing Countries

HCFC-141B

116.9 1.230

32.2

15.7

4.0

1.0

16.2

HCFC-225 ca / cb

202.9 1.550

53.9

14.1

3.2

1.0

14.5

CFC-113

187.4 1.560

47.8

14.7

1.6

0.0

14.8

7248_C011.fm Page 212 Thursday, May 10, 2007 8:00 AM

212

Hansen Solubility Parameters: A User’s Handbook

A group of more than 50 tests with various solvents (and blends) in vapor degreasing operations were examined by this author.16 It was found that the “bright line” between31 acceptable and not acceptable cleaning performance is represented by an RA value of about 8. This means that: If the RA calculated for soil components and a single solvent or azeotrope is less than 8, the solvent cleaning operation has a good chance to be successful. If the RA calculated exceeds 8, the solvent cleaning operation does not have a good chance to be successful. Another solvent or blend, or a process different from vapor degreasing should be considered. The limiting value of 8 will be used for RA in this study.

APPLICATION OF HSP METHODOLOGY TO CLEANING OPERATIONS Figure 11.6 and Figure 11.7, in which the cleanliness performance data are exhibited, should be considered as the two-dimensional spherical solubility plots in Reference 3. Data in Figure 11.7 are similar to that in Figure 11.6. Only the ranges have been changed. These solution cleaning results are consistent with and similar to the teachings of Reference 3, which are about solution of polymers or other materials within solvents (or the reverse). The results are enlisted as shown in the following: 1. Quality of cleaning is related to the distance in HSP space (RA) between solvent and the soil least compatible with the solvent. This means less soil is left on parts in actual cleaning tests when RA is smallest. The basis for Figure 11.6 and Figure 11.7 makes 100% ▲ ▲

% Soiled

80%

▲

60% 40%

▲▲ ▲ ▲ ▲ ▲

▲

▲

▲ ▲ ▲ ▲

20% ▲ ▲▲ ▲ ▲ ▲ ▲▲ ▲ ▲ ▲▲ ▲ ▲ ▲ ▲▲

▲▲▲▲ ▲

0% 0

5

10

▲

15 MAX RA

20

25

30

% Soiled

FIGURE 11.6 10% 9% 8% 7% 6% 5% 4% 3% 2% 1% 0%

▲ ▲

▲ ▲

▲

▲

▲

▲ ▲ ▲

▲ ▲ ▲ ▲

▲▲ ▲ ▲ ▲ ▲ ▲▲ ▲

▲

▲ ▲ ▲ ▲

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 MAX RA

FIGURE 11.7

7248_C011.fm Page 213 Thursday, May 10, 2007 8:00 AM

Use of Hansen Solubility Parameters

213

physical sense; the material not cleaned (percentage soiled) by a solvent is related to the soil least soluble. 2. There is a limit (breakpoint, “bright line,” demarcation, or separation) between one type of behavior (acceptable cleanliness) and the opposite (unacceptable cleanliness). There are very few or negligible anomalies (situations where HSP values suggest that there should be poor compatibility, and there is good compatibility, as well as the reverse). There are no results of poor cleaning when RA is 8 or below. 3. The value of RA (8), chosen to differentiate acceptable and unacceptable cleanliness, is similar to the interaction radius between polymers and solvents typical for selected correlations given in Appendix Table A.2 of Reference 3 and Appendix Table A.2 of this Handbook. The average value in Appendix Table A.2 is about 10. Results of calculation of the HSP values for each binary32 azeotrope blend based on the solvents in Table 11.3 and in Table 11.433 through Table 11.15 are given and illustrated in Figure 11.8 through Figure 11.20.

ANALYSIS OF CAPABILITY OF DESIGNER SOLVENTS The limited utility34 of neat designer solvents has been experienced in the decade since they were commercialized and is characterized via HSP methodology in Figure 11.5. As viewed within the HSP space, the value of the components from which these azeotropes are formed is to add hydrogen bonding and polar intermolecular forces to designer solvents that generally do not display those forces. This is illustrated schematically in Figure 11.21, and is essentially the approach described in the section on “An Open Question — Answered,” to find overlap between Figure 11.4 and Figure 11.5. Each azeotropic blend noted above35 is a fluid with a single boiling point, but with different physical properties, safety, health, environmental impacts, and economic consequences than are manifested by the 13 chosen designer solvents. The proper metric36 for discrimination among these azeotropes is the effective completion of the proposed cleaning work. That metric is the magnitude of the mismatch between intermolecular forces of the solvent (single or azeotropic blend) and composite soil. Quantization of this mismatch is produced for specific soils by the parameter RA. As described in the section on limiting RA value for good cleaning performance (at least 95% removal of all soils) is around 8. The capability to clean various soils with azeotropes of designer solvents is given in Table 11.16. The information is sorted to meet the needs of users by giving the various designer solvents that can be blended to clean the stated soil. Users can use HSP to differentiate one solvent or solvent blend from another based on the values of the soil found in the specific application. Information in the preceding tables can be used to screen azeotropes formulated from designer solvents and reduce the needed burden of feasibility testing. Reference 3 or Reference 15, as well as the appendices in this book, can be used as sources of HSP values when the six soils chosen to be representative in this study are not representative of those in the current application. This analysis, using HSP methodology, clearly shows the superiority of the fluoroether structure (HFE) as a feedstock for azeotropes. Four to five units (MPa1/2) of polar HSP value can be added through addition of secondary solvents to form constant-boiling37 azeotropes. HSP methodology further allows recognition of the limits of azeotropic blends based on designer solvents. No composition is expected to be capable of cleaning soil number 6 (tricresyl phosphate). This is because the amount of polar and hydrogen bonding intermolecular forces needed to produce a compatible solution cannot be obtained by adding a second solvent rich in those forces to designer solvents that are nearly devoid of them.

7248_C011.fm Page 214 Thursday, May 10, 2007 8:00 AM

214

Hansen Solubility Parameters: A User’s Handbook

TABLE 11.4 Azeotropes with HFE-7100 Primary Azeo Component HFE-7100

Type of Solvent

Wt (%) HFE-7100 100%

BP (ºC) 60.0

SpG (g/cc) 1.520

Isobutyl alcohol

Azeotrope

99.0%

58.0

1.507

2-Butanol

Azeotrope

98.4%

58.0

1.499

1-Propanol (nPA)

Azeotrope

97.9%

56.0

1.492

t-Butyl alcohol

Azeotrope

94.0%

56.0

1.438

1,2-Dichloropropane

Azeotrope

94.8%

59.0

1.496

2-Propanol (IPA) 1-Chlorobutane

Azeotrope Azeotrope

95.1% 87.1%

54.0 57.0

1.453 1.390

HCFC-225 ca/cb

Azeotrope

28.5%

53.0

1.541

Ethanol

Azeotrope

93.0%

52.0

1.428

Methanol

Azeotrope

93.0%

46.0

1.427

1,2-Dichloroethylene (CIS) n-Propyl bromide Ethyl formate 1,2-D ichloroethylene (TRANS) Methyl acetate Methyl formate 2-Chloropropane

Azeotrope

65.7%

55.0

1.430

Azeotrope Azeotrope Azeotrope

25.8% 64.0% 44.1%

54.0 31.2 40.0

1.376 1.232 1.361

Azeotrope Azeotrope NearAzeotrope

39.1% 31.4% 22.0%

52.6 50.2 35.0

1.100 1.103 0.950

HSP Reference WO 96/36689, U.S.P. 6,426,327, U.S.P. 6,008,179 WO 96/36689, U.S.P. 6,426,327, U.S.P. 6,008,179 WO 96/36689, U.S.P. 6,426,327, U.S.P. 6,008,179 WO 96/36689, U.S.P. 6,426,327 WO 96/36689, U.S.P. 6,426,327 U.S.P. 6,426,327 WO 96/36689, U.S.P. 6,426,327 WO 96/36689, U.S.P. 6,426,327 WO 96/36689, U.S.P. 6,426,327 WO 96/36689, U.S.P. 6,426,327 WO 96/36689, U.S.P.6,426,327 U.S.P. 6,689,734 U.S.P. 6,753,304 U.S.P. 6,426,327 U.S.P. 6,753,304 U.S.P. 6,753,304 U.S.P. 6,646,020, U.S.P. 6,426,327

δD 13.7

δP 2.2

δH 1.0

Soils Cleaned RA < 7.93

13.7

2.3

1.3

1, 2, 5

13.8

2.3

1.4

1, 2, 5

13.8

2.4

1.6

1, 2, 5

13.9

2.5

2.5

1, 2, 5

13.9

2.5

1.1

1, 2, 5

13.9 14.2

2.6 2.9

2.4 1.2

1, 2, 5 1, 2, 5

14.0

2.9

1.0

1, 2, 5

14.0

3.0

3.3

1, 2, 3, 5

13.9

3.5

3.7

1, 2, 3, 5

15.0

4.4

1.8

1, 2, 5

15.5 14.6 15.7

5.0 5.2 5.7

3.5 4.6 2.3

1, 2, 4, 5 1, 2, 3, 4, 5 1, 2, 4, 5

15.0 14.9 15.3

5.8 7.0 7.5

5.7 8.1 1.9

2, 3, 4, 5 2, 3, 4, 5 2, 4, 5

Figure 11.8 through Figure 11.20 are cluttered, rich in information, and capable of effective communication. At a single glance, one can estimate: In Figure 11.8, that no azeotrope of HFE-7100 can clean soils based on castor oil or tricresyl phosphate. In Figure 11.18, that HCFC-141b is a likely candidate to be a feedstock from which one or more azeotropes can be formulated to clean soils based on castor oil. In Figure 11.15 and Figure 11.16, that HFC-245fa and HFC-365mfc, respectively, are not likely candidates from which azeotropes can be formulated to clean any but single hydrocarbon soils. In Figure 11.12, that OS-10 is a very useful feedstock from which one or more azeotropes can be formulated to clean a variety of soils.

7248_C011.fm Page 215 Thursday, May 10, 2007 8:00 AM

Use of Hansen Solubility Parameters

215

TABLE 11.5 Azeotropes with HFE-7200 Primary Azeo Component HFE-7200

Type Solvent

wt % HFE-7200 100%

BP, °C 76.1

SpG, g/cc 1.430

Ethanol

Azeotrope

88.0%

62.0

1.417

1,2-Dichloropropane

Azeotrope

87.0%

73.0

1.388

1-Chlorobutane

Near Azeo

87.1%

69.0

1.293

1-Butyl Alcohol

Azeotrope

84.0%

67.0

1.262

n-Propyl Bromide

Azeotrope

56.0%

63.0

1.385

Methanol

Azeotrope

84.0%

53.0

1.266

1,2-Dichloroethylene (TRANS)

Azeotrope

68.0%

48.0

1.307

HSP Reference WO 96/36688 & USP 6,288,018 WO 96/36688 & USP 6,288,018 WO 96/36688 & USP 6,288,018 WO 96/36688 & USP 6,288,018 WO 96/36688 & USP 6,288,018 WO 96/36688 & USP 6,288,018 USP 6,699,829

δD 13.3

δP 2.0

δH 1.0

Soils Cleaned RA < 7.93

13.8

2.8

1.6

1, 2, 5

13.9

2.8

1.3

1, 2,

13.8

2.8

4.0

1, 2, 3

13.8

2.8

4.5

1,. 2, 3

14.5

3.7

2.5

1, 2,

13.8

4.6

6.5

2, 3, 4

15.9

6.2

2.6

1, 2, 4

It would be difficult to imagine how another technical approach could allow efficient identification of solvent capability for industrial cleaning from a table of azeotropes in a handbook.

CONCLUSIONS The developments described in this chapter are original and support the following conclusions: 1. Soils are chemicals. They can be described by the component chemicals of which they are comprised. As chemicals, their intermolecular forces can be characterized by HSPs. And as a mixture of chemical components, their overall HSP value can be computed through a conventional volumetric blend rule. This outcome allows soils to be characterized in a quantitative manner. In this chapter, six soil materials representative of common industrial soils were so characterized. 2. The same approach can be followed for a solvent and multiple component solvent blends with the same outcome — that is, characterization in a quantitative manner. In this chapter, thirteen “designer” solvents were so characterized. These solvents are notable for being similar to simple hydrocarbons with low-polar and hydrogen bonding forces and being considered to foster limited concerns about environmental, safety, or health hazards. 3. In a previous publication (Reference 16), it has been shown that the effectiveness of solvent cleaning systems (vapor degreasers) can be predicted by the similarity of intermolecular forces between soil composites and solvent blends. Similarity means the distance in HSP space between the soil and the solvent materials. The quantitative term is RA, and good cleaning is observed when RA is about 8 or less. 4. In the same previous publication (Reference 16), two-component azeotropes of these designer solvents were identified. 5. In this chapter, the blend HSP values for solvent azeotropes were compared to those of soil composites with the standard for comparison being that the RA value between them be less than 8. These comparisons were tabulated and plotted.

7248_C011.fm Page 216 Thursday, May 10, 2007 8:00 AM

216

Hansen Solubility Parameters: A User’s Handbook

TABLE 11.6 Azeotropes with HFE-8200 Primary Azeo Component HFE-8200

Type of Solvent

Wt (%) HFE-8200 100%

BP (ºC) 76.0

SpG (g/cc) 1.430

HCFC 225 ca/cb

Azeotrope

30.6%

53.0

1.511

Isobutyl alcohol

Azeotrope

99.0%

58.0

1.419

2-Butanol

Azeotrope

98.8%

58.0

1.417

t-Butyl alcohol

Azeotrope

94.0%

56.0

1.362

1-Propanol (nPA)

Azeotrope

97.4%

56.0

1.402

1,2-Dichloropropane

Azeotrope

94.5%

59.0

1.412

2-Propanol (IPA)

Azeotrope

93.3%

55.0

1.356

1-Chlorobutane

Azeotrope

87.8%

57.0

1.329

Ethanol

Azeotrope

95.2%

52.0

1.376

Ethyl formate

Azeotrope

79.9%

31.3

1.287

n-Propyl bromide

Azeotrope

25.8%

54.0

1.356

Methanol

Azeotrope

89.6%

46.0

1.319

1,2-Dichloroethylene (CIS) Methyl acetate

Azeotrope

65.7%

55.0

1.376

Azeotrope

44.3%

52.7

1.104

Azeotrope

50.0%

40.0

1.338

Azeotrope

40.1%

50.4

1.122

NearAzeotrope

22.0%

35.0

0.942

1,2-Dichloroethylene (TRANS) Methyl formate

2-Chloropropane

HSP Reference U.S.P. 6,426,327 U.S.P. 6,426,327 U.S.P. 6,426,327 U.S.P. 6,426,327 U.S.P. 6,426,327 U.S.P. 6,426,327 U.S.P. 6,426,327 U.S.P. 6,426,327 U.S.P. 6,426,327 U.S.P. 6,753,304, U.S.P. 6,426,327 U.S.P. 6,426,327 U.S.P. 6,426,327 U.S.P. 6,426,327 U.S.P. 6,753,304, U.S.P. 6,426,327 U.S.P. 6,426,327 U.S.P. 6,753,304, U.S.P. 6,426,327 U.S.P. 6,646,020, U.S.P. 6,426,327

δD 13.0

δP 4.0

δH 1.0

Soils Cleaned RA < 7.93

13.7

3.5

1.0

1, 2, 5

13.0

4.0

1.3

1, 2, 5

13.1

4.0

1.3

1, 2, 5

13.2

4.1

2.4

1, 2, 5

13.1

4.1

1.7

1, 2, 5

13.3

4.2

1.1

1, 2, 5

13.3

4.2

2.8

1, 2, 3, 5

13.6

4.3

1.2

1, 2, 5

13.2

4.4

2.5

1, 2, 5

13.7

5.2

3.1

1, 2, 3, 5

15.3

5.4

3.4

1, 2, 3, 4, 5

13.4

5.4

4.7

1, 2, 3, 5

14.5

5.5

1.8

1, 2, 5

14.6

6.1

5.3

2, 3, 4, 5

15.1

6.1

2.2

1, 2, 4, 5

14.6

7.0

7.3

2, 3, 4, 5

15.1

7.8

1.9

2, 4, 5

7248_C011.fm Page 217 Thursday, May 10, 2007 8:00 AM

Use of Hansen Solubility Parameters

217

TABLE 11.7 Azeotropes with HFC-4310mee Primary Azeo Component HFC-4310mee

Type of Solvent

Wt (%) HFC-4310mee 100%

BP (ºC) 54.4

SpG (g/cc) 1.580

2-Propanol (IPA) Ethanol Methanol n-Propyl bromide

Azeotrope Azeotrope Azeotrope Azeotrope

97.4% 97.1% 95.3% 23.0%

45.5 43.4 39.9 52.0

1.540 1.535 1.509 1.515

1,1-Dichloroethane

Azeotrope

73.0%

43.0

1.445

Acetone

Azeotrope

83.9%

57.6

1.361

1,2-Dichloroethylene (CIS) 1,2-Dichloroethylene (TRANS)

Azeotrope

67.9%

42.3

1.471

Azeotrope

61.7%

39.0

1.443

HSP Reference Vertrel XP Vertrel XE Vertrel XM U.S.P. 6,689,734 U.S.P. 5196,137 U.S.P. 5824,634 U.S.P. 5196,137 Vertrel MCA

δD 11.6

δP 0.0

δH 0.0

Soils Cleaned RA < 7.93

11.8 11.8 11.9 12.8

0.3 0.5 1.1 1.5

0.8 1.1 2.0 1.1

1, 5 1, 2, 5 1, 2, 5 1, 2, 5

13.2

2.6

1.0

1, 2, 5

12.7

2.9

1.9

1, 2, 5

13.6

2.9

1.2

1, 2, 5

14.0

3.5

1.4

1, 2, 5

TABLE 11.8 Azeotropes with OS-10 Primary Azeo Component OS-10 Hexamethyld isiloxane

Type of Solvent

Wt (%) OS-10 Hexamethyld isiloxane 100%

BP (ºC) 100.6

SpG (g/cc) 0.764

HSP

Sec-butyl acetate

Azeotrope

96.0%

100.5

0.768

2-Pentanol

Azeotrope

87.0%

97.8

0.770

Propylene glycol methyl ether n-Propyl acetate

Azeotrope

82.0%

95.7

0.788

Azeotrope

61.0%

96.7

0.808

2-Propanol (IPA)

Azeotrope

54.3%

76.4

0.774

Ethanol

Azeotrope

63.8%

71.4

0.773

Methanol

Azeotrope

59.1%

58.7

0.775

Reference U.S.P. 5834,416 U.S.P. 5478,493 U.S.P. 5478,493 U.S.P. 5834,416 U.S.P. 5773,403 U.S.P. 5773,403 U.S.P. 5773,403

δD 12.4

δP 0.0

δH 0.0

Soils Cleaned RA < 7.93

12.5

0.1

0.3

1, 5

12.8

0.8

1.6

1, 2, 5

12.9

1.0

2.4

1, 2, 5

13.4

1.5

2.7

1, 2, 5

13.9

2.7

7.4

1, 2, 3, 5

13.6

3.1

6.9

1, 2, 3, 5

13.5

4.9

8.9

2, 3, 4, 5

6. The comparison showed that expected cleaning performance of designer solvents can be enhanced by combining it with other solvents into binary azeotropes. That enhancement is a gain of about four or five units (MPa1/2) of total (Hildebrand) solubility parameter, which allows prediction of effective cleaning of several common industrial soils. However, soils that display high levels of polar and hydrogen bonding forces, such as tricresyl phosphate, are not expected to be removed in a solvent cleaning process employing azeotropes of these designer solvents.

7248_C011.fm Page 218 Thursday, May 10, 2007 8:00 AM

218

Hansen Solubility Parameters: A User’s Handbook

TABLE 11.9 Azeotropes with OS-20

Type of Solvent

Wt (%) OS-20 Octamethyltrisiloxane 100%

BP (ºC) 152.8

SpG (g/cc) 0.820

Propylene glycol n-butyl ether

Azeotrope

89.0%

151.8

0.826

Propylene glycol n-propyl ether

Azeotrope

60.0%

141.8

0.844

Ethyl lactate

Azeotrope

63.0%

139.4

0.888

Primary Azeo Component OS-20 Octamethyltrisiloxane

HSP

Reference U.S.P. 5454,970, 5628,833 U.S.P. 5516, U.S.P. 5628,833 U.S.P. 5454, U.S.P. 5628,833

δD 11.7

δP 0.0

δH 0.0

Soils Cleaned RA < 7.93

12.1

0.4

1.3

1, 2, 5

13.2

1.9

5.1

1, 2, 3, 5

13.1

2.4

4.0

1, 2, 3, 5

TABLE 11.10 Azeotropes with OS-30 Primary Azeo Component OS-30 Decamethyltrisiloxane Dipropylene glycol n-propyl ether Dipropylene glycol methyl ether acetate Dipropylene glycol methyl ether (DPM, DPGME) Propylene glycol monobutyl ether (PGBE, PnB)

Type of Solvent

Wt (%) OS-30 DecamethylBP SpG trisiloxane (ºC) (g/cc) 100% 193.9 0.854

Azeotrope

91.0%

186.7

Azeotrope

89.0%

193.8

Azeotrope

61.0%

170.4

Azeotrope

15.0%

170.6

HSP Soils Cleaned RA < 7.93

δD 12.2

δP 0.0

δH 0.0

0.859 U.S.P. 5824,632 0.866 U.S.P. 5824,632 0.889 U.S.P. 5824,632

12.4

0.4

1.0 1, 2, 5

12.5

0.5

0.8 1, 2, 5

13.4

2.1

4.1 1, 2, 3, 5

0.874 U.S.P. “5824,632

14.8

3.8

7.8 2, 3, 4, 5

Reference

The methodology used in this chapter can be easily employed with other soil composites and solvent blends. This is a crucial point because additional solvents other than the designer solvents can be used as a feedstock to formulate azeotropes. Such additional solvents will not be as limited with regards to polar and hydrogen bonding intermolecular forces as are the designer solvents chosen for this study. It appears possible16 to identify an azeotropic solvent pair that is capable of being expected to clean any proposed soil composite.

7248_C011.fm Page 219 Thursday, May 10, 2007 8:00 AM

Use of Hansen Solubility Parameters

219

TABLE 11.11 Azeotropes with HFC-245fa Primary Azeo Component HFC-245fa Trichloroethylene (TCE) HCFC-123

Type of Solvent

Wt (%) HFC-245fa 100%

BP (ºC) 15.3

SpG (g/cc) 1.320

Near-Azeotrope

98.1%

14.8

1.322

Near-Azeotrope

98.0%

15.0

1.323

Methylene chloride 2-Propanol (IPA)

Near-Azeotrope

98.3%

15.2

1.320

Near-Azeotrope

96.8%

14.5

1.291

1-Propanol (nPA)

N ear Azeo

96.8%

14.4

1.293

Methanol

Near-Azeotrope

97.0%

14.2

1.294

Ethanol

Near-Azeotrope

95.8%

14.4

1.283

1,2Near-Azeotrope Dichloroethylene (TRANS) Water Near-Azeotrope

37.5%

—

1.295

83.0%

7.0

1.251

HSP Reference U.S.P. 6,100,229 U.S.P. 6,362,153 U.S.P. 6,100,229 U.S.P. 5683,974 U.S.P. 5683,974 U.S.P. 5683,974 U.S.P. 5683,974 U.S.P. 5851,977 U.S.P. 6,514,928

δD 15.7

δP 0.0

δH 0.0

Soils Cleaned RA < 7.93

15.7

0.1

0.1

1, 5

15.7

0.1

0.0

1, 5

15.7

0.1

0.1

1, 5

15.7

0.3

0.9

1, 5

15.7

0.4

0.9

1, 5

15.7

0.6

1.1

1, 5

15.7

0.6

1.3

1, 5

16.2

3.1

1.2

1, 5

15.7

3.4

0.9

1, 5

TABLE 11.12 Azeotropes with HFC-365mfc Primary Azeo Component HFC-365mfc

HSP Type of Solvent

Wt (%) HFC-365mfc

BP, (ºC)

SpG, (g/cc)

Ethanol

Azeotrope

98.4%

39.2

1.258

Water

Azeotrope

98.0%

38.0

1.263

Methanol

Azeotrope

96.2%

—

1.241

HCF2OCF2OCF2H

NearAzeotrope Azeotrope

60.1%

32.6

1.450

46.0%

33.0

1.009

2-Chloropropane

Reference U.S.P. 5445757 U.S.P. 6,365566 U.S.P. 6,743,765 U.S.P. 6,255273 U.S.P. 6,646,020

Soils Cleaned RA < 7.93

δD 16.4

δP 0.0

δH 0.0

16.4

0.2

0.5

1, 5

16.4

0.4

0.

1, 5

16.3

0.7

1.3

1, 5

13.0

3.8

2.5

1, 2, 5

16.1

5.3

1.3

1, 5

7248_C011.fm Page 220 Thursday, May 10, 2007 8:00 AM

220

Hansen Solubility Parameters: A User’s Handbook

TABLE 11.13 Azeotropes with PFC-5060 Primary Azeo Component Perfluorohexane (PFC 5060) HCFC-141b HCFC-123

HSP

Type of Solvent

wt % Perfluorohexane (PFC 5060) 100%

BP, °C 56.0

SpG, g/cc 1.680

Azeotrope Azeotrope

50.0% 50.0%

26.0 26.5

1.420 1.561

Reference

δD 12.1

δP 0.0

δH 0.0

Soils Cleaned RA < 7.93

USP 5560,861 USP 5560,861

14.2 13.7

2.3 2.9

0.6 0.5

1, 2, 5 1, 2, 5

TABLE 11.14A Azeotropes with HCFC-141b HSP Primary Azeo Component HCFC-141b

Type of Solvent

wt % HCFC-141b 100%

BP, °C 32.1

SpG, g/cc 1.230

Reference

δD 15.7

δP 4.0

δH 1.0

Soils Cleaned RA < 7.93

Azeotrope

50.0%

20.0

1.414

USP 5494,601

13.8

2.3

0.6

1, 2, 5

Azeotrope

50.0%

26.0

1.420

USP 5560,861

14.2

2.3

0.6

1, 2, 5

Perfluoropentane (PFC 5050) Perfluorohexane (PFC 5050) Cyclopentane 1,2-Dichloroethylene (TRANS) 2-Chloropropane Ethanol Methanol

Near Azeo Near Azeo

98.5% 98.1%

32.2 32.2

1.218 1.230

USP 5085798 USP 5126,067

15.7 15.7

3.9 4.1

1.0 1.0

1, 5 1, 5

Near Azeo Near Azeo Blend

98.5% 98.0% 95.6%

32.2 31.9 30.0

1.222 1.216 1.201

15.7 15.7 15.7

4.1 4.1 4.6

1.0 1.6 2.4

1, 5 1, 5 1, 2, 5

HCFC-123 Methylene Chloride

Near Azeo Blend

50.0% 70.0%

31.5 31.0

1.334 1.257

15.4 16.4

4.6 4.7

1.0 2.5

1, 2, 5 1, 4, 5

2-Propanol (IPA)

Near Azeo

50.0%

31.5

0.959

USP 5085797 USP 4,836,947 PROMOSOL 141b MS USP 5194,169 PROMOSOL 141b MC USP 5318,716

15.8

5.3

10.4

2, 3, 4, 5

TABLE 11.14B Azeotropes with HCFC-225 ca/cb

Type of Solvent

Wt (%) HCFC-225 ca/cb 100%

BP (ºC) 54.0

SpG (g/cc) 1.550

HFE-7100

Azeotrope

28.5%

53.0

1.541

HFE-8200

Azeotrope

30.6%

53.0

1.511

Primary Azeo Component HCFC-225 ca/cb

HSP

Reference WO 96/36689, U.S.P. 6,426,327 U.S.P. 6,426,327

δD 14.1

δP 3.2

δH 1.0

Soils Cleaned RA < 7.93

14.0

2.9

1.0

1, 2, 5

13.7

3.5

1.0

1, 2, 5

7248_C011.fm Page 221 Thursday, May 10, 2007 8:00 AM

Use of Hansen Solubility Parameters

221

TABLE 11.15 Azeotropes with CFC-113 Primary Azeo Component 1,1,2-Trichlorotrifluoroethane (CFC-113) Nitromethane

Wt (%) 1,1,2-Trichlorotrifluoroethane BP (CFC 113) (ºC) 100% 47.6

Type of Solvent

HSP SpG (g/cc) 1.560

Azeotrope

97.1%

46.8

1.544

Methylene chloride Azeotrope

85.8%

37.0

1.522

Acetone

87.5%

45.0

1.391

Azeotrope

Reference U.S.P. 3,573,213 U.S.P. 2,999,817 U.S.P. 2,999,815

δD 14.7

δP 1.6

Soils δH Cleaned 0.0 RA < 7.93

14.7

2.3

0.2 1, 5

15.3

2.4

1.0 1, 5

14.9

3.5

1.5 1, 2, 5

HSP OF BINARY AZEOTROPES 17 Azeotropes with HFE-7100 Hydrogen Bonding HSP, MPaˆ (½)

14

TRICRESYL PHOS

12 CASTOR OIL

10

Methyl Formate ETHYL CINNAMATE

8 6 4

BUTYL STEARATE Methyl Acetate Ethyl Formate LINSEED OIL Ethanol

2

ASTM 0FUEL “A” 0 2

4 6 8 10 Polar HSP, MPaˆ (½)

12

14

FIGURE 11.8

HSP OF BINARY AZEOTROPES

Hydrogen Bonding HSP, MPa^(1/2)

7 Azeotropes with HFE-7200 14

CASTOR OIL

10 8 6

ETHYL CINNAMATE

Methanol BUTYL STEARATE

t-Butyl Alcohol 1-Chlorobutane LINSEED OIL ?????????????? ?????????????? 2 Ethanol 1,2-Dichloropropane 4

ASTM 0 FUEL “A” 0 2

FIGURE 11.9

TRICRESYL PHOS

12

4

8 10 6 Polar HSP, MPa^(1/2)

12

14

7248_C011.fm Page 222 Thursday, May 10, 2007 8:00 AM

222

Hansen Solubility Parameters: A User’s Handbook

HSP OF BINARY AZEOTROPES 17 Azeotropes with HFE-8200 Hydrogen Bonding HSP, MPaˆ (½)

14

TRICRESYL PHOS

12 CASTOR OIL

10 8 6 4

ETHYL CINNAMATE Methyl Formate BUTYL STEARATE Methyl Acetate Methanol LINSEED OIL

2

ASTM 0FUEL “A” 0 2

4 6 8 10 Polar HSP, MPaˆ (½)

12

14

FIGURE 11.10

FIGURE 11.11 HSP OF BINARY AZEOTROPES 7 Azeotropes with OS-10 Hexamethyldisiloxane Hydrogen Bonding HSP, MPaˆ (½)

14

TRICRESYL PHOS

12 10 8

CASTOR OIL 2-Propyl (PA) Ethanol ETHYL CINNAMATE BUTYL STEARATE

6 4

LINSEED OIL

2

ASTM 0FUEL “A” 0 2

FIGURE 11.12

4 6 8 10 Polar HSP, MPaˆ (½)

12

14

7248_C011.fm Page 223 Thursday, May 10, 2007 8:00 AM

Use of Hansen Solubility Parameters

223

HSP OF BINARY AZEOTROPES 3 Azeotropes with OS-20 Octamethyltrisiloxane Hydrogen Bonding HSP, MPaˆ (½)

14

TRICRESYL PHOS

12 CASTOR OIL

10 8 6 4

ETHYL CINNAMATE BUTYL STEARATE Propylene Glycol n-Propyl Ether LINSEED OIL

2

ASTM 0FUEL “A” 0 2

4 6 8 10 Polar HSP, MPaˆ (½)

12

14

FIGURE 11.13 HSP OF BINARY AZEOTROPES 4 Azeotropes with OS-30 Decamethyltrisiloxane Hydrogen Bonding HSP, MPaˆ (½)

14

TRICRESYL PHOS

12 CASTOR OIL

10 8 6 4

ETHYL CINNAMATE BUTYL STEARATE Propylene Glycol n-Propyl Ether LINSEED OIL

2

ASTM 0FUEL “A” 0 2

4 6 8 10 Polar HSP, MPaˆ (½)

12

14

FIGURE 11.14 HSP OF BINARY AZEOTROPES 9 Azeotropes with HFC-245fa Hydrogen Bonding HSP, MPaˆ (½)

14

TRICRESYL PHOS

12 CASTOR OIL

10 8

ETHYL CINNAMATE BUTYL STEARATE

6 4

LINSEED OIL

2

ASTM 0FUEL “A” 0 2

FIGURE 11.15

4 6 8 10 Polar HSP, MPaˆ (½)

12

14

7248_C011.fm Page 224 Thursday, May 10, 2007 8:00 AM

224

Hansen Solubility Parameters: A User’s Handbook

HSP OF BINARY AZEOTROPES 5 Azeotropes with HFC-365mfc Hydrogen Bonding HSP, MPaˆ (½)

14

TRICRESYL PHOS

12 CASTOR OIL

10 8

ETHYL CINNAMATE BUTYL STEARATE

6 4 2

LINSEED OIL Methanol

ASTM 0FUEL “A” 0 2

2-Chloropropane 4 6 8 10 Polar HSP, MPaˆ (½)

12

14

FIGURE 11.16

HSP OF BINARY AZEOTROPES 2 Azeotropes with Perfluorohexane (PFC) Hydrogen Bonding HSP, MPaˆ (½)

14

TRICRESYL PHOS

12 CASTOR OIL

10 8

ETHYL CINNAMATE BUTYL STEARATE

6 4

LINSEED OIL

2

ASTM 0FUEL “A” 0 2

FIGURE 11.17

4 6 8 10 Polar HSP, MPaˆ (½)

12

14

7248_C011.fm Page 225 Thursday, May 10, 2007 8:00 AM

Use of Hansen Solubility Parameters

225

HSP OF BINARY AZEOTROPES 10 Azeotropes with HCFC-141b Hydrogen Bonding HSP, MPaˆ (½)

14

TRICRESYL PHOS

12 CASTOR OIL

10 8

ETHYL CINNAMATE BUTYL STEARATE

6 4

LINSEED OIL

2

ASTM 0FUEL “A” 0 2

4 6 8 10 Polar HSP, MPaˆ (½)

12

14

FIGURE 11.18

HSP OF BINARY AZEOTROPES 2 Azeotropes with HCFC-225 ca/cb Hydrogen Bonding HSP, MPaˆ (½)

14

TRICRESYL PHOS

12 CASTOR OIL

10 8

ETHYL CINNAMATE BUTYL STEARATE

6 4

LINSEED OIL

2

ASTM 0FUEL “A” 0 2

FIGURE 11.19

4 6 8 10 Polar HSP, MPaˆ (½)

12

14

7248_C011.fm Page 226 Thursday, May 10, 2007 8:00 AM

226

Hansen Solubility Parameters: A User’s Handbook

HSP OF BINARY AZEOTROPES 3 Azeotropes with 1,1,2-Trichlorotrifluoroethame Hydrogen Bonding HSP, MPaˆ (½)

14

TRICRESYL PHOS

12 CASTOR OIL

10 8

ETHYL CINNAMATE BUTYL STEARATE

6 4

LINSEED OIL

2

ASTM 0FUEL “A” 0 2

4 6 8 10 Polar HSP, MPaˆ (½)

12

14

Soils

Soils

iti

on

of

Co

m po

ne

nt s

Soils Soils Azeotropes

Ad d

Polar HSP (soil or solvent)

FIGURE 11.20

“Designer” Solvents

H2 Bonding HSP (soil or solvent) FIGURE 11.21

TABLE 11.16 Capability of Azeotropes of Designer Solvents Soil Number

Soil Name

1

ASTM fuel “A”

2

Butyl stearate

3 4 5

Castor oil Ethyl cinnamate Linseed oil

6

Tricresyl phosphate

Designer Solvents from which Azeotropes Can Be Blended to Clean the Stated Soil (RA < 8) HFE-7100, HFE-7200, HFE-8200, OS-10, OS-20, OS-30, HFC-245fa, HFC365mfc, PFC-5060, HCFC-141b, HCFC-225 ca/cb, CFC-113 HFE-7100, HFE-7200, HFE-8200, OS-10, OS-20, OS-30, HFC-365mfc, PFC5060, HCFC-141b, HCFC-225 ca/cb, CFC-113 HFE-7100, HFE-7200, HFE-8200, OS-10, OS-20, OS-30, HCFC-141b HFE-7100, HFE-7200, HFE-8200, OS-10, OS-30, HCFC-141b HFE-7100, HFE-7200, HFE-8200, OS-10, OS-20, OS-30, HFC-245fa, HFC365mfc, PFC-5060, HCFC-141b, HCFC-225 ca/cb, CFC-113 None

7248_C011.fm Page 227 Thursday, May 10, 2007 8:00 AM

Use of Hansen Solubility Parameters

227

NOTES 1. In Chapter 11, cleaning solvents are assumed to be used in a vapor degreasing process. Subboiling (cold cleaning) is not the process for which use of the solvents mentioned in this chapter is envisioned. 2. Durkee, J.B., Management of Industrial Cleaning Technology and Processes, 2006 by Elsevier Science, Oxford and Amsterdam. 3. Hansen, C.M., Hansen Solubility Parameters — A User’s Handbook, CRC Press, Boca Raton, FL, 1999, Equation 1.9 and Equation 1.10. 4. Durkee, J.B., What Would the ‘Perfect’ Cleaning Solvent Be?, Cleaning Times column in Metal Finishing Magazine, November 2005, p. 61. 5. See Chapter 1, Chapter 2, and Chapter 8 of Reference 3. 6. Note that thorough solutioning of a soil by a cleaning solvent is no more a sufficient condition for a successful cleaning application than is thorough solutioning of a polymer in a coating carrier solvent, a sufficient condition for a successful coating application. However, thorough solutioning is usually a necessary condition for either type of application to be successful. 7. An efficient approach to making this observation is to carefully apply some of the liquid soil composite to a glass slide. When the coated slide is illuminated from below, regions that are multiple phases can usually be delineated. One must be cognizant of Heisenberg’s uncertainty principle in preparing to make this observation, as the soil sample must be equivalent to the actual residue. 8. Solubility behavior of the two solvents is represented in Figure 11.1 as circles. The center point of the circle for each solvent is the locus of their polar and hydrogen bonding HSP values. Because these two solvents are different but miscible, there is a combined effect. Equation 11.1 is used to calculate that effect for any blend of solvents. That is the center point for the combined circle. The radius (that will be identified as R ) of each circle (solvent) represents their sphere of similarity in two dimensions. This is a region or zone in HSP space within which HSP values are sufficiently similar to those at the center point. Similarity has a practical meaning. It is that solvents are miscible, and soils are soluble in solvents. In two dimensions, the region is visualized in Figure 11.1 as a circle. Figure 11.1 is an idealized representation. There are two aspects of this idealized representation of solvent character as graphical areas: (1) solvents or soils or polymers are compatible (likely miscible) when their areas overlap (2) whereas the center point HSP values can be computed from molecular characteristic (properties), the extent of graphical area from the center point must be measured. The latter will be shown in the section on “An Open Question — Answered.” 9. If the soil is a commercial product, its material safety data sheet (MSDS) will likely contain information that either is the recipe or from which a suitable estimate of the recipe can be inferred. 10. See Equation 1.9 in Chapter 1 in Reference 3, and Equation 11.3 in “Method for Calculating HSP of Composites (Soils or Solvents)” in this chapter. 11. Refers to the total or Hildebrand solubility parameter. 12. Reference 3, Chapter 1. 13. Disperse, polar, hydrogen bonding, or total. 14. In Equation 11.1, φ is the volume fraction of component 1, and δ is any solubility parameter. It is understood that φcomp 1 + φcomp 2 = 1. Volume fraction is easy to compute because solvents are stored in pails or drums and used by volume, although they are sold by weight. The capacity of a vapor degreaser sump is given in gallons or liters. 15. The appendix to Reference 3 is an excellent source of HSP values for solvents (Appendix Table A.1), polymers, and some soils (Appendix Table A.2). Another useful reference is Barton, A.F.M., CRC Handbook of Solubility Parameters and Other Cohesion Parameters, 2nd ed., CRC Press, Boca Raton, FL, 1991. Data found here are one feedstock for implementation of Equation 11.1. 16. The other feedstock is composition information about azeotropic solvent composition. This can be found in Durkee, J.B., Management of Industrial Cleaning Technology and Processes, 2006 by Elsevier Science, Oxford and Amsterdam. 17. The reason for inclusion of molecular weight values in the basic data is because the mixture concentrations are often given in molar concentration, though that is not the case in this example. Composition is normally given on a weight basis in this chapter, because it is that basis by which solvents are normally sold.

7248_C011.fm Page 228 Thursday, May 10, 2007 8:00 AM

228

Hansen Solubility Parameters: A User’s Handbook

18. In this way, cleaning is unlike construction of coatings. In a coating, all components are generally compatible with the carrier solvent. In cleaning, where there can be destruction of coatings, all components must be manageable. In solvent cleaning, soil components can be partially soluble, swollen, insoluble, or soluble. Generally, a cleaning process can be designed to manage removal of soil components from surfaces without regard to their relationship to the solvent. In cleaning, management involves hydrodynamic (mechanical) force, heat, and solvency. 19. Said in a colloquial manner, “A chain is only as strong as its weakest link.” 20. Hansen, C.M. and Skaarup, K., The three dimensional solubility parameter — key to paint component affinities, J. Paint Technol., Vol. 305, No. 511, 511–514, 1967. 21. Finding the number 4 in Equation 11.3 may be surprising. There is considerable discussion in the literature about the need for it. It has been found empirically useful over several decades of practical experience. There are two justifications for it: (1) it converts three-dimensional spheroidal plots with dispersion solubility parameters to spherical ones, and (2) theoretical considerations associated with the Prigogine theory of corresponding states. See Chapter 5 in this book. 22. Unfortunately, selection of a suitable cleaning solvent based on similarity of intermolecular forces is but a necessary first step in development of cleaning process applications. Certain selection criteria are listed as follows: • A solvent that has a strong affinity for a soil but a low holding capacity for it (mass solubility) would be a poor choice. • A solvent selected to do cleaning work must be able to dissolve the soil to the extent desired in the time allotted under the prevailing conditions. Cleaning is a transport process. • Also, a poor choice is a solvent that only gradually penetrates and swells the soil and allows it to be removed by rinse fluids. • Finally, a solvent that efficiently dissolves an adequate mass of soil in an efficient time but only at a temperature above its boiling point is nearly useless. Pressurized contacting equipment is expensive. Yet, without successful completion of a screening evaluation via use of HSP, development of cleaning process applications becomes either resource intensive or based on hearsay evidence. 23. Note that Figure 11.1 is a generalized presentation. Soils can be matched to solvents for the purpose of solvent selection; solvents can be matched to other solvents for the purpose of producing a blend with certain physical properties, or soils can be matched to other soils for the purpose of evaluating if solution-based or force-based (aqueous technology) should be used, depending upon whether the soils are expected to combine in a single or multiple phases. 24. HFE-8200 is actually a component of the product sold as HFE-7100. The latter is mixture of Perfluoron-butyl methyl ether (35%) and perfluoroisobutyl methyl ether (65%). HFE-8200 is the pure perfluoroisobutyl methyl ether. United States Pharmacopeia (USP) 6,426,327 teaches that azeotropic behavior of mixtures of the two isomers and another component is relatively independent of distribution of isomers in the HFE material. Consequently, in this analysis, HFE-7100 and HFE-8200 are considered equivalent in terms of the boiling point and composition of azeotropes. However, the HSP properties of HFE-7100 and HFE-8200 are slightly different. This difference has been retained in the calculations of solubility behavior of the azeotropes. 25. Owing to its inertness as a solvent, high specific gravity, and low surface tension, it finds use in cleaning work as a rinsing agent. Yet, because of its inertness, PFC-5060 is a designer solvent that on emission migrates without degradation to the Earth’s stratospheric layer and acts as a global warming agent. Hence, its use is limited “by design.” 26. The disperse HSP value is momentarily neglected in the interests of graphical clarity when a twodimensional graph is used. These values are relatively similar. 27. The structure shown represents the most common (89%) component (ricinolenic acid, 12-hydroxyoleic acid) of castor oil. Castor oil is also referred to as the triglyceride of this acid. 28. The structure shown, linoleic acid, is the most common (52%) acidic component of linseed oil. The other major acidic component of the triglyceride is linolenic acid that has one fewer double bond.

7248_C011.fm Page 229 Thursday, May 10, 2007 8:00 AM

Use of Hansen Solubility Parameters

229

29. In the main, these azeotropes are claimed in U.S. patents whose grant date is during the 1990s. The patent assignee is most often the manufacturer of the designer solvent. There is no published table of these binary azeotropes except that found in Reference 15, which lists around 450 of them. They were identified in 2004 during an extensive search of the world patent literature. Readers are reminded that the normal interpretation of U.S. patent law is that when a third component is added to a claimed binary composition, the ternary mixture is still covered by the U.S. patent. In other words, if these binary azeotropes are not commercial products, users cannot formulate them without permission from the patent holder. 30. The parts used are those from the business operation. The soils are those identified by the business as being of concern and are applied as pure components to previously cleaned parts (or coupons). The method of assaying cleanliness is gravimetric. All information is publicly available. See http://www.cleanersolutions.org. Most work was done around 2004. There are several uncertainties about use of these 53 data points: (1) a variety of vapor degreasing processes, test conditions, and test exposure times were used, (2) soil components were identified by chemical abstracts service (CAS) number from the manufacturer’s MSDS that is not always a complete listing of ingredients, and (3) whereas the soil components were identified, their concentration in the applied soil mix was not. Yet, there was considerable level of replicates available through the use of multiple runs where the same solvent was used under a different product identification. Hence, one can say with 95% confidence that if the RA value is less than 8, then the worst level of cleanliness of the least compatible soil component would be 94%. The expected value of cleanliness would be around 97%. 31. Normally, the limit of acceptability regarding cleaning performance is set by the requirements of the next processing or use step. For these laboratory trials, no such information was available. 32. Both ternary and quaternary azeotropes containing designer solvents have been identified and patented. Some find applications in aerosol-powered cleaning products. Here, there is little concern about composition change in use, as the entire volume applied evaporated quickly after application. But in vapor degreasing operations, which is the venue of this chapter, the concern about composition change of ternary and quaternary (vs. binary) azeotropes with use cannot be neglected. 33. Azeotropes in Table 11.4 through Table 11.14 are sorted by increasing value of polar HSP with that of the base designer solvent given in the initial row. Boiling points are also given, but not the rows that are not sorted by that parameter. In this study, all azeotropes noted are minimum-boiling azeotropes. As all of these designer solvents have commercial value as unique compounds, the azeotropes they form with other solvents are claimed in patents. U.S. patent numbers are given as references where available. 34. No denigration of designer solvents is intended in this work. Rather, these solvents represent the result of significant and successful chemical research into identification and manufacture of chemical structures, which alleviate many of the safety, health, and environmental concerns about solvents commonly used in cleaning (and other) work in the late 20th century. Elimination of those concerns was roundly and justifiably applauded by users. In a sense, that research was too successful. Seemingly, as a class of product, designer solvents were those whose use raised fewer concerns, but whose value in use was more limited. Binary azeotropes based on designer solvents are an attempt to span the gap between absence of concern and absence of value in use. 35. Only blends defined by measured physical property and boiling point data are included. Theoretical or predicted azeotropes are not included because they cannot be purchased as blends for cleaning work. A good source of information about predicted azeotropes is Harding, S.T., Locating All Heterogeneous and Reactive Azeotropes, distributed by the American Institute of Chemical Engineers, (January 1, 1998), ASIN: B0006R4J04. 36. A secondary standard is to fulfill the conventional requirements upon which solvent cleaning work (vapor degreasing) is based. See Reference 2 for details. 37. This phrase refers to the identifying characteristic of an azeotrope, that the composition of the vapor and the composition of the boiling liquid are identical. In other words, as the mixture boils, the composition (like that of a single component solvent) remains constant.

7248_C011.fm Page 230 Thursday, May 10, 2007 8:00 AM

7248_C012.fm Page 231 Wednesday, May 23, 2007 11:14 AM

— Chemical 12 Applications Resistance Charles M. Hansen ABSTRACT Hansen solubility parameters (HSP) can correlate differences in physical behavior observed in chemical resistance testing of polymers and polymer-containing systems when a sufficiently large number of different organic solvents are included in a study. These correlations can then be used to predict the chemical attack expected in systems that have not yet been tested. Examples of HSP correlations included here are for solubility, degree of surface attack, tensile strength reduction, and simple evaluations of chemical resistance of the suitable-for-use or not type. Environmental stress cracking is discussed in more detail in Chapter 14. In each case, the molecular size of the liquids used can affect the result and should be considered in some way. Chapter 16 treats absorption and diffusion in polymers with this in mind. A common problem is that tests with larger molecular weight liquids have not reached equilibrium absorption within the timeframe of the exposure. HSP correlations are presented for chemical resistance studies of epoxy and zinc silicate tank coatings, PET, POM, PA6/66, PUR, PPS, PEI, Neoprene®, etc.

INTRODUCTION HSP are widely used in the coatings industry to select solvents for dissolving polymers and binders. This has been discussed in Chapter 8 and also in References 1 through 12, as well as elsewhere. The fact of solution is in itself clearly one simple form of chemical attack of the polymers they dissolve. This means that chemical resistance for some polymers can be partly inferred from HSP correlations of their solubility and/or swelling. HSP correlations of this type have been discussed in Chapter 5. An example is that if a chemical does not dissolve an epoxy component or the curing agent, then it is quite unlikely that it will attack a fully crosslinked epoxy coating or glue. The HSP correlations of surface phenomena, which have been discussed in Chapters 6 and 7 and elsewhere,13–18 can also provide some insight into chemical resistance. Liquids not wetting a surface are not as likely to attack it as those that do wet it, although there are no guarantees. A relation between spontaneous spreading and dewetting of liquids and environmental stress cracking has been found.19 This is discussed in more detail in Chapter 14. Liquids that spontaneously spread were found to induce environmental stress cracking at lower critical strains than those liquids that do not. Some surface studies may involve evaluation of a more direct form of chemical attack, such as the attack/whitening of PET coated with “amorphous” PET to improve weldability. This example is discussed in more detail later. Whatever is being correlated, the general considerations of the HSP characterizations discussed in earlier chapters are the same for the HSP correlations of chemical resistance reported here. However, there are certain additional pitfalls to be aware of when correlating chemical resistance. These include (lack of) attainment of equilibrium, the effects of molecular size of the test chemical, difference in local segments of polymers (even in homopolymers), and acid/base reactions. Once a reliable HSP characterization of chemical resistance is available, it can be used to calculate the behavior of other systems that have not been tested. Obtaining a good HSP correlation of chemical resistance that allows reliable predictions depends very much on careful treatment of the available data or generation of data with such a correlation in mind. Unfortunately, very few 231

7248_C012.fm Page 232 Wednesday, May 23, 2007 11:14 AM

232

Hansen Solubility Parameters: A User’s Handbook

studies of chemical resistance have been designed with the purpose of generating HSP correlations. Also, it must be clear that the chemical attack discussed in this chapter does not include true chemical reactions leading to covalent bonding or destruction, such as with acids and bases, whether they are organic or inorganic. Chemical reactions forming new compounds are often found with amines and organic acids. These reactions often lead to discoloration in systematic solubility parameter testing with one or more of the amines used as test solvents. Discolored systems should simply be neglected in HSP correlations of physical (reversible) solubility. The products of reactions of well-defined organic bases with well-defined organic acids have actually been studied systematically from a solubility parameter point of view.20 Some of the results of this study are discussed in Chapters 15 and 18.

CHEMICAL RESISTANCE — ACCEPTABLE-OR-NOT DATA Additional sources of data for solubility parameter characterizations include chemical resistance tables reported by raw material suppliers21–26 or collected in books27–29 and other sources such as those supplied on electronic media by the Plastics Design Library.30 Although these data are certainly valuable in themselves, it has been found that given data sets are not always as reliable/consistent/coherent as could be desired for solubility parameter correlations. Attainment of equilibrium may not have been achieved, and this effect is rarely confirmed or even considered. Solvents with low diffusion coefficients will appear to be less aggressive than they might become at longer exposure times or at higher concentrations. As discussed earlier, true chemical attack with acids and bases must sometimes be sorted out. Likewise, the data are often limited in number and scope, and the chemical reagents have not been chosen for the purpose of solubility parameter correlations. Nevertheless, with the use of due caution, it is often possible to find excellent solubility parameter correlations using chemical resistance data of the acceptable-or-not type, particularly when the list of agents is long. Additional precautions with regard to data of the acceptable-or-not type include whether a molecular size effect is present as discussed in the following. Also, it can be assumed that if a chemical attacks a polymer at, say, 20°C, then it will also attack it at, say, 70°C. It can then be included as data in a 70°C correlation, even though it may not have been tested at that temperature (and in principle the HSP are only valid at room temperature).

EFFECTS OF SOLVENT MOLECULAR SIZE It has been emphasized in Chapters 1, 2, and 5 that the size of solvent molecules is important for polymer solubility. HSP correlations have confirmed that this effect is even more important when chemical resistance is being considered. Smaller molecules are expected to be better, that is, more aggressive from a thermodynamic point of view than larger ones, all else being equal. This is known from the theories of polymer solubility discussed in Chapters 1 through 4, and also from the discussion of barrier polymers and diffusion found in Chapter 13 and 16, respectively. So it is not surprising that solvent molecular size can be an important fourth parameter in correlations of chemical resistance. An appropriate way to check this is sorting output data from a computer (or other) HSP optimization according to the molecular volume of the test solvents. What appear to be errors in the correlation may become systematically arranged. It can easily be seen, if the top of the list includes the type of “error” where the smaller molecular species are “better” than expected by comparison with all the other solvents. This may take the form of unexpectedly dissolving, being more aggressive than expected, penetrating more rapidly, or reducing mechanical properties more severely. Larger molecular species which are “poorer” than expected by comparison with the data for the other solvents are often seen at the bottom of the list. One can focus upon the molecular size range of greatest interest in such cases and repeat the correlation, neglecting those species which are outside of this range of molecular volumes. The correlation is then strictly valid only

7248_C012.fm Page 233 Wednesday, May 23, 2007 11:14 AM

Applications — Chemical Resistance

233

for the size range specified. Some indication of the behavior of the solvents with V larger than the upper limit is possible if their RED numbers are greater than 1.0. These would not be expected to attack under any circumstances. Likewise, the solvents with V less than the lower limit can be expected to attack if their RED numbers are less than 1.0. Larger numbers of solvents are needed in the study if this is to be done with any benefit. As stated above, the size and shape of solvent molecules are very important for kinetic phenomena such as diffusion, permeation, and attainment of equilibrium. Chapters 13 and 16, respectively, discuss correlations of HSP and elaborate on diffusion phenomena in more detail. However, it will be repeated here that smaller and more linear molecules diffuse more rapidly than larger and more bulky ones. The diffusion coefficient may be so low that equilibrium is not attained for hundreds of years at room temperature in common solvent exposures of rigid polymers like polyphenylene sulfide (PPS) with thicknesses of several millimeters.31 Such effects lead to comparisons where some systems may have reached an equilibrium uptake, whereas others have not. Likewise, the second stage in the two-stage drying process in polymer film formation by solvent evaporation can last for many years.4,31 See Chapter 16. Polymer samples used for solubility parameter or chemical resistance testing may contain retained solvent or monomer for many years, and this may also affect the evaluations. However, the effects of water can be extremely rapid as discussed in the following.32 Attempts to include the molecular volume into a new composite solubility parameter and size parameter have not been particularly successful.33,34 This may be because the size effects are not necessarily caused through the thermodynamic considerations on which the solubility parameters are based, but rather through a kinetic effect of diffusion rate.

CHEMICAL RESISTANCE — EXAMPLES Chemical resistance means different things in different contexts. Various examples of HSP correlations of chemical resistance are included in the following. The HSP data for the correlations discussed are included in Table 12.1. Experimental data are always preferred over predicted behavior based on a correlation. However, a good HSP correlation can be used to find many chemicals that will clearly attack or that will clearly not attack. There are also situations where the attack is mild, and whether or not satisfactory results are found with a product depends on its use. Data in chemical resistance tables are often of the type +, +/–, –, or satisfactory/questionable/unsatisfactory, recommended/not recommended (R/NR), or something similar. The liquids which attack are clearly good solvents for the material in question and will be located within the HSP spheres with RED numbers being successively lower for more severe attack, all other things being equal. The correlations can include the solvents with mild attack (+/, questionable) either in the attacking (NR) group or in the nonattacking (R) group. They can also be neglected, not knowing which group to put them into. The treatment used in the individual correlations presented here is indicated in the following. Unless otherwise specified, the results are for room temperature.

TANK COATINGS Chemical resistance is important for tank coatings used in the transport of bulk chemicals. The data in Table 12.1 include two older HSP correlations for chemical resistance for two types of tank coatings supplied by Hempel’s Marine Paints. These are for a two-component epoxy type and for a zinc silicate type. The data and correlations are about 20 years old. They are included here for purposes of demonstration. Improvements in chemical resistance are known to have been implemented in a newer epoxy tank coating, but no HSP correlation has been made. A HSP correlation of the solubility of a lower molecular weight epoxy, Epon® 1001 (Shell Chemical Corp.), is included for comparison. The numbers are not too different from those of the HSP correlation for chemical

7248_C012.fm Page 234 Wednesday, May 23, 2007 11:14 AM

234

Hansen Solubility Parameters: A User’s Handbook

TABLE 12.1 Hansen Solubility Parameter Correlations for Selected Materials Material

δD

δP

δH

Ro

FIT

G/T

Epoxy tank coat (two component) Epoxy solubility (Epon 1001) Zinc silicate coating PET-amorphous coating PET-CR (+/– Rb) PUR-CR (+/– Rb) POMC/POMH (+/ Rb) (+/– Rb) POMC (+/– NRb) PA6/PA66 (+/– NRb) Halar 300 ECTFE 23°C Halar 300 ECTFE 50°C Halar 300 ECTFE 100°C Halar 300 ECTFE 120/149°C Neoprene-CR (+/– Rb) PPS tensile strength <60% 93°C PEI ULTEM 1000 600 psi PEI ULTEM 1000 1200 psi PEI ULTEM 1000 2500 psi PEI ULTEM 1000 solubility PES mechanical properties PES solubility

18.4 18.1 23.5 17.0 18.2 18.1 17.1 17.9 18.9 16.8 18.1 18.1 18.3 18.1 18.7 17.3 17.0 17.4 19.6 17.1 19.6

9.4 11.4 17.5 11.0 6.4 9.3 3.1 5.9 7.9 8.4 7.5 7.9 8.7 4.3 5.3 5.3 6.0 4.6 7.6 9.9 10.8

10.1 9.0 16.8 4.0 6.6 4.5 10.7 8.3 11.7 7.8 8.5 7.9 7.9 6.7 3.7 4.7 4.0 9.0 9.0 6.3 9.2

7.0 9.1 15.6 9.0 5.0 9.7 5.2 6.6 8.7 2.7 5.2 6.7 7.5 8.9 6.7 3.3 4.0 7.2 6.0 6.3 6.2

—a —a —a 1.000 0.865 0.981 0.955 0.609 0.950 0.993 0.700 0.710 0.800 0.937 0.991 1.000 1.000 0.967 0.952 0.931 0.999

—a —a —a 7/11 7/34 16/26 2/28 11/28 9/31 2/102 18/92 49/91 48/74 30/48 9/16 3/20 4/20 9/20 8/45 6/25 5/41

Note: The symbols G for good solvents and T for total solvents are maintained, with the understanding that G solvents are within the HSP correlation spheres and are not recommended for use. a b

Data not available. See text: NR — not resistant, R — resistant.

resistance for the two component epoxy tank coating. These three correlations have been reported earlier.8 The fact of a successful HSP correlation for a completely inorganic type of coating like zinc silicate is surprising. This is still another demonstration of the universality of the applications possible with the HSP concept. Although the data fit numbers were not recorded at the time, the two chemical resistance correlations reported here were clearly considered reliable.

PET FILM COATING Another example of chemical resistance, or lack of the same, is the attack of the amorphous, modified PET coating on PET films to improve their weldability. This correlation is based on only 11 well-chosen data points but clearly shows that attack for many chemicals can be expected. Among those chemicals not attacking are hydrocarbons, glycols, and glycol ethers and higher alcohols which have a reasonably high hydrogen bonding character.

ACCEPTABLE

OR

NOT — PLASTICS

Several examples of HSP correlations of data reported in the form acceptable-or-not are included in Table 12.1. The data for these are all found in Reference 21. Other data sources for these are also available. HSP correlations of this type are included for PET, PUR, POMH, POMC, and

7248_C012.fm Page 235 Wednesday, May 23, 2007 11:14 AM

Applications — Chemical Resistance

235

PA6/PA66 using data from Reference 21. The first three correlations of this type consider reported evaluations of minor attack which will require further evaluation (+/–) as if these systems were suitable for use (resistant). The last two correlations consider this condition as not suitable for use (not resistant). This is to demonstrate that differences are found, depending on how the data are considered and that outliers are often found when correlating this type of data. It is for this reason that more extensive tables of HSP correlations of chemical resistance are not reported here. Too much space is required to try to explain why given results are outliers. However, some reasons have been given earlier and others are in the following. The HSP values for PET based on chemical resistance are somewhat different from those of the amorphous PET coating, which is readily attacked by far more solvents. The compositions are also different. The POM correlations have typical problems in that the correlation considering all minor attack as negligible is based on only two severely attacking solvents among the 28 solvents tested. When the solvents demonstrating minor attack are considered as being in the attacking group, the data fit shows that there are many outliers. A systematic analysis of why this is found will not be attempted for reasons of space, even if all the outliers could be explained in one way or another. The HSP found for a polymer in this type of correlation may not be representative of that polymer in all aspects of its behavior. There is a question as to coverage over the whole range of solvent HSP possible by the test solvents. Additional solvents may be required to make an improved correlation based on the improved coverage possible. There is also a question as to which segments of the polymer may be subject to attack by which solvents. Block copolymers may demonstrate two separate (overlapping) correlations that cannot be reasonably force fitted into a single HSP correlation. Viton® is an example of this. The most severe attack or swelling may occur in one region or another of the polymer or maybe even on a third component, such as a cross-linking segment. Viton® is discussed in Chapter 5 and Chapter 13 in more detail. In spite of these pitfalls, it is strongly suggested that those generating this type of resistance data should try HSP correlations to evaluate the consistency of the data before reporting it. Outliers can be reconsidered; whether or not equilibrium has been attained can be inferred, and the probable effects of solvent molecular size may become apparent. The effects of temperature on the chemical resistance of poly(ethylene co-chlorotrifluoroethylene) (ECTFE) Halar® 300 can be seen in Figure 12.1. The data on which these correlations are based are of the recommended-or-not type and were found in Reference 26. This figure has affectionately been dubbed a “bullseye,” as there appears to be symmetry about a central point, although this is not strictly true as the HSP data confirm. The radius of the chemical resistance spheres increases with increasing temperature, as expected, as more solvents then become more severe in their attack. The HSP data for these correlations are also included in Table 12.1. The data fits are not particularly good at the higher temperatures. To complete this section, a correlation of chemical resistance data for Neoprene® rubber (Du Pont)22 is included. Solvents in the intermediate category, i.e., that of a questionable-for-use recommendation, are considered as being in the nonattacking group for this correlation. Previously it was indicated why HSP correlations of this type lead most often to guidance rather than to a firm recommendation. There are many pitfalls to be aware of both in generating such correlations as well as in using them, but their usefulness becomes clearer with some experience. A suitable goal for a future project is to determine the effective HSP for various media frequently encountered in resistance lists, such as mustard, certain detergents and oils, etc. This could perhaps be done by composition in some cases. In other cases, one could see whether behavior paralleled that of a known chemical. A third method is to determine these parameters by recognizing a similarity to all materials attacked and a difference from those not attacked. This approach has been used to assign HSP to some liquids when calculations were uncertain. A computer program was developed similar to the SPHERE program as described in Chapter 1, but working in the

7248_C012.fm Page 236 Wednesday, May 23, 2007 11:14 AM

236

Hansen Solubility Parameters: A User’s Handbook

SOLVENT RESISTANCE OF HALAR 300 AT 4 DIFFERENT TEMPERATURES

14 13 12 11 HANSEN POLAR PARAMETER, δH

120° = 149° 10 100° 9

23°

120°✚ 23° ● 100°▲ 50° ✳

8

50°

7 6 5 4 3 2 1 0 0

1

2

3

4

5

6

7

8

9

10

11

12

13

14

HANSEN HYDROGEN BONDING PARAMETER, δH

TEMP.

D

P

H

R

FIT

NO(NR)

23°

16.8

8.4

7.8

2.7

0.993

102 (2)

50°

18.1

7.5

8.5

5.2

0.70

92 (18)

100°

18.1

7.9

7.9

6.7

0.71

91 (49)

120/149°

18.3

8.7

7.9

7.5

0.80

74 (48)

FIGURE 12.1 Chemical resistance of Halar® 300 ECTFE at various temperatures. Liquids within the spheres (circles) are not recommended at the given temperatures. HSP data given in Table 12.1.

7248_C012.fm Page 237 Wednesday, May 23, 2007 11:14 AM

Applications — Chemical Resistance

237

opposite manner. The solubility of a number of polymers was evaluated in the solvent. The solvent parameters were then systematically varied by the program to reduce the collective error, that is, to locate the best possible set of HSP for the solvent. In general, the data fits for this procedure were comparable to those found for polymer solubility using the SPHERE program. In other words, not all the predictions based on the HSP thus assigned to the solvent agreed with the experimental data, but the errors were small.

TENSILE STRENGTH The long-term exposure of polymers or polymer composites to solvents normally leads to changes in mechanical properties. One of the more direct techniques to measure such effects is to determine the tensile strength. The tensile strength reduction for glass fiber reinforced polyphenylene sulfide (PPS) after exposure to a number of solvents at 93°C for 12 months has been reported.23 A HSP correlation of these data using the “good” solvents as those which reduce tensile strength under these conditions to less than 60% of the initial value is found in Table 12.1 and Figure 12.2. More extensive correlations for PPS are found in Reference 31. Additional HSP tensile strength correlations have been generated for polyetherimide, ULTEM® 1000, using data reported by General Electric.24 It is clear that the chemical resistance is dependent on the stress level. Higher stress levels lead to more severe attack by a larger number of chemicals. The solvents considered as being those which attack led to cracking during the study, which lasted 336 h. Some led to earlier cracking than others, which could be treated in a separate correlation, but this has not been done. A more rapid attack is expected from the better solvents with the smallest size and shape. The correlations all have high data fits.

δP

δD

PPS

δH

5

FIGURE 12.2 HSP correlation of the tensile strength reduction of Ryton® PPS. Within the sphere are liquids which reduce the tensile strength to less than 60% of the original value after exposure for 1 year at 93°C. (From FORCE Institute, Solvent Resistance of Polymer Composites, Glass Fibre Reinforced Polyether Sulphone (PES), 1st ed., Center for Polymer Composites, 1994, 31. With permission.)

7248_C012.fm Page 238 Wednesday, May 23, 2007 11:14 AM

238

Hansen Solubility Parameters: A User’s Handbook

The next entry in Table 12.1 is for true solubility of ULTEM 1000. These data were generated in a standard solubility parameter study. This correlation is not directly comparable with the previous ones for ULTEM 1000 as the number and range of solubility parameters included in the test solvents are different. The parameters for the polymer found in this correlation are those expected to reflect its (thermodynamic) affinities most closely. The previous study24 did not include a sufficient number of solvents having widely different HSP to give a true total picture of the ULTEM 1000. A final HSP correlation of the suitable-for-use or not type is presented in Table 12.1. This is for polyethersulfone (PES) based on mechanical properties after exposure to various liquids. The HSP correlation for the recommendation from the supplier25 for glass fiber reinforced PES (Ultrason® E, BASF) can be compared with the HSP correlation for simple solubility of the polymer in standard test liquids, also given in Table 12.1. There is a difference, but it is not large.

SPECIAL EFFECTS WITH WATER As stated elsewhere in this book (Chapter 1 and Chapter 15, in particular), the seemingly unpredictable behavior of water has often led to its being an outlier in HSP correlations. For this reason, it is suggested that data for water used as a test solvent not be included in HSP correlations. Water can be a very aggressive chemical. Water uptake in most polymers increases with increased temperature. This is because the solubility parameters of the water and polymer are closer at higher temperatures. The very high δH parameter for water decreases more rapidly with increasing temperature than the δH parameter for most polymers. This has been discussed in Chapter 1 and Chapter 8, but it is repeated here with examples for those interested in chemical resistance. Water is an exceptionally good plasticizer because of its small molecular size. The presence of water not only softens (reduces the glass transition temperature) a polymer as such, but it also means diffusion rates of other species will be increased. Therefore, the presence of water in a film can influence the uptake of other materials, with hydrophilic materials, in particular, being more prone to enter the film. The increase of water uptake with increased temperature can cause special problems with blistering if the temperature of a water-saturated polymer falls rapidly to a lower temperature. The previously soluble water can no longer be truly dissolved. Some of the water already in the polymer is now in excess and suddenly appears as small clusters or droplets of freed liquid water within the polymer itself (see Chapter 8, Figure 8.3). These droplets can quickly collect into blisters, especially if there are hydrophilic sites in the polymer or at an interface to which water will rapidly diffuse. This special type of failure has been discussed in more detail elsewhere32 (see also Chapter 1 and Chapter 8). The phase separated water has been called SWEAT (soluble water exuded at lowered temperatures). The author has observed this phenomenon as a mechanism of failure for epoxies, polyesters, alkyds, polyethersulfone (PES), polyphenylene sulfide (PPS), and even EPDM rubber. This mechanism can be confirmed experimentally by cycling samples continually exposed to water between two relevant temperatures using a quench from the higher one to the lower one. One follows weight gain by rapidly weighing samples that are surface dry. Typical results for the SWEAT phenomena for EPDM are seen in Figure 12.3 and for PPS in Figure 12.4. Control samples that are not cycled reach equilibrium and stay there, whereas the cycled samples suddenly begin to gain weight well beyond the equilibrium value. The extra weight is phase-separated water within the samples. This has been discussed in detail in Reference 35 for PPS and PES. A related problem can be encountered in chemical resistant coatings for tanks that have been in contact with methanol. If a coated tank has been used to store methanol, and perhaps hot methanol in particular, the coating is more than likely saturated with methanol. It may take several days of exposure to fresh air (to reduce the amount of methanol to acceptable levels) before subsequent direct contact with water or seawater can be tolerated. If there is too much methanol retained in the coating, the water diffusing into the coating will associate with the methanol. The increasing

7248_C012.fm Page 239 Wednesday, May 23, 2007 11:14 AM

Applications — Chemical Resistance

239

WATER ABSORPTION, CYCLING 120°-15° EPDM-GASKET 5.00 4.50

WATER UPTAKE, W/W%

4.00 3.50 3.00 2.50 2.00 1.50 1.00 0.50 0.00

0

1

2

3 4 EXPOSURE TIME, √DAYS

5

6

7

FIGURE 12.3 A rapid quench to a lower temperature can free water already dissolved in a polymer in the form of SWEAT. SWEAT can lead to blistering, cracking, and delamination. The data in the figure are weight gain for EPDM with cycling in water between 120 and 15°C. Water in excess of the equilibrium value at longer cycling times is SWEAT.

water content in the mixture of methanol and water will ultimately cause the solubility parameters of the mixture to be sufficiently high so that it becomes incompatible with the coating. Blisters form and total delamination can occur. These blisters are often near the substrate, as this is where the retained methanol will be found at highest concentrations.

CONCLUSION HSP can correlate differences in physical behavior observed in chemical resistance testing of polymers and polymer containing systems when a sufficiently large number of different organic solvents are included in a study. HSP correlations including systematic consideration of the solvent molar volume (or other suitable size parameter[s]) should be an inherent part of all future studies of chemical resistance. These correlations aid in the determination of whether equilibrium has been attained, as well as provide insight into the behavior expected from untested solvents whose HSP are stored in a solvent database or can be calculated. Examples include HSP correlations of true solubility and swelling, degree of surface attack, tensile strength reduction, and correlations for simple evaluations of chemical resistance of the suitable-or-not type. It is reemphasized that, in each case, the molecular size of the liquids evaluated will affect the result, and this should be considered in some way. Data for true acidic or basic chemical attack must not be included in HSP correlations, as HSP correlations reflect physical attack and not chemical attack. It is strongly suggested that data for water not be included in these correlations as well. Its behavior is too unpredictable compared with other test liquids, and if it is included as an outlier, this fact will force a correlation with less predictive ability than had it been neglected.

7248_C012.fm Page 240 Wednesday, May 23, 2007 11:14 AM

240

Hansen Solubility Parameters: A User’s Handbook

WATER ABSORPTION VS.TEMPERATURE RYTON R4 XT-HV, plate thickness 1 and 2 mm 2 90 C, 1 mm ■

1.5

90 C, 1 mm

.5

▲ ▲ ▲ ▲ ▲ ■ ▲ ▼

▲ ▲ ▲

■

■ ■

▲ ▲

90 C, 2 mm ▲ ▲ ■▲ ▲

■ ■

50 C, 1 mm

■

✕

■

▼ ▼ ✕ ▲ ▲ ▲ ▼ ▲ ▲ ▲ ✕ ✛ ▼ ■ ✛ ▼ ▼ ▼ ▼ ▼ ▼ ▼

▲

0

▲ ▲ ▲ ▲ ▲ ▲

▲▲ ▲ ▲ ▲ ▲▲ ▲

✕ ✛

▲ ▲ ▲ ▲ ▲ ▲ ▲ ▲ ▲ ▲ ▲

WEIGHT GAIN (%)

■

1

✕ ✛

✕ ✛

✕

23 C, 1 mm

✛

✛

▼

▼

-.5

▼ ▼

140 C, 2 mm ▲▲ ▼▼

-1 0

10

20 30 40 50 60 70 SQUARE ROOT OF TIME / THICKNESS SQRT (HRS)/mm

80

FIGURE 12.4 A rapid quench to a lower temperature can free water already dissolved in a polymer in the form of SWEAT. SWEAT can lead to blistering, cracking, and delamination. The data in the figure are weight gain for PPS with cycling in water between 90 and 23°C. Water in excess of the equilibrium value is SWEAT. (From Hansen, C.M., The Resistance of PPS, PES and PA Polymer Composites to Temperature Cycling During Water Exposure, Center for Polymer Composites (Denmark), Danish Technological Institute, Taastrup, 1994, 11. With permission.)

REFERENCES 1. Hansen, C.M., The three dimensional solubility parameter — key to paint component affinities I. Solvents, plasticizers, polymers, and resins, J. Paint Technol., 39(505), 104–117, 1967. 2. Hansen, C.M., The three dimensional solubility parameter — key to paint component affinities II. Dyes, emulsifiers, mutual solubility and compatibility, and pigments, J. Paint Technol., 39(511), 505–510, 1967. 3. Hansen, C.M. and Skaarup, K., The three dimensional solubility parameter — key to paint component affinities III. Independent calculation of the parameter components, J. Paint Technol., 39(511), 511–514, 1967. 4. Hansen, C.M., The Three Dimensional Solubility Parameter and Solvent Diffusion Coefficient, Their Importance in Surface Coating Formulation, Doctoral dissertation, Danish Technical Press, Copenhagen, 1967. 5. Hansen, C.M., The universality of the solubility parameter, Ind. Eng. Chem. Prod. Res. Dev., 8(1), 2–11, 1969. 6. Hansen, C.M. and Beerbower, A., Solubility parameters, in Kirk-Othmer Encyclopedia of Chemical Technology, Suppl. Vol., 2nd ed., Standen, A., Ed., Interscience, New York, 1971, pp. 889–910. 7. Hansen, C.M., 25 years with solubility parameters (25 År med Opløselighedsparametrene, in Danish), Dan. Kemi, 73(8), 18–22, 1992. 8. Hansen, C.M., Solubility parameters, in Paint Testing Manual, Manual 17, Koleske, J.V., Ed., American Society for Testing and Materials, Philadelphia, PA, 1995, pp. 383–404.

7248_C012.fm Page 241 Wednesday, May 23, 2007 11:14 AM

Applications — Chemical Resistance

241

9. Barton, A.F.M., Handbook of Solubility Parameters and Other Cohesion Parameters, CRC Press, Boca Raton, FL, 1983. 10. Barton, A.F.M., Handbook of Polymer-Liquid Interaction Parameters and Solubility Parameters, CRC Press, Boca Raton, FL, 1990. 11. Anonymous, Brochure: Co-Act — A Dynamic Program for Solvent Selection, Exxon Chemical International Inc., 1989. 12. Dante, M.F., Bittar, A.D., and Caillault, J.J., Program calculates solvent properties and solubility parameters, Mod. Paint Coat., 79(9), 46–51, 1989. 13. Hansen, C.M., Characterization of surfaces by spreading liquids, J. Paint Technol., 42(550), 660–664, 1970. 14. Hansen, C.M., Surface dewetting and coatings performance, J. Paint Technol., 44(570), 57–60, 1972. 15. Hansen, C.M. and Pierce, P.E., Surface effects in coatings processes, Ind. Eng. Chem. Prod. Res. Dev., 13(4), 218–225, 1974. 16. Hansen, C.M. and Wallström, E., On the use of cohesion parameters to characterize surfaces, J. Adhes., 15(3/4), 275–286, 1983. 17. Shareef, K.M.A., Yaseen, M., and Reddy, O.J., Suspension interaction of pigments in solvents. Characterization of pigments surfaces in terms of three-dimensional solubility parameters of solvents, J. Coat. Technol., 58(733), 35–44, February 1986. 18. Beerbower, A., Surface free energy: a new relationship to bulk energies, J. Colloid Interface Sci., 35, 126–132, 1971. 19. Nielsen, T.B. and Hansen, C.M., Surface wetting and the prediction of environmental stress cracking (ESC) in polymers, Polym. Degradation Stability, 89, 513–516, 2005. 20. Hansen, C.M., Some aspects of acid/base interactions (Einige Aspekte der Säure/Base-Wechselwirkung, in German), Farbe und Lack, 7, 595–598, 1977. 21. Anonymous, Plastguide, SCS Dukadan AS, Randers, Denmark, 1990. 22. Anonymous, Fluid Resistance of Viton®, Du Pont Company, Polymer Products Department, Elastomers Division, Wilmington, DE, 1989. 23. Anonymous, RYTON® PPS Polyphenylene Sulfide Resins — Corrosion Resistance Guide, Philips Petroleum Co., U.S. 24. Anonymous, Ultem® Resin Design Guide, GE Plastics, Pittsfield, MA, 1989. 25. Anonymous, Verhalten von Ultrason® gegen Chemikalien — BASF Technische Information TI-KTE/TH-01 d 82132, October 1991. 26. Anonymous, Expanded List — Chemical Resistance of Halar® Fluoropolymer, Ausimont, USA, Inc. 27. Carlowitz, B., Thermoplastic Plastics (in German), Thermoplastische Kunststoffe, Zechner und Hüthig, Speyer am Rhein, 1980. 28. Anonymous, Chemical Resistance Data Sheets Volume 1 Plastics; Volume 2 Rubbers, new ed., Rapra Technology Limited, Shawbury, Shrewsbury, Shropshire, 1993. 29. Anonymous, Chemical Resistance of Plastics and Elastomers used in Pipeline Construction, George Fischer +GF+, 1992. 30. Plastics Design Library, Chemical resistance data, William Andrew, Inc., Norwich, NY. 31. Hansen, C.M., Solvent Resistance of Polymer Composites — Glass Fibre Reinforced Polyphenylene Sulfide, Centre for Polymer Composites (Denmark), Danish Technological Institute, Taastrup, 1993, pp. 1–62. 32. Hansen, C.M., New developments in corrosion and blister formation in coatings, Prog. Org. Coat., 26, 113–120, 1995. 33. Van Dyk, J.W., Paper presented at the Fourth Chemical Congress of America, New York, August 25–30, 1991. 34. Anonymous [Note: This was, in fact, Van Dyk, J.W., but this does not appear on the bulletin], Using Dimethyl Sulfoxide (DMSO) in Industrial Formulations, Bulletin No. 102, Gaylord Chemical Corp., Slidell, LA, 1992. 35. Hansen, C.M., The Resistance of PPS, PES and PA Polymer Composites to Temperature Cycling During Water Exposure, Centre for Polymer Composites (Denmark), Danish Technological Institute, Taastrup, 1994.

7248_C012.fm Page 242 Wednesday, May 23, 2007 11:14 AM

7248_C013.fm Page 243 Wednesday, May 23, 2007 11:18 AM

— Barrier 13 Applications Polymers Charles M. Hansen ABSTRACT The permeation coefficient, P, of a liquid or a gas through a polymer is given by the product of the diffusion coefficient, D, and the solubility coefficient, S: P = DS. S correlates with the Hansen solubility parameters (HSP). At low permeant concentrations D is a constant. However, as the permeant concentration increases, its plasticizing effect on the polymer becomes significant, and the diffusion coefficient increases markedly. This effect can be very significant. The successful correlations of permeation phenomena with HSP are thought to be largely a result of this exceptional dependence of D on the dissolved permeant. As the amount of permeant being dissolved increases with closer matches of the HSP for permeant and barrier polymer, the end result is that both S and D, and therefore P, are functions of the HSP match. HSP correlations are given for breakthrough times in chemical protective clothing, permeation rates through barrier polymers, and barrier polymer swelling. Both liquids and gases are treated. Absorption and diffusion in polymers is treated extensively in Chapter 16.

INTRODUCTION The permeation of a liquid or a gas through a polymer can be described by the relation P = DS

(13.1)

P, the permeation coefficient, is the product of the diffusion coefficient, D, and the solubility coefficient, S. The diffusion coefficient indicates how fast the permeant molecules can move through the polymer. The solubility coefficient indicates how much of the permeant can be dissolved in the polymer. The amount dissolved in the polymer determines the concentration gradient over a film, and the concentration gradient is the driving force for mass transport. When solubility is higher, the concentration gradient is correspondingly higher, and, assuming the same diffusion coefficient, mass transport will be proportionately higher. S will be lower when the HSP of the barrier film and a solvent are very different. A significant factor affecting D is the molecular size and shape of the permeant molecules. Larger molecular size and more complex and bulky molecular shape are major factors that lead to lower diffusion coefficients. The diffusion coefficient for oxygen in polyvinyl chloride (PVC) is well over a million times greater than that of n-hexane (at low concentrations) in the same polymer.1 This difference in diffusion coefficients is a result of differences in molecular size. Likewise, it has been found that the rate of diffusion at the same concentration is about the same for different solvents with approximately the same size and shape, even though they may have different solubility parameters (but not so different that both are able to dissolve in the polymer at the level of comparison).2–4 The polymers in these studies were a copolymer of 87% vinyl chloride and 13% vinyl acetate, polyvinyl acetate, and polymethyl methacrylate.

243

7248_C013.fm Page 244 Wednesday, May 23, 2007 11:18 AM

244

Hansen Solubility Parameters: A User’s Handbook

CONCENTRATION-DEPENDENT DIFFUSION Low molecular weight liquids are plasticizers for polymers if they can be dissolved in them. Water, for example, can significantly soften many polymers even though it is dissolved to only a few percent. The low molecular weight materials can greatly reduce the glass transition temperature of their mixtures with a polymer as they have considerably more free volume associated with them than the polymers themselves. This extra free volume allows easier polymer segmental motion. The diffusion of the smaller species (and other species) becomes faster as their local concentration and plasticizing effect become greater. The solvent diffusion coefficient data in Figure 13.1 were first presented in Reference 3. See also Chapter 16. This figure shows diffusion coefficients for several solvents in polyvinyl acetate (PVAc) at 25°C. The diffusion coefficient for water shown in the figure was found by absorption and desorption experiments in thin films where a correction for the surface resistance was also required.5 See Chapter 16. It can be seen in this figure that for moderate solvent concentrations in this rigid polymer, the local diffusion coefficient increases by a factor of about 10 for an increase in solvent concentration of about 3 to 4 vol%. As this behavior is general for solvents in polymers, a rule of thumb indicates that the local diffusion coefficient for solvents in rigid polymers can increase by a factor of about one million when about 20 vol% solvent is present compared with the solvent-free state.3–7 This rule of thumb assumes that the polymer behaves as a rigid polymer over the concentration range being considered. This difference corresponds to the speed of a snail in the woods compared with a modern jet airliner.

7

-LOG DIFFUSION COEFFICIENT

CM2/SEC

A 8 B C

9

D 10 11 12 13 14 15

0

8 16 24 VOLUME FRACTION PENETRANT

32

FIGURE 13.1 Diffusion coefficients in polyvinyl acetate at 25°C for methanol (A), ethylene glycol monomethyl ether (B), chlorobenzene (C), and cyclohexanone (D). Original data are in Reference 3. The data point for water (*) is included for comparison. (From Hansen, C.M., Permeability of polymers, Pharmaceutical and Medical Packaging 98, 1998, 7.12 With permission.)

7248_C013.fm Page 245 Wednesday, May 23, 2007 11:18 AM

Applications — Barrier Polymers

245

Concentration-dependent diffusion coefficients are also found for elastomers. Here, the rule of thumb is that the diffusion coefficient increases by a factor of about 10 for an increase in solvent concentration of about 15 vol%.7 This shows that liquid contact with chemical protective clothing, for example, leads to concentration-dependent diffusion coefficients because the amount taken up at the contact surface on liquid contact is very often more than 15%. Concentration-dependent diffusion has been discussed at length by Crank.8 It is also discussed here because it is a major factor in the success of HSP correlations of permeation phenomena. The Crank-Nicholsen finite difference treatment for concentration-dependent diffusion8 was extended by Hansen3 and used to describe film formation by solvent evaporation,4 to explore what is termed anomalous diffusion,5 to develop an easy method to evaluate data leading to concentration-dependent diffusion coefficients,6 and to account for the effects of concentration-dependent diffusion and surface boundary resistance simultaneously.5–7 Klopfer9 developed analytical solutions involving concentration-dependent diffusion for many situations found in practical building applications, particularly with respect to transport of water in building materials. Concentration-dependent diffusion can be handled properly without great difficulty for most situations of practical interest. Neglect of this effect can lead to errors, the significance of which will increase with increasing amounts of the dissolved materials. In addition to demonstrating concentration dependence, the diffusion coefficient data for PVAc in Figure 13.1 also show the well-established relations that those solvents with larger and more complicated chemical structures are those with lower diffusion coefficients. Water has one “significant” atom, methanol has two, and ethylene glycol monomethyl ether (EGMME) has five. The diffusion coefficient for water in PVAc at low concentration, Do, is 10,000 times larger than that for the latter. An example of how to estimate diffusion coefficients in PVAc for other liquids, such as methylene chloride, is as follows. The diffusion coefficients in PVAc for methylene chloride, with three significant atoms, can be expected to be somewhat lower than those for methanol, but much higher than those for EGMME. Planar chlorobenzene diffuses more rapidly than nonplanar cyclohexanone, even though the number of significant atoms is the same. Another type of comparison which is possible is to state that the diffusion coefficients for toluene are expected to be close to those for chlorobenzene because of a similarity in molecular size and shape. This was confirmed by solvent retention studies where toluene and chlorobenzene were retained in identical amounts in a film of VYHH® (87 wt% vinyl chloride, 13 wt% vinyl acetate, Union Carbide). Toluene, which does not dissolve this polymer, was introduced by placing a completely dry polymer film in a closed container over toluene vapors. Diffusion can be expected to be slower in more rigid polymers, i.e., those with higher glass transition temperatures, unless the rigidity is such as to allow decided holes of suitable size to enable quite rapid diffusion of much smaller molecules. These considerations lead to the best combination of properties for a barrier polymer as being one with a high glass transition temperature and with HSP far removed from those of the permeant. If, in practice, this leads to water sensitivity, an alternate strategy, such as a laminated system, may be required.

SOLUBILITY PARAMETER CORRELATIONS BASED ON PERMEATION PHENOMENA SOLUBILITY PARAMETER CORRELATIONS

OF

BREAKTHROUGH TIMES

Extensive permeation studies and collections of permeation data are available within the chemical protective clothing industry.10,11 Such data can also be used to establish correlations with HSP. A list of HSP for barrier polymers used in chemical protective clothing has been published12 based on data by Forsberg and Olsson.10 Some of these correlations have been improved in most instances by correlating the more extensive data of Forsberg and Keith.11 The definition of a “good” solvent which was used for these correlations was that the breakthrough time was less than some selected

7248_C013.fm Page 246 Wednesday, May 23, 2007 11:18 AM

246

Hansen Solubility Parameters: A User’s Handbook

TABLE 13.1 HSP Correlations of Breakthrough Times for Barrier Polymers Typically Used in Chemical Protective Clothing. Units are MPa1/2 Type

δD

δP

δH

Ro

V Limits

FIT

No.

Neoprene® Neoprene Butyl Viton® Nitrile Challenge 5100® PE

16.0 19.0 17.0 15.6 19.8 16.6 16.5

8.8 8.0 1.5 9.6 13.3 7.0 2.7

4.0 0.0 0.0 7.8 2.2 3.8 6.1

10.1 13.2 7.3 7.1 13.6 2.3 7.9

None 71.0/172 71.0/175.8 72.6/148.9 84.3/177.2 None None

0.574 1.000 0.902 0.896 0.907 0.925 0.969

66 50 86 77 58 116 32

Note: “Good” solvents in these correlations have breakthrough times of less than 1 h.

value, either 20 min, 1 h, or 4 h. Table 13.1 includes some of these improved HSP correlations based on a 1-h breakthrough time for commercial film thickness. HSP alone cannot always correlate barrier properties unless comparisons are limited to solvent molecules with approximately the same size (and shape). This, of course, means that the diffusion coefficients at the reasonably low concentrations expected in better barrier polymers do not vary too greatly from each other. In many cases, satisfactory correlations could only be found when the differences in HSP between the permeant and the barrier polymer were combined with a size (and shape) parameter(s). The molecular volume, V, was found to be a reasonably successful single parameter for this purpose. Printing the correlation data arranged in increasing order of permeant V clearly showed whether the molecular size was important. With regard to the protective ability of the different garments, it was found that, in general, and as expected, the solvents with larger, more complicated structures required much longer times for breakthrough for a given protective membrane type than comparison with other solvents would indicate. Outliers were usually these larger molecular species and permeants with smaller or more linear structure, where diffusion is much more rapid than expected in average comparisons. This size effect is in agreement with what has been known about solvent retention in coatings2–4,12,13 and what has been discussed previously. An excellent example of this type of improved correlation is included in Table 13.1 for the breakthrough times of less than 1 h for neoprene rubber used in chemical protective clothing. The first correlation for this material listed in Table 13.1 gives a very poor data fit (0.574). There were 46 out of 66 liquids which had breakthrough times shorter than 1 h. It is clear from closer analysis of the details of the correlation that the outliers are methanol, carbon disulfide, and alkyl alcohol with shorter breakthrough times than predicted and the phthalate plasticizers which have longer breakthrough times than predicted. A perfect fit is found when the molecular volume range of the permeants included in the correlation is abbreviated to between 71 and 172 cc/mol. This eliminates these “outliers.” This correlation is based on 39 liquids with breakthrough times of less than 1 h out a total of 50. The HSP correlations for 1-h breakthrough times for other barrier polymers discussed in Table 13.1 give polymer HSP in the range of those expected from their composition. This includes butyl rubber, Viton® (The Du Pont Company), nitrile rubber, Challenge® 5100 (Chemical Fabrics Corporation, Merrimack, NH), and polyethylene (PE). The thicknesses of all of the films discussed here are those commonly used in chemical protective clothing. HSP correlations of the swelling of Viton are discussed here as well as in Chapter 5 and by Zellers,14,15 Evans and Hardy,16 and by Nielsen and Hansen.17

7248_C013.fm Page 247 Wednesday, May 23, 2007 11:18 AM

Applications — Barrier Polymers

247

D, P, H, R = 16.6, 5.4, 4.0, 3.8 FIT = 0.997 FOR 68 < MV < 98

150

BREAKTHROUGH TIME “=” “NO=”

■ BTC ● BCN

NONE

< 3 HR

■

> 3 HR

■

Evaluation uncertain

● ● x

● MIK

MOLAR VOLUME

BCL

100

MSO ■ ● STY ● CHA ● EAC ■ TCE EET MMA ■ ●● NTB ● TCR ●CHK DEN●● CRB ● ETA● TTE ■ ATC DOUB– EVE ■ ● ACA MAM BNZ ■ LE CRPVAM ■TCL■●MEK● BTR ■ ● ANL● ● DOX MAN ■ THF● PRA BOND ■ALC ● FFA EPC ● PYR ■ MVK ■ CRF●● VOC ■ EDC● ● ACI ALN ● DMF ACB● ● ■ ALM ● ACE POX●■ ACN ■ DCM● ARL MIC● CBB● ● AAC ● AAD ● MBR ALL

50

0.0

1.0

2.0 3.0 RED NUMBER

■ ALA ATN ● ● NME

4.0

5.0

FIGURE 13.2 Graphical method to present HSP correlations. The data are plotted using permeant molar volume vs. RED number. HSP correlation for breakthough times of less than 3 h in Challenge 5100. Symbols used are explained in Table 13.2. (Reprinted from Hansen, C.M. et al., The Performance of Protective Clothing: Fourth Volume, ASTM STP 1133, McBriarty, J.P. and Henry, N.W., Eds., American Society for Testing and Materials, Philadelphia, 1992, 906. With permission. Copyright ASTM.)

The RED number (Chapters 1 and 2) is a key parameter to judge solvent quality. This is given in HSP correlations using Chapter 1, Equation 1.10 as Ra (Chapter 1, Equation 1.9), the difference in HSP between a solvent and a polymer, divided by the radius of the correlating sphere, Ro. The radius of the sphere is actually determined as the difference in HSP of the “worst” good solvent(s) and the HSP for the polymer (which is the center point of the sphere). RED numbers near zero indicate very good solvents (rapid breakthrough). RED numbers increase as the solvent quality decreases. For RED numbers greater than 1.0, the solvent quality is considered “bad,” although swelling may still occur. Figure 13.2 shows one way to graphically use RED numbers to present data from HSP correlations of permeation phenomena. In this correlation, “good” permeants have breakthrough times of less than 3 h. The data are plotted using V vs. solvent–polymer affinity, i.e., the RED number.18 The barrier material, Challenge 5100, is a fluoropolymer supported by fiberglass. The abbreviations for the permeants in Figure 13.2 are explained in Table 13.2. This correlation shows that molecules with molar volumes greater than about 100 cc/mol will not have breakthrough times of less than 3 h, regardless of RED number. Molecules with molar volumes greater than about 75 cc/mol require a terminal double bond and lower RED numbers to breakthrough under these conditions. Molecules with still lower molar volumes appear to come through with only a slight dependence on the RED number. The effect with the terminal double bonds clearly indicates the preferential direction of motion for this type of molecule. The molecules in effect worm their way through the barrier polymer.

7248_C013.fm Page 248 Wednesday, May 23, 2007 11:18 AM

248

Hansen Solubility Parameters: A User’s Handbook

TABLE 13.2 List of Symbols Used in Figure 13.2 Symbol AAC AAD ACA ACB ACE ACI ACN ALA ALC ALM ALN ANL ARL ATC ATN BCM BCN BNZ BTC BTR BUT BZN CAC CBB CBT CCF CHA CHK CLA CLB CRB CRF CRP DCM DEN DMF DOX DSO EAC

Compound

Symbol

Acetic acid Acetaldehyde Acetic anhydride Acetyl bromide Acetyl chloride Acetone Acrylonitrile Allyl alcohol Allyl chloride Allyl amine Allyl cyanide Aniline Acrolein Allyl isothiocyanate Acetonitrile Bromochloromethane Butyl acetate Benzene Butyl acrylate Butyraldehyde Butane Benzonitrile Chloroacetylchloride Carbon disulfide Carbon tetrachloride Dichloromonofluoromethane (Freon 21) Cyclohexylamine Cyclohexanone Chloroacetone 1-Chlorobutane Chlorobenzene Chloroform Chloroprene Dichloromethane Diethylamine Dimethyl formamide 1,4-Dioxane Dimethyl sulfoxide Ethyl acrylate

EBR EDC EET EIM EPC ESH ETA EVE EVK F12 FFA HXA MAL MAM MAN MAT MBR MEK MES MIC MIK MMA MSO MVK NEE NME NTB POX PRA PYR STY TCE TCL TCR THF TOL TTE VAM VDC

Compound Ethyl bromide Ethylene dichloride Diethyl ether Ethyleneimine Epichlorohydrin Ethanethiol Ethyl acetate Ethyl vinyl ether Ethyl vinyl ketone Dichlorodifluoromethane (Freon 12) Furfural Hexane Methanol Methyl acrylate Methacrylonitrile Methyl acetate Methyl bromide Methyl ethyl ketone Methyl sulfide Methyl isocyanate Methyl isobutyl ketone Methyl methacrylate Mesityl oxide Methyl vinyl ketone Nitroethane Nitromethane Nitrobenzene Propylene oxide Propylamine Pyridine Styrene 1,1,2,2,-Tetrachloroethylene Trichloroethylene 1,1,1-Trichloroethane Tetrahydrofuran Toluene Tetrachloroethylene Vinyl acetate 1,1-Dichloroethylene

Source: Reprinted from Hansen, C.M. et al., The Performance of Protective Clothing: Fourth Volume, ASTM STP 1133, McBriarty, J.P. and Henry, N.W., Eds., American Society for Testing and Materials, Philadelphia, 1992, 903. With permission. Copyright ASTM.

SOLUBILITY PARAMETER CORRELATION

OF

PERMEATION RATES

Permeation rates for different permeants in a polymer can also be correlated to find HSP for the polymer. This is done by dividing a data set into two groups. The “good” solvents will have permeation rates greater than an arbitrarily selected value, and the “bad” solvents will have

7248_C013.fm Page 249 Wednesday, May 23, 2007 11:18 AM

Applications — Barrier Polymers

249

TABLE 13.3 HSP Correlations Related to Barrier Polymers Material

δD

δP

δH

Ro

FIT

G/T

LDPE permeation coefficient 21.1°C Permeation viable skina PP swellingb ACLAR® 22 >5 wt% swelling ACLAR® 22 2%<swelling<5% Psoriasis scales swelling (Chap. 14) Viton® swellc >10 wt% 20°C EVOH sol/swelld Polyvinyl chloride swelle Cellophane — >25% swell PETP chemical resistance (+/– OK)f PA6/PA66 chemical resistance (+/– OK)g PA6/PA66 chemical resistance (+/– bad)g

16.3 17.6 18.0 14.7 18.0 24.6 13.1 20.5 18.7 16.1 18.2 17.0 18.9

5.9 12.5 3.0 3.9 1.0 11.9 13.7 10.5 9.7 18.5 6.4 3.4 7.9

4.1 11.0 3.0 6.7 2.0 12.9 3.9 12.3 7.7 14.5 6.6 10.6 11.7

8.2 5.0 8.0 6.8 4.0 19.0 14.7 7.3 6.4 9.3 5.0 5.1 8.0

1.000 1.000 1.000 1.000 1.000 0.927 0.742 0.925 1.000 0.955 0.865 0.984 0.950

26/47 4/13 13/21 6/26 4/21 35/50 36/57 5/24 13/47 4/22 7/34 2/31 9/31

Note: Data fit and the number of good liquids (G) and total number of liquids (T) in the correlations are also indicated. Units are MPa1/2. a

This correlation is discussed in more detail in Chapter 15 and is based on limited data.20 This correlation is based on data by Lieberman and Barbe22 and is discussed in more detail in Chapter 5. c This correlation is discussed in Chapter 5. Swelling data is from Reference 23. d Ethylene vinyl alcohol copolymer (EVOH), four liquids dissolved and one (morpholine) swelled very strongly. e Visual observation of very strong swelling and/or solubility. f Polyethylene terephthalate (PETP) chemical resistance based on rather uncertain data 24 (see discussion in Chapter 12). Recommendation of uncertain-for-use is considered as acceptable for use. Attacking solvents are within correlating HSP sphere. g Polyamide 6/66 chemical resistance based on rather uncertain data 24 (see discussion in Chapter 12). Recommendation of uncertain-for-use is used as indicated. Attacking solvents are within the correlating HSP sphere.

b

Source: From Hansen, C.M., Permeability of polymers, Pharmaceutical and Medical Packaging 98, 7.6, 1998. With permission.

permeation rates lower than this value. Such a correlation based on permeation coefficients for various liquids in PE is included in Table 13.3. The permeation coefficient data, (g x mm)/(m2 x d), are reported by Pauly19 for low density polyethylene (LDPE). “Good” solvents are arbitrarily considered as those which have permeation coefficients in these units which are greater than 1.5 at 21.1°C. The parameters reported correlate the data well but are somewhat different from those which might be expected for a polyolefin. Reasons for this are not evident but may include additives in the polymer, local oxidation, or some other local variation in the composition of the polymer. It should be remembered that permeation occurs in the amorphous regions only. This is why high density PE is a better barrier polymer than low density PE; the higher densities are attributable to a higher percentage of crystallinity. A problem of some concern is the permeation through buried water pipes by chemicals or oil products which somehow reach them, either by general pollution or by gasoline or oil spills. One clearly expects more extensive permeation by chemicals that have HSP not too different from those of the polymer from which the pipe is made, all other things being equal. These pipes are often made from polyolefins.

7248_C013.fm Page 250 Wednesday, May 23, 2007 11:18 AM

250

Hansen Solubility Parameters: A User’s Handbook

A HSP correlation has been possible in a very special case of polymer permeability where the barrier polymer is viable human skin.20 This is discussed in more detail in Chapter 15. Human skin is a polymeric barrier with several functions, one of which is to help keep undesirable chemicals out of the body. Some chemicals readily permeate this boundary, and this fact has been used to establish a tentative HSP correlation for the permeation rate of viable human skin. This correlation also has a relation to the HSP correlation for the swelling of psoriasis scales,21 which is also discussed in Chapter 15.

SOLUBILITY PARAMETER CORRELATION OF POLYMER SWELLING Solubility is a major factor in the equation P = DS, so correlations of solvent uptake in polymers are important to understand their barrier properties. The correlation for swelling of polypropylene reported in Table 13.3 is based on solvent uptake data reported by Lieberman and Barbe.22 The limit of 0.5% weight gain was arbitrarily set to differentiate “good” solvents from “bad” ones. A different limit might give different parameters. The HSP found in a given correlation of swelling depends on which polymer segments the smaller amounts of permeant prefer to associate with. The predictive ability of the correlation will depend on the number of test liquids used in the study and their given HSP values. How different are the HSP of the test liquids? What are their values compared with the predictions desired? The parameters reported in Table 13.3 for polypropylene seem to accurately reflect what is expected in terms of low polarity and low hydrogen bonding properties for this type of polymer. As stated previously, a problem of some significance in any study of solvents at low concentrations in polymers is that the smaller amounts of solvent relative to the polymer can lead to preferential association of solvent with those local regions/segments/groups in the polymer that have energies (HSP) most similar to their own. Like seeks like. These local regions may not necessarily reflect the same affinities as the polymer as a whole, such as are indicated by the totally soluble-or-not approach most commonly used in HSP evaluations. These local association effects can influence results on swelling studies at low solvent uptakes in both good and bad solvents, for example. Copolymers, such as Viton, are particularly susceptible to this problem. Swelling data for Viton23 are correlated by the HSP values included in Table 13.3. A poor data fit can be anticipated when a single HSP sphere is used to describe what should be represented by (at least) two overlapping HSP spheres (see also Figure 13.3). Zellers14,15 also had difficulty correlating the swelling of Viton. Other types of studies carried out at low solvent concentrations can also be influenced by this segregation/association phenomena. An extension of this type of situation can be cited in the tendencies of water to associate with itself as well as with local hydrophilic regions within polymers. The amount of water taken up at equilibrium is not reflected by an overall HSP correlation of polymer solubility or swelling. As little as 1% of hydrophilic additive can effectively destroy the water barrier properties of a polymer film, but this small amount cannot be measured in swelling or solubility studies leading to HSP correlations. This fact, among other things, has made simple predictions of the behavior of water very difficult. Correlations of polymer solubility and swelling have led to several of the HSP data sets reported in Table 13.3 (see also the data reported in Chapter 5). The HSP correlations for chemical resistance based on data of the acceptable-for-use or recommended/not recommended type are not as reliable as those usually found for solubility and swelling where a suitably large number of liquids are used in the testing. The reasons for this are discussed in depth in Chapter 12. The data used in the chemical resistance correlations reported in Table 8.3 were taken from Reference 24.

7248_C013.fm Page 251 Wednesday, May 23, 2007 11:18 AM

Applications — Barrier Polymers

251

ACLAR®22 Bimodal Uptake HSP Correlations Uptake

δD

δP

δH

R

5 - 10 % 2-5% > 10 %

14.7 3.9 6.7 6.8 18.0 1.0 2.0 4.0 Trichloroethylene within both

FIT 1.000 1.000

● ■ ✕

12

δP, Polar Parameter

10

8

6

Ro = 4

6.8

(14.7)

2

(18.0)

Ro = 4.0

0 0

2

4

6

8

10

12

14

δH, Hydrogen Bonding Parameter FIGURE 13.3 Bimodal HSP correlation(s) for uptake of liquids in ACLAR® 22. Trichloroethylene uptake is the largest among the test solvents because it is the only solvent found within both regions. It has RED numbers of 0.999 for the >5% correlation and 0.978 for the correlation of uptake between 2 and 5%. As the data fit is 1.0 for both correlations, other sets of parameters can also give data fits of 1.0. However, the numbers are approximately correct. Units are MPa1/2.

SOLUBILITY PARAMETER CORRELATION OF PERMEATION COEFFICIENTS FOR GASES Gases can also be assigned δD, δP, and δH parameters. For strictly nonpolar gases, the values of δP and δH will be zero, but other gases, such as carbon dioxide, hydrogen sulfide, etc., will have significant values for all three parameters. Table 13.4 gives the δD, δP, and δH parameters for a number of gases. It is not surprising that there are HSP correlations of permeation coefficients for gases in different polymers as a function of their solubility parameter differences. One such correlation using the total solubility parameter has been given by König and Schuch,25 who showed

7248_C013.fm Page 252 Wednesday, May 23, 2007 11:18 AM

252

Hansen Solubility Parameters: A User’s Handbook

TABLE 13.4 HSP for Common Gases of Interest in Permeation Phenomena Gas

δD

δP

δH

Water Ammonia Chlorine Sulfur dioxide Carbon dioxidea Carbon monoxide Ethane Ethylene Helium Hydrogen Hydrogen sulfide Methane Nitrogen oxide Nitrogen Nitrous oxide Oxygen Acetylene

15.5 13.7 17.3 15.8 15.7 11.5 15.6 15.0 1.0 5.1 17.0 14.0 11.5 11.9 12.0 14.7 14.4

16.0 15.7 10.0 8.4 6.3 4.9 0 2.7 0 0 6.0 0 20.0 0 17.0 0 4.2

42.3 17.8 0.0 10.0 5.7 0 0 2.7 0 0 10.2 0 0 0 0 0 11.9

Note: Units are MPa1/2. a

Values changed from 1st Edition. See Chapter 10 Addendum.

that the better barrier polymers for oxygen, i.e., those with low oxygen permeation coefficients, were those whose solubility parameters were most different from the solubility parameters of oxygen. The better barrier polymers for oxygen include polyacrylonitrile and polyvinyl alcohol, whereas the poorer barriers include polyolefins and polytetrafluoroethylene (PTFE). The amounts of gases dissolved at low pressures are usually low, and constant diffusion coefficients are expected. This may not be true at higher pressures where solubility parameters for the gases increase more rapidly than those of the polymers and polymers can absorb them to a greater extent. See Chapter 10. An example of how HSP principles can be applied to interpreting the behavior of gas barrier films can be found in the performance of poly(chlorotrifluoroethylene). The data on which the example is based are taken from the commercial literature supplied by Allied Signal concerning their barrier films under the tradename of ACLAR®.26 These films are excellent barriers for water and oxygen, and various laminating possibilities exist, including polyethylene, polyvinyl chloride, and polyethylene terephthalate. The barrier properties of films made from this material are not nearly as good for carbon dioxide as they are for nitrogen or oxygen. A contributing factor in this is that the HSP of the polymer is somewhat different from the HSP of oxygen and nitrogen, but close to the HSP of carbon dioxide. A HSP correlation for the swelling of ACLAR 22 to greater than 5 wt% is included in Table 13.3. The RED numbers for water, nitrogen, oxygen, and carbon dioxide based on this correlation are 5.5, 1.4, 1.1, and 0.48, respectively. Nitrogen has slower permeation than oxygen, and both are much slower than carbon dioxide, in general agreement with this ranking. One might have expected the permeation rate of carbon dioxide to be lower than that of nitrogen and oxygen as it is a larger molecule, but the enhanced solubility of carbon dioxide in the polymer overrides this expectation.

7248_C013.fm Page 253 Wednesday, May 23, 2007 11:18 AM

Applications — Barrier Polymers

253

Figure 13.3 shows that there are two distinct spherical characterizations possible for ACLAR 22. The first of these, as discussed earlier, is for liquid uptake to greater than 5% by weight. The second of these is for uptake between 2 and 5% by weight. This second correlation is also reported in Table 13.3. There is one liquid in the data which is common to both of the HSP regions pictured in Figure 13.3. This is trichloroethylene (which was assumed to be a good solvent in both of the perfect correlations in spite of being absorbed to over 10% by weight). Even though trichloroethylene has high RED numbers in both correlations, this solvent is absorbed more than any of the other solvents tested because of this property. The primary uptake region has HSP that might be expected from a fluoropolymer, whereas the secondary HSP region is what might be expected from a chlorinated species. Such secondary regions can potentially allow higher permeation rates and greater absorption of unpredictable materials based on a single HSP correlation. Searching a database of solvents, plasticizers, aromatic compounds, etc., would show which of these could behave in an unexpected manner. Sometimes an indirect approach allows prediction of the uptake of a gas in a polymer. This involves determining the uptake of the gas in a liquid having solubility parameters that are similar to those of the polymer. This approach expands the usefulness of gas–liquid equilibrium data. Correlations of gas–liquid solubility with the solubility parameter are included in Figure 13.4 for the equilibium values for water27 and in Figure 13.5 for the equilibrium values for nitrogen.28 The quantity P*y/x is given by the total pressure, P*, the mole fraction in the gas phase, y, and the mole fraction in the liquid phase, x. The abbreviations used in Figure 13.5 are explained in Table 13.5. The solubility parameters for gases not found in Table 13.4 may be found in standard references.29–31 HSP for gases can be calculated using the procedures outlined in Chapter 1 with the special figure for gases included in Chapter 18 (Figure 18.2). There can be problems of dividing the total cohesive energy into three parts. Sulfur trioxide is a good example of how one cannot come further in dividing the energy of vaporization into components without experimental data. The techniques of Chapter 3 have not been explored in this context however. The total (Hildebrand) solubility parameter, δt, indicated from the total energy of vaporization is 31.3 MPa1/2. δD found by the usual techniques is 15.6 MPa1/2. This leaves a residual corresponding to a solubility parameter value of 27.2 MPa1/2 to be distributed between permanent dipole and hydrogen bonding (electron transfer) effects. There is no dipole moment, and neither is there a hydrogen atom. This clearly requires experimental data to resolve the distribution of the energy of vaporization into components for all of these effects, even if there are supporting estimates from the techniques of Chapter 3. It might be noted that the scale in Figure 13.4 for the uptake of water in various liquids is exponential with data covering almost five decades in concentration. The phenomena correlated in this figure confirm the expectation that nonpolar polymers, with solubility parameters far different from those of water, will be good barrier polymers for water because of low water solubility. This is generally true, of course. As mentioned earlier, such polymers include the polyolefins as well as chlorinated and fluorinated polymers. These comments and generalities are not necessarily valid for polymers containing additives. Depending on the nature of the additive and the amounts present, some of these can totally change the barrier performance of the base polymer.

LAMINATES Laminated barrier polymer systems are designed to make the best use of the properties of each of the individual layers, as well as to optimize cost with performance. The most common type has a polyolefin on the exterior surfaces to protect the inner barrier polymer from water. These interior barrier polymers often have relatively high solubility parameters, such as ethylene vinyl alcohol copolymers (EVOH), polyamides (PA), or polyethylene terephthalate (PET). If the inner barrier polymer takes up water, it will be plasticized, and its barrier properties will be reduced. A polyolefin

7248_C013.fm Page 254 Wednesday, May 23, 2007 11:18 AM

254

Hansen Solubility Parameters: A User’s Handbook

D P H 1 % IN WATER 15.1 20.4 16.5 H from Perry: “CHEMICAL ENGINEERS’ HANDBOOK” (1963). 106

H6

105

N2  CO

H2S  NO

C2H4

104

H = P+ y/x

H2

O

 2 CH4  C2H6



 N2O  CO2

 C2H2

103

 C12

102  SO2

101

 NM3

100 0

10

20

30

40

DISTANCE TO “1% IN WATER” MPa½ FIGURE 13.4 HSP correlation of gas–water equilibrium data using water HSP values found with a correlation using a limit of >1% liquid soluble in water as a “good” solvent,21 and gas–water equilibrium data from Perry et al.27 P* is the total pressure, y is the mole fraction in the gas phase, and x is the mole fraction in the liquid phase. (From Hansen, C.M., Dan. Kemi, 73(8), 21, 1992. With permission.)

laminated to such a potentially water-sensitive barrier film can significantly delay this uptake and loss of barrier properties and maintain reasonable costs. Depending on the performance desired, various combinations of laminates can be systematically designed using HSP considerations as one of the design parameters.

7248_C013.fm Page 255 Wednesday, May 23, 2007 11:18 AM

Applications — Barrier Polymers

255

5000 NITROGEN

D = 11.9, P = 0.0 H = 0.0, RAD. = 1.0

4000

MTA

3000

t

P y⁄x

ETA

2000 BEN ETY

MTC HEP

1000

PEN

ISA

DEC HEX

PTF PFM PFH

0 0

1

RED ⁄ 10

2

3

FIGURE 13.5 HSP correlation of nitrogen–liquid equilibrium data at temperatures near 25°C and low pressure. P* is the total pressure, y is the mole fraction in the gas phase, and x is the mole fraction in the liquid phase. (From Hansen, C.M., Dan. Kemi, 73(8), 20, 1992. With permission.)

GENERAL CONSIDERATIONS HSP correlations have been possible for many phenomena where differences in behavior on contact with different solvents have been studied. The HSP correlations are preferably based on systems in thermodynamic equilibrium, although the correlations presented previously on breakthrough times are an exception to this. These correlations were possible because of the exceptional dependence of the permeation phenomena on the amount of permeant being dissolved. There is a strong dependence of the diffusion coefficient of permeants in polymers on their size and shape. This can clearly affect HSP correlations of permeation coefficients, as two permeants with identical HSP will have different D if their sizes and shapes are significantly different. This

7248_C013.fm Page 256 Wednesday, May 23, 2007 11:18 AM

256

Hansen Solubility Parameters: A User’s Handbook

TABLE 13.5 Key to Symbols Used in Figure 13.5 Symbol PFH PTF PFM PEN HEX HEP MYC ISA ETY BEN ETA MTA

Compound Perfluoroheptane Perfluorotributylamine Perfluoromethylcyclohexane Pentane Hexane Heptane Methyl cyclohexane Isobutyl acetate Ethyl acetate Benzene Ethanol Methanol

differential in diffusion rate based on solvent size and shape can also give apparent errors in HSP correlations of polymers for their chemical resistance, for example, where not enough exposure time has been allowed for attainment of equilibrium. This is clearly a problem in the determination of equilibrium degree of swelling and low amounts of uptake. This problem has also been found, for example, for exposures of thick samples (3 to 4 mm) of rigid polymers used for tensile testing after solvent exposure for given times. Crystalline polymers also have a tendency to be more readily soluble in solvents with lower V, all other parameters being equal, but this is explained by thermodynamic considerations rather than a relatively faster diffusion process. In all of these cases, the majority of the outliers in the correlations are the test liquids with higher V. The time required for attainment of equilibrium with the larger diffusing molecules can be so long as to be prohibitive for their reasonable inclusion in HSP correlations. It is suggested that diffusion rates be carefully considered when liquids with very high V are outliers in HSP correlations. Neglecting such data for the sake of a correlation can be justified, but this is a warning that the diffusion process for these liquids has not yet achieved equilibrium and that the effects of such liquids can be expected to be more severe at still longer times than those used in the study. HSP correlations can be used in this way to find those exposure liquids which have not reached equilibrium at the exposure time chosen for evaluations.

CONCLUSION Successful HSP correlations for the permeation and solubility behavior of selected barrier polymers have been presented to demonstrate the use of simple principles to arrive at optimum barrier systems, as well as to determine reliable HSP for the polymers studied. Selection of polymer–permeant combinations with widely different solubility parameters will ensure low solubility of the permeant in the polymer. This reduces the concentration gradient and prevents significant self-plasticization of the polymer. The self-plasticization leads to concentration-dependent diffusion coefficients, an effect which becomes more significant with increasing amounts of permeant being dissolved, i.e., a closer HSP match. See Chapter 16 for more discussion of diffusion in polymers. HSP correlations have been presented for breakthrough times in chemical protective clothing, permeation rates in barrier polymers, and swelling of various types of polymers. Both gases and liquids are treated.

7248_C013.fm Page 257 Wednesday, May 23, 2007 11:18 AM

Applications — Barrier Polymers

257

REFERENCES 1. Rogers, C.E., Permeation of gases and vapours in polymers, in Polymer Permeability, Comyn, J., Ed., Elsevier Applied Science, London, 1985, pp. 11–73. 2. Hansen, C.M., Some aspects of the retention of solvents in high polymer films, Färg och Lack, 10(7), 169–186, 1964. 3. Hansen, C.M., The Three Dimensional Solubility Parameter and Solvent Diffusion Coefficient, Doctoral dissertation, Danish Technical Press, Copenhagen, 1967. 4. Hansen, C.M., A mathematical description of film drying by solvent evaporation, J. Oil Colour Chem. Assoc., 51(1), 27–43, 1968. 5. Hansen, C.M., Diffusion in polymers, Polym. Eng. Sci., 20(4), 252–258, 1980. 6. Hansen, C.M., The measurement of concentration-dependent diffusion coefficients — the exponential case, Ind. Eng. Chem. Fundam., 6(4), 609–614, 1967. 7. Hansen, C.M., Diffusion coefficient measurements by solvent absorption in concentrated polymer solutions, J. Appl. Polym. Sci., 26, 3311–3315, 1981. 8. Crank, J., The Mathematics of Diffusion, Oxford University Press, Oxford, 1956. 9. Klopfer, H., Water Transport by Diffusion in Solid Materials (Wassertransport durch Diffusion in Feststoffen, in German), Bauverlag GMBH, Wiesbaden, 1974. 10. Forsberg, K. and Olsson, K.G., Guidelines for Selecting Chemical Protective Gloves, (Riktlinjer för val av kemiskyddshandskar, in Swedish), Förening Teknisk Företagshälsovård (FTF), Stockholm, 1985. 11. Forsberg, K. and Kieth, L.H., Chemical Protective Clothing Performance Index, 4th ed., Instant Reference Sources, Austin, TX, 1991. 12. Hansen, C.M. and Hansen, K.M., Solubility parameter prediction of the barrier properties of chemical protective clothing, Performance of Protective Clothing: Second Symposium. ASTM STP 989, Mansdorf, S.Z., Sager, R., and Nielsen, A.P., Eds., American Society for Testing and Materials, Philadelphia, PA, 1988, pp. 197–208. 13. Hansen, C.M., The free volume interpretation of plasticizing effectiveness and diffusion in high polymers, Off. Dig., 37(480), 57–77, 1965. 14. Zellers, E.T., Three-dimensional solubility parameters and chemical protective clothing permeation. I. Modelling the solubility of organic solvents in Viton gloves, J. Appl. Polym. Sci., 50, 513–530, 1993. 15. Zellers, E.T., Three-dimensional solubility parameters and chemical protective clothing. II. Modelling diffusion coefficients, breakthrough times, and steady-state permeation rates of organic solvents in Viton gloves, J. Appl. Polym. Sci., 50, 531–540, 1993. 16. Evans, K.M. and Hardy, J.K., Predicting solubility and permeation properties of organic solvents in Viton Glove material using Hansen’s solubility parameters, J. Appl. Polym. Sci., 93, 2688–2698, 2004. 17. Nielsen, T.B. and Hansen, C.M., Elastomer swelling and Hansen solubility parameters, Polym. Testing, 24, 1054–1061, 2005. 18. Hansen, C.M., Billing, C.B., Jr., and Bentz, A.P., Selection and use of molecular parameters to predict permeation through fluoropolymer-based protective clothing materials, in The Performance of Protective Clothing: Fourth Volume, ASTM STP 1133, McBriarty, J.P. and Henry, N.W., Eds., American Society for Testing and Materials, Philadelphia, PA, 1992, pp. 894–907. 19. Pauly, S., Permeability and diffusion data, in Polymer Handbook, 3rd ed., Branderup, J. and Immergut, E.H., Eds., Wiley-Interscience, New York, 1989, pp. VI/445–446. 20. Ursin, C., Hansen, C.M., Van Dyk, J.W., Jensen, P.O., Christensen, I.J., and Ebbehoej, J., Permeability of commercial solvents through living human skin, Am. Ind. Hyg. Assoc. J., 56, 651–660, 1995. 21. Hansen, C.M. and Andersen, B.H., The affinities of organic solvents in biological systems, Am. Ind. Hyg. Assoc. J., 49(6), 301–308, 1988. 22. Lieberman, R.B. and Barbe, P.C., Polypropylene polymers, in Encyclopedia of Polymer Science and Engineering, 2nd ed., Vol. 13, Mark, H.F., Bikales, N.M., Overberger, C.G., Menges, G., and Kroschwitz, J.I., Eds., Wiley-Interscience, New York, 1988, pp. 482–483. 23. Anonymous, Fluid Resistance of Viton®, Du Pont Company, Polymer Products Department, Elastomers Division, Wilmington, DE, 1989. 24. Anonymous, Plastguide, SCS Dukadan AS, Randers, Denmark, 1990.

7248_C013.fm Page 258 Wednesday, May 23, 2007 11:18 AM

258

Hansen Solubility Parameters: A User’s Handbook

25. König, U. and Schuch, H., Structure and permeability of polymers (Konstitution und Permeabilität von Kunststoffen, in German), Kunststoffe, 67(1), 27–31, 1977. 26. Anonymous, ACLAR® Barrier Films, AlliedSignal — Advanced Materials, Allied Signal Inc., Morristown, NJ. 27. Perry, J.H., Chilton, C.H., and Kirkpatrick, S.D., Eds., Chemical Engineers’ Handbook, 4th ed., McGraw-Hill, New York, 1963, Section 14, pp. 2–7. 28. Hansen, C.M., 25 years with solubility parameters (25 År med Opløselighedsparametrene, in Danish), Dan. Kemi, 73(8), 18–22, 1992. 29. Hildebrand, J. and Scott, R.L., The Solubility of Nonelectrolytes, 3rd ed., Reinhold, New York, 1950. 30. Hildebrand, J. and Scott, R.L., Regular Solutions, Prentice-Hall Inc., Englewood Cliffs, NJ, 1962. 31. Barton, A.F.M., Handbook of Polymer-Liquid Interaction Parameters and Solubility Parameters, CRC Press, Boca Raton, FL, 1990.

7248_C014.fm Page 259 Wednesday, May 23, 2007 11:24 AM

— Environmental 14 Applications Stress Cracking in Polymers Charles M. Hansen ABSTRACT Hansen solubility parameters (HSP) can be used to help predict which chemicals can cause environmental stress cracking (ESC) in polymers. ESC requires tensile stress and correlates when the RED number (relative energy difference in the polymer-solvent interaction) found from HSP considerations is plotted vs. a molecular size parameter, the molar volume, V. There are three distinct regions on this plot. There is a region at low RED including those challenge liquids that dissolve the polymer or are very aggressive, and ESC is not found as such. There is a region at high RED where the absorption is zero or not great enough to matter or else the absorption rate is slow enough to allow relaxation of the polymer in preference to ESC. ESC can occur in an intermediate region where there is some absorption of challenge liquid, although examples are given where ESC takes place without measurable absorption for good matches in HSP at relatively high stress/strain. The ESC region on these plots increases in size with increased tensile stress and/or increased critical strain.

INTRODUCTION The previous chapter dealt with the chemical resistance of polymers and briefly touched on environmental stress cracking (ESC) as one aspect of chemical attack. This chapter expands the previous discussion of this special type of physical chemical attack on polymers. This form of failure represents at least 25% of all failures in plastics and therefore deserves special attention.1 A failure by ESC can appear almost immediately, after minutes, after hours, or even after years, and often occurs without prior warning. There is a considerable literature on ESC. An excellent general source of the ESC literature is Wright’s encompassing book on failures in polymers in general.1 The present chapter is in many ways a supplement to this excellent work. It is now clear that the polymer and the chemical that initiates the stress cracking must have similar or reasonably similar HSP. The ESC initiator need not dissolve the polymer as such. It has generally been assumed that some similarity in HSP is required such that some absorption occurs. More recently it has been recognized that measurable absorption is not always necessary for ESC to occur in some polymers. It is theorized that some physical movement, such as rotation of the polymer chain segments at the contact surface, can initiate the cracking process. In the latter case the similarity of the HSP of the challenge liquid may be to the HSP of an entity in the polymer chain that otherwise might be oriented away from the polymer surface. Under given conditions of strain it may prefer the environment of the ESC initiator once there is contact with the polymer. An understanding of this phenomenon seems to be evolving, but it is not complete as discussed below. As stated above, the generally accepted mechanism for ESC has been that some absorption of the active chemical weakens the polymer structure locally, such that the tensile stress increases in adjacent regions. The tensile stress must be sufficient to locally pull polymer chains or segments of chains from the bulk. As the structure weakens locally, there is added stress in adjacent regions. This may be sufficient to cause a craze or crack, but in many cases further absorption or further chemical penetration seems necessary to repeat the same process until finally the stress becomes 259

7248_C014.fm Page 260 Wednesday, May 23, 2007 11:24 AM

260

Hansen Solubility Parameters: A User’s Handbook

too high and a craze or crack occurs. The number of chemicals giving ESC in a given polymer increases with increases in the level of stress/strain. The critical strain is that minimum value of strain at which the given challenge chemical will cause crazing/cracking in the given polymer. Any strain levels above this will result in ESC on contact with the given chemical. In some cases absorption of liquid permits stress relaxation, and expected crazing and cracking does not occur. This is presumably for a more or less massive uptake but not enough to completely dissolve the polymer. On the other hand, the polymer is plasticized and/or weakened by the absorption of these chemicals, and there is potentially a different type of problem to cope with. If a chemical reaction is involved, such as with acids and bases, then the phenomenon is more properly called stress corrosion cracking (SCC). Wright indicates that about 90% of ESC failures involve amorphous thermoplastic polymers with the remaining being found with partly crystalline thermoplastic polymers. Lustiger2 lucidly describes the mechanism of ESC in partly crystalline polymers with polyethylene as an example. There are three types of polymer chains to consider in this case. Those with free ends (cilia) extending into the amorphous phase, those with chains extending into the amorphous phase but that loop back into the same lamella (loose loops), and tie-molecules or chains that extend from one lamella and anchor into an adjacent one. It is such tie-molecules that give the ultimate resistance to ESC. One must either break the tie molecules or pull them out of one of the lamella. In a sense these act like strong physical cross-links. Similar considerations are valid for amorphous thermoplastic polymers. Higher molecular weight enhances polymer chain entanglements, and thus reduces the tendency for ESC when solvent is absorbed. There are balances in properties and processing ability that are required for both amorphous thermoplastic polymers as well as partly crystalline polymers in order to achieve maximum ESC resistance and still maintain other desired behavior. It is beyond the scope of this chapter to discuss models for the fracture mechanics mechanisms of ESC. Emphasis is placed on the consequences of similarity of HSP for the polymer and challenge chemical.

ESC INTERPRETED USING HSP Wright1 has discussed ESC data in terms of HSP for polycarbonate (PC), polyvinylchloride (PVC), polymethyl methacrylate (PMMA), and polystyrene (PS). ESC data are discussed for many other polymers including polyethylene (PE), polyamides (PA), polyether ether ketone (PEEK), polytetrafluoroethylene (PTFE), and styrene-acrylonitrile copolymer (SAN). There are many practical examples of how not to do things. Barton3 has discussed in detail the theory and application of the cohesion (solubility) parameters including HSP for many systems, as witnessed by the 739 pages in his book and hundreds of references. Environment-induced degradation is also discussed with various plots for poly(2,6dimethyl-1,4-phenylene oxide), PS, PMMA, PVC, and polysulfone. The latter three are found originally in Vincent and Raha4. Wyzgoski5 constructed several plots to help interpret ESC data for nylon (PA) 6,6 using HSP. Hansen and Just6 studied ESC in the COC-type polymer called Topas® 6013 from Ticona. Figure 14.1, shows two essentially concentric HSP spheres resulting from this work. The inner sphere encompasses those solvents that dissolve the polymer, whereas the outer sphere encompasses these as well as those giving ESC for the given samples. Additional HSP correlations for ESC in PET, PCTG, and PC were reported for ESC (critical strain) data from Moskala and Jones7. HSP correlations for ESC in PEI using data from8 were also reported. All of these HSP correlations are included in Table 14.1. The HSP correlations for ESC reported in Table 14.1 must be considered with care. In the first place there are other factors than HSP that are important, including the size and shape of the given challenge molecule in addition to the state of stress/strain, which is not always well defined. Another point to be remembered is that the RED number (Equation 1.10) refers to the correlation in question. The same solvent polymer pair will have different RED numbers for a HSP correlation of true

7248_C014.fm Page 261 Wednesday, May 23, 2007 11:24 AM

Applications — Environmental Stress Cracking in Polymers

Dispersion Solubility Parameter, δD

261

40

30

20 15

15 10

10 Polar 5 Solubility Parameter, δP

10 Hydrogen Bonding Solubility Parameter, δH

5 0

0

FIGURE 14.1 Three-dimensional HSP plot showing those solvents that dissolve the COC polymer, Topas 6013, Ticona, in the shaded region, and those that induce ESC in the clear shell. The two HSP spheres are almost concentric as can be seen from the data in Table 14.1. (Reprinted with permission from Hansen, C.M. and Just, L., Ind. Eng. Chem. Res., 40(1), 21–25, 2001. Copyright 2001 American Chemical Society.)

TABLE 14.1 HSP Correlations for ESC in Polymers Polymer

δD

δP

δH

R0

FIT

G/Ta

Topas 6013 Solubility Topas 6013 Sol. + cracks PC critical strain <0.6%b PVC crit.str. <0.6%c PET crit. str. <0.6% PCTG crit. str. <0.55% PCTG crit. str. <0.50% PC crit. str. <0.31% PEI Ultem® 1000 600 psi PEI Ultem 1000 1200 psi PEI Ultem 1000 2500 psi PEI Ultem 1000 solubility

18.0 17.3 21.5 16.0 21.3 18.3 18.8 18.0 17.3 17.0 17.4 19.6

3.0 3.1 9.5 10.0 4.5 9.3 8.8 9.0 5.3 6.0 4.6 7.6

2.0 2.1 5.1 5.0 12.3 11.3 10.8 6.0 4.7 4.0 9.0 9.0

5.0 6.4 12.9 10.7 13.9 8.0 7.9 10.0 3.3 4.0 7.2 6.0

1.000 0.974 0.857 1.000 1.000 1.000 0.989 1.000 1.000 1.000 0.967 0.952

8/43 15/43 18/47 5/16 12/19 8/19 6/19 9/18 3/20 4/20 9/20 8/45

Note: Units are MPa1/2. All correlations from Reference 6, except as noted. See also Chapter 12; Ultem is a registered trademark of the General Electric Company. a

G/T is the number of “good” or attacking solvents relative to the total number of solvent data points used in this correlation. b Data from Reference 10 through Reference 14. c Data from Mai, Y.-W., J. Mater. Sci., 21, 904–916, 1986. With permission.

7248_C014.fm Page 262 Wednesday, May 23, 2007 11:24 AM

262

Hansen Solubility Parameters: A User’s Handbook

solubility than it will have for a correlation for ESC where the “good” solvents have critical strains less than, say, 0.6%. HSP correlations can be made with many different types of data (barrier properties, wetting behavior, swelling, solubility, sedimentation rates, etc.) as discussed elsewhere in this book. The term RED number is intimately connected with a given HSP correlation. Hansen9 used data from various sources 6,10–13 to construct a new type of plot to correlate ESC data. This plot uses the RED number, Equation 1.10, and the molar volume, V, of the test chemicals. Figure 14.2 is such a plot for a large number of liquids in contact with small, injection-molded cylinders of a COC-type polymer called Topas® 6013 from Ticona.6,9 There are three distinct regions on this plot. There is a region at low RED including those challenge liquids that dissolve the polymer or are very aggressive, and ESC is not found as such. There is a region at high RED where the absorption is zero or not great enough to matter, or else the absorption rate is slow enough to allow relaxation of the polymer in preference to ESC. ESC can occur in an intermediate region where there is some absorption of challenge liquid, although examples are given below for other polymers where ESC takes place without measurable absorption for good matches in HSP. The ESC region on these plots increases in size with increased tensile stress and/or higher values of strain. Closer study of Figure 14.2 shows that there are several liquids that do not give ESC whereas this would normally be expected from their position on this plot. From Table 14.2, it can be seen that these include methyl isobutyl ketone, acetophenone, and nitrobenzene. These are surrounded on the plot by liquids that do result in ESC including n-hexane, n-butyl acetate, ethyl acetate, and diethyl ether. The RED numbers and molecular volumes for all seven of these liquids are comparable. The explanation lies in the fact that the four liquids giving ESC have measurable absorption, whereas the three not giving ESC apparently do not absorb at all under the test conditions. It has been determined that 1,4-dioxane, methyl isobutyl ketone, acetophenone, and phenyl acetate do not absorb into this polymer at room temperature.14 The methyl “side-group” on the methyl isobutyl ketone, and the benzene rings in acetophenone, nitrobenzene, and phenyl acetate, are evidently sufficient to provide the steric hindrance that prevents absorption. Methyl isobutyl ketone does give ESC at higher stress levels. All of the test liquids causing ESC failure in the immersed COC cylinders had measurable surface resistances retarding absorption, but they did absorb.14 Surface resistance phenomena may play an important part in the ESC process itself. Delayed absorption can also potentially lead to postponing a catastrophic ESC failure beyond normal testing times. Chapter 16 in this book is therefore dedicated to absorption and diffusion in polymers, and especially emphasizes the frequently overlooked surface resistance. This surface resistance is thought to originate primarily from the rate at which adsorbing molecules can locate a hole in the polymer surface large enough to accommodate them. Larger and more structurally complicated molecules have much more difficulty finding such a suitable hole, so the surface transport coefficient is inversely proportional to the molecular cross-section. Molecules that are too large can simply not enter the polymer. Once an adsorbed molecule locates in a suitable hole, the rate of motion into the bulk is dependent on the local diffusion coefficient. Therefore the surface transport coefficient is directly proportional to the diffusion coefficient. See the discussion in Chapter 16 for more details and examples. Figure 14.3 and Figure 14.4 use the same parameters, RED vs. V, for correlating ESC in the polymers PC and PVC, respectively. The data used to construct these were taken from references 10–13, that is, from the older literature. It can be seen that additional liquids will give ESC as the critical stress levels become higher. Before proceeding to perhaps still more complicated theories and situations in the next section, one should be reminded that there are large variations in the critical strain even for the hexane isomers, which all have very similar total (Hildebrand) or Hansen dispersion solubility parameters. For PC, these isomers have critical strains that increase from 0.85% to 1.68% as branching increases.12 These data clearly reinforce the need for consideration of shape.

7248_C014.fm Page 263 Wednesday, May 23, 2007 11:24 AM

Applications — Environmental Stress Cracking in Polymers

TOPAS 6013 solubility

●

3.0 2.8

δD

δP

δH

RO

18.0

3.0

2.0

5.0

Soluble

C ESC  No ESC DMF

D Severe deformation

NEE

2.6 No problem at stress level

2.4 HSP, Red number for true solubility

263

CHK DAA NMP

2.2 ACI

2.0 1.8 ESC possible

1.6

2NP THF

1.4

D

1.2

MCHL

C

1.0

EET ETA C C NTB

ISOPH HXA

C

ACETOPH MIK C BCN

DOX EDC

C

●●

0.8

●

Solubility

● ●

0.6

● ●

0.4

●

0.2 0

10

20

30

40

50

60

70 80 90 100 110 120 130 140 150 160 170 Size parameter, V

FIGURE 14.2 Plot of ESC and solubility data for injection-molded cylinders of a COC polymer (Topas 6013, Ticona) using RED number vs. molar volume. There is a region of solubility (RED less than 1.0), an intermediate region for ESC, and a region at higher RED where ESC is not found. The ESC solvents all have linear molecular structure. Data from Reference 6. Symbols are explained in Table 14.2. (Reproduced from Hansen, C.M., Polym. Degradation Stability, 77, 43–53, 2002. With permission from Elsevier Science.)

ESC WITH NONABSORBING STRESS CRACKING INITIATORS Recent research has shown that even nonabsorbing chemicals can induce ESC, in some cases very rapidly.15–17 It is presumed that there are many additional cases of this kind, but they have not been reported as such. One possible explanation for this is HSP-induced motion of the polymer chain segments in the surface. This phenomenon is called surface mobility here and has also been discussed in Chapter 15 and Chapter 18. An example of surface mobility is contact with water converting surfaces like those of peat moss from hydrophobic to hydrophilic. An applied water droplet initially beads up but soon soaks readily in. The surface again becomes hydrophobic when the water is gone. This might be considered as Nature’s valve to conserve water within a system. There are other examples given in Chapter 18. Rotation or other form of polymer chain segment

7248_C014.fm Page 264 Wednesday, May 23, 2007 11:24 AM

264

Hansen Solubility Parameters: A User’s Handbook

TABLE 14.2 Symbols Used in Figure 14.2 Symbol

Compound

ACI ACETOPH BCN CHK DAA DMF DOX EDC EET ETA HXA MCI MIK 2NP NEE NMP NTB THF

Acetone Acetophenone Butyl acetate Cyclohexanone Diacetone alcohol Dimethyl formamide 1,2-Dioxane Ethylene dichloride Diethyl ether Ethyl acetate n-Hexane Methylene dichloride Methyl isobutyl ketone 2-Nitropropane Nitroethane N-methyl-2-pyrrolidone Nitrobenzene Tetrahydrofuran

motion may also be sufficient to start the cracking process. All of the ESC cases known to the author where absorption has not been measurable have had relatively high stress/strain imposed on the samples. There have been many other undocumented cases of ESC where one would not suspect absorption of the ESC initiator. These include contact with chemicals in hair spray, deodorants, hand creams, butter, and various kinds of oils including essential oils. In all cases, it is thought that the affinity of the active chemical must be high or moderately high, such that even though it cannot absorb, it is still capable of inducing motion in the polymer chains at the surface. A related study of surface phenomena by Nielsen and Hansen18 explored whether ESC could be predicted by the wetting behavior of the challenge chemicals on PC, COC, and ABS type polymers. A large number of challenge chemicals were divided into three groups. Group A included those that spontaneously spread when applied as a droplet. Group B included those that would not spontaneously spread, but which would not retract either, when they were applied as films. Group C liquids did not spontaneously spread and retracted when they were applied as films. It was found that all A and B liquids gave ESC, although the B types had higher critical strains in general. Some type C liquids gave ESC and others did not. Retraction of an applied film is not an indication that ESC will not occur. Contamination of the test surface may also lead to retraction of an applied film, for example, where this would not occur otherwise. Care must be taken, and this test is only an indication of a potential problem. ESC was found for PC polymers with polydimethylsiloxanes having molecular weights of 340 and below but not for those with 410 and above. Exactly why this happens is not known, but surface entry or lack of same (see Chapter 16) and surface mobility of polymer molecules may both be involved.

DISCUSSION HSP correlations of ESC phenomena have been presented for a number of common polymers. As stated above, the mechanisms for crazing and crack initiation and growth have not been discussed in detail here. The basis of the failures is the pulling of polymer chains from each other in all cases,

7248_C014.fm Page 265 Wednesday, May 23, 2007 11:24 AM

Applications — Environmental Stress Cracking in Polymers

265

CALCULATED ESC FOR PC AT CRITICAL STRAIN = 0.6 2.5

▲

2 ●

Glycerol 0.77, 1.99

●

● ● 1.5

●  

RED NO.

 ● 2-Propanol 1.02  ◆  ◆ ▲



● 





●

●

 

▲ ●  ⴐ ⴐ Cyclohexanol 0.98, 1.48

◆●  ◆ ● ⴐ ⴐ Dimethylformamide 1.55 ⴐⴐ 

◆▲ ◆ ◆

1

0.5

●

◆

ⴐ ◆

0

0

50

100

150

200

250

300

350

MOL VOLUME - CC/MOL  0-0.2ⴐ0.2-0.4 ◆ 0.4-0.6 ▲ 0.6-0.8  0.8-1 ●>1 FIGURE 14.3 Plot of ESC data for polycarbonate (PC) using data from Mai10 and others11–13 to establish a HSP correlation based on critical strain since true solubility data was lacking. The limit for critical strain was chosen as 0.6%. Liquids with critical strains below this value should have RED less than 1.0, just as liquids with critical strains above this should have RED larger than 1.0. Specific data and a discussion regarding the correlation are included in the text. Higher critical strains lead to larger ESC regions on this plot. The critical strains are indicated in the caption. (Reproduced from Hansen, C.M., Polym. Degradation Stability, 77, 43–53, 2002. With permission from Elsevier Science.)

however. A more detailed discussion of the phenomena involved is considered beyond the scope of this chapter. HSP has been used as a correlating parameter, necessarily with other parameters such as molecular size and shape of the challenge chemical and the stress/strain condition of the samples, to provide improved predictability, and to ask new questions regarding ESC. Figure 14.1 shows that the ESC solvents have RED numbers slightly larger than those required for solubility. The data for this figure are found in.6 Figure 14.2 clearly shows that the molecular shape of the challenge molecules is important in addition to V. ESC may occur for a challenge chemical with a linear molecular structure, but not

7248_C014.fm Page 266 Wednesday, May 23, 2007 11:24 AM

266

Hansen Solubility Parameters: A User’s Handbook

HSP CORRELATION FOR ESC IN PVC AT CRITICAL STRAIN = 0.6 2.5 ●

2

●

▲

1.5 PENTANE

RED NO.

▲

CYCLOHEXANE

METHYL CYCLOHEXANE HEPTANE

▲ ▲ ▲



1 ⴐ



ⴐ

● ●

0.5

●

●

 ▲

ETHYL BENZENE

●

NITROMETHANE

●

ⴐ

●

0 0

20

40

60

80

100

120

140

160

MOL VOLUME - CC/MOL ● DISSOLVES  0-0.2 Ⳮ 0.2-0.4 ◆ 0.4-0.6 ▲ 0.6-0.8  0.8-1.0 ● >1 FIGURE 14.4 Plot of ESC data for polyvinylchloride (PVC) using data from Mai10 to establish a HSP correlation based on critical strain. ESC data were used for the correlation since specific solubility data on these samples were not available. The limit for critical strain was chosen as 0.6%. Solvents with critical strains below this value should have RED less than 1.0, just as solvents with critical strains above this should have RED larger than 1.0. Higher critical strains lead to larger ESC regions on this plot. The critical strains are indicated in the caption. (Reproduced from Hansen, C. M., Polym. Degradation Stability, 77, 43–53, 2002. With permission from Elsevier Science.)

for one with the same RED and V, but with a cyclic or more branched structure. The absorption rate of the bulkier molecules is too slow or they may not even be able to absorb altogether. The samples providing the data for this figure were small cylinders, so it was impossible to determine critical strains for the given liquids with these samples. Figure 14.3 clearly shows larger ESC regions for higher critical strain limits for polycarbonate. More solvents logically give ESC as the strain increases. The data used in this figure were found in.10–14

7248_C014.fm Page 267 Wednesday, May 23, 2007 11:24 AM

Applications — Environmental Stress Cracking in Polymers

267

Figure 14.4 shows the ESC behavior of PVC using data taken from the literature.10 This figure confirms the ability to use this type of plot to correlate and predict ESC in polymers. It is not unusual for suppliers of polymers to change compositions without indicating what has been done and to continue to use the general indication of polymer type, such as PC. Thus, the PC data in Figure 14.3 and the PVC data in Figure 14.4 may not be valid for all polymers of these nominal types. This is a significant problem in further research and developing fundamental understanding. Researchers either must work with commercial materials they do not have full knowledge of, or else take the path of making test materials to be certain of what is being dealt with. In the latter case, the information may not apply directly in practice because the practical materials may have added components. This dilemma apparently has no fully satisfactory solution.

CONCLUSION HSP have been used to develop correlations of ESC. It would appear that testing is still required to ascertain what behavior is to be expected, but these correlations clearly indicate where the problems will be greatest. It is usually the unexpected events that cause the catastrophic failures, but designs or changes in designs that lead to increased tensile stress have also been the causes of ESC in practice. Care is especially required whenever the HSP of the polymer gets too close to the HSP of potential challenge chemicals. The required affinity between active chemical and polymer allows correlations of ESC with HSP, most often with the simultaneous need to consider the size and shape of the active chemical and the stress/strain condition of the polymer.

REFERENCES 1. Wright, D., Failure of Plastics and Rubber Products, Rapra Technology Limited, Shawbury, Shrewsbury, Shropshire, U.K., 2001. 2. Lustiger, A., Understanding Environmental Stress Cracking in Polyethylene, Medical Plastics and Biomaterials Magazine, MPB Article Index (originally published July 1996), http://www.devicelink.com/mpb/archive/96/07/001.html. 3. Barton, A.F.M., Handbook of Solubility Parameters and Other Cohesion Parameters, 2nd ed., CRC Press, Boca Raton, FL, 1991. 4. Vincent, P.I. and Raha, S., Influence of hydrogen bonding on crazing and cracking of amorphous thermoplastics, Polymer, 13, 283–287, 1972. 5. Wyzgoski, M.G., The role of solubility in stress cracking on nylon 6,6, in Macromolecular Solutions, Seymour, R.B. and Stahl, G.A., Eds., Pergamon Press, New York, 1982, pp. 41–60. 6. Hansen, C.M. and Just, L., Prediction of environmental stress cracking in plastics with Hansen solubility parameters, Ind. Eng. Chem. Res., 40(1), 21–25, 2001. 7. Moskala, E.J. and Jones, M., Evaluating Environmental Stress Cracking of Medical Plastics, Medical Plastics and Biomaterials Magazine, May, 1998, pp. 34–45. 8. Anonymous, Ultem Resin Design Guide, GE Plastics, Pittsfield, MA, 1989. 9. Hansen, C.M., On Predicting Environmental Stress Cracking in Polymers, Polym. Degradation Stability, 77, 43–53, 2002. 10. Mai, Y.-W., Environmental stress cracking of glassy polymers and solubility parameters, J. Mater. Sci., 21, 904–916, 1986. 11. Kambour, R.P., Gruner, C.L., and Romagosa, E.E., Bisphenol-A polycarbonate immersed in organic media. Swelling and response to stress, Macromolecules, 7, 248–253, 1974. 12. Jacques, C.H.M. and Wyzgoski, M.G., Prediction of environmental stress cracking of polycarbonate from solubility considerations, J. Appl. Polym. Sci., 23, 1153–1166, 1979. 13. Henry, L.F., Prediction and evaluation of the susceptibilities of glassy thermoplastics to environmental stress cracking, Polym. Eng. Sci., 14(March), 167–176, 1974. 14. Nielsen, T.B. and Hansen, C.M., Significance of surface resistance in absorption by polymers, Ind. Eng. Chem. Res., 44(11), 3959–3965, 2005.

7248_C014.fm Page 268 Wednesday, May 23, 2007 11:24 AM

268

Hansen Solubility Parameters: A User’s Handbook

15. Al-Saidi, L.F., Mortensen, K., and Almdal, K., Environmental stress cracking resistance. Behaviour of polycarbonate in different chemicals by determination of the time-dependence of stress at constant strains, Polym. Degradation Stability, 82, 451–461, 2003. 16. Hansen, C.M., Environmental stress cracking of PTFE in kerosene, Polym. Degradation Stability, 77, 511–513, 2002. 17. Kjellander, C.K., Publications being prepared. 18. Nielsen, T.B. and Hansen, C.M., Surface wetting and the prediction of environmental stress cracking (ESC) in polymers, Polym. Degradation Stability, 89, 513–516, 2005.

7248_C015.fm Page 269 Wednesday, May 9, 2007 9:37 AM

Solubility 15 Hansen Parameters — Biological Materials Charles M. Hansen and Tim S. Poulsen ABSTRACT The Hansen solubility parameters (HSP) of many biological materials can be found from correlations of how they interact with well-defined liquids. The three HSP parameters, δD, δP , and δH quantitatively account for the cohesion energy density arising from atomic, dispersion type interactions (D), molecular, dipolar interactions (P), and molecular, hydrogen bonding interactions (H). Examples of HSP correlations included in this chapter are DNA, cholesterol, chlorophyll, wood chemicals, polypeptides (proteins), human skin, nicotine, lard, and urea. The often-quoted “like dissolves like” has been expanded to “like seeks like” (self-association) to discuss the implications of these correlations. The ability of HSP to correlate surface phenomena has made this change mandatory. Biological materials such as proteins and DNA have well defined structures in a given environment. DNA adopts double helices, whereas proteins consist of a combined shape of the secondary, tertiary, and in some cases quaternary structure that together determine the conformation of the protein. The ultrastructure of wood is another example of Nature’s way of establishing order in complex systems. The proper function of a protein requires that certain functional groups are at precise locations within its tertiary and/or in some cases quaternary structure. The conformation of proteins and DNA can be influenced, and in many cases controlled, by solvent quality. The solvent quality in a given environment is expected to determine whether a protein is dissolved or not, and also to control the way it adsorbs onto other materials or interacts with itself. Controlled changes in solvent quality can lead to controlled changes in conformation. Solvents can change not only the ability of noncovalent interactions such as van der Waals, hydrogen bonding, and ionic bonding, but also induce chiral rotation. The key to the importance of noncovalent interaction is that such interactions can continually be broken and reformed under physiological conditions. The portion of the molecule with energy properties most similar to the surrounding liquid will be oriented toward the liquid (“like seeks like”). The term hydrogen bonding is generally used to describe the noncovalent interactions in DNA, proteins, and other biological molecules, implying that this is the dominating interaction. The HSP correlation based on solvent interactions with DNA resulted in δD;δP;δH values equal to 19.0;20.0;11.0. These numbers clearly show that hydrogen bonding provides by far the smallest contribution of the three types of interaction, representing only about 14% of the cohesive energy involved (using Chapter 1, Equations 1.6–1.8). The term hydrogen bonding must be considered as an insufficient description of the interactions that determine the structure in such molecules.

269

7248_C015.fm Page 270 Wednesday, May 9, 2007 9:37 AM

270

Hansen Solubility Parameters: A User’s Handbook

INTRODUCTION HSP have been used to characterize many biological materials.1–7 Most of the materials discussed in these references are also included in the present discussion, but many more can be added by experiment or calculation. There are many simple experimental methods to determine the HSP for biological materials. These involve contacting a material of interest with a series of well-chosen liquids. The fact of solubility, differences in degree of equilibrium swelling, rapid permeation or not, significant surface adsorption or not, or other measurable quantity significantly influenced by physical affinity relations can be observed and used to find the HSP for a material being studied. These methods have been discussed in more detail in earlier chapters. The basis of the division of the cohesive energy density into three parts accounting for the atomic dispersion (D), molecular dipolar (P), and molecular hydrogen bonding (H) interactions, respectively, is given in detail in Chapter 1. The HSP for simpler compounds can be calculated according to the methods given in Chapter 1. HSP values for nicotine, skatole, wood chemicals, etc., that are discussed in this chapter were calculated using these methods. Figure 15.1 shows a typical HSP sphere correlating experimental solubility data for lignin.1 The good solvents are located within the sphere which is based on Chapter 1, Equation 1.9. Again, as stated in previous chapters, this equation is in agreement with the Prigogine corresponding states theory of polymer solutions as discussed in Chapter 2. The statistical thermodynamics approach presented in Chapter 3 also shows agreement with the concepts to which this book is dedicated. Furthermore, this equation has also been shown to be correct for such complex materials as asphalt and bitumen, as described in Chapter 9, and carbon dioxide solubility in solvents, as described in Chapter 10. A HSP correlation can, of course, be used to predict the behavior of solvents not included in the experimental work. It is convenient to print the solvent database in order from best solvent to worst solvent to aid in finding alternatives. This is a quantitative application of the generally used statement “Like Dissolves Like.” In the following discussion, this concept is expanded to “Like Seeks Like” (self-association). This implies that segments of molecules seek regions of similar HSP if this is possible. This may result in solutions or in selective orientation of segments of molecules in more complicated systems. Table 15.1 contains HSP data for several biologically interesting materials. These are discussed in the following in more detail with an indication of how such data may be used. The data included in this table are the δD, δP , and δH parameters; the radius of interaction for the HSP correlation, Ro; if appropriate, the data fit (where a fit of 1.000 is perfect as discussed in Chapter 1). G is number of “good” solvents and the total number of solvents in a given correlation is T. The units for the solubility parameters and Ro are MPa1/2. Plots of the kind given in Figure 15.1 for lignin are sometimes used to interpret relations among different materials. RED numbers indicate solvent quality with lower values being indicative of better solvents (see Chapter 1, Equation 1.10). The correlations reported here are a result of data processing with the SPHERE program described in Chapter 1. The output is often arranged with the best solvent (lowest RED number) at the top of the list. A most interesting and important class of molecules are called amphipathic. These exhibit both hydrophilic and hydrophobic properties simultaneously. An example from biology is the amphipathic molecules (lipids) that form the basis of the biological membrane bilayers that surround cells. Such amphipathic molecules have a head group that is strongly hydrophilic, coupled to a hydrophobic tail – usually hydrocarbon in nature. When one attempts to dissolve these molecules in water, they form special structures. These may be monolayers on the water surface, with only the head groups immersed. Alternatively, if the mixture is vigorously stirred, micelles (spherical structures stabilized by a single layer of molecules at the water interface) or bilayer vesicles may form. Another example is amino acid side chains. These are by nature not only different in size and shape, but also in the charge they carry, their general affinity for water (hydrophilicity) and/or their general aversion to water (hydrophobicity). The native conformation of proteins is a strong

7248_C015.fm Page 271 Wednesday, May 9, 2007 9:37 AM

Hansen Solubility Parameters — Biological Materials

271

Hansen Solubility Parameters

Lignin

δD

δP

δH

RO

21.9

14.1

16.9

13.7

40

Dispersion Solubility Parameter, δD

30

Lignin

20 15 Polar Solubility 10 Parameter, δP

15 10

10

5

5 0

0

Hydrogen Bonding Solubility Parameter, δH

FIGURE 15.1 HSP correlation showing the solubility of lignin. Good solvents are located within the sphere. Units are MPa1/2. (From Hansen, C.M. and Björkman, A., Holzforschung, 52, 339, 1998. With permission.)

function of the interactions that take place within and between polypeptide chains. This is also highly dependent on the interaction that takes place with water, as proteins exist in an aqueous environment. These general concepts of hydrophilic and hydrophobic entities can also be quantified using HSP.

HYDROPHOBIC BONDING AND HYDROPHILIC BONDING (SELF-ASSOCIATION) The concept of “like seeks like” offers a general explanation of hydrophobic bonding. An aliphatic hydrocarbon chain on a protein, for example, is not soluble in water and ultimately finds another aliphatic hydrocarbon chain with which to associate. This same type of process leads to micelle formation when the solubility limit of surface active agents is exceeded. Hydrophobic bonding is found when the HSP for the associating segments are too low to allow solubility in the continuous phase. When it is immersed in water a polypeptide chain will not stay in an elongated form. It will instead fold up into secondary structures according to the polarity of the side chains it contains and the rotation of peptide backbone bond angles that are largely determined by Van der Waals radii of side chains. This can be called hydrophilic bonding. Hydrophilic bonding is formed when the HSP for the associating segments are too high to allow solubility in the continuous phase. If the continuous phase is a hydrocarbon liquid, the associating segments may be characterized by high δH, for example, because of the presence of an alcohol, acid, or amide group.

7248_C015.fm Page 272 Wednesday, May 9, 2007 9:37 AM

272

Hansen Solubility Parameters: A User’s Handbook

TABLE 15.1 Hansen Solubility Parameter Correlations for Biologically Interesting Materials, MPa1/2 Material

δD

δP

δH

Ro

FIT

G/T

DNA Cholesterol solubility Lard 37°C solubility Lard 23°C solubility Olive oil solubility Psoriasis scales swelling Human skin — permeation Nicotine — calculation Skatole — calculation Chlorophyll — solubility Sinapyl alcohol calculation Coniferyl alcohol calculation p-Coumaryl alcohol calculation Lignin — solubility Dextran C (= amorphous cellulose) See Chapter 5 Sucrose solubility N-methyl-morpholine-N-oxide calculation Blood serum — swelling Zein — solubility Urea — solubility Water — >1% soluble in Water — totally miscible Water — single molecule

19.0 20.4 15.9 17.7 15.9 24.6 17.6 18.8 20.0 20.2 19.2 19.0 19.1 21.9 24.3 23.4 19.0 25.5 22.4 22.9 15.1 18.1 15.5

20.0 2.8 1.2 2.7 1.2 11.9 12.5 7.8 7.1 15.6 7.3 7.0 7.0 14.1 19.9 18.4 16.1 10.3 9.8 14.9 20.4 17.1 16.0

11.0 9.4 5.4 4.4 5.4 12.9 11.0 6.4 6.2 18.2 16.1 16.3 17.3 16.9 22.5 20.8 10.2 22.1 19.4 21.3 16.5 16.9 42.3

11.0 12.6 12.0 8.0 12.0 19.0 5.0 — — 11.1 — — — 13.7 17.4 16.0

1.000 1.000 1.000 1.000 1.000 0.927 1.000 — — 0.864 — — — 0.990 0.999 0.981

6/12 25/41 29/50 21/50 29/50 35/50 4/13 — — 7/35 — — — 16/82 5/50 6/50

17.8 11.9 16.2 18.1 13.0 —

0.980 0.964 0.984 0.856 0.880 —

4/51 4/50 14/50 88/167 47/166 —

Note: The units for the solubility parameters and Ro are MPa1/2. G/T represents the number of good liquids (G) and the total number of liquids (T) in the correlation.

Figure 15.2 demonstrates how hydrophilic bonding between versamid polymer blocks reacted into an alkyd (polyester) polymer gives a thixotropic alkyd paint with its special nondrip properties. Agitation of the paint is enough to break the hydrophilic bonds allowing easy spreading, but they reform quickly again after application. The most common secondary structures are alpha helices and beta sheets that are stabilized by local inter-residue interactions mediated by hydrogen bonds. An alpha helix can take the form of an amphipathic helix with a polar and a nonpolar side. This plays a crucial role in helix–helix interactions and in the interaction of small peptides that have a helical conformation with membranes, air–water interfaces, and self-assembly processes. Beta sheets are alternative secondary structure to the alpha-helix in proteins. Like alpha-helices, beta-sheet backbones are stabilized by hydrogen bonds between two beta sheets, but the bonds occur between neighboring strands. If the beta−strand contains alternating polar and non-polar residues it forms an amphipathic beta sheet. This distribution of hydrophilic and hydrophobic residues has been observed in the membrane protein porin that forms a beta-barrel structure. Here the nonpolar residues stick into the hydrophobic part of the lipid membrane and the hydrophilic residues form part of the channel interior responsible for the passage of small molecules across the membrane. Hydrophobic bonding is a major effect that drives proper protein folding. Hydrophobic sidechains are oriented to minimize the energy lost by the intrusion of amino acids into the water solvent, which disrupts lattices of water molecules. Hydrophobic bonding forms an interior,

7248_C015.fm Page 273 Wednesday, May 9, 2007 9:37 AM

Hansen Solubility Parameters — Biological Materials

273

A Mineral spirits B A δP

B

B regions are not soluble and “Precipitate” together

B

A

Polymer B (versamid)

Butanol

M.S. Polymer A (alkyd) δH FIGURE 15.2 HSP relations for establishing thixotropy in an alkyd-type paint. The solid circle represents the solubility of the alkyd (A) and the dotted circle that of the Versamid (B). The Versamid segments associate because they are not soluble in mineral spirits. Addition of n-butanol destroys the thixotropic effect, since the solvent then becomes too good. Similar relations exist for the true solution of some proteins by additions of urea to water. This denatures them, by effectively dissolving them in a solvent mixture that is better than water itself.

hydrophobic protein core, where most hydrophobic sidechains can closely associate and are shielded from interactions with solvent water. Formation of “hydrogen bonds” within proteins is based on the lack of solvency in the continuous media, water, because the HSP of these segments is too high. Additions of urea, as discussed later in more detail, increase the HSP of the continuous media to such an extent that it can now dissolve the “hydrogen bonded” segments. The protein is denatured, which in fact means that these segments are dissolved in a good solvent. Additions of salts can also improve solvency for a given material or segments of materials. Additions of salts can also reduce solvency. These phenomena must also have their explanation in the “like seeks/dissolves like” phenomena, but more research is required to quantify them. Such mechanisms of controlling solvent quality can be expected to be used by Nature in many biological systems to control adsorption and/or transport of various types of materials as in self-association.

DNA The double helix structure of DNA suggested by Watson and Crick is stabilized by hydrogen bonding between bases on opposite strands when the bases are paired in one particular way (A+T or G+C). In the Watson–Crick model the base pairs are stacked on one another with their planes nearly perpendicular to the helix axis where the hydrophilic phosphate–deoxyribose backbones are on the outside, in contact with the aqueous environment. This complementary base pairing (hybridization) is central to all processes involving nucleic acids. In cells it occurs in, e.g., DNA replication,

7248_C015.fm Page 274 Wednesday, May 9, 2007 9:37 AM

274

Hansen Solubility Parameters: A User’s Handbook

TABLE 15.2 Hansen Solubility Parameter Correlation for DNA Solvent

δD

δP

δH

RED

V

Dimethyl sulfoxide 2,6-Dichloro-7-methyl purine Coumarin Purine Caffeine Formamide Pyrimidine Phenol Urea Cyclohexanol Methyl riboside Adonitol

18.4 20.5 20.0 20.5 19.5 17.2 20.5 18.0 20.9 17.4 17.0 18.0

16.4 11.7 12.5 11.7 10.1 26.2 9.4 5.9 18.7 4.1 12.0 12.0

10.2 14.2 6.7 14.2 13.0 19.0 11.3 14.9 26.4 13.5 32.8 36.0

0.353 0.651 0.807 0.853 0.923 0.977 1.002 1.342 1.447 1.492 2.142 2.393

71.3 162.4 156.3 100.0 157.9 39.8 78.8 87.5 45.8 106.0 117.2 95.1

DNA

D = 19.0 P = 20.0 H = 11.0 Ro = 11.0 FIT = 1.000 NO = 12

Note: Units of D, P, H and Ro are MPa1/2. V is in cc/mole. The order in the table is from expected best at the top to expected worst at the bottom.

transcription, rRNA, and tRNA structure, but it is also used in laboratories in RNA and DNA gel blots, PCR, sequencing, genotyping, microarrays, in situ hybridization, etc. DNA melts (denatures) at 90-100°C in 0.1-0.2 M Na+. This may lead to deterioration of morphology. Fortunately, organic solvents reduce the thermal stability of double-stranded polynucleotides, so that hybridization can be performed at lower temperatures in the presence of formamide, for example. Formamide is often used in connection with DNA.8 For in situ hybridization this implies that microscopic preparations must be hybridized at 65–75° for prolonged periods. The melting temperature, Tm, is found when a population of particular DNA sequences is at a midpoint between fully double-stranded and single-strand. Formamide reduces the Tm of DNA-DNA and DNA-RNA duplexes in a linear fashion by about 0.65°C for each volume percent of the solvent that is present. Other common solvents can also reduce Tm, including dimethyl sulfoxide. An article in the older literature9 reports aspects of the interaction of different low molecular weight materials with DNA. The summary of this article states that the order of increasing activity was found to be: adonitol, methyl riboside (both negligible) < cyclohexanol < phenol, pyrimidine, uridine < cytidine, thymidine < purine, adenosine, inosine, deoxyguanosine < caffeine, coumarin, 2,6-dichloro-7-methylpurine. Urea was ineffective with poly A and only slightly effective with DNA. At a concentration of 0.3M, purine lowered the Tm of DNA by about 9°C. The HSP for several of these having reasonably simple structures were estimated by the methods of Chapter 1. These HSP data were divided into two arbitrary groups of “good” and “bad” with a dividing line between purine as good and pyrimidine as bad. The compounds intermediate in the above list were structurally too complicated to allow a reliable calculation. Formamide and dimethyl sulfoxide were also considered as “good” and added to the data for the correlation reported in Table 15.2. The encouraging correlation reported in Table 15.2 ranks the given solvents in approximately the same order as that given in Reference 9. All the solvents from pyrimidine and lower were considered as being “bad” and all those above this were considered as being “good.” Even urea, where performance may be affected significantly by the presence of water, seems to be placed correctly. Formamide is not at the top of the list, but is the preferred solvent of use today in many cases. The effectiveness of formamide is primarily because of its low molecular volume, but it will also be a good solvent for phosphate salts, which may also contribute some effect. Dimethyl

7248_C015.fm Page 275 Wednesday, May 9, 2007 9:37 AM

Hansen Solubility Parameters — Biological Materials

275

sulfoxide will also be a reasonably good solvent for phosphate salts. Low molecular volume is very conducive to dissolving polymers with structure or crystallinity, as the small molecules can reach the critical sites more readily than larger ones. Smaller molecules are also predicted to be thermodynamically better, all else being equal. The radius is arbitrary and depends on the criterion used for good and bad. If pyrimidine had been considered as being good, then the D, P, and H could be maintained with a slightly larger Ro, and the data fit would still be 1.000. There are many different D, P, and H, combinations possible when the data fit is 1.000, but the present correlation, in spite of the very few solvents, is still considered reasonably reliable because of the essentially correct ranking. Other supporting evidence that the correlation is reasonable can be found in the estimated HSP for adenine and thymine. These can be considered as single relevant portions of DNA. The HSP are δD;δP;δH equal to 20.0;16.0;14.9 for adenine and 19.0;20.5;13.0 for thymine. Both of these are reasonably close to HSP equal to 19.0;20.0;11.0, the estimated values for DNA based on its interaction with a number of solvents as reported in Table 15.2. All units are MPa1/2. The δH value for DNA is only 11.0 MPa1/2 compared with δD equal to 19.0, and δP equal to 20.0. This clearly shows that the hydrogen-bonding interactions are far less important than the other two types of interaction. The cohesive energy derived from hydrogen bonding is about 14% of the total using Chapter 1, Equations 1.6 to 1.8.

CHOLESTEROL Cholesterol has been characterized with HSP based on its solubility in a large number of solvents. δD, δP, and δH and Ro for cholesterol solubility were found as 20.4;2.8;9.4 and 12.6, all values having units of MPa1/2. The test method involved placing 0.5 g of cholesterol in test tubes together with 5 ml of each of 41 different solvents. The temperature was 23°C. Total solution or not at this concentration was evaluated visually. The 25 “good” solvents dissolved the entire amount of cholesterol added. These data were analyzed by the SPHERE computer program described in Chapter 1 to find the HSP for cholesterol. This has also been reported in Reference 10. Figure 15.3 shows this HSP correlation for cholesterol. This figure also includes several solvents that are discussed in the following. The data fit of 1.0 indicates that there are other sets of parameters for spheres which can be expected to give a perfect separation of the good solvents from the bad ones by a “spherical” HSP correlation. Continued testing with additional test solvents located in the boundary region of the sphere is possible to define it more precisely. This was not warranted under the present circumstances, but is recommended if more extensive use of these data is planned. A general confirmation of the HSP correlation for cholesterol was done by studying mixtures of nonsolvents. Many mixtures of two nonsolvents which dissolve polymers when admixed have been reported in the literature.1 Such synergistic mixtures can be predictably found when they are pairwise on opposite sides of an HSP sphere. The 50:50 vol mixtures of n-hexane with 2-nitropropane and n-hexane with ethanol predictably dissolved cholesterol at 0.5 g/5 ml. During the course of this study, it also became obvious that the solubility of cholesterol in hydrocarbons was limited and quite temperature dependent, being considerably higher at slightly elevated temperatures. This behavior in hydrocarbon solvents relates to the interactions of cholesterol in the hydrocarbon (hydrophobic) portions of lipid layers. The limited solubility in hydrocarbon media and very low solubility in water favors a location at an aqueous interface with the alcohol group of the cholesterol molecule oriented toward the high energy aqueous phase, where it is more compatible, and the hydrocarbon portions oriented into the lipid layer. Changes toward lower temperature will tend to force more cholesterol out of a hydrocarbon matrix. The δH parameter of alcohol solvents decreases relatively more rapidly with increasing temperature than for solvents where the δH parameter is low (or zero), such as with the hydrocarbon solvents. This brings the HSP of the alcohol solvent closer to the HSP of the hydrocarbon solvents, and miscibility improves markedly as temperature increases.

7248_C015.fm Page 276 Wednesday, May 9, 2007 9:37 AM

276

Hansen Solubility Parameters: A User’s Handbook

MATERIAL

δD

δP

δH

CHOLESTEROL

20.4

2.8

9.4

2 – NITROPROPANE

16.2

12.1

4.1

HEXANE

14.9

0.0

0.0

ETHANOL

15.8

8.8

19.4

RO 12.6

DISSOLVING MIXTURES OF NON–SOLVENTS

2–NITRO– PROPANE

15

10 ETHANOL δP 5

(9.4, 2.8, 20.4) 0 20

15

10

5

0

HEXANE

δH

FIGURE 15.3 HSP sphere correlating the solubility of cholesterol. Nonsolvents which synergistically interact to become improved solvents when mixed are indicated. These can predictably be found by selecting pairs located on opposite sides of the HSP solubility parameter sphere. Units are MPa1/2.

One can also surmise what might happen when ethanol or other organic solvent is present in the body. Organic solvents with HSP resembling those of the lipid layer may be found due to occupational exposure or for other reasons, such as drinking alcohol-containing beverages. The presence of ethanol or other organic solvent in the lipid layer allows greater cholesterol miscibility in its hydrocarbon portions. The reason for this is the synergistic effect of ethanol and hydrocarbon segments described earlier. The simple experiments described previously indicate that the cholesterol uptake in hydrocarbon portions of a lipid layer will be greatly enhanced when ethanol is present. This, of course, preferentially removes some of the cholesterol from the blood stream. The solubility of cholesterol in an essentially nonsolvent such as water can be enhanced by additions of a solvent improver such as ethanol. The average HSP for these mixtures are closer to those of cholesterol itself. Therefore, those persons with alcohol in their blood can anticipate a

7248_C015.fm Page 277 Wednesday, May 9, 2007 9:37 AM

Hansen Solubility Parameters — Biological Materials

277

slightly higher solubility of cholesterol in their blood because the continuous phase has solubility parameters closer to those of cholesterol. This effect and that discussed earlier should help to reduce cholesterol levels in the blood and blood vessels of those who ingest small to moderate amounts of alcohol on a regular basis.

LARD Experimental data and HSP correlations for the solubility of refined lard at 23°C and 37°C have been reported.2 The criterion for a good solvent is that it totally dissolves the sample at the given temperature. The concentrations chosen were 10%. The results of the correlations are given in Table 15.1. The refined lard is a semisolid with a melting point of 42°C. The composition of refined lard is very similar to that of human depot fat, so the conclusions drawn for the solubility of lard will also be generally valid for depot fat. Olive oil is a convenient material to use at room temperature to study the behavior of depot fat (lard), as the same solvents that dissolve it at room temperature also dissolve lard at 37°C. This is reported in Table 15.1. The best room temperature solvents for lard include trichloroethylene, styrene, toluene, and methyl methacrylate. Octyl alcohol does not have a strong affinity for lard at room temperature with a RED number (see Chapters 1 and 2) of 0.96. The good solvents reflect the crystalline nature of the lard, as toluene, for example, is an excellent swelling solvent for partly crystalline polyethylene. Esters are among the best solvents for lard at 37°C, reflecting the presence of the ester groups in the lard, which is very nearly a liquid at this temperature.

HUMAN SKIN A first attempt to characterize human skin with HSP was made by visually evaluating the swelling of psoriasis scales immersed for a prolonged time in different solvents.2 Uptake could clearly be seen by dimensional changes and a marked enhancement of clarity. It was anticipated that the solubility parameter correlation for the psoriasis scales (keratin) would to some extent reflect permeation in human skin but that other factors, such as the presence of water and lipids, for example, would also be important. The data fit for this correlation (0.927) indicates that a reasonably reliable correlation for swelling of the psoriasis scales (keratin) has been found. However, the δD parameter is thought to be too high. Permeation data generated in an extensive study allowed placement of the tested solvents into groups according to actual permeation rates through viable human skin.4 Figure 15.4 graphically shows the HSP correlation that resulted. There are too few data to establish a reliable correlation, but a sphere with center at δD, δP , and δH of 17.6, 12.5, and 11.0, which has a radius of 5.0, encompasses the parameters for the four solvents with the highest permeation rates while excluding the others. The units for these parameters are MPa1/2. n-octyl acetate has a near zero permeation rate. This correlation cannot be considered precise because of insufficient data, and there are, in fact, numerous spheres with somewhat similar but different combinations of the parameters that also can accomplish this. Nevertheless, there is a good guideline for future work, whether it be an expanded correlation or formulation of products designed for a prescribed compatibility with human skin. Calculations for skatole and nicotine predict that moderate rates of skin permeation can also be expected for these. It might be noted that the four solvents with high permeation rates also have very high affinity for psoriasis scales according to the correlation previously noted. Likewise, the cyclic solvents propylene carbonate, gamma-butyrolactone, and sulfolane have, or are predicted to have, high affinity for psoriasis scales, but they are placed in the low permeation rate group for actual permeation through viable human skin. These all have high δP and low δH. n-butyl acetate and toluene are also in this group. This reflects the complexity of actual skin permeation and the importance of using viable skin for testing. The cyclic nature of the solvents, however, is also expected to slow the rate

7248_C015.fm Page 278 Wednesday, May 9, 2007 9:37 AM

278

Hansen Solubility Parameters: A User’s Handbook

SOLUBILITY PARAMETER PLOT FOR SKIN PERMEATION RATE δD

δP

δH

MV

DMSO DMF DMAC NMP

18.4 17.4 16.8 18.0

16.4 16.7 11.5 12.3

10.2 11.3 10.2 7.2

71.3 77.0 92.5 96.5

MCL MEK ETH

18.2 16.0 15.8

6.3 9.0 8.8

6.1 5.1 19.4

63.9 90.1 58.5

BAC PPC TOL BTA SUL

15.8 20.0 18.0 19.0 18.4

3.7 18.0 1.4 16.6 16.6

6.3 4.1 2.0 7.4 7.4

132.5 85.0 106.8 76.8 95.3

OAC

15.8

2.9

5.1

196.0

l HIGH

l MODERATE

ⴛ LOW

à “0”

CIRCLE: δ P = 12.5, δ H = 11.0, RO = 5.0

20 HANSEN POLAR SOLUBILITY PARAMETER, δ H

PERMEATION RATE

PPCⴛ

BTA ⴛ ⴛ SUL

15

lDMSO lDMF

lNMP 10

l

MEK l

5

TOLⴛ

lDMAC

l

ETH

MCL

BACⴛ Ã OAC

0 0

5 10 15 20 HANSEN HYDROGEN BONDING PARAMETER, δ H

FIGURE 15.4 Permeation rates of selected solvents through viable human skin show a correlation with the HSP4 although the data are not extensive. Units are MPa1/2.

7248_C015.fm Page 279 Wednesday, May 9, 2007 9:37 AM

Hansen Solubility Parameters — Biological Materials

279

of permeation relative to linear solvents of comparable affinity. Factors affecting permeation have been discussed at length in Chapter 13. Of course, the presence of water and/or other skin components can also have an effect on the permeation rate. Finally, the swelling of the psoriasis scales involved equilibrium swelling of the individual systems, whereas the permeation rate studies did not have this uniformity. Concentration gradients are required for permeation to occur.

PROTEINS — BLOOD SERUM AND ZEIN HSP correlations for the swelling of blood serum and for the solubility of zein, a protein derived from corn, are included in Table 15.1. The data used in these correlations are found in Reference 3. Solvents with the lowest RED numbers in the correlation for the solubility of zein are listed in Table 15.3. The HSP parameters for blood serum and zein are not too different. The blood serum data are based on visual observation of swelling, while the zein data are for visual observation of true solution. It is noteworthy that there are only four good solvents in the data set reported in Table 15.3, and that the HSP parameters for the proteins are much higher than for any liquid which can be used in such testing. These HSP parameters are found by a form of extrapolation, where all of the good solvents are located in the boundary region of the respective spheres. The values are very much dependent on the mathematical model which includes the coefficient “4” (see Chapter 1 and Chapter 2). The saturated solution of urea and water is also a (predictably) good solvent in that it swells blood serum and dissolves zein, but it was not included as a data point in the correlations as such. Mixtures of solvents, water, and mixtures of solvents with water have been avoided as test solvents to the extent possible because of too many interactions, which are apparently not always predictable by these simple considerations. The general prediction that additions of urea to water will improve solvency of proteins is discussed below.

CHLOROPHYLL AND LIGNIN5 The results of HSP correlations of solubility for lignin and chlorophyll are given in Table 15.1. More specific information on the lignin correlation is found in Table 15.4A and Table 15.4B. It can be seen that these indeed have high affinity/physical resemblance to each other, with the HSP values not being too different. A major difference is that chlorophyll is soluble in ethanol, whereas lignin is not. This indicates a higher hydrophilicity, of course, and gives a higher δH parameter to chlorophyll compared with lignin. It can be presumed that the HSP for these materials are the result of natural selection by nature for optimum compatibility relations with immediate surroundings and function. A discussion of this is beyond the scope of this work, but this point has been studied in more detail for the relations among wood chemicals and wood polymers as outlined in the next section. Here, the HSP for lignin have a demonstrated clear importance with regard to compatibility relations.

WOOD CHEMICALS AND POLYMERS The results of HSP calculations and correlations for several wood chemicals and polymers are given in Table 15.1. These results are part of a study considering the ultrastructure of wood from a solubility parameter point of view.6 The study is based on the principle of “like seeks like” and leads to a proposed configuration of the ultrastructure. The HSP for amorphous cellulose are presumed to be similar to those of Dextran (Dextran C, British Drug Houses). The crystallinity in cellulose will require that good solvents have higher affinity/HSP than most of those dissolving Dextran, however. N-methyl-morpholine-N-oxide is an example. The HSP for Dextran are higher than those of sucrose (which values are similar to the other sugars as well). It is common for polymers to have higher HSP than the monomers from which they are made. It is also common

7248_C015.fm Page 280 Wednesday, May 9, 2007 9:37 AM

280

Hansen Solubility Parameters: A User’s Handbook

TABLE 15.3 Calculated Solubility Sphere for Zein Solvent

δD

δP

δH

1,3-Benzenediol Benzyl alcohol Diethanolamine Phenol o-Methoxyphenol Furfuryl alcohol Hexamethylphosphoramide 3-Chloro-1-propanol 1,3-Butanediol Propylene glycol Diethylene glycol Ethylenediamine m-Cresol Aniline Dipropylene glycol 1,1,2,2-Tetrabromoethane Ethanolamine Succinic anhydride 2-Pyrolidone Allyl alcohol Ethylene glycol Ethylene glycol monomethyl ether Cyclohexanol Diethylenetriamine Benzoic acid Triethyleneglycol 1,1,2,2-Tetrachloroethane Ethanol 1-Propanol Morpholine Ethylene glycol monoethyl ether Dimethylformamide Propylene glycol monophenyl ether Quinoline Hexylene glycol Dimethyl sulfone Dimethyl sulfoxide Ethylene cyanohydrin 1-Butanol 2-Propanol Ethylene dibromide Tetramethylurea Glycerol Diethylene glycol monomethyl ether Diethylene glycol monoethyl ether N,N-Dimethylacetamide Bromoform

18.0 18.4 17.2 18.0 18.0 17.4 18.5 17.5 16.6 16.8 16.6 16.6 18.0 19.4 16.5 22.6 17.0 18.6 19.4 16.2 17.0 16.2 17.4 16.7 18.2 16.0 18.8 15.8 16.0 18.8 16.2 17.4 17.4 19.4 15.7 19.0 18.4 17.2 16.0 15.8 19.2 16.7 17.4 16.2 16.1 16.8 21.4

8.4 6.3 10.8 5.9 8.2 7.6 8.6 5.7 10.0 9.4 12.0 8.8 5.1 5.1 10.6 5.1 15.5 19.2 17.4 10.8 11.0 9.2 4.1 13.3 6.9 12.5 5.1 8.8 6.8 4.9 9.2 13.7 5.3 7.0 8.4 19.4 16.4 18.8 5.7 6.1 3.5 8.2 12.1 7.8 9.2 11.5 4.1

21.0 13.7 21.2 14.9 13.3 15.1 11.3 14.7 21.5 23.3 20.7 17.0 12.9 10.2 17.7 8.2 21.2 16.6 11.3 16.8 26.0 16.4 13.5 14.3 9.8 18.6 9.4 19.4 17.4 9.2 14.3 11.3 11.5 7.6 17.8 12.3 10.2 17.6 15.8 16.4 8.6 11.0 29.3 12.6 12.2 10.2 6.1

SOLUB

0* 0* 1 1* 0 0 0

1* 1* 0

0

0 0

0 0

RED

V

0.761 0.876 0.891 0.893 0.910 0.933 0.950 0.976 0.991 0.997 0.998 0.999 1.001 1.004 1.004 1.021 1.037 1.043 1.061 1.068 1.068 1.073 1.087 1.090 1.099 1.101 1.108 1.112 1.117 1.127 1.128 1.130 1.136 1.137 1.140 1.155 1.165 1.166 1.169 1.179 1.180 1.198 1.198 1.200 1.221 1.226 1.228

87.5 103.6 95.9 87.5 109.5 86.5 175.7 84.2 89.9 73.6 94.9 67.3 104.7 91.5 130.9 116.8 59.8 66.8 76.4 68.4 55.8 79.1 106.0 108.0 100.0 114.0 105.2 58.5 75.2 87.1 97.8 77.0 143.2 118.0 123.0 75.0 71.3 68.3 91.5 76.8 87.0 120.4 73.3 118.0 130.9 92.5 87.5

7248_C015.fm Page 281 Wednesday, May 9, 2007 9:37 AM

Hansen Solubility Parameters — Biological Materials

281

TABLE 15.3 (CONTINUED) Calculated Solubility Sphere for Zein Solvent 2-Butanol 1-Octanol Ethyl lactate Methyl salicylate Zein

δD

δP

δH

15.8 17.0 16.0 16.0

5.7 3.3 7.6 8.0

14.5 11.9 12.5 12.3

SOLUB

RED

V

1.232 1.233 1.236 1.239

92.0 157.7 115.0 129.0

D = 22.4 P = 9.8 H = 19.4 RAD. = 11.9 FIT = 0.964 NO = 50

Note: Units are MPa1/2. This table contains the first entries in a much larger database to show which solvents are most likely to affect proteins. The SOLUB column indicates good solvents with a 1, bad solvents with a 0, and untested solvents with a blank. The “*” points out those solvents that do not conform exactly with the correlation.

that the solubility of crystalline polymers requires good solvents to have higher HSP than otherwise expected and that smaller molecular volume is an advantage. The relatively high HSP for cellulose, which also includes a large number of –OH groups, provides a proper energetic environment for the backbones of hemicelluloses, as well as those of their side groups which contain –OH groups. The hemicellulose side groups with acetyl and ether linkages can be expected to orient toward the lower HSP lignin. Neither lignin nor hemicelluloses are compatible with cellulose in the usual sense, but the hemicelluloses can form oriented configurations in connection with cellulose and with lignin. The monomers for lignin, sinapyl alcohol, coniferyl alcohol, and p-coumaryl alcohol all have HSP which are on the boundary of the solubility sphere for solubility of Dextran (amorphous cellulose), so their affinities indicate they will seek the lower HSP domain of the lignin. Hemicelluloses act like surfactants, with some side groups favoring the cellulose environment and others favoring the lignin environment. If one considers the HSP for higher ketones, esters, and ethers in Table 15.4, it can be seen that none of these simple liquids will dissolve lignin. This indicates that the acetyl- and ether-containing side groups on the hemicelluloses may not penetrate lignin as such but prefer to remain on its surface, probably finding a local (interface) site with closest possible HSP. A sketch of these predicted relations is found in Figure 15.5. This is a clear example of self-association in nature. In addition to those previously mentioned, one can deduce which chemicals are most prone to penetrate directly through wood. These will dissolve lignin. Included are chlorinated phenols and other wood impregnation materials. It is known that pentachlorophenol, for example, readily diffuses into and through wood specimens. Still another question is how wood transports its own chemicals at various stages of the life of a tree. The same principles are valid. A preferred pathway is where HSP are similar. This can be made possible by molecular rotation and orientation. This can perhaps change with time and local environment. Other types of predictions are possible from comparisons of the HSP correlations in Table 15.1. For example, it can be determined that all the solvents dissolving lignin are also predicted to swell psoriasis scales. This generality then suggests special care is in order when handling woodimpregnating chemicals. The protective clothing chosen should have HSP quite different from the HSP of the chemical involved, as discussed in Chapter 13. An important effect that may have been overlooked in the solubility of wood and wood components is that there are acid groups present in hemicelluloses, for example, and these can be neutralized by bases. This gives an organic salt with high HSP.11 (See also Chapter 18.) Such a salt is hydrophilic and will collect water. This may lead to phase separation, and some destruction of ultrastructure is possible. This is an effect which is known to have caused blistering in coatings.

7248_C015.fm Page 282 Wednesday, May 9, 2007 9:37 AM

282

Hansen Solubility Parameters: A User’s Handbook

TABLE 15.4A Calculated Solubility Sphere for Lignin Solubility Solvent

δD

δP

δH

SOLUB

RED

V

Acetic acid Acetic anhydride Acetone Acetonitrile Acetophenone Aniline Benzaldehyde Benzene 1-Bromonaphthalene 1,3-Butanediol 1-Butanol Butyl acetate Butyl lactate Butyric acid gamma-Butyrolactone Butyronitrile Carbon disulfide Carbon tetrachloride Chlorobenzene 1-Chlorobutane Chloroform m-Cresol Cyclohexane Cyclohexanol Cyclohexanone Cyclohexylchloride Diacetone alcohol o-Dichlorobenzene 2,2-Dichlorodiethyl ether Diethylamine Diethylene glycol Diethylene glycol monobutyl ether Diethylene glycol monomethyl ether Diethyl ether Diethyl sulfide Di(isobutyl) ketone Dimethylformamide Dimethyl sulfoxide 1,4-Dioxane Dipropylamine Dipropylene glycol Ethanol Ethanolamine Ethyl acetate Ethylbenzene 2-Ethyl-1-butanol Ethylene glycol Ethylene glycol monobutyl ether Ethylene glycol monoethyl ether Ethylene glycol monoethyl ether acetate

14.5 16.0 15.5 15.3 19.6 19.4 19.4 18.4 20.3 16.6 16.0 15.8 15.8 14.9 19.0 15.3 20.5 17.8 19.0 16.2 17.8 18.0 16.8 17.4 17.8 17.3 15.8 19.2 18.8 14.9 16.6 16.0 16.2 14.5 16.8 16.0 17.4 18.4 19.0 15.3 16.5 15.8 17.0 15.8 17.8 15.8 17.0 16.0 16.2 15.9

8.0 11.7 10.4 18.0 8.6 5.1 7.4 0.0 3.1 10.0 5.7 3.7 6.5 4.1 16.6 12.4 0.0 0.0 4.3 5.5 3.1 5.1 0.0 4.1 6.3 5.5 8.2 6.3 9.0 2.3 12.0 7.0 7.8 2.9 3.1 3.7 13.7 16.4 1.8 1.4 10.6 8.8 15.5 5.3 0.6 4.3 11.0 5.1 9.2 4.7

13.5 10.2 7.0 6.1 3.7 10.2 5.3 2.0 4.1 21.5 15.8 6.3 10.2 10.6 7.4 5.1 0.6 0.6 2.0 2.0 5.7 12.9 0.2 13.5 5.1 2.0 10.8 3.3 5.7 6.1 20.7 10.6 12.6 5.1 2.0 4.1 11.3 10.2 7.4 4.1 17.7 19.4 21.2 7.2 1.4 13.5 26.0 12.3 14.3 10.6

0 0 0 0 0 0 0 0 0 1 0 0 0 0 1 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 1 0 1* 0 0 0 1 1 0 0 1 1 1 0 0 0 1* 0 1 0

1.195 1.006 1.212 1.277 1.096 0.897 1.044 1.582 1.254 0.895 1.060 1.403 1.158 1.337 0.833 1.298 1.586 1.683 1.369 1.506 1.293 0.917 1.761 1.013 1.193 1.424 1.085 1.210 1.006 1.552 0.836 1.105 1.001 1.605 1.543 1.480 0.774 0.727 1.211 1.631 0.831 0.988 0.788 1.306 1.615 1.169 1.002 1.134 0.925 1.204

57.1 94.5 74.0 52.6 117.4 91.5 101.5 89.4 140.0 89.9 91.5 132.5 149.0 110.0 76.8 87.3 60.0 97.1 102.1 104.5 80.7 104.7 108.7 106.0 104.0 118.6 124.2 112.8 117.2 103.2 94.9 170.6 118.0 104.8 107.4 177.1 77.0 71.3 85.7 136.9 130.9 58.5 59.8 98.5 123.1 123.2 55.8 131.6 97.8 136.1

7248_C015.fm Page 283 Wednesday, May 9, 2007 9:37 AM

Hansen Solubility Parameters — Biological Materials

283

TABLE 15.4A (CONTINUED) Calculated Solubility Sphere for Lignin Solubility Solvent Ethylene glycol monomethyl ether Furan Glycerol Hexane Isoamyl acetate Isobutyl isobutyrate Isooctyl alcohol Isophorone Mesityl oxide Methanol Methylal Methyl ethyl ketone Methyl isoamyl ketone Methyl isobutyl carbinol Methyl isobutyl ketone Morpholine Nitrobenzene Nitroethane Nitromethane 2-Nitropropane 1-Pentanol 1-Propanol Propylene carbonate Propylene glycol Pyridine Styrene Tetrahydrofuran Tetrahydronaphthalene Toluene 1,1,1-Trichloroethane Trichloroethylene Xylene Lignin

δD

δP

δH

SOLUB

RED

V

16.2 17.8 17.4 14.9 15.3 15.1 14.4 16.6 16.4 15.1 15.0 16.0 16.0 15.4 15.3 18.8 20.0 16.0 15.8 16.2 15.9 16.0 20.0 16.8 19.0 18.6 16.8 19.6 18.0 16.8 18.0 17.6

9.2 1.8 12.1 0.0 3.1 2.9 7.3 8.2 6.1 12.3 1.8 9.0 5.7 3.3 6.1 4.9 8.6 15.5 18.8 12.1 4.5 6.8 18.0 9.4 8.8 1.0 5.7 2.0 1.4 4.3 3.1 1.0

16.4 5.3 29.3 0.0 7.0 5.9 12.9 7.4 6.1 22.3 8.6 5.1 4.1 12.3 4.1 9.2 4.1 4.5 5.1 4.1 13.9 17.4 4.1 23.3 5.9 4.1 8.0 2.9 2.0 2.0 5.3 3.1

1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0* 1 0 0 0 0 0 0 0

0.906 1.372 1.128 1.904 1.447 1.516 1.237 1.125 1.268 1.076 1.479 1.274 1.411 1.279 1.464 0.986 1.054 1.254 1.286 1.260 1.143 1.117 1.513 0.944 0.987 1.421 1.163 1.392 1.538 1.500 1.298 1.524

79.1 72.5 73.3 131.6 148.8 163.0 156.6 150.5 115.6 40.7 169.4 90.1 142.8 127.2 125.8 87.1 102.7 71.5 54.3 86.9 108.6 75.2 85.0 73.6 80.9 115.6 81.7 136.0 106.8 99.3 90.2 123.3

D = 21.9 P = 14.1 H = 16.9 Ro = 13.7 FIT = 0.990 NO = 82

UREA Data for the HSP correlation for urea solubility in organic solvents are given in Table 15.1. All of the parameters are rather high, which is characteristic of a low molecular weight solid. The data fit is very good. Perhaps the most interesting thing about this correlation is that it clearly shows that additions of urea to water will improve solubility for a variety of materials including proteins. This is the reason for the improved solubility discussed previously in connection with the destruction of hydrophilic bonding in proteins. The saturated solution of urea and water is also the best physically acting solvent for whole, dried blood that the author could locate in a previous (unpublished) study. The fact of high HSP for urea/water mixtures has led to its use in many varied types of products.7 The saturated solution of urea in water has found particular successes in the following examples.

7248_C015.fm Page 284 Wednesday, May 9, 2007 9:37 AM

284

Hansen Solubility Parameters: A User’s Handbook

TABLE 15.4B Calculated Solubility Sphere for Lignin Solubility Solvent 2-Pyrolidone Succinic anhydride Dimethyl sulfone Dimethyl sulfoxide Hexamethylphosphoramide o-Methoxyphenol 1,3-Butanediol Ethylene cyanohydrin Dimethyl formamide Diethylenetriamine Ethanolamine Diethanolamine Benzyl alcohol Furfuryl alcohol Dipropylene glycol gamma-Butyrolactone Diethylene glycol Phenol Ethylenediamine Allyl alcohol Triethyleneglycol 1,3-Butanediol Aniline 3-Chloro-1-propanol Ethylene glycol monomethyl ether N,N-Dimethyl acetamide Trimethylphosphate Benzoic acid m-Cresol Methyl-2-pyrrolidone 1,1,2,2-Tetrabromoethane Ethylene glycol monoethyl ether Quinoline Propylene glycol Triethylphosphate 1,1,2,2-Tetrachloroethane Tetramethylurea Diethylene glycol monoethyl ether Morpholine Pyridine Ethanol Furfural Hexylene glycol Propylene glycol monophenyl ether Diethylene glycol monomethyl ether Ethylene glycol 2,2-Dichlorodiethyl ether Acetic anhydride Tricresyl phosphate Cyclohexanol

δD

δP

δH

19.4 18.6 19.0 18.4 18.5 18.0 18.0 17.2 17.4 16.7 17.0 17.2 18.4 17.4 16.5 19.0 16.6 18.0 16.6 16.2 16.0 16.6 19.4 17.5 16.2 16.8 16.7 18.2 18.0 18.0 22.6 16.2 19.4 16.8 16.7 18.8 16.7 16.1 18.8 19.0 15.8 18.6 15.7 17.4 16.2 17.0 18.8 16.0 19.0 17.4

17.4 19.2 19.4 16.4 8.6 8.2 8.4 18.8 13.7 13.3 15.5 10.8 6.3 7.6 10.6 16.6 12.0 5.9 8.8 10.8 12.5 10.0 5.1 5.7 9.2 11.5 15.9 6.9 5.1 12.3 5.1 9.2 7.0 9.4 11.4 5.1 8.2 9.2 4.9 8.8 8.8 14.9 8.4 5.3 7.8 11.0 9.0 11.7 12.3 4.1

11.3 16.6 12.3 10.2 11.3 13.3 21.0 17.6 11.3 14.3 21.2 21.2 13.7 15.1 17.7 7.4 20.7 14.9 17.0 16.8 18.6 21.5 10.2 14.7 16.4 10.2 10.2 9.8 12.9 7.2 8.2 14.3 7.6 23.3 9.2 9.4 11.0 12.2 9.2 5.9 19.4 5.1 17.8 11.5 12.6 26.0 5.7 10.2 4.5 13.5

SOLUB

1

1 1

1 1 1

1 1 1

1

1 0*

1 1 1

1* 1 0 0 0

RED

V

0.599 0.609 0.665 0.727 0.758 0.761 0.766 0.769 0.774 0.785 0.788 0.792 0.800 0.821 0.831 0.833 0.836 0.839 0.865 0.866 0.878 0.895 0.897 0.902 0.906 0.911 0.913 0.915 0.917 0.918 0.919 0.925 0.929 0.944 0.965 0.968 0.973 0.981 0.986 0.987 0.988 0.989 0.998 1.000 1.001 1.002 1.006 1.006 1.008 1.013

76.4 66.8 75.0 71.3 175.7 109.5 87.5 68.3 77.0 108.0 59.8 95.9 103.6 86.5 130.9 76.8 94.9 87.5 67.3 68.4 114.0 89.9 91.5 84.2 79.1 92.5 115.8 100.0 104.7 96.5 116.8 97.8 118.0 73.6 171.0 105.2 120.4 130.9 87.1 80.9 58.5 83.2 123.0 143.2 118.0 55.8 117.2 94.5 316.0 106.0

7248_C015.fm Page 285 Wednesday, May 9, 2007 9:37 AM

Hansen Solubility Parameters — Biological Materials

285

TABLE 15.4B (CONTINUED) Calculated Solubility Sphere for Lignin Solubility Solvent 1-Propanol Propylene carbonate Triethyanolamine Nonyl phenoxy ethanol Methyl salicylate Dimethyl phthalate Ethyl lactate Benzaldehyde Trifluoroacetic acid Di-(2-Chloro-isopropyl) ether Nitrobenzene Ethylene dibromide 1-Butanol 2-Propanol Methanol Bromoform Diacetone alcohol Ethylene carbonate Epichlorohydrin 2-Butanol Acetophenone Diethylene glycol monobutyl ether Methylene dichloride Benzyl butyl phthalate Acrylonitrile Formic acid Isophorone 1-Octanol Glycerol Formamide Ethylene glycol monobutyl ether Ethylene dichloride 1-Pentanol 1-Nitropropane Bromobenzene Ethylene glycol monomethyl ether acetate Ethyl cinnamate Propylene glycol monomethyl ether Diethyl phthalate Diethyl sulfate Butyl lactate Diethylene glycol hexyl ether Propylene glycol monoethyl ether Propylamine Tetrahydrofuran 2-Octanol 2-Ethyl-1-butanol Isobutyl alcohol 2-Methyl-1-propanol 1-Decanol

δD

δP

δH

SOLUB

RED

V

16.0 20.0 17.3 16.7 16.0 18.6 16.0 19.4 15.6 19.0 20.0 19.2 16.0 15.8 15.1 21.4 15.8 19.4 19.0 15.8 19.6 16.0 18.2 19.0 16.4 14.3 16.6 17.0 17.4 17.2 16.0 19.0 15.9 16.6 20.5 15.9 18.4 15.6 17.6 15.7 15.8 16.0 15.7 16.9 16.8 16.1 15.8 15.1 15.1 17.5

6.8 18.0 22.4 10.2 8.0 10.8 7.6 7.4 9.9 8.2 8.6 3.5 5.7 6.1 12.3 4.1 8.2 21.7 10.2 5.7 8.6 7.0 6.3 11.2 17.4 11.9 8.2 3.3 12.1 26.2 5.1 7.4 4.5 12.3 5.5 5.5 8.2 6.3 9.6 14.7 6.5 6.0 6.5 4.9 5.7 4.9 4.3 5.7 5.7 2.6

17.4 4.1 23.3 8.4 12.3 4.9 12.5 5.3 11.6 5.1 4.1 8.6 15.8 16.4 22.3 6.1 10.8 5.1 3.7 14.5 3.7 10.6 6.1 3.1 6.8 16.6 7.4 11.9 29.3 19.0 12.3 4.1 13.9 5.5 4.1 11.6 4.1 11.6 4.5 7.1 10.2 10.0 10.5 8.6 8.0 11.0 13.5 15.9 15.9 10.0

0 0

1.013 1.015 1.018 1.021 1.026 1.028 1.034 1.044 1.044 1.052 1.054 1.059 1.060 1.066 1.076 1.077 1.085 1.088 1.090 1.095 1.096 1.105 1.112 1.113 1.116 1.121 1.125 1.125 1.128 1.129 1.134 1.136 1.143 1.144 1.144 1.145 1.149 1.149 1.149 1.154 1.158 1.160 1.160 1.162 1.163 1.163 1.169 1.169 1.169 1.171

75.2 85.0 133.2 275.0 129.0 163.0 115.0 101.5 74.2 146.0 102.7 87.0 91.5 76.8 40.7 87.5 124.2 66.0 79.9 92.0 117.4 170.6 63.9 306.0 67.1 37.8 150.5 157.7 73.3 39.8 131.6 79.4 108.6 88.4 105.3 121.6 166.8 93.8 198.0 131.5 149.0 204.3 115.6 83.0 81.7 159.1 123.2 92.8 92.8 191.8

0

0 0 0 0

0 0

0 0 0 0

0

0 0

7248_C015.fm Page 286 Wednesday, May 9, 2007 9:37 AM

286

Hansen Solubility Parameters: A User’s Handbook

TABLE 15.4B (CONTINUED) Calculated Solubility Sphere for Lignin Solubility Solvent

δD

δP

δH

Propylene glycol monopropyl ether Dibutyl phthalate Dipropylene glycol methyl ether Cyclohexanone Acetic acid Ethyl formate Trichlorobiphenyl Anisole Ethylene glycol monoethyl ether acetate o-Dichlorobenzene 1,4-Dioxane Acetone Nonyl phenol Acetaldehyde 1,1-Dimethyl hydrazine bis-(m-Phenoxyphenyl) ether 2,4-Pentanedione Ethyl chloroformate Dibenzyl ether 2-Ethyl hexanol Isooctyl alcohol Tetrachloroethylene 2-(Diethylamino) ethanol Benzyl chloride Benzonitrile Ethyl bromide Nitroethane 1-Bromonaphthalene Naphthalene 2-Nitropropane Methyl acetate 2,2,4-Trimethyl 1,3-pentanediol monoisobutyrate Methylene diiodide Butylamine Mesityl oxide 1,1-Dichloroethylene Propionitrile Methyl ethyl ketone Acetonitrile Methyl isobutyl carbinol Ethanethiol Methyl methacrylate Nitromethane Chloroform Diethylene glycol butyl ether acetate Butyronitrile Trichloroethylene Cyclohexylamine Methyl acrylate Ethyl acetate

15.8 17.8 15.5 17.8 14.5 15.5 19.2 17.8 15.9 19.2 19.0 15.5 16.5 14.7 15.3 19.6 17.1 15.5 17.3 15.9 14.4 19.0 14.9 18.8 17.4 16.5 16.0 20.3 19.2 16.2 15.5 15.1 17.8 16.2 16.4 17.0 15.3 16.0 15.3 15.4 15.7 17.5 15.8 17.8 16.0 15.3 18.0 17.2 15.3 15.8

7.0 8.6 5.7 6.3 8.0 8.4 5.3 4.1 4.7 6.3 1.8 10.4 4.1 8.0 5.9 3.1 9.0 10.0 3.7 3.3 7.3 6.5 5.8 7.1 9.0 8.0 15.5 3.1 2.0 12.1 7.2 6.1 3.9 4.5 6.1 6.8 14.3 9.0 18.0 3.3 6.5 5.5 18.8 3.1 4.1 12.4 3.1 3.1 9.3 5.3

9.2 4.1 11.2 5.1 13.5 8.4 4.1 6.7 10.6 3.3 7.4 7.0 9.2 11.3 11.0 5.1 4.1 6.7 7.3 11.8 12.9 2.9 12.0 2.6 3.3 5.1 4.5 4.1 5.9 4.1 7.6 9.8 5.5 8.0 6.1 4.5 5.5 5.1 6.1 12.3 7.1 4.3 5.1 5.7 8.2 5.1 5.3 6.5 5.9 7.2

SOLUB

0 0

0 0 0 0

0

0 0 0

0

0 0 0

0 0 0 0

0

RED

V

1.174 1.180 1.192 1.193 1.195 1.196 1.200 1.202 1.204 1.210 1.211 1.212 1.212 1.212 1.213 1.224 1.226 1.232 1.232 1.236 1.237 1.237 1.241 1.247 1.247 1.250 1.254 1.254 1.257 1.260 1.260 1.263 1.267 1.267 1.268 1.271 1.273 1.274 1.277 1.279 1.280 1.286 1.286 1.293 1.295 1.298 1.298 1.301 1.302 1.306

130.3 266.0 157.4 104.0 57.1 80.2 187.0 119.1 136.1 112.8 85.7 74.0 231.0 57.1 76.0 373.0 103.1 95.6 192.7 156.6 156.6 101.1 133.2 115.0 102.6 76.9 71.5 140.0 111.5 86.9 79.7 227.4 80.5 99.0 115.6 79.0 70.9 90.1 52.6 127.2 74.3 106.5 54.3 80.7 208.2 87.3 90.2 113.8 113.8 98.5

7248_C015.fm Page 287 Wednesday, May 9, 2007 9:37 AM

Hansen Solubility Parameters — Biological Materials

287

TABLE 15.4B (CONTINUED) Calculated Solubility Sphere for Lignin Solubility Solvent Propylene glycol monoisobutyl ether Propylene glycol monobutyl ether Di(2-Methoxyethyl) ether Ethylene glycol butyl ether acetate 1-Methyl naphthalene Bromochloromethane Butyric acid Diethyl ketone Ethyl acrylate Tributyl phosphate Diethyl carbonate Chlorobenzene Furan Dioctyl phthalate Di-iso-butyl carbinol Methacrylonitrile Tetrahydronaphthalene Butyl acrylate Butyl acetate Stearic acid Methyl isoamyl ketone Ethyl butyl ketone Octanoic acid Styrene Cyclohexylchloride Amyl acetate Butyraldehyde sec-Butyl acetate Ethyl amyl ketone Isoamyl acetate Biphenyl Dichloromonoflouromethane Propyl chloride Methyl butyl ketone Methyl isobutyl ketone Methyl amyl acetate Isobutyl acetate Methyl chloride Methylal Di(isobutyl) ketone Ethyl chloride Tridecyl alcohol 1,1,1-Trichloroethane 1,1-Dichlorethane 1-Chlorobutane o-Xylene Isobutyl isobutyrate Xylene Oleyl alcohol Toluene

δD

δP

δH

15.1 15.3 15.7 15.3 20.6 17.3 14.9 15.8 15.5 16.3 16.6 19.0 17.8 16.6 14.9 15.3 19.6 15.6 15.8 16.3 16.0 16.2 15.1 18.6 17.3 15.8 14.7 15.0 16.2 15.3 21.4 15.8 16.0 15.3 15.3 15.2 15.1 15.3 15.0 16.0 15.7 14.3 16.8 16.5 16.2 17.8 15.1 17.6 14.3 18.0

4.7 4.5 6.1 4.5 0.8 5.7 4.1 7.6 7.1 6.3 3.1 4.3 1.8 7.0 3.1 10.8 2.0 6.2 3.7 3.3 5.7 5.0 3.3 1.0 5.5 3.3 5.3 3.7 4.5 3.1 1.0 3.1 7.8 6.1 6.1 3.1 3.7 6.1 1.8 3.7 6.1 3.1 4.3 8.2 5.5 1.0 2.9 1.0 2.6 1.4

9.8 9.2 6.5 8.8 4.7 3.5 10.6 4.7 5.5 4.3 6.1 2.0 5.3 3.1 10.8 3.6 2.9 4.9 6.3 5.5 4.1 4.1 8.2 4.1 2.0 6.1 7.0 7.6 4.1 7.0 2.0 5.7 2.0 4.1 4.1 6.8 6.3 3.9 8.6 4.1 2.9 9.0 2.0 0.4 2.0 3.1 5.9 3.1 8.0 2.0

SOLUB

0

0 0

0 0 0

0 0

0

0

0 0

0 0 0 0 0

RED

V

1.313 1.317 1.318 1.330 1.331 1.336 1.337 1.346 1.351 1.356 1.366 1.369 1.372 1.372 1.374 1.389 1.392 1.395 1.403 1.408 1.411 1.417 1.418 1.421 1.424 1.427 1.428 1.432 1.434 1.447 1.450 1.451 1.462 1.464 1.464 1.465 1.470 1.473 1.479 1.480 1.485 1.486 1.500 1.502 1.506 1.512 1.516 1.524 1.535 1.538

132.2 132.0 142.0 171.2 138.8 65.0 110.0 106.4 108.8 345.0 121.0 102.1 72.5 377.0 177.8 83.9 136.0 143.8 132.5 326.0 142.8 139.0 159.0 115.6 118.6 148.0 88.5 133.6 156.0 148.8 154.1 75.4 88.1 123.6 125.8 167.4 133.5 55.4 169.4 177.1 70.0 242.0 99.3 84.8 104.5 121.2 163.0 123.3 316.0 106.8

7248_C015.fm Page 288 Wednesday, May 9, 2007 9:37 AM

288

Hansen Solubility Parameters: A User’s Handbook

TABLE 15.4B (CONTINUED) Calculated Solubility Sphere for Lignin Solubility Solvent Diethyl sulfide Diethyl amine Benzene Naphtha.high-flash Carbon disulfide Oleic acid Diethyl ether Triethylene glycol monooleyl ether Ethylbenzene Methyl oleate Dipropylamine Dibutyl stearate Triethylamine Trimethylbenzene Isopropyl palmitate Dibutyl sebacate cis-Decahydronaphthalene para-Diethyl benzene Mesitylene Carbon tetrachloride trans-Decahydronaphthalene Chlorodiflouromethane Cyclohexane Methyl cyclohexane Eicosane Trichlorofluoromethane Hexadecane Dodecane Mineral spirits Decane Nonane Octane 1,1,2-Trichlorotrifluoroethane Heptane Hexane Pentane Tetraethylorthosilicate Butane 2,2,2,4-Trimethylpentane Isopentane 1,2-Dichlorotetrafluoroethane Dichlorodifluoromethane Water Perfluoro(dimethylcyclohexane) Perfluoromethylcyclohexane Perfluoroheptane Bromotrifluoromethane Lignin

δD

δP

δH

SOLUB

RED

V

16.8 14.9 18.4 17.9 20.5 14.3 14.5 13.3 17.8 14.5 5.3 14.5 17.8 17.8 14.3 13.9 18.8 18.0 18.0 17.8 18.0 12.3 16.8 16.0 16.5 15.3 16.3 16.0 15.8 15.7 15.7 15.5 14.7 15.3 14.9 14.5 13.9 14.1 14.1 13.7 12.6 12.3 15.5 12.4 12.4 12.0 9.6

3.1 2.3 0.0 0.7 0.0 3.1 2.9 3.1 0.6 3.9 1.4 3.7 0.4 0.4 3.9 4.5 0.0 0.0 0.0 0.0 0.0 6.3 0.0 0.0 0.0 2.0 0.0 0.0 0.1 0.0 0.0 0.0 1.6 0.0 0.0 0.0 0.4 0.0 0.0 0.0 1.8 2.0 16.0 0.0 0.0 0.0 2.4

2.0 6.1 2.0 1.8 0.6 5.5 5.1 8.4 1.4 3.7 4.1 3.5 1.0 1.0 3.7 4.1 0.0 0.6 0.6 0.6 0.0 5.7 0.2 1.0 0.0 0.0 0.0 0.0 0.2 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.6 0.0 0.0 0.0 0.0 0.0 42.3 0.0 0.0 0.0 0.0

0 0 0

1.543 1.552 1.582 1.585 1.586 1.603 1.605 1.614 1.615 1.628 1.631 1.643 1.645 1.645 1.647 1.652 1.669 1.673 1.673 1.683 1.704 1.719 1.761 1.774 1.790 1.797 1.803 1.823 1.823 1.844 1.844 1.858 1.860 1.873 1.904 1.936 1.944 1.969 1.969 2.003 2.042 2.065 2.081 2.122 2.122 2.161 2.340

107.4 103.2 89.4 181.8 60.0 320.0 104.8 418.5 123.1 340.0 136.9 382.0 138.6 133.6 330.0 339.0 156.9 156.9 139.8 97.1 156.9 72.9 108.7 128.3 359.8 92.8 294.1 228.6 125.0 195.9 179.7 163.5 119.2 147.4 131.6 116.2 224.0 101.4 166.1 117.4 117.6 92.3 18.0 217.4 196.0 227.3 97.0

0 0 0 0

0

0

0

D = 21.9 P = 14.1 H = 16.9 RAD. = 13.7 FIT = 0.990 NO = 82

7248_C015.fm Page 289 Wednesday, May 9, 2007 9:37 AM

Hansen Solubility Parameters — Biological Materials

289

--M1β4M1β4M1β4G1β4M1β4G1β4M1β4M1β4G– 2 3 3 2 3 6 Ac Ac β Ac Ac α (LIGN) 1 (LIGN) 1 M Ga (CELL) (CELL) FIGURE 15.5 Expected generalized sketch of the configuration of cellulose, hemicelluloses, and lignin in wood cell walls. See text or Reference 6 for further details. The sketch is for glucomannan. M is mannose monomer; G is glucose monomer; Ga is galactose monomer; Ac is an acetyl group; (LIGN) is a region similar in HSP to lignin (or acetal etc.); (CELL) is a region similar in HSP to cellulose, being any of cellulose, hemicellulose backbone, or hemicellulose side chain with an alcohol group (M, Ga).

1. Lithographic stones were previously conditioned to make them more receptive to ink by application of this liquid to change wetting behavior. 2. The saturated solution of urea and water, which swells and softens wood, has been used to give wood flexibility so that it can easily be formed. 3. It has been used by Eskimos to soften seal skins by swelling and softening them. A similar application in Mexico involves curing leather. This application probably originated in prehistoric times. 4. It has been used to improve the flow of house paints on cold days or when no other source of liquid has been available (such as on a scaffold), as it is a good solvent miscible in many paints. 5. It is reported to have been used to set hair, as it also softens and swells it. 6. It was used in the early manufacture of gunpowder as a dispersion medium during grinding because of improved wetting for the powder. 7. Amazonian Indians used this liquid to coagulate latex prior to sale and shipment. This was practiced particularly during World War II. Other unspecified and undocumented uses include those possible because the liquid has the ability to soften human skin, thus allowing easier transport of medicinal chemicals into the body. Urea itself has HSP very close to those of sugar and proteins. As all of these are biocompatible materials, it is clear that the incorporation of significant numbers of urea groups in, for example, polyurethane polymers or other products, can greatly enhance biocompatibility.

WATER Water has been discussed in detail in Chapter 1. Briefly stated, one can use the HSP for water or a correlation for water solubility to get a general explanation for observed phenomena. Accurate calculation of the HSP for solvent–water mixtures cannot be expected because of the irregularities of water associating with itself, the solvent, and a potential solute. Lindenfors12 described the association of two molecules of water with one molecule of dimethyl sulfoxide, a solvent frequently mentioned in connection with biological systems. A simplistic approach based on the ratio of δH for water as a single molecule vs. that in the correlation(s) for water solubility suggests that (42.3/16.5)2 or about six water molecules are linked by hydrogen bonding into some type of entity. Various structures for assemblies of water molecules have been discussed in the literature. The clusters with six water molecules are among the more probable ones.13 The data on water solubility used in the HSP correlations are reported by Wallström and Svenningsen.14

7248_C015.fm Page 290 Wednesday, May 9, 2007 9:37 AM

290

Hansen Solubility Parameters: A User’s Handbook

SURFACE MOBILITY Surface mobility allows given segments of molecules to orient at surfaces in a direction where their HSP match more closely. The surfaces of hydrophobic polymers (peat moss) can become hydrophilic when contacted with water. One can speculate as to why this occurs. One possibility is that this phenomenon conserves water within the structure. Whenever water is present on an otherwise hydrophobic surface, it can become hydrophilic if the surface molecules can rotate or move hydrophilic entities toward the water. This allows the water to spontaneously spread and potentially enter the structure if there are suitable passages. When this is accomplished, and contact with water ceases, the surface dries and becomes hydrophobic once more. The molecules rotate with a lower energy moiety toward the air. This hydrophobic surface helps prevent evaporation of water, as water is not particularly soluble in it, and the hydrophilic segments oriented toward the interior of the structure will help bind the water where it is. The basis of the orientation effects described earlier for hemicelluloses is another example of orientation toward regions where HSP matches better. These phenomena are also discussed in Chapter 18. It is also appropriate to repeat that solvent quality has a great deal to do with pigment dispersion stability, in that the adsorbed stabilizing polymer should remain on the pigment surface. A solvent which is too good can remove it. This is discussed in detail in Chapter 5. The implication of these examples is that solvent quality is very important for the orientation of molecules at interfaces. A change in solvent quality can easily lead to a change in the configuration of molecules at surfaces. It is not surprising that Nature has used this to advantage in various ways.

CHIRAL ROTATION, HYDROGEN BONDING, AND NANOENGINEERING It has been found that anthracene units appended to a single screw-sense helical polyguanidine changed orientation when the temperature was increased beyond 38.5°C.15 The configuration found above 38.5°C was the same as that found in tetrahydrofurane. At temperatures lower than this, the orientation of the appended anthracene was that found in toluene. A mixture of tetrahydrofurane/toluene equal to 90/10 vol% approximated the conditions at the critical temperature. For those who have diligently read this handbook, it would appear obvious that it is the cohesive energy density just above or just below the critical temperature that controls the structure. More specifically it is the set of HSP values that do this, as these reflect the mix of sources of the cohesive energy density according to Equation 1.6 to Equation 1.8. There is also massive evidence showing that the interactions can be interpreted as the difference in HSP using Equation 1.9. It is well known that solubility limits can be passed by lowering the temperature in some cases and by increasing it in other cases. When the cohesive energy density of the solvent is higher than that of the polymer, solvency increases with increased temperature. When the cohesive energy density of the solvent is lower than that of the polymer, solvency decreases with increases in temperature. This is discussed in Chapter 2 and has been thoroughly treated by Patterson.16,17 In the present case the HSP of toluene are comparable to those of anthracene whereas tetrahydrofuran has much higher values. δD, δP, and δH equal to 18.7, 4.1, and 3.3 for anthracene have been reported by a multiple regression technique based on its solubility in a large number of solvents.18 The corresponding values are 18.0, 1.4, and 2.0 for toluene and 16.8, 5.7, and 8.0 for tetrahydrofurane. Thus increasing the temperature will increase the solvency in tetrahydrofurane to the point where it becomes able to cause the appended anthracene to adopt the same orientation as it has in toluene. As toluene has lower HSP than anthracene, the solvency will decrease with increases in temperature. Extending this way of thinking to the problem of moving very large biological molecules — while they are being assembled, for example — is presumably one of controlling the local solubility. When the molecule is locally able to reside in the surrounding fluid, it can move much more readily

7248_C015.fm Page 291 Wednesday, May 9, 2007 9:37 AM

Hansen Solubility Parameters — Biological Materials

291

than when it is not. An insoluble molecule or molecular segment will adsorb at a location where the energies (HSP) match, and where the geometry is also accommodating. This is most often called hydrogen bonding, but it must be all three (or more) types of cohesive energy that are collectively active. The molecule or molecular segment can be removed again when it and the surrounding liquid have a favorable energy relation.

CONCLUSION Many materials of biological significance have been assigned HSP based on their interaction with a large number of solvents whose HSP are known. A correlation for solvent effects on DNA has ranked the extent of these effects for different solvents in essentially the same order as that reported in an older study.9 This correlation for DNA can presumably be improved by additional data, but still reflects the magnitudes of the types of energies that are involved in forming/destroying the double helices. The δD;δP;δH found for DNA are 19.0;20.0;11.0, all in MPa1/2. This clearly shows that hydrogen bonding is by far the smallest of the energies involved in the noncovalent interactions that determine the DNA structure. A HSP correlation has been used to find predictably synergistic solvent mixtures where two nonsolvents dissolve cholesterol when mixed. The ethanol/aliphatic hydrocarbon synergistic mixture is discussed as being of particular interest to the fate of cholesterol in lipid layers. The HSP of chlorophyll and lignin are quite similar, indicating they will be compatible with very much the same kind of surroundings. The physical interrelationships for wood chemicals and wood polymers (lignin, hemicelluloses, and cellulose) are discussed. The side chains on hemicelluloses which contain alcohol groups and the hemicellulose backbone will be most compatible with cellulose and will orient toward this. The hemicellulose side chains without alcohol groups (acetal, acid) are closer in HSP to lignin and will orient in this direction. The acetal side chains actually have lower HSP than will dissolve lignin, for which reason they are expected to lie on the surface of the lignin or perhaps penetrate slightly into the lignin at very special local points where the HSP match is better than the average values seen over the lignin molecule as a whole. Molecular design of molecules or structures that change conformation with slight changes in the cohesive energy characteristics of given continuous media seems possible using HSP concepts. The changes are caused by preferred orientation of segments of one conformation toward the continuous phase, where its HSP match better, thus reducing the free energy of the system. If the cohesive energy characteristics of the continuous media change in a direction that no longer favors this orientation, the molecule will change configuration to one where the free energy is lower. The attraction of the segments not oriented toward the continuous phase to neighboring molecules is commonly called hydrogen bonding in proteins and similar materials. This attraction is caused collectively by all the types of energy involved through the prevailing difference in HSP and is a result of insolubility (rejection by) the continuous media. Geometrical considerations are clearly also a major factor in addition to the cohesive energy density focused upon here. HSP analyses of relative affinities can be applied to a large number of other biological materials and may provide insights into relationships which are not readily obvious or cannot be studied otherwise. The best situation is where the materials in question can be tested directly, otherwise the calculation procedures described in Chapter 1 can be used with some loss of reliability in the predictions.

REFERENCES 1. Hansen, C.M., The three dimensional solubility parameter — key to paint component affinities I. Solvents, plasticizers, polymers, and resins, J. Paint Technol., 39(505), 104–117, 1967.

7248_C015.fm Page 292 Wednesday, May 9, 2007 9:37 AM

292

Hansen Solubility Parameters: A User’s Handbook

2. Hansen, C.M. and Andersen, B.H., The affinities of organic solvents in biological systems, Am. Ind. Hyg. Assoc. J., 49(6), 301–308, 1988. 3. Hansen, C.M., The universality of the solubility parameter, Ind. Eng. Chem. Prod. Res. Dev., 8(1), 2–11, 1969. 4. Ursin, C., Hansen, C.M., Van Dyk, J.W., Jensen, P.O., Christensen, I.J., and Ebbehoej, J., Permeability of commercial solvents through living human skin, Am. Ind. Hyg. Assoc. J., 56, 651–660, 1995. 5. Hansen, C.M., 25 years with the solubility parameter (25 År med Opløselighedsparametrene, in Danish), Dan. Kemi, 73(8), 18–22, 1992. 6. Hansen, C.M. and Björkman, A., The ultrastructure of wood from a solubility parameter point of view, Holzforschung, 52(4), 335–344, 1998. 7. Hansen, C.M., Solvents for coatings, Chem. Technol., 2(9), 547–553, 1972. 8. Blake, R.D. and Delcourt, S.G., Thermodynamic effects of formamide on DNA stability, Nucl. Acid Res., 24, 2095–2103, 1996. 9. Ts’o, P.O.P., Helmkamp, G.K., and Sander, C., Interaction of nucleosides and related compounds with nucleic acids as indicated by the change of helix-coil transition temperature, Proc. Natl. Acad. Sci. U S A, 48, 686–698, 1962. 10. Hansen, C.M., Cohesion energy parameters applied to surface phenomena, Handbook of Surface and Colloid Chemistry, Birdi, K.S., Ed., CRC Press, Boca Raton, FL, 1997, pp. 313–332. 11. Hansen, C.M., Some aspects of acid/base interactions (Einige Aspekte der Säure/Base-Wechselwirkung, in German), Farbe und Lack, 7, 595–598, 1977. 12. Lindenfors, S., Solubility of cellulose ethers (Löslichkeit der Celluloseeäther, in German), Das Papier, 21, 65–69, 1967. 13. Gregory, J.K., Clary, D.C., Liu, K., Brown, M.G., and Saykally, R.J., The Water Dipole Moment in Water Clusters, Science, 275, 1997, pp. 814–817. 14. Wallström, E. and Svenningsen, I., Handbook of Solvent Properties, Report T1-84, Scandinavian Paint and Printing Ink Research Institute, Hoersholm, Denmark, 1984. 15. Tang, H.-Z., Novak, B.M., He, J., and Polavarapu, P.L., A thermal and solvocontrollable cylindrical nanoshutter based on a single screw-sense helical polyguanidine, Angew. Chem. Int. Ed., 44, 7298–7301, 2005. 16. Patterson, D. and Delmas, G., New aspects of polymer solution thermodynamics, Off. Dig. Fed. Soc. Paint Technol., 34(450), 677–692, 1962. 17. Delmas, D., Patterson, D., and Somcynsky, T., Thermodynamics of polyisobutylene-n-alkane systems, J. Polym. Sci., 57, 79–98, 1962. 18. Wu, P.L., Beerbower, A., and Martin, A., Extended Hansen approach: calculating partial solubility parameters of solid solutes, J. Pharm. Sci., 71(11), 1285–1287 (1982).

7248_C016.fm Page 293 Wednesday, May 23, 2007 11:36 AM

and Diffusion in 16 Absorption Polymers Charles M. Hansen ABSTRACT Predicting whether or not a given chemical will attack a given polymer is important. Hansen solubility parameters (HSP) have been used for this purpose as discussed elsewhere in this book. Consideration of the absorption and diffusion of the chemical in the polymer is often required in addition to HSP in order to make reliable predictions, however. This has been discussed in particular in Chapter 12 through Chapter 14, where chemical resistance, barrier properties, and environmental stress cracking are treated in detail. Chemicals with smaller and more linear molecules absorb and diffuse more readily than those with larger and more bulky structures. Surface resistance to absorption is sometimes so dominating that absorption does not occur in some cases, even though this might be expected based on simple HSP considerations. This chapter examines surface resistances in connection with absorption and diffusion in polymers in order to help improve understanding of these factors and to emphasize the necessity of simultaneous consideration of surface resistance when absorption rates and diffusion within the bulk of the polymer itself are of interest. Methods to measure surface resistance and concentration-dependent diffusion coefficients are discussed. Solving the diffusion equation with simultaneous consideration of surface resistance and with a concentration dependent diffusion correctly models absorption, desorption, film formation by solvent evaporation, and various forms of so-called anomalous diffusion such as “time-dependent,” Case II, and Super Case II. Surface phenomena such as surface resistance to absorption deserve far more attention than has been given in the past.

LIST OF SYMBOLS USED IN THIS CHAPTER (Please note that these are different from those used in the other chapters.) A B C CA Cs D D0 D1 D2 Dapp Dav Dmax Dv F FM FB Fa

Minimum cross-sectional area of molecule in Equation 16.20 Ratio of diffusion resistance to surface resistance. See Equation 16.13 Dimensionless concentration. See Equation 16.5 Concentration at break in curve in Figure 16.1 Dimensionless surface concentration Diffusion coefficient. Preferred units are cm2/s Diffusion coefficient at zero concentration or lowest concentration in an experiment Diffusion coefficient on exposed side of film Diffusion coefficient on exit side of film Apparent diffusion coefficient Average diffusion coefficient Maximum diffusion coefficient in an experiment Increase in the diffusion coefficient over zero conditions for a given situation Mass flux Correction factor for concentration dependent diffusion in Equation 16.11 Correction factor for surface resistance in Equation 16.11 Correction factor for concentration dependent diffusion in absorption experiments 293

7248_C016.fm Page 294 Wednesday, May 23, 2007 11:36 AM

294

Fd L Mt M∞ P Papp P∞ R Rd Ri Rs T X Vf Vt V2 b c cs c0 c1 c2 c∞ h hav k k2 l t t1/2 w x Δp ΦR

Hansen Solubility Parameters: A User’s Handbook

Correction factor for concentration dependent diffusion in desorption experiments Film thickness Absorbed mass at time t Mass at equilibrium conditions Permeation coefficient Apparent permeation coefficient True permeation coefficient Radius of cylindrical sample in Equation 16.24 Resistance to mass transport from diffusion Resistances to permeation from sources 1, 2, 3, etc., in Equation 16.17 to 16.19 Resistance to mass transport by surface resistance(s) Dimensionless time given by Equation 16.3 Dimensionless distance Volume fraction of diffusing material Constant in Figure 16.4 used to find local diffusion coefficients at Vf greater than 0.20 Diffusion coefficients at Vf equal 0.20 relative to diffusion coefficient at zero concentration (Figure 16.1 and Figure 16.4) Thickness of cylindrical sample in Equation 16.24 Concentration of diffusing material Surface concentration Initial concentration Concentration on exposed side of sample Concentration on exit side of sample Concentration at equilibrium conditions Surface mass transfer coefficient Average surface mass transfer coefficient Constant in Equation 16.6 Constant in Equation 16.20 Length of sample in Equation 16.23 Time Time required to absorb (or desorb) one half of the equilibrium amount Width of sample in Equation 16.23 Distance Pressure difference of diffusing material across membrane Local concentration is Figure 16.6

INTRODUCTION Chapters 12 through 14 have dealt with chemical resistance, environmental stress cracking, and barrier polymers, respectively. Absorption and diffusion of the chemicals into polymers are important in each of these. Factors affecting absorption and diffusion in polymers are therefore of considerable importance and must frequently be included along with the Hansen solubility parameters (HSP) to make correlations and predictions for these phenomena. This chapter emphasizes the importance of surface resistance for absorption in polymers as this has been largely overlooked in the literature. The focus of much of the relevant literature has been on anomalous diffusion, but even in this context the influence of surface resistance has largely been neglected. Surface resistance can delay or prevent the absorption of solvents that should absorb readily based on simple HSP considerations. This could lead to a false sense of security based on short time testing only. Absorption requires some degree of solubility. Therefore, equilibrium absorption can be expected to correlate with HSP. In studies of the absorption of a chemical into a polymer, its surface concentration is usually assumed to reach the equilibrium value immediately. As discussed in the

7248_C016.fm Page 295 Wednesday, May 23, 2007 11:36 AM

Absorption and Diffusion in Polymers

295

following, this is not always true. Whatever the rate at which the surface concentration increases, absorption will proceed according to the laws of diffusion, Fick’s First Law, Equation 16.1, and Fick’s Second Law, Equation 16.2. The latter is often called the diffusion equation. Its derivation can be found in Crank.1 For the sake of simplicity, these equations are given here for diffusion in one dimension (x) only. For a constant diffusion coefficient, F = - D0(∂c/∂x)

(16.1)

∂c/∂t = ∂/∂x (D0∂c/∂x)

(16.2)

The diffusion equation is derived in a very general way and also accounts for concentration dependent diffusion coefficients whenever this is encountered. F is the mass flux, D0 is the (constant) diffusion coefficient or the diffusion coefficient at the lowest concentration if there is concentration dependence (see below), c is the local concentration, x is the distance in the x dimension, and t is the time. The solutions to the diffusion equation given in this chapter use the dry film thickness as reference. This is because it is far simpler to keep track of what is going on instead of continually adjusting local film thickness as a function of the amount of solvent present. When a diffusing solvent is present, for example, then the actual local thickness should be increased proportionately according to its local volume fraction. The use of dimensionless variables to solve these equations makes the solutions more useful by making them applicable to all values of the combined variables. For this purpose the following are defined: Dimensionless time: T = D0t/L2

(16.3)

X = x/L

(16.4)

C = (c – c0)/(c∞ – c0)

(16.5)

Dimensionless distance:

Dimensionless concentration:

L is the thickness of the plane sheet being considered. c0 is the initial uniform concentration in the film. The (local) dimensionless concentrations rise from 0 to 1.0 in an absorption experiment where equilibrium absorption, c∞, is finally obtained. For the sake of completeness the predicted and experimentally confirmed exponential dependence of the diffusion coefficient on concentration, D(c), will be introduced at this early stage. A more detailed discussion of its significance is given in a special section below. D(c) = D0ekc = D0Dv

(16.6)

“k” is a constant that is valid up to a given concentration as discussed below. Dv is the increase in the diffusion coefficient over that at zero conditions for a given concentration. At the maximum concentration this becomes Dmax, and for a constant diffusion coefficient it is 1.0. Using these variables, Equation 16.2 can be rewritten in dimensionless form as Equation 16.7. The dimensionless diffusion equation for an exponential diffusion coefficient is: ∂C/∂T = ∂/∂X (Dv∂C/∂X)

(16.7)

7248_C016.fm Page 296 Wednesday, May 23, 2007 11:36 AM

296

Hansen Solubility Parameters: A User’s Handbook

The earliest edition of Crank’s monumental work1 has the advantage of many graphical solutions to the dimensionless diffusion equation being presented with plots that can be used with a high degree of accuracy. These plots include numerous reference lines not found in later editions.

STEADY STATE PERMEATION In the absence of significant surface resistances, solving Equation 16.1 (or Equation 16.7) for steady state permeation in one direction only for a constant diffusion coefficient gives Equation 16.8. F = –D0(c1 – c2)L

(16.8)

The surface concentration on the exposed side, c1, is usually assumed to be c∞, and c2 is usually assumed to be zero. Initially, c1 will be less than c∞ when there is a significant surface resistance on the exposed side, as discussed in the following. Likewise, c2 will be greater than c0 if there is a significant surface resistance on the low concentration side of the film. The preferred units for these quantities are F in g/(cm2×s), L in cm, c in g/cm3, and D0 in cm2/s. It can be seen from Equation 16.8 that the surface concentration on the exposed side determines the concentration gradient over the film, assuming c2 is zero. This is a situation that seems to prevail in general, but exceptions are discussed below, and there will be significant surface resistance in presumably all cases as film thickness approaches zero. The surface concentrations, c1 (or cs), will be higher for closer matches in HSP between challenge chemicals and polymers. Therefore, it is not surprising that HSP correlations can be made when permeation rates for a large number of chemicals have been measured for a given polymer, as reported in Chapter 13. This is particularly true when the molecules involved are all relatively small. When the molecular sizes of the challenge chemicals become too large and/or their shape becomes sufficiently complicated with side groups and cyclic structures, simple HSP correlations are no longer possible. The diffusion coefficient is affected by these size and shape factors, and the HSP can no longer be used as a single correlating parameter. In addition, surface resistances can also become very significant, as discussed below and in Chapter 14. The molecular volume, V, of the challenge chemical has been used with some success to account for size effects, but this does not directly account for differences in shape. Examples of correlations for diffusion through chemical protective clothing, for example, demonstrated that molecular size had to be taken into account2 (see also Figure 13.2 and the discussion in Chapter 13). The most reliable HSP correlations in these cases do not immediately consider the solvents with smaller molecules that may permeate faster than expected by comparison with all the others. Likewise, improved understanding and correlations are obtained by initially neglecting the solvents with larger molecules that do not permeate as fast as expected, in spite of close matches in HSP with those solvents that do permeate rapidly. Once a reliable HSP correlation is established without these obvious outliers, their behavior is better understood. Predictions then often become possible for other very small and/or very large molecular species as well.

THE DIFFUSION EQUATION CONSTANT DIFFUSION COEFFICIENTS Relevant solutions to Equation 16.2 or Equation 16.7 are used to measure diffusion coefficients. The diffusion equation must be solved with two boundary conditions and an initial condition. This discussion will consider a plane film exposed to absorbing chemical on two sides. The initial condition is chosen as a uniform concentration within a film, so that C is 0 for all X, regardless of whether the initial concentration is zero or not. Diffusion is also usually assumed to take place in one direction only, but side effects can become important in thicker samples as discussed below. The chemical concentration at the exposed surface(s) is assumed to immediately rise to the

7248_C016.fm Page 297 Wednesday, May 23, 2007 11:36 AM

Absorption and Diffusion in Polymers

297

equilibrium value. The second boundary condition is found at the middle of the free film exposed on two sides where there is no transfer in either direction. In other words Equation 16.1 is set equal to zero at this point. For a constant diffusion coefficient, solving the diffusion equation with these conditions gives an initial straight-line absorption curve as a function of the square root of time. This is generally called Fickian diffusion. The straight line passes through the origin, and the time required to absorb one-half of the equilibrium amount, t1/2, can be used in Equation 16.9 to find D0. D0 = 0.049 L2/t1/2

(16.9)

This equation is based on T having a value of 0.049 when half of the equilibrium amount has been absorbed. There is an identical result for desorption experiments where the time required for half of a uniformly absorbed material to leave the film also requires the same T value.

CONCENTRATION DEPENDENT DIFFUSION COEFFICIENTS Solutions for the diffusion equation have also been generated for concentration dependent diffusion coefficients for the same boundary and initial conditions as described above. In this case it can be shown that Equation 16.7 reduces to Equation 16.10.3–5 ∂Dv/∂T = Dv(∂2Dv/∂X2)

(16.10)

This equation has been solved numerically many times for different values of Dmax for both absorption and desorption with a uniform initial concentration.3–5 Dmax is the ratio of the maximum diffusion coefficient found at the maximum concentration encountered in an experiment to D0. The half-times calculated for both absorption and desorption were converted to Fa or Fd, respectively, their ratio to the value 0.049. These values are given in Table 16.1. Fa and Fd are the FM for use in Equation 16.11. The correction factors for desorption experiments were also found for the time required for only one fourth of the material to leave the film in desorption experiments. This was necessary because of the extremely long experimental times (months) required for even this amount to leave, and also holds true for very thin films. These quarter-time correction factors for desorption were used to generate the results reported in Figure 16.1 and are discussed in the following.6 The results reported in this figure confirm that Fa and Fd are correct and useful. The same diffusion coefficients were found by both absorption and quarter-time desorption measurements. The correction factors for the absorption measurements, Fa, were close to 2.0, whereas those for the quartertime desorption measurements, Fd, were in the range of 40 to 144.

D (c ) = FM × FB

0.049 × L2 tE

(16.11)

The factor FB in Equation 16.11 is a related correction accounting for any surface resistance as discussed below. FB will always be greater than 1 as a surface resistance slows the transport process and leads to an apparent diffusion coefficient that is too low. The exponential increase in diffusion coefficient with concentration is expected based on free volume theory.7 It is beyond the scope of this chapter to include this theory in detail. The main feature of diffusion in polymers is that the macromolecular chains are barriers to transport. Factors that either promote mobility of the chain segments or increase the distance between them will enhance the movement of smaller molecules. The transport occurs as very small movements rather than larger jumps. An increase in concentration of plasticizing smaller molecules leads to an increase in the free volume of the system, as the smaller molecules have more free volume associated with

7248_C016.fm Page 298 Wednesday, May 23, 2007 11:36 AM

298

Hansen Solubility Parameters: A User’s Handbook

TABLE 16.1 Correction Factors for the Measurement of Concentration Dependent Diffusion Coefficients for Use in Equation 16.11 Desorption Dmax

(Fd)1/2

(Fd)1/4

100 2 5 101 102 103 104 105 106 107 108

1.00 1.56 2.70 4.00 13.40 43.30 138.7 443 1,370 4,300 13,670

1.00 1.55 2.61 3.84 10.20 23.10 47.40 89.0 160.5 290 506

Absorption (Fa)1/2 1.00 1.30 1.70 2.01 3.30 4.85 6.14 7.63 8.97 10.60 12.10

Note: (Fd)1/2 is for desorption half-times, (Fd)1/4 is for desorption quarter-times, and (Fa)1/2 is for absorption half-times.

them than do polymers. Thus diffusion coefficients increase as solvent concentration increases. As can be seen in Figure 16.1 and Figure 13.1, the diffusion coefficient for common solvents increases by a factor of about 10 for an increase of solvent concentration 0.03 volume fraction at lower solvent concentrations. This corresponds to slightly more than doubling the diffusion coefficient for each added 0.01 volume fraction of solvent. Concentration dependence in measuring diffusion coefficients by absorption can also be accounted for by the method of integrals given by Crank.1 Treatment of the experimental data given by Crank with the correction factors given in Table 16.1 leads to exactly the same result for the exponential diffusion coefficients of chloroform in polystyrene as was found by the method of integrals. Both procedures require iterations as the true value of Dmax is not known initially and must be estimated to find Fa (and a new Dmax) until convergence is obtained.

SURFACE RESISTANCE MATHEMATICAL BACKGROUND The diffusion equation must be solved with the appropriate boundary conditions at the surfaces when significant surface resistances are encountered. For a film exposed on two sides, the absorption process then can be modeled with the following boundary condition at the surfaces: F = h (c∞ − cs )

(16.12)

The surface mass transfer coefficient, h, has preferred units of cm/s. The surface concentration at any given time is cs. One estimate of h can be obtained by plotting the weight gain against time. The limiting slope at time approaching zero is used to find F. As cs is zero at time equal to zero, h can be estimated from this initial flux divided by c∞.

7248_C016.fm Page 299 Wednesday, May 23, 2007 11:36 AM

Absorption and Diffusion in Polymers

299

- LOG diffusion coefficient at 20 °C, cm²/sec

6 Isotope

8 Absorption Fa = 1.8

F = Fa x FB = 1.2 x 250 = 300

F = Fa x FB = 1.3 x 1.25 = 1.63

Fd = 144 10

Desorption (to vacuum) 12 Fd = 40

14 0.1

0.2

0.3

0.4

0.5

0.6

Vf FIGURE 16.1 Diffusion coefficients for chlorobenzene in polyvinylacetate measured by absorption halftimes, desorption quarter-times, and isotope experiments as a function of the volume fraction of chlorobenzene, Vf.6 The lower curve is based on dry polymer content. The upper curve is based on total thickness. Corrections for surface resistance, FB, are also required for Vf above about 0.2 volume fraction. In an extreme case for absorption with Vf near 0.5, the correction for surface resistance was a factor of 250.21 (Reprinted from Hansen, C.M., Prog. Org. Coat., 51(1), 55–66, 2004. With permission from Elsevier.)

It is useful to rewrite this boundary condition in dimensionless terms using the quantity B. This is the ratio of diffusion resistance, Rd, to that of surface resistance Rs. Thus, B = Rd/Rs = (L/D0)/(1/h) = hL/D0

(16.13)

Large B is indicative of a diffusion-controlled process. For B = 1, Rd = Rs, and for lower B surface resistance dominates. Surface resistance becomes increasingly important as the film thickness decreases. The dimensionless boundary condition corresponding to Equation 16.12 for use with Equation 16.7, considering the x direction only is ∂C/∂X = B(1 – Cs)

(16.14)

For an exponential concentration dependence of the diffusion coefficient, this boundary condition can be used with Equation 16.10 as3,5 ∂Dv/∂X = BlnDv

(16.15)

7248_C016.fm Page 300 Wednesday, May 23, 2007 11:36 AM

300

Hansen Solubility Parameters: A User’s Handbook

TABLE 16.2 Correction Factors, FB, for Use with a Constant Diffusion Coefficient in Equation 16.11 B

1/B

FB

∞ 10 2 1 0.5 0.1

0 0.1 0.5 1 2 10

1.0 1.45 3.14 4.95 6.8 37.5

Note: This means that FM is equal to 1.0 in this case.

SURFACE RESISTANCE

IN

ABSORPTION EXPERIMENTS

Solutions to the diffusion equation have been presented for various surface resistances (surface conditions) by Crank for a constant diffusion coefficient.1 FB has been evaluated from these graphical results as the ratio of the half-time for a given B value to 0.049. These results are reported in Table 16.2. Equation 16.16 can be used for B less than about 0.5. At higher B values it is not exact. FB = (3.75/B) +1

(16.16)

As stated above, when surface resistance can be neglected, a plot of the (relative) uptake vs. the square root of time initially is a straight line that passes through the origin. When surface resistance becomes important, the delayed uptake can be seen as a form of time-lag phenomena, with a clear “S” shape. It is also interesting to note at this point that solutions to the diffusion equation with the boundary condition of an exponential increase of the surface concentration with time give absorption curves with exactly the same “S” shapes. Numerical solutions to the diffusion equation of the type described in Reference 3 and Reference 5 confirm that significant surface resistance leads to an exponential increase in the surface concentration. The factors leading to and controlling the prevailing surface concentrations are of major interest and not necessarily the fact that these increase in an exponential manner with time. Figure 16.2 shows S-shaped curves for absorption in the COC polymer, Topas® 6013 from Ticona. Surface resistance is significant in all three cases shown, as can be seen by the S-shaped curves that do not pass through the origin.8 The apparent B values for these cases are 10 for ethylene dichloride, 0.5 for diethyl ether, and 20 for propyl amine. Solvent absorption was followed in injection-molded samples for 13 solvents in cyclic olefinic copolymer (COC), 4 solvents in two different grades of polycarbonate (PC), and 2 solvents in the terpolymer acrylonitrile/butadiene/styrene (ABS). It was discovered that a surface resistance to absorption was significant in 19 of these 23 cases. Approximate surface mass transfer coefficients and approximate diffusion coefficients were determined where possible. There is no significant surface resistance to absorption in those cases where the absorbing molecules are smaller and linear such as for tetrahydrofurane, n-hexane, and 1,3-dioxolane in the COC polymer, and butyric acid in PC (Lexan® 104R, General Electric). When the challenge molecules are too large or bulky, no absorption occurs. Such systems included 1,4-dioxane, methyl isobutyl ketone, acetophenone, and phenyl acetate in the COC polymer. There

7248_C016.fm Page 301 Wednesday, May 23, 2007 11:36 AM

Absorption and Diffusion in Polymers

301

Ethylenedichloride in COC

Diethylether in COC

Propylamine in COC

Weight change [mg/g]

300 250 200 150 100 50 0 0

25

50

75

100 125 Time [sqrt(min)]

150

200

200

FIGURE 16.2 Absorption of ethylene dichloride, diethyl ether, and n-propylamine in a COC polymer, Topas® 6013, Ticona, at 23°C. (Reprinted with permission from Nielsen, T.B. and Hansen, C.M., Ind. Eng. Chem. Research, 44(11), 3959–3965, 2005. Copyright 2005 American Chemical Society.)

was no weighable absorption, even though HSP would predict this. The molecules can simply not get through the surface layer, and its resistance is therefore effectively infinitely large. Other examples of lack of absorption are oleic acid that simply does not absorb in the PC polymers or ABS. Between these extremes are situations where surface resistance clearly affects the absorption process. Surface resistance becomes significant when molecules can be transported away from the surface into the bulk of the polymer faster than they can be adsorbed/absorbed just at/in the surface.

SURFACE RESISTANCE

IN

PERMEATION EXPERIMENTS

As stated above, Equation 16.13 clearly shows that the importance of surface resistance will increase as film thickness decreases, and vice versa. This can be used in permeation measurements to find the true permeation coefficient as well as the sum of all other resistances to the transport process. One measures apparent permeation coefficients Papp, for different film thicknesses and extrapolates the inverse of the apparent transport coefficient versus the inverse of the film thickness to zero, corresponding to infinite film thickness. This is portrayed in Figure 16.3 for the permeation of water through an acrylic coating.9 If a single measurement had been made at a film thickness of 40 microns, which is a normal film thickness, the apparent permeation coefficient would have been one-half that of the true permeation coefficient. The data in this type of figure can be interpreted using the following set of equations: F = Δp/(L/Papp) = Δp/(L/P∞ + R1 + R2 + R3 ….)

(16.17)

L/Papp = L/P∞ + R1 + R2 + R3 ….

(16.18)

1/Papp = 1/P∞ + (R1 + R2 + R3 ….)/L

(16.19)

The extrapolation to 1/L equal to zero gives the inverse of the true permeation coefficient, 1/P∞, and the slope gives the sum of the resistances. Δp is the overall vapor pressure difference in the system, and the given R represent different sources of resistance. Other extrapolations are possible to gain more information, depending on the situation.10 One can find the vapor diffusion coefficient (resistance), for example, by varying the amount of liquid

7248_C016.fm Page 302 Wednesday, May 23, 2007 11:36 AM

302

Hansen Solubility Parameters: A User’s Handbook

1 x 10-12 Papp 20

15

10

P∝

5

05

10

15

20

25

1 x 10-3 L

FIGURE 16.3 A plot of the inverse of the apparent permeability coefficient (Papp in kg Pa1 s1 m1) vs. the inverse of the film thickness (L in m) for an acrylic coating. Extrapolation to 1/L to zero gives the inverse of the true permeability coefficient. (Reprinted from Huldén, M. and Hansen, C.M., Prog. Org. Coat., 13(3/4), 171–194, 1985. With permission from Elsevier.)

in a cup-type experiment. It has been found that these surface and vapor diffusion effects must be accounted for in cup-type and related experiments with thinner polymer films and porous types of materials. These include thinner paint films, wood in the fiber direction, and paper, for example.9–11 This type of experiment using paper as a film separated resistances for the permeation of the paper by water as well as for diffusion of water in air, (heat transfer) evaporation of the water, and an estimate of the surface resistances on the two sides of the paper.10

SURFACE RESISTANCE — A DISCUSSION In the literature surface resistance has alternatively been called skin layer effect, surface condition, interfacial resistance, or boundary layer resistance, and leads to what is often called an induction time and “time-dependent” diffusion. In addition to the discussion and literature cited above, surface effects have been noted in many studies of diffusion in polymers. The following also deal with surface effects in the absorption of solvents into polymers. A skin layer was found on the surface of injection molded polypropylene.12 NMR microimaging was used to confirm that carbon tetrachloride absorption was retarded by this layer. The rapid surface cooling in the injection molding process gives a surface different from the interior, where the cooling rate (and orientation) is different. This effect is presumably found with many injection molded polymers. McDonald et al. also arrived at the conclusion that surface flux limited diffusion of solvent into polymer could explain observed behavior such as Case II and transitions between Fickian

7248_C016.fm Page 303 Wednesday, May 23, 2007 11:36 AM

Absorption and Diffusion in Polymers

303

diffusion and Case II diffusion.13 Toluene diffusion in polystyrene is discussed in detail. Case II diffusion involves a linear absorption curve when time is used rather than the customary square root of time. This is discussed in more detail below. Shankar studied interfacial resistance in absorption experiments.14 The conclusions were that non-Fickian characteristics can result from slow transfer to the surface layer, and that some of the observed features of anomalous sorption can be explained with the help of this model. Surface resistance was found to be particularly important in thinner sheets. Systematic variations in absorption phenomena with film thickness are a clear indication of a significant surface resistance. When there is a surface resistance, the surface concentration only slowly rises to the equilibrium value. Surface resistance in the methyl iodide-cellulose acetate system produced the characteristic Sshaped absorption curve. Characteristic S-shaped absorption curves were found in the methylene chloride-PEEK system by Gryson et al.15 The initial rate of absorption was found to be strongly dependent on the surface condition but the equilibrium values were not. Surface resistance was shown to affect permeation through polymer films by Kim and Kammermeyer who measured actual concentration profiles in Nylon-6, cellulose acetate, and polyethylene membranes.16 Water permeation through Nylon-6 films was studied for different film thickness. It was clearly shown that the surface concentration of water did not reach the equilibrium value at the equilibrium permeation rate unless the film thickness was greater than 0.05 cm at 35°C. Similar studies showed that there were also significant surface resistances for dioxane, benzene, and n-hexane in polyethylene, and for water in cellulose acetate. Hwang and Kammermeyer showed significant surface resistance by studying permeation as a function of film thickness for water in acetyl cellulose acetate and Nylon-6, hydrogen through stainless steel, and p-dioxane through Nylon-6 and polyethylene.17 Skaarup18 performed many permeation experiments on Polyamide 6 (PA 6) (BASF Ultramid B4) and polyvinylacetate (Hoechst, Mowilith 50) where surface resistance was significant. Of particular interest are the studies on PA 6 where the permeation of ethyl laurate, 2,4-dimethyl-2pentanol, benzyl alcohol, n-butanol, ethyl acetate, and n-pentane was studied. Skaarup arrived at the following general relation for the permeation of these liquids through PA hav = k2(Dav/A1/2)

(16.20)

hav (in cm/s) is the average surface mass transfer coefficient, Dav (in cm2/s) is the average diffusion coefficient in the film (see Equation 16.21), and A (in cm2) is the (minimum) cross-sectional area of the molecule in question. k2 is a constant with the value approximately 4.5(10)-6. This equation indicates that a molecule in the surface of a polymer will proceed inward in direct proportion to the diffusion coefficient. It likewise indicates that the greater the cross-sectional area of the molecule, the more difficulty it will have to reside at the surface in a condition where it can take a small jump into the bulk. h on the exposed side of the films was approximately 1/3 of that on the exit side. The reason for this is thought to be that the orientation of molecules leaving the film is directed more toward the exit surface, whereas the orientation of molecules approaching the entry side is more random. A molecule landing “sideways” and hitting a polymer segment will be rejected. The right orientation at the right place allows adsorption/absorption. The significance of the surface resistances can be demonstrated by the film thickness at which the resistance from diffusion within the film is equal to the sum of the two surface resistances. This was about 180 microns or higher for the linear aliphatic molecules, and increased for more complicated structures. Dav can be found from Equation 16.21 as the logarithmic mean where D1 is the diffusion coefficient on the exposed side and D2 is the diffusion coefficient on the exit side. Dav = (D1/D2 – 1)/ln(D1/D2)

(16.21)

7248_C016.fm Page 304 Wednesday, May 23, 2007 11:36 AM

304

Hansen Solubility Parameters: A User’s Handbook

This equation shows how much the average diffusion coefficient differs from that at essentially zero concentration corresponding to D2. If the ratio D1/D2 is 10 then Dav is 3.9 times larger than D1. This implies that concentration dependent diffusion can lead to significant errors when diffusion and permeation coefficients at very low concentrations are needed. D1/D2 equal to 10 implies that the concentration difference is only about 0.03 volume fraction across the membrane when rigid polymers are involved. For elastomers this same ratio implies a concentration difference near 0.15 volume fraction, as judged from the diffusion coefficient data reported in Figure 16.1 at concentrations above about 0.20 volume fraction where the system is above its glass transition temperature. These same considerations are valid for permeation coefficients since: P = DS

(16.22)

P is the permeation coefficient measured at the given steady state concentration difference, (c1 – c2), and S is the solubility coefficient. If D1 is 10 times greater than D2 then P is approximately 3.9 times larger than would have been measured for an extremely low concentration difference across the membrane.

SIDE EFFECTS Corrections for absorption or desorption from the sides of thicker samples should be applied to measured diffusion coefficients if there is significant diffusion through the sides of thicker samples. Equation 16.23 for diffusion in plane samples of isotropic media can be used for this purpose.19 D0 = Dapp/(1 + L/l + L/w)2

(16.23)

Dapp is the apparent diffusion coefficient found from initial slope measurements1 or Equation 16.9 if there is a linear absorption curve as far as t1/2 when the square root of time is used. The length and width of the film are l and w, respectively, with L being retained as the thickness. It can be seen that D0 and Dapp are equal to the extent that the second and third terms in Equation 16.18 are not significant. Conventional tensile bars, for example, have significant side effects when they are used to measure diffusion coefficients. The uptake is more rapid than otherwise anticipated, so the apparent diffusion coefficients must be reduced accordingly. Tensile bars that are 4 mm thick and 10 mm wide will require corrections that are at least as great as a factor of 1.96. A square sample that is 10 mm on each side and 1 mm thick requires a correction for end effects equal to 1.44. For cylindrical geometry a derivation similar to that used to find Equation 16.23 results in Equation 16.24. D0 = Dapp/(1 + b/R)2

(16.24)

R is the radius of the cylinder and b is its thickness. These equations are based on the initial uptake (or loss) to evaluate Dapp. Here again, an easy procedure is to extrapolate the initial slope on a plot of uptake versus the square root of time to a fictive t1/2 found when the relative uptake is equal to 0.5. This value for t1/2 can be used in Equation 16.9 to find Dapp.

MEASURING DIFFUSION COEFFICIENTS CONCENTRATION DEPENDENCE

WITH

SURFACE RESISTANCE

AND

It was necessary to measure diffusion coefficients over as wide a range as possible in order to solve the diffusion equation to simulate film formation by solvent evaporation.3,20 This was done by comparison of experimental results with solutions to the diffusion equation accounting for both

7248_C016.fm Page 305 Wednesday, May 23, 2007 11:36 AM

Absorption and Diffusion in Polymers

305

concentration dependence and surface resistance at the same time. These techniques are reported elsewhere.3–5,21 The technique used was to interpret the experiments initially assuming that the diffusion coefficient was a constant at the concentration assigned to the experiment. Corrections to this estimate (multiplying factors) were then made by comparison with suitable solutions to the diffusion equation using the data in Table 16.1 in Equation 16.11. The correction factors, Fa, at low concentrations were close to 1.5–2.0 (corresponding to increases of from 50 to 100%) for the usual step-wise absorption experiments from one concentration to a slightly higher one. The correction factors, Fd, were well above 100 for desorption measurements from an equilibrium state near 15–20%vol solvent to vacuum. The results in Figure 16.1 show that the same diffusion coefficients were found in both cases. Absorption experiments only were relevant above about 20%vol because there is a marked change in diffusion behavior at concentrations above and below this value (for the system chlorobenzene and polyvinylacetate). When concentrations reach higher than about 20%vol, the correction factors for surface resistance, FB, increase rapidly. They can become as high as 100 or more.21 The total diffusion coefficient curve from 0 to 100% solvent (chlorobenzene) was completed with a self-diffusion coefficient and several isotope experiments.3 The isotope experiments are for very high solvent concentrations that correspond to liquid lacquer formulations. The solvent-in-polymer diffusion coefficient for chlorobenzene in polyvinylacetate increases more than 9 decades as the solvent concentration increases from essentially zero to 100%, where the self-diffusion coefficient is 1.65(10)5 cm2/s. These coordinated absorption and desorption experiments were necessary to determine the true diffusion coefficients at all concentrations so that the solutions to the diffusion equation for film drying by solvent evaporation were correct.3,20 It is also necessary to consider concentration dependence simultaneously with surface resistance in many cases of practical importance. Surface resistance has been present in many studies reported in the literature and clearly affects many results involving diffusion in polymers, but only rarely has there been mention of this fact. This situation naturally led to solving the diffusion equation with relevant values for diffusion coefficients and surface resistances for additional situations of interest. The overall result was a simple explanation for the various types of so-called anomalous diffusion. It is the balance between the (concentrationdependent) diffusion resistance and the surface resistance that determines whether the diffusion is "Fickian" or whether it is presumed to be anomalous. This is discussed in the next sections.

FILM FORMATION BY SOLVENT EVAPORATION The process of film formation by solvent evaporation has been fully described using the diffusion equation with a significant surface resistance and local diffusion coefficients found from the data in Figure 16.1. Loss of solvent was followed experimentally for about 2 years from an initial Vf of 0.75 until it was totally lost.3,20 Surface resistance was significant at concentrations above about 0.2 volume fraction solvent in the system that was studied most extensively (chlorobenzene in polyvinylacetate). This was also confirmed by the calculations since the surface concentration fell to essentially zero when this amount was reached, and the continued loss of solvent was controlled by internal diffusion to the surface. In this situation h is affected by factors such as vapor pressure and latent heat of the solvent, heat transfer to the surface, and air velocity past the surface. The local diffusion coefficient changed from 5(10)–7 cm2/s initially to 1(10)–14 cm2/s at zero concentration. The experimental and calculated curves are reported in Figure 16.4. The process of solvent loss takes place in two distinct stages. Surface resistance controls the first stage, whereas the second stage is controlled by the rate at which solvent molecules can diffuse to the film/air surface in order to evaporate. The quantity V2 in Figure 16.4 is given as 106. As can be seen in Figure 16.1, where diffusion coefficients are reported for the same system, this is the extent of the variation of the diffusion coefficient from 0 concentration up to CA, the concentration at the break in the curve. Vt is a fictitious diffusion coefficient to calculate the diffusion coefficients at concentrations above CA.

7248_C016.fm Page 306 Wednesday, May 23, 2007 11:36 AM

306

Hansen Solubility Parameters: A User’s Handbook

Volume Solvent / Volume Polymer

101

V2= 106 Vt= 1010 CA= 0.2 B as indicated Exptl. 165 microns

10

5

10-1

B=10 Exptl. B=106 7 22 microns B=10 ~MO –CA –CA CS= O CS= O For B=107 For B=106

CS= O For B=105 Calculated Experimental Effect of water - a steeper slope One day L=30 microns

-2

10

10-7

10-6

10-5

10-4 Dot Dimensionsless T, (L)2

10-3

10-2

FIGURE 16.4 Calculated and experimental drying curves for the evaporation of chlorobenzene from polyvinyl acetate.3,20 See the discussion in the accompanying text.

Diffusion coefficients were calculated for each local site in the films. The iterative procedure at each successive time interval was that described in Crank1 as the Crank–Nicolson method. The films were divided into a sufficient number of finite difference elements to assure correct results. Film drying in a climatized room was faster than film drying in a vacuum apparatus, where diffusion coefficients were measured. Absorbed water plasticizes the film. The calculated and measured desorption curves under vacuum coincided at long times (several months). Vacuum does not hasten release of solvent by diffusion at longer times. It only makes certain that the surface concentration is zero, which it will be in almost all cases anyway.

ANOMALOUS DIFFUSION (CASE II, SUPER CASE II) For absorption with a constant diffusion coefficient one normally finds that the initial weight gain is linear with the square root of time as stated above. This is called Fickian or normal diffusion. In Case II diffusion, the weight increases linearly with time. This is largely a result of concentration dependent diffusion coefficients with a large jump in concentration. This is true even when there is very little surface resistance of significance.5 Super Case II emerges as the surface resistance becomes more significant relative to the diffusion resistance.5 Uptake is linear with time early in the process, but at some longer time the rate of gain increases markedly. See Figure 16.5. This change occurs when the diffusing material reaches the middle of a sheet exposed on both sides, for example. This can be seen in Figure 16.6 where concentration gradients have been calculated for Super Case II behavior. Simultaneous diffusion and surface resistance combine in a special way to produce this behavior. The chronology of the events is that surface concentration increases as time goes on. There is an advancing front into the film that is somewhat more pronounced than that shown in Figure 16.6, because the distances in this figure are based on dry film thickness. When the concentration at the center of the film starts to increase above zero, the rate of absorption also begins to increase. The concentration profiles ultimately become flat at moderate elapsed time, as diffusion within the film is now rapid compared to the surface resistance. The diffusion resistance

7248_C016.fm Page 307 Wednesday, May 23, 2007 11:36 AM

Absorption and Diffusion in Polymers

307

1.0 B:

0.8 Mt / M∝

109 108

0.6

107

0.4 0.2 106 0.0 0.0

0.2

0.4

0.6 0.8 6 T x 10

1.0

1.2

1.4

FIGURE 16.5 Solvent uptake curves for various values of B and concentration dependent diffusion coefficients equal to those reported for chlorobenzene in polyvinylacetate (See Figure 16.1). B is the ratio of diffusion resistance to surface resistance, including concentration dependence. (Reproduced from Hansen, C.M., Diffusion in Polymers, Polym. Eng. Sci., 20(4), 252–258, 1980. With permission from the American Chemical Society.)

1.0 0.869 0.8

ØR

0.6 0.4 0.2 0.0 0.0

0.037

0.125

0.098

0.250

0.146

0.375

0.216

X

0.500

0.319

0.625

0.386

0.750

0.467

0.875

0.562

1.0

FIGURE 16.6 Concentration gradients for conditions corresponding to Super Case II type (anomalous) diffusion. Values on the curves are relative weight gains. ∅R is the local concentration. (Reproduced from Hansen, C.M., Polym. Eng. Sci., 20(4), 252–258, 1980. With permission from the American Chemical Society.)

becomes less and less as the concentration in the middle of the film increases, and the rate of uptake increases. There is still another effect at very long times where the surface resistance begins to be more important again relative to the diffusion resistance. The rate of weight gain decreases just before the equilibrium value is attained. The concentration gradients in Figure 16.6 are for the curve in Figure 16.5 for B = 107. This is a typical Super Case II type curve with initial absorption linear with time followed by a sudden increase in the absorption rate as the solvent reaches the center of a film exposed on two sides. The final leveling off at very long times is also characteristic of Super Case II. The absolute value of B depends on the degree of concentration dependence over the selected concentration interval

7248_C016.fm Page 308 Wednesday, May 23, 2007 11:36 AM

308

Hansen Solubility Parameters: A User’s Handbook

in the given case. See Reference 5 for greater detail. The parameters chosen to calculate this curve are realistic.

GENERAL COMMENTS Discussion of the various other explanations for anomalous diffusion is beyond the scope of this chapter. These discussions are focused on time-dependent replication of the observed diffusion phenomena, such as an advancing front,22 and on explanations based on stress relaxation phenomena at the head of this advancing front.23,24 Whereas these may have aspects of validity, they can only be convincing when the verifiable surface resistances and the verifiable exponential concentration dependence of the diffusion coefficient are also taken into account. It would appear that enhanced stress relaxation, like the increase of diffusion coefficient with solvent concentration, is dependent on the increase in free volume brought locally by the solvent molecules themselves. The measurement of surface resistances and concentration-dependent diffusion coefficients did not require use of any mathematical tools or explanations other than the diffusion equation solved with appropriate initial and boundary conditions. There were advancing fronts involved both in the mathematics and in the samples in the experiments. The simple approach of solving the diffusion equation with a verifiable (exponential) concentration-dependent diffusion coefficient, along with appropriate and verifiable parameters for the boundary conditions, explains and replicates both the absorption and the desorption of solvents in polymers over an extremely large concentration range. This is true with and without significant surface resistance. The methodology and concepts presented in this chapter on diffusion in polymers combined with the methodology and concepts in the rest of this handbook should provide insight into many situations of industrial and theoretical interest. These include controlled release of drugs, absorption and transport through polymeric packaging of various kinds, improved prediction of the behavior of chemical protective clothing, absorption into coatings, release of absorbed chemicals from plastics, release of sterilization gas, etc.

CONCLUSION Many organic liquids are aggressive with respect to many types of polymers. Predicting aggressive behavior is important, and the absorption of the organic liquids into the polymers is a key factor in this respect. This absorption can be strongly affected by both surface phenomena and the rate of diffusion within the bulk of the polymer. This chapter has emphasized the importance of surface resistance in this process. Additional material on diffusion in polymers including simple mathematical descriptions of film drying by solvent evaporation and so-called anomalous diffusion have been included to present several aspects of the importance of surface resistance to contribute to its better understanding. There is no surface resistance to absorption where the absorbing molecules are small enough and/or linear. When the challenge molecules are too large or bulky, no absorption occurs, even though Hansen solubility parameter considerations lead one to predict that this is clearly expected. The molecules can simply not get through the surface layer, and its resistance is therefore effectively infinitely large. Between these extremes are situations where surface resistance necessarily affects the absorption process. Surface resistance becomes significant when molecules can be transported away from the surface into the bulk of the polymer faster than they can be adsorbed/absorbed just at/in the surface. The simple approach of solving the diffusion equation with appropriate and verifiable parameters for the boundary conditions explains and replicates both the absorption and desorption of solvents in polymers over an extremely large concentration range. It is not yet possible to predict

7248_C016.fm Page 309 Wednesday, May 23, 2007 11:36 AM

Absorption and Diffusion in Polymers

309

which systems will give significant surface resistance to absorption. Experimental studies are still required, but some similarity in HSP must be present for absorption to occur in any event.

REFERENCES 1. Crank, J., The Mathematics of Diffusion, Oxford University Press, Oxford, 1956. 2. Hansen, C.M. and Hansen, K.M., Solubility parameter prediction of the barrier properties of chemical protective clothing, Performance of Protective Clothing: Second Symposium, ASTM STP 989, Mansdorf, S.Z., Sager, R., and Nielsen, A.P., Eds., American Society for Testing and Materials, Philadelphia, PA, 1988, pp. 197–208. 3. Hansen, C.M., The Three Dimensional Solubility Parameter and Solvent Diffusion Coefficient, Their Importance in Surface Coating Formulation, Doctoral dissertation, Danish Technical Press, Copenhagen, 1967. 4. Hansen, C.M., The measurement of concentration-dependent diffusion coefficients — the exponential case, Ind. Eng. Chem. Fundam., 6(4), 609–614, 1967. 5. Hansen, C.M., Diffusion in polymers, Polym. Eng. Sci., 20(4), 252–258, 1980. 6. Hansen, C.M., Aspects of solubility, surfaces, and diffusion in polymers, Prog. Org. Coat., 51(1), 55–66, 2004. 7. Korsmeyer, R.W., Von Meerwall, E., and Peppas, N.A., Solute and penetrant diffusion in swellable polymers. II. Verification of theoretical models, J. Polym. Sci.: Polym. Phys. Ed., 24, 409–434, 1986. 8. Nielsen, T.B. and Hansen, C.M., Significance of surface resistance in absorption by polymers, Ind. Eng. Chem. Res., 44(11), 3959–3965, 2005. 9. Huldén, M. and Hansen, C.M., Water permeation in coatings, Prog. Org. Coat., 13(3/4), 171–194, 1985. 10. Nilsson, E. and Hansen, C.M., Evaporation and vapor diffusion resistance in permeation measurements by the cup method, J. Coat. Technol., 53(680), 61–64, 1981. 11. Hansen, C.M., Potential errors in water/water vapor permeation measurements using the cup method, Färg och Lack, 39(3), 57–60, 1993. 12. Abbott, R.J., Chudek, J.A., Hunter, G., and Squires, L., Skin layer effects on the diffusion of carbon tetrachloride into injection moulded polypropylene studied by 1H NMR microimaging, J. Mater. Sci. Lett., 15, 1108–1110, 1996. 13. McDonald, P.J., Godward, J., Sackin, R., and Sear, R.P., Surface flux limited diffusion of solvent into polymer, Macromolecules, 34, 1048–1057, 2001. 14. Shankar, V., Influence of interfacial resistance on kinetics of sorption, Polymer, 22, 748–752, 1981. 15. Grayson, M.A., Pao, P.S., and Wolf, C.J., Transport of methylene chloride in poly(aryl-ether-etherketone) (PEEK), J. Polym. Sci.: Part B: Polym. Phys., 25, 935–945, 1987. 16. Kim, N.K. and Kammermeyer, K., Actual concentration profiles in membrane permeation, Sep. Sci., 5(6), 679–697, 1970. 17. Hwang, S.T. and Kammermeyer, K., Effect of thickness on permeability, in Permeability of Plastic Films and Coatings, Hopfenberg, H.B. Ed., Plenum, New York, 1974, pp. 197–205. 18. Skaarup, K., Abstract, Lecture at Nordic Polymer Days, 1988; Grænseflademodstand ved Diffusionsprocesser I Polymerer, Danmarks Ingeniør Akadami, Kemiafdelingen (in Danish) 1981, Surface Resistance in Diffusion Processes in Polymers (in English). 19. Marom, G., The role of water transport in composite materials, in Polymer Permeability, Comyn, J., Ed., Elsevier, London, 1985, chap. 9. 20. Hansen, C.M., A mathematical description of film drying by solvent evaporation, J. Oil Colour Chem. Assn., 51(1), 27–43, 1968. 21. Hansen, C.M., Diffusion coefficient measurements by solvent absorption in concentrated polymer solutions, J. Appl. Polym. Sci., 26, 3311–3315, 1981. 22. Windle, A.H., Case II sorption, in Polymer Permeability, Comyn, J., Ed., Elsevier, London, 1985, chap. 3. 23. Petropoulos, J.H., Interpretation of anomalous sorption kinetics in polymer-penetrant systems in terms of a time-dependent solubility coefficient, J. Poly. Sci. Polym. Phys. Ed., 22, 1885–1900, 1984.

7248_C016.fm Page 310 Wednesday, May 23, 2007 11:36 AM

310

Hansen Solubility Parameters: A User’s Handbook

24. Sanopoulou, M., and Petropolous, J.H., Systematic analysis and model interpretation of micromolecular non-fickian sorption kinetics in polymer films, Macromolecules, 34, 1400–1410, 2001.

7248_C017.fm Page 311 Wednesday, May 23, 2007 11:40 AM

— Safety and 17 Applications Environment Charles M. Hansen ABSTRACT Hansen solubility parameters (HSP) can be used to gain insight into many safety and environmental issues. These include substitution to more desirable materials, products, and processes, where a listing of solvents having HSP similarity to the one(s) to be substituted provides an overview of the potential choices for improvement. Selection of suitable chemical protective clothing can be improved by considering HSP correlations of breakthrough time. Evaluating risks for inadvertent chemical uptake in plastic can be helped by HSP correlations of chemical resistance and/or permeation phenomena. Similarity of HSP suggests which chemicals are most likely to be rapidly absorbed into given plastic types. These same approaches can be used to evaluate the potential for uptake of chemicals through human skin.

INTRODUCTION Many organic materials are potential safety hazards. They can also be harmful to the environment. Unfortunately, it is often a matter of experience before the risks are uncovered because of damage being done. Thus, over the years, there have been a series of substitutions with or without the aid of HSP to eliminate or to at least reduce such problems. An example is the lack of emphasis on the use of some ethylene glycol ethers as solvents because of their teratogenic effects, whereas they were used in massive quantities earlier. The problem of replacing ozone-depleting chemicals is a case involving the external environment. This is discussed a great deal in Chapter 11. Other large-scale substitutions can also be cited where HSP can aid, but a list of this type is not the purpose of this chapter. The emphasis here is on the use of sound formulating principles to reduce the potential hazard in terms of reformulation or substitution. When a satisfactory substitution cannot be found, personal protection of one type or another may be required. Here, again, HSP can help. Evaluating other forms of environmental risks can be aided by using HSP. An example is the occasional misuse of plastic containers normally used for soft drinks to store chemicals such as herbicides and pesticides. These are likely to diffuse into the plastic container wall itself, making customary washing insufficient. HSP can indicate which chemicals can do this, thus providing information on the means to improve handling of the problem. This type of information can be generated for any polymer where HSP correlations of chemical resistance, weight gain, etc., can be generated. All of these situations are discussed in more detail in the following.

SUBSTITUTION Substitution involves the replacement of a potentially dangerous process or chemical with a new process or chemical having less hazardous properties. The hazards can be judged using accepted approaches — for example, labeling requirements, toxicology assessments, biodegradability, and physical properties for the chemical or products. The volatility of products is also a significant

311

7248_C017.fm Page 312 Wednesday, May 23, 2007 11:40 AM

312

Hansen Solubility Parameters: A User’s Handbook

factor with lower volatility being preferred due to reduced workplace concentrations and reduced replacement requirements for cleaners and the like which often recirculate in nearly closed systems. On the other hand, the problem should not just be transferred from the air exhaust system to the sewer. The use of technologies involving water or mechanical methods, such as mechanical joints rather than the use of solvents, are preferred. Examples of preferred coatings technologies are the use of powders which flow at higher temperatures or polymerization by radiation, both of which use solvent-free base products to provide the coating. Other product types which may be targeted include cutting fluids, cleaners of various types, adhesives, sealers, and fillers. In general, one primarily wishes to substitute for • • • • •

Carcinogens or suspected carcinogens Substances with risk phrases for being very toxic, toxic, allergenic, carcinogenic, teratogenic, mutagenic, or causing cumulative or irreversible effects Substances with moderate or serious aquatic toxicity Nonbiodegradable substances Substances with high predicted aquatic effects — for example, chemicals which preferentially distribute to a nonaqueous phase to a very high degree

Efforts should be made to develop products with the lowest possible hazard. Those who have read the earlier chapters in this book will immediately recognize HSP as a tool to aid in the substitution and systematic formulation for reduced safety and environmental risks. A key element in this is a listing of solvents where those most resembling the candidate for substitution are at the top of the list. The program described in Chapter 1 can do this by entering the HSP for the solvent to be replaced and requesting a listing with those solvents most similar to it (i.e., the lowest RED numbers as defined in Chapter 1, Equations 1.10) at the top of the list. One must then sort through these potential replacement candidates using other information to arrive at a better alternative. It is clear that much more data than HSP are required to make the desired substitutions. However, a further discussion of this is beyond the scope of this chapter, which emphasizes HSP only. The currently used HSP techniques and correlations can aid in some aspects of substitution, and it is anticipated that future correlations will help in this endeavor. Many cases of substitutions in practice have been listed by Goldschmidt,1 Olsen,2 Soerensen and Petersen,3 and Filskov et al.4 A long list of references for the Danish experience with occupational risks and solutions is given by Soerensen and Petersen.3

ALTERNATIVE SYSTEMS Alternative systems with less solvent or no solvent have been focused on by the coatings and printing ink industries for many years. Examples of such systems are coatings with higher solids, radiation-curable inks and coatings, powder coatings, electrodeposition coatings, and other waterreducible products. It might appear that solvent technology and use of HSP will not be as important as it has been in the past. This is not the case, however, as demonstrated in earlier chapters. For example, HSP principles can be used to aid in improved stability and adhesion, to predict polymer/filler interactions, to improve barrier polymers, and to aid in understanding some biological phenomena. The use of solvents in alternative coatings systems has been the topic of several previous publications by the author.5–8 Some general principles of solvent selection have been discussed in Chapter 8 and earlier,9 as well as elsewhere more recently.10

7248_C017.fm Page 313 Wednesday, May 23, 2007 11:40 AM

Applications — Safety and Environment

313

SOLVENT FORMULATION AND PERSONAL PROTECTION FOR LEAST RISK Solubility parameter principles have been used in formulating alternative, low VOC (volatile organic compound) products. A number of the general formulation principles can be briefly stated for the sake of completeness. These include the following: 1. Solvents with lower viscosity most often lead to polymer solutions with lower viscosity. Such a change allows the use of higher solids at the original viscosity. However, these may evaporate more rapidly and can be expected to have a lower flash point. 2. Solvents with linear and smaller structures diffuse more rapidly than those with branched and larger structures. Inclusion of slower evaporating, more linear solvents can hasten the through-drying of a coating. 3. Two (or more) mixed solvents with lower labeling requirements may be able to replace a single solvent. HSP can be used in this type of endeavor. 4. The surface tension of water-reducible coatings can often be significantly reduced by relatively small additions of ethanol or other alcohol-type solvent. These can, of course, also be used in conjunction with other surface active materials. Materials with least potential risk are to be used in the Nordic countries wherever possible. The risk must be indicated by the seller/producer in terms of a labeling code. The risk can then be assessed by users or, perhaps more specifically, by primarily professional users. Such labeling is required on paints, printing inks, cleaners, or for any product containing significant amounts of solvent or hazardous chemical. The labeling code dictates the personal protection required for the product, depending on the way it is used. Spraying product in a smaller room with limited ventilation requires much more protection than applying paint with a brush outdoors. Tables have been published which give the protection required (gloves, dust mask, fresh air mask, body suit, etc.) for a given set of application conditions for a wide variety of products from paints and printing inks through cleaners.11 A key element in these tables is the labeling code developed in Denmark according to the MAL (in Danish: Maletekniske Arbejshygieniske Luftbehov) system. For present purposes, this is translated as the FAN (fresh air number). Higher MAL/FAN dictate that more extensive personal protection is required.

THE DANISH MAL SYSTEM — THE FAN12 As indicated previously, the quality of the working environment must be considered in all cases where organic solvents are being used. The Danish MAL system or other labeling system can be systematically used for this purpose. The Danish MAL reflects the cubic meters of fresh air required for ventilation of 1 l of product to below the threshold limit value (TLV). This number is modified by a constant, depending on the evaporation rate (or vapor pressure). Higher evaporation rates imply greater hazard, so the multiplier is larger. The concept behind the MAL system can be better understood in English by translating the MAL number as the FAN. Other numbers in addition to the TLV/GV/OEL (occupational exposure limit) and FAN have been generated to help evaluate risks by inclusion of evaporation rate/vapor pressure considerations. The vapor pressure divided by the TLV is called the vapor hazard ratio (VHR) and the actual calculated vapor composition (using activity coefficients) divided by the TLV has been called SUBFAC. A Danish publication comparing several of these is available.4 To demonstrate the principle, the simple MAL = FAN has been tabulated in Table 17.1 for several solvents.

7248_C017.fm Page 314 Wednesday, May 23, 2007 11:40 AM

314

Hansen Solubility Parameters: A User’s Handbook

TABLE 17.1 Fresh Air Numbers (FAN/MAL) for Selected Solvents from the Danish MAL Labeling Systema FAN/MAL

Solvent

1400 1100 880 110 88 78 74 58 54 48 46 29 28 26 25 24 23

Chloroform Tetrachloromethane Benzene Dichloromethane Trichloroethylene n-Hexane Toluene C9 Aromatics Methanol Methyl ethyl ketone Xylene 2-Propanol Propylene glycol monomethyl ether 1,1,1-Trichloroethane C>9 Aromatics Tetrahydrofurane Acetone

FAN/MAL 20 19 17 15 14 14 13 13 13 12 7 6 5 4b 0 0

Solvent n-Propanol Propylene glycol monomethyl ether acetate Propyl acetate Propylene glycol monopropyl ether Mineral spirits/white spirit Butyl acetates Ethyl acetate Cyclohexane Benzin/petroleum ether (as heptane) Heptane Ethanol Propylene glycol monobutyl ether Dipropylene glycol monomethyl ether DBE (dibasic esters) Ethylene glycol Propylene glycol

a

These numbers are developed primarily with regard to health hazards from vapors. The second number in the FAN code is added for hazard for skin contact, eye contact, respiratory system contact, and/or ingestion. In addition to these, the European Union requires use of Xi, Xn, C, T, etc. Several of the solvents require such labeling as well. One must also consider R (risk) and S (protective measure) labeling requirements. b Estimated from composition of the mixed solvent.15

Each product containing solvent is assigned a two-digit number to place it into a potential hazard category. This number is a summation of the hazards possible for the components which are considered potentially hazardous. The first number relates to the potential hazard from the vapors and will vary from 00 through 0, 1, 2, 3, 4, 5, to 6 as the potential hazard increases. The second number varies similarly and relates to the potential hazard from direct contact with the skin, eyes, breathing system, and by ingestion. This second number will not be less than 1 if organic solvents are included in the product in significant amounts. The following is a list of several solvents which are considered less desirable, based on this second number being 3 (or higher for higher concentrations in some cases): Toluene and Xylene at >10%, all common ethylene glycol based ethers and their acetates (including diethylene glycol monobutyl ether, for example), terpenes, monomers at rather low concentrations, amines at moderate concentrations, and the most common chlorinated solvents. A “3” in this category places the protection required in a significantly higher category with requirements for gloves as a minimum and frequently fresh air masks as well. As indicated, a two-digit MAL code defines which safety precautions are required for each of a large number of processing operations and conditions, including interior and exterior painting and gluing, whether or not large surfaces are involved, the quality of ventilation provided, surface preparation, painting of ships, larger construction sites, each of the printing processes, and industrial coating (spray boxes, cabinets, etc.).11 The protective measure required may be a face guard, eye protection, a dust mask, a gas filter mask, a combination filter mask, a fresh air supplied mask, or a body suit, in order of increasing requirements.

7248_C017.fm Page 315 Wednesday, May 23, 2007 11:40 AM

Applications — Safety and Environment

315

Examples of complete labeling of products and solvents are beyond the scope of this chapter. The purpose of the discussion is to suggest that possible substituting solvents can be listed, such as in Table 17.1, in an attempt to find a substitute with a lower labeling requirement. Systematic consideration of labeling requirements is becoming a significant parameter in commercial applications of solvents and products containing solvents. This is happening all over the world, both with regard to worker safety as well as to the external environment. Such a procedure has been used to arrive at optimum commercially useful solvent compositions with the lowest possible risk for workers in the serigraphic printing industry as described in Danish patents DK 153797B (1989) and DK 160883 (1991) which correspond to European patents EP 0 205 505 B1 and EP 0 270 654 B13,14 The preferred compositions reduce the MAL number to a minimum and also consider lowest possible internationally required labels as a requirement. The low label requirements of DBE (dibasic esters) have been emphasized in comparison with other solvents.15 Systematic solvent selection procedures have also been strongly suggested for use in the selection of solvents for restoring older paintings.16 This is discussed in Chapter 5.

SELECTION OF CHEMICAL PROTECTIVE CLOTHING HSP correlations for barrier properties of some types of chemical protective clothing are given in Chapter 13, Table 13.1. These correlations are based on data presented by Forsberg and Kieth.17 Other examples of HSP correlations of barrier properties of protective clothing are discussed in Chapter 13. Earlier publications also include HSP correlations of barrier properties of chemical protective clothing.18–21 The procedure for using these correlations requires knowledge of the HSP of the chemicals involved. These may be found in a suitable table or can be calculated according to the procedures outlined in Chapter 1. One then evaluates the RED number for the situation of interest. The RED number is discussed in Chapter 1 (Equation 1.10). If this number is less than 1, the system in not expected to be suitable for use. If the RED number is close to 1.0, there may be some doubt about the recommendation. RED numbers significantly greater than 1.0 can be considered for use. As discussed in Chapter 13, the molecular size of the chemical involved is important in these evaluations. The major use of such correlations is to evaluate potential barrier types for chemicals where test results are not available. One can usually divide the results into groups of clearly not acceptable, questionable, and worthy of further consideration. There have been recent attempts to improve on the direct correlation of breakthrough times and permeation rates with HSP by trying to estimate the solubility and diffusion coefficients separately using HSP.22–25 These efforts have been discussed in Chapter 2 and Chapter 13.

UPTAKE OF CONTENTS BY A PLASTIC CONTAINER Plastic containers have become increasingly popular in recent years. They have many advantages (which will not be discussed here), but there is also one disadvantage that HSP can shed more light on. This is the fact that plastic materials are able to absorb various liquids to some extent. The extent of absorption clearly depends on the HSP of the plastic used in the container compared with the HSP of the liquid which is in contact with it. Containers in contact with food have been tested well for suitability for this purpose, including barrier properties relative to the contents. This is not the point of the present discussion. A problem exists with the inadvertent storage of hazardous liquids in the plastic container prior to its expected recycling as a container for a food or beverage. Many types of liquids can be temporarily stored in such containers. Whereas the earlier glass or metal containers could not absorb potentially dangerous materials, a plastic container can do this. A simple washing operation cannot be expected to remove all of the absorbed material. Washing

7248_C017.fm Page 316 Wednesday, May 23, 2007 11:40 AM

316

Hansen Solubility Parameters: A User’s Handbook

only removes what is on the surface or what can diffuse to the surface during the washing process, which presumably takes place at some higher temperature. HSP concepts can focus attention on the types of chemicals that can absorb into a given type of plastic container. This is useful information in terms of what analyses should be performed prior to recycling. The principles discussed here can possibly contribute in other ways to improve the recycling process based on the increased level of knowledge. There may be other ways to reduce the problem.

SKIN PENETRATION Human skin is a complicated system. Nevertheless, it has been possible to characterize some aspects of the behavior of human skin by HSP. The HSP found in a correlation of permeation rates of liquids in contact with viable skin26 are similar to those found for the swelling of psoriasis scales.27,28 This has been discussed in Chapter 15 in more detail, but also relates to worker safety. The HSP for these correlations are included in Chapter 13, Table 13.1 and Chapter 15, Table 15.1. A skin penetration warning has been attached to many liquids taken up in the lists of limiting values for workplaces which are published in different countries. It was found earlier based on the HSP correlation with the swelling of psoriasis scales that this practice could be misleading, as HSP predicted many liquids without this warning also swelled psoriasis scales (keratin) and could therefore be expected to penetrate the skin.27 The lack of a skin penetration warning for these liquids is partly attributable to the fact that this warning is based on experience. The bad experience giving the warning includes a combination of all effects, most notably the combination of dose and toxicity, rather than the potential dose effect only which is indicated by similarity of HSP. Earlier discussions also led to the impression that those involved in this area did not consider the swelling of psoriasis skin as having relevance to the permeation of living skin. The finding that comparable δP and δH are found from correlating the permeation rates of solvents through living skin is a new input into this discussion. It is recognized that the δD parameter is different, but reasons for this are not clear. An improved HSP correlation of the permeation rates of solvents through living skin based on a larger number of solvents than the 13 included in the work of Ursin et al.26 is perhaps required to give improved predictions in marginal cases, i.e., those near the boundaries of the HSP sphere describing the situation. The size and shape of the penetrating liquid molecules must also be considered. Predictions of the barrier properties of viable human skin should receive more attention. In addition, there is some discussion of the use of HSP in this respect in Chapter 15.

TRANSPORT PHENOMENA Many chemicals have been the subject for concern in the past for various environmental reasons. Among these is the presence in artic regions of chemicals that do not readily break down. Some chlorinated materials, such as pentachlorophenol, have the ability to penetrate skin and wood, and to be transported by animals, birds, or aquatic species after they have taken them up. The HSP of given chemicals can give a clue as to whether or not they can follow the same pathways in the environment. An example is given in Table 17.2 where the HSP for tetrabromobisphenol A (TBBPA) are reported along with the similarity of these with other relevant materials. TBBPA has a distance from Pentachlorophenol (PCP) of only 3.5 units. This is very close and means that where PCP is soluble, TBBPA will also be soluble. TBBPA has a distance from the center of the spherical HSP correlation for Lignin solubility of only 6.8 units. Dividing this by the radius of the sphere to find the RED number (relative energy difference) shows that it is well within the solubility region with a RED of 0.5. TBBPA is readily soluble in lignin (wood).

7248_C017.fm Page 317 Wednesday, May 23, 2007 11:40 AM

Applications — Safety and Environment

317

TABLE 17.2 The Affinities of Tetrabromobisphenol A (TBBPA) for Selected Biological Materials Material

δD

δP

δH

Ro

Dist. to TBBPA

TBBPA Pentachlorophenol Lignin solubility correlation Rapid skin penetration correlation Swelling of psoriasis scales correlation Depot fat (37°C) total solubility Blood serum

20.2 21.5 21.9 17.6 24.6 15.9 25.5

9.1 6.9 14.1 12.5 11.9 1.2 10.3

13.8 12.8 16.9 11.0 12.9 5.4 22.1

— — 13.7 5.0 19.0 12.0 17.8

0.0 3.5 6.8 (RED = 0.50) 6.5 (RED = 1.36) 9.3 (RED = 0.49) 14.3 (RED) = 1.20 13.1 (RED = 0.73)

Note: Units are MPa1/2. Source: Reprinted with permission from Hansen, C.M., Conference Proceedings, Pharmaceutical and Medical Packaging 2001, Skov, H.R., Ed., Hexagon Holding, Copenhagen, 2001, pp. 20.1–20.10.

TBBPA has a distance from the center of the spherical correlation for rapid skin permeation of 6.5. This is just outside the region for rapid permeation, but means that permeation will certainly take place at a moderate rate. The size and shape of the molecule will dictate the rate of permeation, not the solubility relations. TBBPA will readily absorb into the outer layer of the skin (keratin) here described by swelling of psoriasis scales. The rate of absorption is dictated by the size and shape of the molecule, and not by the solubility relations. These data confirm that TBBPA is readily soluble where pentachlorophenol is soluble. If the chemical stability is comparable or better, then it can be presumed to appear at the same places if it gets into the environment. The same is true of related brominated compounds in general. Both pentachlorophenol and TBBPA can readily penetrate wood and wood products. They will also be readily taken up at lower concentrations without delay by human skin as shown by the correlations of rapid skin penetration and the swelling of psoriasis scales. Similar analyses can be done with other chemicals. HSP are available in Appendix Table A.1 for many phthalate plasticizers, mono 2-ethylhexyl phthalate, bisphenol A, N-methyl-2-pyrrolidone, and various glycol ethers based on ethylene oxide. In principle any compound of interest can be assigned HSP for use in predicting behavior in connection with a large number of environmental subjects. The correlations of the solubility of depot fat at 37°C, the solubility of a protein (human blood serum), and swelling of keratin (psoriasis scales swelling) indicate that TBBPA prefers to reside in the nonfatty tissue, but that it will have some solubility in the fatty tissues as well. The collection of chemicals in fatty tissue is another topic that could be explored. A simple rule of thumb is that lack of water solubility will encourage collection in the fatty tissues, but this could be given more precision with HSP. The major problem with this collection is that the central nervous system is almost completely fatty in nature, and an excess of foreign materials can shortcircuit, misdirect, or stop messages, leading to memory problems, etc.

CONCLUSION In conclusion, it can be noted that HSP provides a tool to aid in substitution and in systematic formulation of less hazardous products and processes. One can also use HSP to more rapidly arrive at an optimum choice of chemical protective clothing. HSP provides other insights with regard to

7248_C017.fm Page 318 Wednesday, May 23, 2007 11:40 AM

318

Hansen Solubility Parameters: A User’s Handbook

uptake of undesirable chemicals in the human skin, in packaging materials, and perhaps even in a wide variety of other materials such as those found in nature.

REFERENCES 1. Goldschmidt, G., An analytical approach for reducing workplace health hazards through substitution, Am. Ind. Hyg. Assoc. J., 54, 36–43, January 1993. 2. Olsen, E., Substitution: a method to fulfill the working environment law for air quality (Substitution: En Metode til at Overholde Arbejdsmiljoelovens Krav til Luftkvaliteten, in Danish), Dan. Kemi, 67(5), 146–153, 1986. 3. Soerensen, F. and Petersen, H.J.S., Substitution of hazardous chemicals and the Danish experience, Occup. Hyg., 1, 261–278, 1995. 4. Filskov, P., Goldschmidt, G., Hansen, M.K., Höglund, L., Johansen, T., Pedersen, C.L., and Wibroe, L., Substitution in Practice — Experience from BST (Substitution i Praksis — Erfaringer fra BST), Arbejsmiljøfondet, in Danish), Copenhagen, 1989. 5. Hansen, C.M., Solvents in water-borne coatings, Ind. Eng. Chem. Prod. Res. Dev., 16(3), 266–268, 1977. 6. Hansen, C.M., Organic solvents in high solids and water-reducible coatings, Prog. Org. Coat., 10(3), 331–352, 1982. 7. Holten-Andersen, J. and Hansen, C.M., Solvent and water evaporation from coatings, Prog. Org. Coat., 11(3), 219–240, 1983. 8. Saarnak, A. and Hansen, C.M., Evaporation from high solids coatings (Avdunstningen Från LFFårger), Färg och Lack, in Swedish, 30(5), 100–105, 1984. 9. Hansen, C.M., Solvents for coatings, Chem. Technol., 2(9), 547–553, 1972. 10. Wu, D.T., Formulating solvents to remove hazardous air pollutants, Polym. Paint Colour J., 185, 20–23, December 1995. 11. Anonymous, Directive on Work with Products Having Codes (Bekendtgørelse om arbejde med kodenummererede produkter, Arbejdstilsynets Bekendtgørelse nr.302 af 13. maj 1993, in Danish), Danish Directorate for Labor Inspection. 12. Anonymous, Directive on Determination of Codes (Bekendtgørelse om fastsættelse af kodenumre, Arbejdstilsynets bekendtgørelse nr. 301 af 13 maj 1993, in Danish), Danish Directorate for Labor Inspection. 13. Madsen, C.H. and Hansen, C.M., EP 0 205 505 B1, 1988. Assigned to CPS Kemi Aps (Now a part of the Autotype/MacDermid Concern). 14. Madsen, C.H. and Hansen, C.M., EP 0 270 654 B1, 1991. Assigned to CPS Kemi Aps (Now a part of the Autotype/MacDermid Concern). 15. Altnau, G., Risikopotentiale von Lösemitteln systematisch bewerten, Farbe+Lack, 103(9), 34–37, 1997; Systematic Evaluation of Risk Potentials of Solvents, Eur. Coat. J., 6/98, 454–457, 1998. 16. Hansen, C.M., Conservation and Solubility Parameters, Nordic Conservation Congress Preprints, Copenhagen, 1994, pp. 1–13. 17. Forsberg, K. and Kieth, L.H., Chemical Protective Clothing Performance Index, 4th ed., Instant Reference Sources, Inc., Austin, TX, 1991. 18. Hansen, C.M., The systematic choice of material in personal protection equipment for organic solvents in safety and health aspects of organic solvents, Progress in Chemical and Biological Research 220, Riihimäki, V. and Ulfvarson, U., Eds., Alan R. Liss, New York, 1986, pp. 297–302. 19. Hansen, C.M. and Hansen, K.M., Which gloves should I put on? (Hvilke Handsker Skal Jeg Tage På?, in Danish), Färg och Lack, 33(3), 45–49, 1987. 20. Hansen, C.M. and Hansen, K.M., Solubility parameter prediction of the barrier properties of chemical protective clothing, Performance of Protective Clothing: Second Symposium, ASTM STP 989, Mansdorf, S.Z., Sager, R., and Nielsen, A.P., Eds., American Society for Testing and Materials, Philadelphia, PA, 1988, pp. 197–208.

7248_C017.fm Page 319 Wednesday, May 23, 2007 11:40 AM

Applications — Safety and Environment

319

21. Hansen, C.M., Billing, C.B., and Bentz, A.P., Selection and use of molecular parameters to predict permeation through fluoropolymer-based protective clothing materials, The Performance of Protective Clothing; Fourth Volume, ASTM STP 1133, McBriarty, J.P. and Henry, N.W., Eds., American Society for Testing and Materials, Philadelphia, PA, 1992, pp. 894–907. 22. Zellers, E.T., Three-dimensional solubility parameters and chemical protective clothing permeation. I. Modeling the solubility of organic solvents in Viton® gloves, J. Appl. Polym. Sci., 50, 513–530, 1993. 23. Zellers, E.T. and Zhang, G.-Z., Three-dimensional solubility parameters and chemical protective clothing permeation. II. Modeling diffusion coefficients, breakthrough times, and steady-state permeation rates of organic solvents in Viton® gloves, J. Appl. Polym. Sci., 50, 531–540, 1993. 24. Zellers, E.T., Anna, D.H., Sulewski, R., and Wei, X., Critical analysis of the graphical determination of Hansen’s solubility parameters for lightly crosslinked polymers, J. Appl. Polym. Sci., 62, 2069–2080, 1996. 25. Zellers, E.T., Anna, D.H., Sulewski, R., and Wei, X., Improved methods for the determination of Hansen’s solubility parameters and the estimation of solvent uptake for lightly crosslinked polymers, J. Appl. Polym. Sci., 62, 2081–2096, 1996. 26. Ursin, C., Hansen, C.M., Van Dyk, J.W., Jensen, P.O., Christensen, I.J., and Ebbehoej, J., Permeability of commercial solvents through living human skin, Am. Ind. Hyg. Assoc. J., 56, 651–660, 1995. 27. Hansen, C.M., The Absorption of Liquids into the Skin, Report No. T 3-82, Scandinavian Paint and Printing Ink Research Institute, Hoersholm, Denmark, 1982. 28. Hansen, C.M. and Andersen, B.H., The affinities of organic solvents in biological systems, Am. Ind. Hyg. Assoc. J., 49(6), 301–308, 1988.

7248_C017.fm Page 320 Wednesday, May 23, 2007 11:40 AM

7248_C018.fm Page 321 Wednesday, May 23, 2007 11:46 AM

18 The Future Charles M. Hansen

ABSTRACT Hansen solubility parameters (HSP) help to quantify the statements “like dissolves like” and “like seeks like.” These parameters have found use in many fields of research and practice, primarily because their unique predictive capabilities are based on sound theoretical principles. HSP have extended the original Hildebrand single solubility parameter approach by quantitatively taking into account the molecular permanent dipole–permanent dipole and molecular hydrogen bonding (electron interchange) interactions. HSP and the Prigogine corresponding states theory of polymer solutions are mutually confirming with regard to treatment of specific interactions, as shown in Chapter 2. This is important, as it confirms that the HSP correlations must continue to include a constant not too different from the currently used “4” (or 0.25). This is necessary to differentiate between the atomic (δD) and the molecular (specific) interactions (δP and δH). Neglecting this differentiation will lead to misinterpretations. The geometric mean average for the interaction of unlike molecules is inherently used in the Hildebrand approach and in the HSP approach as well. This same mean must be used in the Prigogine corresponding states theory if agreement is to be found with the HSP correlations presented in this book. As the agreement is general, the conclusion must be that the geometric mean can be used to average not only dispersion interactions but also those attributable to permanent dipoles and to hydrogen bonding. These findings have been supported more recently by the statistical thermodynamics approach of Panayiotou and coworkers summarized in Chapter 3. This approach allows independent calculation of each of the three parameters. Based on the large number of current uses of HSP, one can easily suppose that there are many more practical uses which remain to be discovered and developed. One need not necessarily extend its theoretical scope to accomplish this. The existing data can be used in a strictly empirical manner if so desired. However, a glimpse has been given of a very general energetic approach to systematically predict and control molecular interactions among many materials of widely different composition. The general predictions possible for these physical interactions have been demonstrated for both bulk phenomena (solubility, swelling, compatibility) and surface phenomena (adsorption, dewetting, spontaneous spreading). In the future, the theory should be explored and used with this general applicability in mind. Problems and situations clearly needing further attention are discussed in the following.

INTRODUCTION There are many matters related to HSP which still need clarification and expansion. Some limitations are clear, but others are not so clear. As this book is primarily directed toward the practitioner, the following discussions will start with more practical topics. The first matter of concern is the availability of data. This handbook attempts to help improve this situation by publishing HSP for a larger number of liquids, about 1200, primarily in the Appendix, Table A.1. This handbook also contains new HSP correlations not present in the first edition. Many of these are given as examples in the text, and others are included in the Appendix, Table A.2. Other sources are discussed below. The second matter of concern is how reliable the HSP data are and how accurately the correlations can predict the behavior of untested systems. Qualitative indications of this for the 321

7248_C018.fm Page 322 Wednesday, May 23, 2007 11:46 AM

322

Hansen Solubility Parameters: A User’s Handbook

data generated by the author are given in the relevant tables. In general correlation coefficients approach 1.0. This indicates perhaps only a few minor outliers in the correlations, as can be seen from those included in this handbook. There are very rarely major outliers, and these usually have another explanation for their behavior, such as very large molecular size, very small molecular size, reactions, or the like. The experience reported in Chapter 4, Table 4.4A, for the reliability of the “original Hansen” approach does not correspond to this experience. Normally there are perhaps 5 or 6 boundary solvents that are not predicted correctly out of about 100 test solvents in a correlation of experimental data. This was the case for the correlations presented in Table 5.1, rather than the ratio of 99 correct answers with 23 incorrect answers indicated in Table 4.4A. The reason for this discrepancy is not known, but if group contribution or other estimates are involved, especially for polymers, then the number of “errors” will increase. A third point which is sometimes irritating is that the scope of the characterizations possible is limited to the cohesive energy spectrum of the test liquids. A situation is often met where only a few solvents having high solubility parameters dissolve a polymer which has still higher HSP. Similarly, only a few solvents may interact intimately with a surface which has very high HSP. These surfaces are clearly wet because of the lower surface tension of all of the liquids, but only a few with high HSP prolong suspension of finer particles, for example. The energy characteristics of such surfaces are apparently higher than those of any liquids which can be used to study them by these techniques. Very high cohesive energies lead to the formation of solids, so there are no pure liquids which can be used to test the very high energy materials. New thinking and new techniques are required to accurately characterize such high energy materials. A full understanding of the behavior of water, organometallic materials, and salt solutions might be helpful in these situations (see the following corresponding sections). The current practice is to extrapolate into the region of very high HSP using Chapter 1, Equation 1.9 which includes the constant “4.” It is assumed that this constant is still valid, even for these very high energy characterizations. The given good solvents are often in the boundary region of the HSP spherical characterizations. The solubility of Dextran C (British Drug Houses)1 is an example of this as shown in Table 18.1 and Table 18.2. See Chapter 5 and Chapter 7 for further discussion of this problem which is present for both polymers and particulate matter. In a sense, the problem is similar to measuring the surface tension of a surface which has such a high value that even water spontaneously spreads on it. One can only conclude that its surface tension is greater than that of water. In the present case, there is a model to extrapolate HSP to higher values than can be measured directly. Another concern related to reliable HSP values is based on the fact that most chemicals in the intermediate molecular weight range, such as that characteristic of plasticizers, are soluble in almost all of the test liquids, except for, for example, glycerin, water, and hexane. It is impossible to establish the three HSP based on such data. One generally has to rely on group contribution methods or other calculations or comparisons, and there will be some uncertainty involved with this. Once the necessary reliable HSP data are available, decisions and ideas are needed on how the data should be used. It is here that the existing theory and future extensions of it are most important. In many cases, engineering approximations leading to a systematic course of action have been possible using data which is currently available. One can often arrive at a prediction for expected behavior using the “like seeks like” principle, even though accurate numbers and an appropriate detailed theory may be lacking. It is hoped that this book will aid in the generation of still more HSP data having a uniformly high quality, such that the interactions among still more materials can be predicted. Logical applications for HSP will be found in self-assembling systems and in what is called nanotechnology, for example. One example is the self-stratifying paints discussed in Chapter 8. Another is the ultrastructure of cell walls in wood discussed in Chapter 15.

7248_C018.fm Page 323 Wednesday, May 23, 2007 11:46 AM

The Future

323

TABLE 18.1 Calculated Solubility Sphere for Dextran C Solubility The Solvents with Their Parameters Solvent

δD

δP

δH

SOLUB

RED

V

Acetone Acetophenone Aniline Benzaldehyde Benzene 1,3-Butanediol 1-Butanol Butyl acetate gamma-Butyrolactone Carbon disulfide Carbon tetrachloride Chlorobenzene Chloroform m-Cresol Cyclohexanol Cyclohexanone Diacetone alcohol o-Dichlorobenzene 2,2-Dichlorodiethyl ether Diethylene glycol Diethyl ether Dimethyl formamide Dimethyl sulfoxide 1,4-Dioxane Dipropylene glycol Ethanol Ethanolamine Ethyl acetate Ethylene dichloride Ethylene glycol Ethylene glycol monobutyl ether Ethylene glycol monoethyl ether Ethylene glycol monomethyl ether Formamide Glycerol Hexane Isophorone Methanol Methylene dichloride Methyl isobutyl carbinol Methyl isobutyl ketone Nitrobenzene Nitromethane 2-Nitropropane Propylene carbonate Propylene glycol

15.5 19.6 19.4 19.4 18.4 16.6 16.0 15.8 19.0 20.5 17.8 19.0 17.8 18.0 17.4 17.8 15.8 19.2 18.8 16.6 14.5 17.4 18.4 19.0 16.5 15.8 17.0 15.8 19.0 17.0 16.0 16.2 16.2 17.2 17.4 14.9 16.6 15.1 18.2 15.4 15.3 20.0 15.8 16.2 20.0 16.8

10.4 8.6 5.1 7.4 0.0 10.0 5.7 3.7 16.6 0.0 0.0 4.3 3.1 5.1 4.1 6.3 8.2 6.3 9.0 12.0 2.9 13.7 16.4 1.8 10.6 8.8 15.5 5.3 7.4 11.0 5.1 9.2 9.2 26.2 12.1 0.0 8.2 12.3 6.3 3.3 6.1 8.6 18.8 12.1 18.0 9.4

7.0 3.7 10.2 5.3 2.0 21.5 15.8 6.3 7.4 0.6 0.6 2.0 5.7 12.9 13.5 5.1 10.8 3.3 5.7 20.7 5.1 11.3 10.2 7.4 17.7 19.4 21.2 7.2 4.1 26.0 12.3 14.3 16.4 19.0 29.3 0.0 7.4 22.3 6.1 12.3 4.1 4.1 5.1 4.1 4.1 23.3

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1* 0 0 0 1 0 0 1* 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0

1.454 1.371 1.241 1.346 1.776 1.054 1.313 1.640 1.077 1.756 1.858 1.601 1.556 1.246 1.312 1.473 1.363 1.474 1.313 1.000 1.795 1.082 1.000 1.485 1.080 1.180 0.880 1.559 1.416 1.003 1.406 1.211 1.170 0.915 0.991 2.037 1.410 1.144 1.411 1.517 1.679 1.336 1.399 1.479 1.172 1.053

74.0 117.4 91.5 101.5 89.4 89.9 91.5 132.5 76.8 60.0 97.1 102.1 80.7 104.7 106.0 104.0 124.2 112.8 117.2 94.9 104.8 77.0 71.3 85.7 130.9 58.5 59.8 98.5 79.4 55.8 131.6 97.8 79.1 39.8 73.3 131.6 150.5 40.7 63.9 127.2 125.8 102.7 54.3 86.9 85.0 73.6

7248_C018.fm Page 324 Wednesday, May 23, 2007 11:46 AM

324

Hansen Solubility Parameters: A User’s Handbook

TABLE 18.1 (CONTINUED) Calculated Solubility Sphere for Dextran C Solubility Solvent Tetrahydrofuran Tetrahydronaphthalene Toluene Trichloroethylene

δD 16.8 19.6 18.0 18.0

δP

δH

5.7 2.0 1.4 3.1

8.0 2.9 2.0 5.3

SOLUB

RED

V

0 0 0 0

1.450 1.618 1.744 1.560

81.7 136.0 106.8 90.2

Note: δD = 24.3 δP = 19.9 δH = 22.5 Ro = 17.4 FIT = 0.999 NO = 50.

HANSEN SOLUBILITY PARAMETER DATA AND DATA QUALITY The author and others including most solvent suppliers and some paint companies (at least) have databases including HSP data for solvents and HSP correlations for polymer solubility etc. Tables of HSP data for many materials are also included in standard reference works.2–5 There is still a tendency to regard the contents of such databases as proprietary information for the benefit of the owner and/or his/her customers. Exxon, for example, has indicated a computer program based on HSP where data for over 500 solvents and plasticizers, 450 resins, and 500 pesticides are included.6,7 The use of these parameters is becoming so commonplace that, in many studies, the δD, δP, and δH parameters often appear without any specific reference to where they came from or what they actually represent. The solvent listing in the Appendix, Table A.1 includes the previously published set of some 240 solvents which have appeared earlier in several sources.2,4,5,8,9 Some of the values have been revised over the years. The materials given in dark type have had some degree of experimental verification. All the others are based on calculations only. The methods described in Chapter 1 were used, although in many cases data was lacking to such an extent that group contributions were used. In some cases data for whole, smaller molecules whose HSP are known can be used to derive group contributions for estimating the HSP of larger molecules wherein they appear as a part. There are many additions to the original set of data. The calculated values have been checked against performance data reported in the literature where this has been possible. An example is the solubility data reported for poly(vinylidene chloride) (PVDC).10 Appendix, Table A.1 also includes HSP for a number of low molecular weight solids. Low molecular weight solids with relatively low melting points have been treated as if they were liquids for extrapolation of latent heats to 25°C. This seems to be satisfactory, and it is consistent with the treatment of high boiling liquids. See Chapter 1 for details of the calculations. The first edition of this handbook contained over 800 HSP values for chemicals. This has been expanded to about 1200 values in the second edition. HSP correlations in addition to those given in connection with examples in the text are included in Appendix, Table A.2. Only data judged (reasonably) reliable are reported. There are limitations on the accuracy of the HSP data derived from Burrell’s solvent range studies reported in standard reference works,2,11,12 but many correlations based on these data are included for reference anyway. The solvent range chosen for the studies does not completely fill out the possibilities selection of different liquids would have allowed. The problem of estimating a sphere based on limited data which do not experimentally define the whole sphere becomes more acute. This problem is greatest for polymers with high HSP, as not only is there a lack of possible data, but much of the volume of the HSP sphere is located where there are no liquids. The cohesion energies are so high that no liquids are possible and only solids are present. An example of a good HSP correlation from the solvent range studies is that of polyethylene sulfide. This polymer has relatively low HSP, and the solvents in the test series provide nonsolvents at higher HSP than those of the polymer to locate the boundaries with sufficient accuracy. This is shown in Tables 18.3 and 18.4. A comparison of

7248_C018.fm Page 325 Wednesday, May 23, 2007 11:46 AM

The Future

325

TABLE 18.2 Calculated Solubility Sphere for Dextran C Solubility The Solvents with Their Parameters Solvent

δD

δP

δH

Succinic anhydride Triethanolamine Dimethyl sulfone Ethylene cyanohydrin 2-Pyrolidone Ethanolamine Formamide Diethanolamine 1,3-Butanediol Glycerol Dimethyl sulfoxide Diethylene glycol Ethylene glycol Propylene glycol 1,3-Butanediol Diethylenetriamine Triethyleneglycol gamma-Butyrolactone Dipropylene glycol Dimethyl formamide Allyl alcohol o-Methoxyphenol Hexamethylphosphoramide Ethylenediamine Methanol Furfuryl alcohol Trimethylphosphate Benzyl alcohol Ethylene carbonate Phenol Ethylene glycol monomethyl ether Propylene carbonate Ethanol 1,1,2,2-Tetrabromoethane Ethylene glycol monoethyl ether N,N-Dimethyl acetamide 3-Chloro-1-propanol Hexylene glycol Methyl-2-pyrrolidone Furfural Aniline m-Cresol 1-Propanol Benzoic acid Triethylphosphate Quinoline Diethylene glycol monoethyl ether Acetic anhydride

18.6 17.3 19.0 17.2 19.4 17.0 17.2 17.2 18.0 17.4 18.4 16.6 17.0 16.8 16.6 16.7 16.0 19.0 16.5 17.4 16.2 18.0 18.5 16.6 15.1 17.4 16.7 18.4 19.4 18.0 16.2 20.0 15.8 22.6 16.2 16.8 17.5 15.7 18.0 18.6 19.4 18.0 16.0 18.2 16.7 19.4 16.1 16.0

19.2 22.4 19.4 18.8 17.4 15.5 26.2 10.8 8.4 12.1 16.4 12.0 11.0 9.4 10.0 13.3 12.5 16.6 10.6 13.7 10.8 8.2 8.6 8.8 12.3 7.6 15.9 6.3 21.7 5.9 9.2 18.0 8.8 5.1 9.2 11.5 5.7 8.4 12.3 14.9 5.1 5.1 6.8 6.9 11.4 7.0 9.2 11.7

16.6 23.3 12.3 17.6 11.3 21.2 19.0 21.2 21.0 29.3 10.2 20.7 26.0 23.3 21.5 14.3 18.6 7.4 17.7 11.3 16.8 13.3 11.3 17.0 22.3 15.1 10.2 13.7 5.1 14.9 16.4 4.1 19.4 8.2 14.3 10.2 14.7 17.8 7.2 5.1 10.2 12.9 17.4 9.8 9.2 7.6 12.2 10.2

SOLUB

1 1

1 1 0 1* 0 0

0 0 0

0

0 0 0 0

0 0

RED 0.739 0.819 0.846 0.866 0.867 0.880 0.915 0.972 0.984 0.991 1.000 1.000 1.003 1.053 1.054 1.063 1.068 1.077 1.080 1.082 1.117 1.121 1.132 1.136 1.144 1.144 1.147 1.152 1.152 1.167 1.170 1.172 1.180 1.199 1.211 1.215 1.216 1.219 1.220 1.230 1.241 1.246 1.250 1.258 1.259 1.265 1.272 1.277

V 66.8 133.2 75.0 68.3 76.4 59.8 39.8 95.9 87.5 73.3 71.3 94.9 55.8 73.6 89.9 108.0 114.0 76.8 130.9 77.0 68.4 109.5 175.7 67.3 40.7 86.5 115.8 103.6 66.0 87.5 79.1 85.0 58.5 116.8 97.8 92.5 84.2 123.0 96.5 83.2 91.5 104.7 75.2 100.0 171.0 118.0 130.9 94.5

7248_C018.fm Page 326 Wednesday, May 23, 2007 11:46 AM

326

Hansen Solubility Parameters: A User’s Handbook

TABLE 18.2 (CONTINUED) Calculated Solubility Sphere for Dextran C Solubility Solvent Tricresyl phosphate Formic acid Tetramethylurea

δD

δP

δH

19.0 14.3 16.7

12.3 11.9 8.2

4.5 16.6 11.0

SOLUB

RED 1.277 1.284 1.285

V 316.0 37.8 120.4

Note: δD = 24.3 δP = 19.9 δH = 22.5 Ro = 17.4 FIT = 0.999 NO = 50.

the solvents included in Table 18.3 with those in Table 18.1 shows which ones are lacking in the high HSP range. An example of a poor correlation using solvent range data is that of the solubility of polyvinyl alcohol. Only two of the solvents, ethanol and 2-propanol, dissolve it. This leads to a correlation with the following data: δD;δP;δH;Ro equal to 17.0;9.0;18.0;4.0 in MPa1/2 with a perfect fit for two good solvents out of 56 in the set of data. The use of these data is not recommended. Ro is clearly too small by comparison with Ro found in HSP correlations for solubility for other water-soluble polymers. One of the problems with some of the reported correlations in the Appendix, Table A.2 is that the data on which they are based were not generated for this purpose. There are shortcomings in terms of lack of full coverage of the HSP space as well as in the total number of liquids for which there are data. Note that a standard set of test solvents such as that used in Table 18.1 takes full coverage into account. However, some of these liquids must be handled with care for reasons of toxicity. Data for chemical resistance, permeation, and other phenomena related to solubility which can be correlated with HSP are practically never accumulated with an HSP correlation in mind. This does not prevent use of such data as demonstrated elsewhere in this book, but it does place some limitations on the reliability of the predictions obtainable from the correlations. A qualitative indication of the reliability of the correlations is given for this reason. Reliable HSP data for many polymers of practical importance are not available at this time. It would seem advisable for raw material suppliers to determine the HSP for their relevant products in a reliable manner and to publish these data on their product data sheets or elsewhere. Including them in a possible future edition of this book may also be a possibility. For the sake of completeness, a couple of warnings are appropriate before proceeding to the next section. As noted in Chapter 1, the three partial solubility parameters tabulated by Hoy13,14 are not compatible with those of the author. As discussed in the next section, the group contribution procedure presented by van Krevelen and Hoftyzer15 does not give satisfactory agreement with the procedures given in Chapter 1. Finally, water (or its mixtures) should not be included currently in any HSP correlations without a very careful analysis of the results. The small molecular volume, exceptionally high δH parameter, and tendency to self-associate depending on the local environment all lead to the likely result that water will be an outlier for the correlation. This results in HSP values which are less reliable, and have lower predictive ability than had water been neglected. Mixtures of organic solvents with water are still more problematic when used as test liquids (see Figure 18.1 and the following discussion). A goal of future work should be to be able to account for the behavior of water in a reliable manner, such that it can be included in studies leading to HSP correlations. The HSP values for water found from the correlation for total water solubility reported in Chapter 1 (Table 1.3) appear promising for some applications where the HSP values for water as a single molecule are clearly not applicable.

7248_C018.fm Page 327 Wednesday, May 23, 2007 11:46 AM

The Future

327

TABLE 18.3 Calculated Solubility Sphere for Polyethylenesulfide The Solvents with Their Parameters Solvent

δD

δP

δH

SOLUB

RED

V

Acetic acid Acetone Acetonitrile Aniline Benzene 1-Butanol sec-Butyl acetate Butyraldehyde Carbon tetrachloride Chlorobenzene p-Chlorotoluene m-Cresol Cyclohexane Cyclopentanone 1,2-Dichloro ethylene (cis) o-Dichlorobenzene 2,2-Dichlorodiethyl ether Dichlorodifluoromethane (Freon 12) Dichloromonofluoromethane Diethyl amine Diethyl ether Diethylene glycol Di-isobutyl ketone N,N-Dimethyl acetamide Dimethyl formamide 1,4-Dioxane Ethanol Ethyl acetate 2-Ethyl hexanol Ethylene carbonate Ethylene glycol Ethylene glycol monobutyl ether Ethylene glycol monoethyl ether Furfural Furfuryl alcohol Glycerol Isoamyl acetate Isoamyl alcohol Isopropyl acetate Methanol Methyl acetate Methyl ethyl ketone Methyl n-amyl ketone Nitroethane Nitromethane Octane

14.5 15.5 15.3 19.4 18.4 16.0 15.0 14.7 17.8 19.0 19.1 18.0 16.8 17.9 17.0 19.2 18.8 12.3 15.8 14.9 14.5 16.6 16.0 16.8 17.4 19.0 15.8 15.8 15.9 19.4 17.0 16.0 16.2 18.6 17.4 17.4 15.3 15.8 14.9 15.1 15.5 16.0 16.2 16.0 15.8 15.5

8.0 10.4 18.0 5.1 0.0 5.7 3.7 5.3 0.0 4.3 6.2 5.1 0.0 11.9 8.0 6.3 9.0 2.0 3.1 2.3 2.9 12.0 3.7 11.5 13.7 1.8 8.8 5.3 3.3 21.7 11.0 5.1 9.2 14.9 7.6 12.1 3.1 5.2 4.5 12.3 7.2 9.0 5.7 15.5 18.8 0.0

13.5 7.0 6.1 10.2 2.0 15.8 7.6 7.0 0.6 2.0 2.6 12.9 0.2 5.2 3.2 3.3 5.7 0.0 5.7 6.1 5.1 20.7 4.1 10.2 11.3 7.4 19.4 7.2 11.8 5.1 26.0 12.3 14.3 5.1 15.1 29.3 7.0 13.3 8.2 22.3 7.6 5.1 4.1 4.5 5.1 0.0

0 0 0 0 1 0 0 0 0 1 0* 0 0 0 1* 1 0 0 0 0 0 0 0* 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

3.352 2.285 3.793 2.125 0.973 3.462 1.898 1.947 1.006 0.600 0.869 2.631 1.155 2.107 1.123 0.954 1.605 2.771 1.308 1.744 1.772 4.970 0.993 2.752 3.286 1.480 4.476 1.604 2.521 4.491 6.077 2.634 3.325 2.825 3.286 6.916 1.699 2.898 2.043 5.483 1.919 1.697 1.019 3.038 3.852 1.551

57.1 74.0 52.6 91.5 89.4 91.5 133.6 88.5 97.1 102.1 118.3 104.7 108.7 89.1 75.5 112.8 117.2 92.3 75.4 103.2 104.8 94.9 177.1 92.5 77.0 85.7 58.5 98.5 156.6 66.0 55.8 131.6 97.8 83.2 86.5 73.3 148.8 109.4 117.1 40.7 79.7 90.1 139.8 71.5 54.3 163.5

7248_C018.fm Page 328 Wednesday, May 23, 2007 11:46 AM

328

Hansen Solubility Parameters: A User’s Handbook

TABLE 18.3 (CONTINUED) Calculated Solubility Sphere for Polyethylenesulfide Solvent 1-Octanol Pentane 1-Pentanol 2-Propanol Propionitrile Propylene carbonate Styrene t-Butyl alcohol Tetrahydronaphthalene Xylene

δD

δP

δH

SOLUB

RED

V

17.0 14.5 15.9 15.8 15.3 20.0 18.6 15.2 19.6 17.6

3.3 0.0 4.5 6.1 14.3 18.0 1.0 5.1 2.0 1.0

11.9 0.0 13.9 16.4 5.5 4.1 4.1 14.7 2.9 3.1

0 0 0 0 0 0 1 0 1 1

2.401 1.933 3.005 3.642 2.948 3.655 0.913 3.317 0.996 0.724

157.7 116.2 108.6 76.8 70.9 85.0 115.6 95.8 136.0 123.3

Note: δD = 17.8; δP = 3.8; δH = 2.2; Ro = 4.1; FIT = 0.981; NO = 56.

GROUP CONTRIBUTION METHODS Suggested calculation procedures to arrive at the HSP for solvents are given in Chapter 1. The group contribution methods need expansion with new groups. New group contributions should be checked for reliability of the predictions in some way, which is not always possible within the timeframe of most projects. The group contribution values consistently used by the author are reported in Chapter 1. Values added over the years are appended to the original table which was attributable to Beerbower.4,17,18 Barton has also collected many tables of group contributions for various purposes.2 As stated previously, the group contributions tabulated by van Kevelen15 have not been found reliable. The δD parameter, in particular, is not predicted well. The author chose not to use these at an early date, although many other authors have chosen to do so. The use of various predictive methods which arrive at different results has always been a problem. Koenhen and Smolders19 evaluated various equations for predicting HSP. Methods for reliable a priori calculation of the HSP for polymers are not available. This is a serious shortcoming. The author has tried several times to calculate the HSP for individual polymers using the same group contributions suggested for the liquids, and almost every time has ultimately resorted to experiment. Calculation of the radius of interaction is a particular problem in this respect. This is definitely an area requiring attention. Chapter 2 discusses some of the factors which must be taken into account when calculating the radius of interaction. If one consistently uses the same method of estimating HSP, it can be assumed that some of the inherent errors will not affect relative evaluations. Utracki and coworkers20 estimated HSP for a number of polymers assuming their δD parameters were not different and group contributions for the δP and δH parameters. This is discussed in Chapter 5.

POLYMERS AS POINTS — SOLVENTS AS SPHERES One way to possibly improve predicting the behavior of polymers is to consider them as points (or more accurately, spheres, with very small radii of interaction that depend on molecular weight) rather than as spheres with large radii, as is presently done. A given solvent is assigned a rather large radius of interaction. This radius is larger for smaller molar volume in this inverted approach. This idea was presented many years ago,8,21 but it has never been fully explored. The first indications were that there seemed to be no real benefit in terms of improved reliability of predictions for polymer solubility in organic solvents, which was of primary interest, so there was no need to start

7248_C018.fm Page 329 Wednesday, May 23, 2007 11:46 AM

The Future

329

TABLE 18.4 Calculated Solubility Sphere for Polyethylenesulfide Solvent

δD

δP

Chlorobenzene Xylene p-Chlorotoluene Styrene o-Dichlorobenzene Benzene Di-isobutyl ketone Tetrahydronaphthalene Carbon tetrachloride Methyl n-amyl ketone 1,2-Dichloro ethylene (cis) Cyclohexane Dichloromonofluoromethane 1,4-Dioxane Octane Ethyl acetate 2,2-Dichlorodiethyl ether Methyl ethyl ketone Isoamyl acetate Diethyl amine Diethyl ether sec-Butyl acetate Methyl acetate Pentane Butyraldehyde Isopropyl acetate Cyclopentanone Aniline Acetone 1-Octanol 2-Ethyl hexanol m-Cresol Ethylene glycol monobutyl ether N,N-Dimethyl acetamide Dichlorodifluoromethane (Freon 12) Furfural Isoamyl alcohol Propionitrile 1-Pentanol Nitroethane Dimethyl formamide Furfuryl alcohol t-Butyl alcohol Ethylene glycol monoethyl ether Acetic acid 1-Butanol 2-Propanol Propylene carbonate

19.0 17.6 19.1 18.6 19.2 18.4 16.0 19.6 17.8 16.2 17.0 16.8 15.8 19.0 15.5 15.8 18.8 16.0 15.3 14.9 14.5 15.0 15.5 14.5 14.7 14.9 17.9 19.4 15.5 17.0 15.9 18.0 16.0 16.8 12.3 18.6 15.8 15.3 15.9 16.0 17.4 17.4 15.2 16.2 14.5 16.0 15.8 20.0

4.3 1.0 6.2 1.0 6.3 0.0 3.7 2.0 0.0 5.7 8.0 0.0 3.1 1.8 0.0 5.3 9.0 9.0 3.1 2.3 2.9 3.7 7.2 0.0 5.3 4.5 11.9 5.1 10.4 3.3 3.3 5.1 5.1 11.5 2.0 14.9 5.2 14.3 4.5 15.5 13.7 7.6 5.1 9.2 8.0 5.7 6.1 18.0

δH 2.0 3.1 2.6 4.1 3.3 2.0 4.1 2.9 0.6 4.1 3.2 0.2 5.7 7.4 0.0 7.2 5.7 5.1 7.0 6.1 5.1 7.6 7.6 0.0 7.0 8.2 5.2 10.2 7.0 11.9 11.8 12.9 12.3 10.2 0.0 5.1 13.3 5.5 13.9 4.5 11.3 15.1 14.7 14.3 13.5 15.8 16.4 4.1

SOLUB

RED

1 1 0* 1 1 1 0* 1 0 0 1* 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

0.600 0.724 0.869 0.913 0.954 0.973 0.993 0.996 1.006 1.019 1.123 1.155 1.308 1.480 1.551 1.604 1.605 1.697 1.699 1.744 1.772 1.898 1.919 1.933 1.947 2.043 2.107 2.125 2.285 2.401 2.521 2.631 2.634 2.752 2.771 2.825 2.898 2.948 3.005 3.038 3.286 3.286 3.317 3.325 3.352 3.462 3.642 3.655

V 102.1 123.3 118.3 115.6 112.8 89.4 177.1 136.0 97.1 139.8 75.5 108.7 75.4 85.7 163.5 98.5 117.2 90.1 148.8 103.2 104.8 133.6 79.7 116.2 88.5 117.1 89.1 91.5 74.0 157.7 156.6 104.7 131.6 92.5 92.3 83.2 109.4 70.9 108.6 71.5 77.0 86.5 95.8 97.8 57.1 91.5 76.8 85.0

7248_C018.fm Page 330 Wednesday, May 23, 2007 11:46 AM

330

Hansen Solubility Parameters: A User’s Handbook

TABLE 18.4 (CONTINUED) Calculated Solubility Sphere for Polyethylenesulfide Solvent Acetonitrile Nitromethane Ethanol Ethylene carbonate Diethylene glycol Methanol Ethylene glycol Glycerol

δD

δP

15.3 15.8 15.8 19.4 16.6 15.1 17.0 17.4

18.0 18.8 8.8 21.7 12.0 12.3 11.0 12.1

δH

SOLUB

RED

V

0 0 0 0 0 0 0 0

3.793 3.852 4.476 4.491 4.970 5.483 6.077 6.916

52.6 54.3 58.5 66.0 94.9 40.7 55.8 73.3

6.1 5.1 19.4 5.1 20.7 22.3 26.0 29.3

Note: δD = 17.8; δP = 3.8; δH = 2.2; Ro = 4.1; FIT = 0.981; NO = 56.

all over again with this inverted system. On the other hand, there may be advantages in terms of more reliable prediction of polymer–polymer miscibility, for example. This was not explored. The requirement of polymer miscibility will be that the respective points (very small spheres) for the polymers must be very close to each other; comparing distances between small spheres is relatively easy. This type of comparison is sometimes difficult to make in the present approach where the degree of overlapping of rather large spheres is used to estimate polymer–polymer miscibility. No fixed rules of thumb have been established to estimate how much overlap is required for miscibility. However, guidelines for improving polymer–polymer miscibility are easily found in the present approach. These include selection of an improved solvent, reduction of polymer molecular weight, and modification of a polymer’s HSP in a desired direction based on the HSP group contributions of its repeating unit or comonomers, for example. Traditionally, solvents are considered as points. This is practical and almost necessary from an experimental point of view as most solvents are so miscible as to not allow any experimental characterization in terms of a solubility sphere. An exception to this is the data for water reported in Table 1.3. The HSP reported here are the center points of HSP spheres where the good solvents are either those that are completely miscible or those that are miscible to only 1% or more.

CHARACTERIZING SURFACES The characterization of surfaces with HSP, or perhaps more correctly cohesion parameters (having exactly the same numerical values), is still in its infancy. This possibility was demonstrated many years ago, however.22 As shown in Chapters 6 and 7, this type of approach can lead to a new understanding of surface phenomena, which in turn allows systematic study and design of surfaces for desired behavior. Data on surface characterizations, in addition to that in Chapter 6 and Chapter 7, are not provided here. This is primarily because such data are lacking but also because surface cohesion parameters may not be reflected by nominal bulk composition. The same basic pigment or filler, for example, can have widely different surface cohesion parameters, depending on how it has been surface treated. Neither has the effect of adsorbed water been clarified. Likewise, a surface characteristic for a polyvinyl chloride or a polyethylene cannot be expected to be valid for all polymers normally said to be of these compositions. There may also be additives which have different compositions and which may have migrated to the surfaces. It appears that the relative simplicity of the surface characterizations discussed in this book would lead to their wider use. One current problem is that blindly entering wetting or spontaneous spreading data into the usual computer routine for finding the HSP values often leads to negative

7248_C018.fm Page 331 Wednesday, May 23, 2007 11:46 AM

The Future

331

40

δh/MPa½

30

20

10

0

0

10

δp/MPa½

20

FIGURE 18.1 HSP plot of characterization of Rhodamin FB (C.I. Basic Violet 10) showing potential problems with incorporation of water mixtures as test solvents (see text for discussion). (From Riedel, G., Farbe und Lack, 82(4), 281–287, 1976. With permission.)

7248_C018.fm Page 332 Wednesday, May 23, 2007 11:46 AM

332

Hansen Solubility Parameters: A User’s Handbook

numbers for one or more of them. This was discussed in Chapter 6. Currently, the best approach is to compare plots or even to just compare tabulated data for the test solvents to determine where two surfaces differ in affinities. Guides for action can also be found by simple comparison of the HSP of those solvents which show a difference in behavior. A more systematic approach for the use of cohesion parameters to describe surface phenomena would be desirable.

MATERIALS AND PROCESSES SUGGESTED FOR FURTHER ATTENTION Examples of the use of HSP for many types of materials and phenomena have been presented in earlier chapters. Some special types of materials are singled out here as worthy of still more attention in the near future. These include surface active agents, water, gases, organic and inorganic salts, organometallic materials, and aromatic (fragrances) materials. The uptake of potentially dangerous chemicals in recyclable packaging also needs attention. An additional area of interest may be found in that many commonly used reaction solvents have similar HSP. These include dimethyl sulfoxide, dimethyl formamide, dimethyl acetamide, and sulfolane, for example. It seems unlikely that this is a coincidence. It could be that the solubility of an activated species having high polarity (δP) and moderate hydrogen bonding (δH) is determining the reaction rate(s). Still another area of major interest is the systematic formulation of filled systems using HSP. This is also still in its infancy. Pigments and fillers need to be characterized. Several of these applications are discussed in more detail below. Surface active materials have remained essentially untouched in terms of HSP, although Beerbower started on this many years ago.8,17,23

SURFACE ACTIVE AGENTS Surface active agents have not been systematically characterized by HSP yet, although Beerbower has developed some aspects of a theory for given situations.8,17,23 The statement “like seeks like” indicates that surface active agents should be extensively treated in terms of HSP. Each end of such molecules will require its own set of HSP, as demonstrated by the example of lithium stearate, discussed later in the Organometallic Compounds section (Figure 18.3). An example to help illustrate the type of prediction possible is to try to answer the question of where the hydrophobic end of a given surfactant will tend to preferentially reside. An aliphatic end group would have lower affinity for polystyrene, for example, than an aromatic one. Octane will not dissolve polystyrene, whereas toluene will. This is reflected by their cohesion energy parameters. This same reasoning applies to other polymers. A surfactant with a fluorinated end will not dissolve in many polymers where a hydrocarbon end will. The cohesion energy parameters characteristic of fluorocarbons are too low. Although these examples are obvious to those skilled in the science of surfaces, they point to the possibility of quantifying affinities of surface active materials in terms of the cohesion energy parameters of their respective end groups. Those familiar with cohesion energy parameters can already discern differences that may improve the chances of success. The data in Chapter 11 confirm that differences in HSP are critical if soils are to be removed effectively. HSP for surfactants can be assigned by experiment or by the methods described in the next paragraph. Each surfactant must be assigned three sets of HSP. The first is for the hydrophobic end, the second is for the hydrophilic end, and the third is for the molecule as a whole. Figure 18.3 confirms the need for the first two characterizations, and the third one is required for predictions when the whole molecule is soluble in the system. Even in a completely dissolved condition, one anticipates some degree of orientation of the ends of the surfactant molecule toward regions of similar HSP. The HSP for the hydrophobic end can be estimated by the methods discussed in Chapter 1, with group contributions, or by simple comparison with similar (usually) hydrocarbon molecules of the same size. Barton has collected group contributions that can also be used in connection with the

7248_C018.fm Page 333 Wednesday, May 23, 2007 11:46 AM

The Future

333

different ends of surface active agents.2 The HSP of the hydrophilic end can be estimated by comparison with existing HSP of, for example, sucrose or other water soluble entity, organic salts (see below), inorganic salts (see below), polyethylene oxide or polyethylene glycol, or whatever resembles this end best. Experiments are preferred, of course, and the estimation for given inorganic salts is still uncertain. When a surfactant denatures a protein, there will be some similarity between the HSP of the protein or urea, if this also denatures the protein (see Chapter 15), and the HSP of the hydrophilic end of the surfactant. If enough data of this kind can be generated, the HSP of the surfactant can be experimentally confirmed. The estimation for the molecule as a whole involves combining the two sets of HSP for the ends. It is thought that this is best done by averaging using estimated molecular volumes for the respective ends relative to the molecular volume of the whole molecule. In closing this section, it can also be repeated that thermodynamic surface and interfacial phenomena correlate with HSP. This has been amply shown in Chapter 6 and Chapter 7. The kinetics of the situation may also be important. Chapter 16 discusses adsorption and absorption in polymer surfaces where there is a surface resistance. There will also be some form of interfacial or boundary layer resistance influencing the adsorption and absorption of surfactants into soils, for example. If the HSP do not match sufficiently well, adsorption and absorption will presumably not occur as readily as when the HSP do match to within some required limit for the desired effect, as shown in Chapter 11. The size and shape of the adsorbing/absorbing entity is also presumed to be important from a kinetic point of view, as demonstrated by the examples in Chapter 16.

SURFACE MOBILITY (SELF-ASSEMBLY) The rule of thumb that “like seeks like” can be very useful in understanding the structure of complicated systems. That this type of consideration can lead to useful results can be seen in the way that the behavior of wood polymers and the ultrastructure of cell walls in wood was treated in Chapter 13 and in much more detail by Hansen and Björkman.24 Hemicelluloses appear to function much like surfactants with the backbone and those side chains containing hydroxyl groups favoring placement toward cellulose (or their own kind). Hemicellulose side chains containing acetyl, acid, or ether groups are expected to favor orientation toward lignin regions. In this example, it is interfacial mobility that is in focus, and it can be expected that the orientations may be changed with the transport or presence of other materials such as water through a given local environment. These predictions and inferences appear to agree with what is expected or has been established by independent measurement, but it is too early to say that confirmation has been obtained independently. The treatment of different segments of block copolymers as separate entities is a related endeavor where more quantitative predictions of compatibility should be possible. It is known that additions of a block of polymer C to both polymer A and polymer B improves their chances of compatibility (at some molecular level). The association of blocks of polymers is also the basis for the thermoplastic elastomers (TPE). These are made with a wide variety of different immiscible (hard and soft) blocks where the phase separation is critical to performance. Typical examples include the styrene/butadiene/styrene block copolymer (SBS), the polyether/polyamide block copolymer (PEBA), and polyurethanes combined with polyethers or polyesters (TPU). Some types are also vulcanized to improve properties. The rotation of some hydrophobic materials to become more hydrophilic when in contact with water is still another example of like seeking like. Peat moss is an example. A drop of water initially pearls on the surface but shortly thereafter disappears into the interior in a spontaneous manner. The peat moss has become hydrophilic (but returns to the hydrophobic state on drying again). This phenomena was actually employed to develop an electrodeposition coating for an evaporator where film-wetting by water was required for good evaporation efficiency.25 After several hours of exposure of a fresh coating to water, the static contact angles with water disappeared and a coherent water film was obtained.

7248_C018.fm Page 334 Wednesday, May 23, 2007 11:46 AM

334

Hansen Solubility Parameters: A User’s Handbook

Many surface phenomena can be understood from the preferences of given segments or materials to seek out regions of similar HSP. Some inferences may be possible from the studies performed on compatible (or nearly compatible) polymers. The HSP data leading to formulation of selfstratifying coatings also provides useful information26 (see also Chapter 6). Systematic studies of these effects are badly needed. One such study27 confirmed that the rotation ability (mobility) of aging polymer surfaces could be followed by measuring the (static) receding contact angle of water. Aging can be expected to lead to increased oxygenation and perhaps also to a decrease in average molecular weight. These effects both contribute to the tendency/ability for oxygenated species attached to an otherwise more hydrophobic polymer to rotate into an applied water droplet. When the (static) receding contact angle for water was measured, it fell with exposure time/aging at shorter times, whereas the (static) advancing contact remained constant. At longer exposure times, when the surface was oxygenated to a greater extent, the advancing contact angle also started to fall. Surface mobility also has an important role in biological processes, as described in Chapter 15. The orientation of molecules to allow given segments to locate in regions of similar HSP is presumed to be a general phenomenon. Hydrophilic bonding (usually referred to in the present context as intermolecular hydrogen bonding) is responsible for the configurations of proteins in water. The proteins that can be dissolved in mixtures of water and urea or given salts, for example, are no longer “hydrogen bonded” in the conventional usage of the term, as they are now truly dissolved by an effectively good solvent that can also dissolve these segments/bonding sites. The usual solvent, water, does not have the correct set of HSP to truly dissolve these segments of the protein molecules. The urea additions correct for this deficiency, and the protein is said to be denatured in the process. The concept of hydrophilic (hyperphilic?) bonding, which is the opposite of hydrophobic bonding, is discussed in more detail with examples in Chapter 15. These phenomena also point to the use of urea or urea groups to improve biocompatibility. Many of the concepts discussed here are directly applicable for self-assembling systems and to procedures and products within the concepts of nanotechnology.

WATER The current treatment of the HSP for water discussed in Chapter 1 and Chapter 15 needs confirmation and/or modification. As noted earlier on several occasions, water is very special because of its low molecular volume, its very high δH parameter for a liquid, and its tendency to selfassociate or to associate with other materials forming special structures. The HSP correlations for the solubility of solvents in water presented in Chapter 1, Table 1.3 have not been tested extensively as yet, but do seem promising. They are clearly useful to make predictions for the solubility of untested solvents in water, but whether or not these HSP data for water can be used in a larger context remains to be determined. General behavior can be predicted, but can specific behavior be predicted? More research is needed in this area, but, in the meantime, water can be considered as having (at least) duality. Sometimes it acts like a single molecule, and sometimes it acts as a cluster of about six molecules (according to the HSP comparison, at least). There may also be other possibilities. The use of the HSP for water found from the correlation of total water solubility appears to be the most promising set of values to work with at the present time. This is especially true for water in lower energy systems. It is not yet advisable to include water in a standard set of test liquids for experimental evaluation of the HSP for polymers or other materials because of its tendency to be an outlier. This means a challenge still exists to understand how to be able to incorporate water into a standard set of HSP test liquids without always being concerned about special interpretations for water, and only for water. An example of how this can lead to oddities is discussed in the following. A characterization problem caused by nonideal mixtures with water is the interpretation of HSP correlations for materials such as the dye Rhodamin FB (C.I. Basic Violet 10).2,16 Use of mixtures of solvent and water as test solvents led to a very nonspherical (noncircular) cohesion

7248_C018.fm Page 335 Wednesday, May 23, 2007 11:46 AM

The Future

335

TABLE 18.5 Calculated Solubility Sphere for Rhodamin FB Solvent

δD

δP

δH

Solubility

RED

V

Acetone Benzene 1-Butanol Butyl acetate gamma-Butyrolactone Cyclohexane Diacetone alcohol Diethylene glycol Diethylene glycol monomethyl ether Diethylenetriamine Dimethyl formamide Dimethyl sulfoxide Dipropylene glycol Ethanol Ethylene glycol Ethylene glycol monoethyl ether 2-Ethyl hexanol Glycerol Methanol Methylisoamyl ketone Methyl-2-pyrrolidone 1-Pentanol 1-Propanol Propylene glycol Tetrahydrofuran Toluene Trichloroethylene Water

15.5 18.4 16.0 15.8 19.0 16.8 15.8 16.6 16.2 16.7 17.4 18.4 16.5 15.8 17.0 16.2 15.9 17.4 15.1 16.0 18.0 15.9 16.0 16.8 16.8 18.0 18.0 15.5

10.4 0.0 5.7 3.7 16.6 0.0 8.2 12.0 7.8 13.3 13.7 16.4 10.6 8.8 11.0 9.2 3.3 12.1 12.3 5.7 12.3 4.5 6.8 9.4 5.7 1.4 3.1 16.0

7.0 2.0 15.8 6.3 7.4 0.2 10.8 20.7 12.6 14.3 11.3 10.2 17.7 19.4 26.0 16.4 11.8 29.3 22.3 4.1 7.2 13.9 17.4 23.3 8.0 2.0 5.3 42.3

0 0 0* 0 0* 0 1* 1 1 1 1 1 1 1 1 1 0 1 1 0 1* 0 0* 1 0 0 0 0

1.125 1.991 0.999 1.517 0.988 2.076 1.001 0.486 0.934 0.487 0.677 0.741 0.570 0.732 0.815 0.707 1.294 0.996 0.589 1.530 1.042 1.138 0.889 0.772 1.295 1.902 1.615 1.965

74.0 89.4 91.5 132.5 76.8 108.7 124.2 94.9 118.0 108.0 77.0 71.3 130.9 58.5 55.8 79.1 156.6 73.3 40.7 142.8 96.5 108.6 75.2 73.6 81.7 106.8 90.2 18.0

Note: δD = 16.7; δP = 17.5; δH = 18.5; Ro = 12.2; FIT = 0.930; NO = 28. Source: Riedel, G., Farbe and Lack, 82(4), 281–287, 1976; Birdi, K.-S., Ed., Handbook of Surface and Colloid Chemistry, CRC Press, Boca Raton, FL, 1997, chap. 10.

energy parameter plot (see Figure 18.1). The irregular plot can presumably still be used as such, but the characterization of the dye in question is not useful in relation to prediction of interactions with other materials. The plot has given several individuals the impression that there are significant problems with the HSP approach when it is applied to this kind of material. This is not true. A computer analysis based on the pure solvent data given by Riedel16 confirms that a good “spherical” characterization of Rhodamin FB is possible using the same data otherwise used in Figure 18.1.5 The HSP data for this correlation are given in Table 18.5. The data fit was 0.93 for 28 data points. Figure 18.1 clearly shows that this HSP sphere covers more space than the data, with a significant portion in the high energy region where there are no liquids. Chapter 1, Equation 1.9 (with the constant 4) was used in this correlation, as it has been used in all the other HSP correlations in this book. The HSP correlations for water-soluble polymers and other high energy materials involved similar extrapolations into domains where there are no liquids. This procedure may be subject to revision at some future point in time, but for the present it seems to be the only procedure possible to maintain consistency in the HSP procedures developed. It should be remembered that many

7248_C018.fm Page 336 Wednesday, May 23, 2007 11:46 AM

336

Hansen Solubility Parameters: A User’s Handbook

25

ENERGY OF VAPORIZATION FOR “NON POLAR” LIQUIDS BUTANE C4H10 (.642) ●

20

●

ISOBUTANE (.639)

TR =0.5

ΔET KJ ∕ MOLE

TR =0.6

15 Xe (.569) ✳

N

E BL O

G

ES AS

● C3H8

(.624)

TR =0.7 ●C

2H6 (.6025)

10 Kr (.578) ✳ O2 (.583) Ar (.577) ✳

5

Ne (.607) ✳

10

●

CH4 (.586)

N2 (.613)

H2 (.613) He (.803)

●

0 0

●

20

✳ 30

40 50 60 MOLAR VOLUME CC/MOLE

70

80

90

100

FIGURE 18.2 Cohesion energy for various low molecular weight materials as a function of molecular volume and reduced temperature (given by curves or in parentheses). (See text for discussion.)

(most) liquids with high HSP (water, methanol, glycols) also have low molecular volumes (V). This makes them “better” solvents than expected by comparison with all the other solvents (whose average V is closer to 100 cc/mol). This fact might give the impression that the constant 4 in Chapter 1, Equation 1.9 should be increased. This is discussed more in Chapter 2. A unifying concept and procedure for the use of water in all testing is needed. The HSP considerations discussed in this book provide help toward reaching this goal.

GASES HSP can also be used to improve understanding of the solubility behavior of gases. Solubility parameters are usually derived from data at the normal boiling points. HSP derived from these numbers seem to be in good qualitative agreement with expectations (even at 25°C), and in many cases quantitative agreement with physical behavior has also been found. Some examples are given by Barton.2 Solubility parameter correlations for oxygen28 (Chapter 13) and nitrogen 29 (Chapter 13) have been used as examples in this book. The δP and δH parameters for these two gases are zero. HSP for many gases where this is not true are reported in Chapter 13, Table 13.4. A specific example where this is not the case is carbon dioxide. Carbon dioxide is extensively discussed in Chapter 10 where the HSP are calculated for large variations in pressure and temperature. This same procedure is applicable to other gases. Chapter 3 also treats a method to calculate the three partial solubility parameters for gases. In the process of calculating the HSP for gases, it was found necessary to extrapolate the data in Chapter 1, Figure 1.1 to lower molar volumes. Figure 18.2 is derived from this. This figure is worthy of some consideration from a theoretical point of view. The basis of the HSP is a corresponding states calculation for ED as the energy of vaporization of a corresponding hydrocarbon solvent (same V and structure) at the same reduced temperature. This is, of course, 298.15 K divided

7248_C018.fm Page 337 Wednesday, May 23, 2007 11:46 AM

The Future

337

by the solvent’s critical temperature. The reduced temperature at the boiling point is indicated in parentheses in Figure 18.2. Questions can be raised as to why the noble gases differ from the hydrocarbon solvents and whether the hydrocarbon solvents were the best choice as reference materials. Also, why is oxygen among the noble gases in cohesion behavior rather than rated with the other gases? At the time of choice for a reference for the dispersion bonding energies, there were ample data on latent heats for the hydrocarbons and the aliphatic hydrocarbons were considered as having δP and δH values equal to zero. This may not quite be true, but the corrections would be minor, and the necessary data for a revised reference are lacking. It appears that the currently chartered course of using hydrocarbon solvents as a basis will be maintained. Some additional considerations may be found in further study of the relation of HSP to the corresponding states theory of Prigogine and coworkers as discussed in Chapter 2. The behavior of the hydrocarbon solvents appears to be included within the Prigogine parameter for differences in size (ρ).

ORGANIC SALTS The HSP of several organic salts have been compared with the HSP for the organic acids and organic bases from which they were made.30 The result was that the organic salts always had considerably higher HSP than either of the components making them up. As examples, the salts made from formic acid and acetic acid combined individually with dimethyl ethanolamine had δD;δP;δH equal to 17.2;21.5;22.5 and 16.8;19.8;19.8, respectively, whereas the δD;δP;δH are 14.3;11.9;16.6 for formic acid, 14.5;8.0;13.5 for acetic acid, and 16.1;9.2;15.3 for dimethyl ethanolamine. All of these values are in MPa1/2. This general relationship was also found for other salts formed by combinations of organic acids with a variety of amines. These values are reported in Appendix Table A.1 in dark type, as there are experimental data to justify the numbers. The HSP for the salts are generally close to those mentioned earlier. These values are high enough to make the salt entities insoluble in most polymers. Their affinities for water will be very high, however, both because of high HSP and also because of the charges associated with the salt groups. There was about 10% shrinkage in volume compared with the original volumes of the acids and bases. In some cases, the cohesion energy of the salts is high enough to make them solids rather than liquids. This study showed that organic salts can indeed be characterized by HSP. More work is necessary, however, with other types of salts. In particular, the acid groups found in nature, such as in hemicelluloses, deserve more attention (see Chapter 15 and Reference 24).

INORGANIC SALTS The solubility of magnesium nitrate [Mg(NO3)2·6H2O] was evaluated in a standard set of solvents1 and later correlated more precisely with HSP. The HSP derived from this are δD;δP;δH, and Ro equal to 19.5;22.1;21.9, and 13.2, respectively, all in MPa1/2. Nitrates are known to be among the most soluble of salts. Somewhat less soluble than the nitrates are chlorides. These are only partly soluble in a few organic liquids with very high HSP. Group contributions to the HSP from the nitrate group are expected to result in lower HSP, and, in particular, lower δD for the nitrate portion of a salt than would be expected from the group contributions from a chloride. This would lead to greater solubility of the nitrates in organic solvents, which is indeed the case. The δD parameter seems to be qualitatively capable of describing the behavior of metals to some extent. It may be possible to arrive at an approximate description of inorganic salt solubility in organic media (perhaps water, too) using HSP or some modification/extension thereof. The salting in and salting out of various polymers can perhaps provide clues to assign HSP in this connection. Finally, it should be noted that an excellent HSP correlation of the chemical resistance of an inorganic zinc silicate coating is reported in Chapter 12, Table 12.1.

7248_C018.fm Page 338 Wednesday, May 23, 2007 11:46 AM

338

Hansen Solubility Parameters: A User’s Handbook

11 cc ABSORBED ? > .8 δH δP

10 ?

9

2

8

1 3.1 1 3.7

?

2.7 8.7

7

δH

6

HANSEN PLOT - LITHIUM STEARATE

5 4 ??

3

?

?

?

?

?

1

2 ?

1 1

2

3

4

5 δP

6

7

8

9

FIGURE 18.3 HSP plot for solvent uptake by lithium stearate.31,32 Units for δP and δH are (cal/cm3)1/2. (From Alan Beerbower, personal communication.)

ORGANOMETALLIC COMPOUNDS No systematic studies of the HSP of organometallic compounds have been made. An exception is perhaps that shown in Figure 18.3 where Beerbower31 used data from Panzer32 to show that lithium stearate does indeed have two distinct regions of solvent uptake and that a HSP plot can show why. This example shows that one can calculate HSP values where the relevant data can be found in the literature and then test these with relevant experiments. Group contributions would be valuable. Metallic bonds differ in nature from those usually discussed in connection with organic compounds. A suspicion is that, at least in practice, the cohesion energy derived from the “metallic” bonding in organometallic compounds can be coupled with the dispersion parameter. There is also a question, for example, of whether metal atoms in the center of more complicated molecules are effectively shielded from any (surface) contact with a solvent. Surface contacts are clearly important, but it appears that the nature of the central atom also has an effect. Finally, it might be noted that Hildebrand and Scott presented a chapter on the solubility parameters of metals.33 Unfortunately, we do not often deal with pure metals in this context, but rather metal oxides, for which no HSP work has been reported, as least not to the author’s knowledge.

AROMAS

AND

FRAGRANCES

Aromas and fragrances are important in connection with packaging materials, foodstuffs, cosmetics, chewing gum, etc. A recent report34 discussed HSP in connection with fragrances and aromas. It is clear that HSP exist for these materials, but very little work has been published in the area. One of the examples included in Reference 34 was the development of an artificial nose based on coated oscillating sensors, which oscillate more slowly when they gain weight. Matching HSP for the coating and material to be detected leads to increased weight gain and increased sensitivity. Other

7248_C018.fm Page 339 Wednesday, May 23, 2007 11:46 AM

The Future

339

examples where HSP could be systematically used include: counteracting undesirable odors using fragrances that have reasonably similar HSP; absorption of odors into plastics, coatings, sealants, etc.; development of packaging with designed HSP to either function as a barrier or as a sink; and an estimate of where a given aromatic material is likely to reside. The key to interpretation is, as usual, that similarity of HSP means higher affinity. It is thought that the masking effect is the result of adsorption of the fragrance on the regions of the sensory system where HSP are similar to those of the undesired odor. There may not be a complete match regarding steric adsorption, but the sensitive areas are covered anyway, and this barrier prevents the odor from arriving at the given sites.

ABSORPTION

OF

CHEMICALS

IN

PLASTICS

HSP correlations exist for chemical resistance, permeation phenomena, and uptake of solvents in many polymers. The recycling of polymeric containers has a potential problem in that the polymers used are able to readily absorb those chemicals whose HSP are not too different from their own. Once a chemical has absorbed into a polymer, and particularly if it is a rigid polymer with a relatively high glass transition temperature, it can be very difficult to get it out again. A relatively slow diffusion process is required to do this. See Chapter 16. It is suggested that an extensive HSP analysis be done for those polymers where potential misuse or contamination of containers prior to recycling is a possibility. This can point out which chemicals are most likely to present the greatest problems.

CHEMICAL RESISTANCE Chemical resistance studies have generally been performed with too few liquids and without the necessary spread of HSP to allow the data to be correlated with confidence. In addition, attainment of equilibrium is not usually confirmed. These shortcomings mean that HSP correlations of chemical resistance must be done with great care. This has been discussed in Chapter 12 in more detail. An additional activity, which should be done for practical reasons, is to assign effective HSP to various test materials such as mustard, ketchup, and other given products that often appear in tests of chemical resistance. Such data will allow greater use of the correlations since guidelines for potential improvements can be obtained.

CONTROLLED RELEASE HSP considerations can provide an extra formulating parameter for the controlled release of drugs. When the HSP relations between a drug and its surroundings are known, predictions of its behavior can be made. When there is a good match in the HSP values, the drug will be more soluble with the ability to move at some rate within a polymer matrix. On the other hand it can be surrounded by a matrix with similar HSP, and this may slow release more than desired. A poor match in HSP may leave holes and expose the drug for more rapid release. When the match is poor, the drug will not be able to permeate through its polymeric surroundings, but it can, of course, pass through open passageways. Drugs will also tend to adsorb at surfaces where there is a good HSP match. It has been shown (personal communication, Andreas Gryczke, Degussa Pharma Polymers, Darmstadt) that calculations of HSP for a number of drugs and for a series of EUDRAGIT® polymers provided a correlation confirming that the drugs are soluble in those polymers where the HSP match closely enough. When the drug and polymer HSP were within 10 MPa1/2 at the concentration studied (20% wgt.), what appeared to be true solubility was found. This was evidenced by x-ray diffraction and DSC measurement that confirmed the drugs were embedded completely amorphous (solid solution). A closer match would presumably be required to do this for higher concentrations of the drugs. Table 18.6 contains HSP for some materials whose solubility in various media may have interest. It is possible to calculate HSP for essentially all drugs, although hydrochlorides and other salts

7248_C018.fm Page 340 Wednesday, May 23, 2007 11:46 AM

340

Hansen Solubility Parameters: A User’s Handbook

TABLE 18.6 Calculated HSP for Various Materials Material

δD

δP

δH

V

Adrenaline 4-Aminopyridine Ascorbic Acid Caffeine Cycloheximide DDT Dopamine Ecstacy Meclofenoxate (base only) Norephedrin 2-Oxopyrrolidinacetamid Quinine Saccharin Serotonin Spermidin

20.5 20.4 18.0 19.5 18.3 20.0 18.2 18.0 16.0 18.0 17.5 19.0 21.0 18.0 16.7

8.7 16.1 12.6 10.1 11.0 5.5 10.3 5.1 6.2 10.7 15.6 4.6 13.9 8.2 11.2

19.9 12.9 27.6 13.0 13.8 3.1 19.5 6.1 9.0 24.1 11.2 11.0 8.8 14.4 12.0

154.5 87.1 106.7 157.9 171.0 268.8 180.0 202.9 198.3 141.9 116.2 310.7 206.8 144.4 155.6

Note: Units are MPa1/2. Source: Hansen, C.M., Conference Proceedings, Pharmaceutical and Medical Packaging 2001, Skov, H. R., Ed., Hexagon Holding, Copenhagen, 2001, pp. 20.1–20.10.

should be measured rather than calculated only, as there is very little experimental data on which to base an estimate with group contributions. There are correlations of HSP with skin permeation presented in Chapter 15 that should allow an estimate of which drugs can enter via this route. Simple calculations confirm that dopamine, nicotine, skatole, and nitroglycerine, for example, have high affinity for skin.

NANOTECHNOLOGY Controlling the orientation of molecules can be a key for switches and the like. An example has been given in Chapter 15 where anthracene units appended to a polymer molecule adopted one orientation in toluene at room temperature, but changed to a new orientation at temperatures above 38.5° in toluene. The configuration at the higher temperatures was also that adopted in tetrahydrofurane. This change of orientation is the result of a poorer solvent at the higher temperatures compared with a good one at room temperature. More details and additional examples are found in Chapter 15. The following link reports what follows for studies of organically modified nanoclays (organoclay) in nanocomposites: http://www.hwi.buffalo.edu/ACA/ACA03/abstracts/text/W0383.html Quote: W0383 Evaluating Organoclays for Nanocomposites by Small Angle Scattering. Derek L. Ho1,2 and Charles J. Glinka1, 1Center for Neutron Research, National Institute of Standards and Technology,

7248_C018.fm Page 341 Wednesday, May 23, 2007 11:46 AM

The Future

341

Gaithersburg, MD 20899, 2Dept. of Materials and Nuclear Engineering, Univ. of Maryland, College Park, MD 20742. Understanding the interaction between organically modified clay (organoclay) platelets and organic solvent molecules as well as the corresponding structure of organoclays in a suspension is a critical step toward tailoring and characterizing nanocomposites formed by organoclays in a polymer matrix. Recently, nanocomposites composed of clays and polymers have been found to have improved mechanical properties as well as enhanced thermal stability. The improved properties are related to the degree of dispersal and exfoliation of the clay platelets in the polymer matrix. In order to understand and optimize potential processing conditions, organoclays were dispersed in a number of organic solvents covering a range of solubility parameters and characterized using small-angle neutron scattering and wide-angle x-ray scattering techniques. With Hansen’s solubility parameters, δo2 = δd2 + δp2 + δh2, the correlation between the degree of exfoliation of organoclays and the solvent in which the clay platelets are dispersed/mixed has been analyzed. It has been found that the dispersion force of the solvent, reflected by δd, is the principal factor determining whether the clay platelets remain suspended in the solvent while the polar (δp) and hydrogen-bonding (δh) forces affect primarily the tactoid formation/structure of the suspended platelets. The organically modified clays studied in this work precipitated in any solvent with molecules with moderately strong hydrogen-bonding groups. The correlation found has been used to correctly identify a solvent, trichloroethylene, which completely exfoliates the organoclays studied in this work.

These and the many other examples in this handbook related to self-assembly (see above), surface adsorption, molecular orientation, and affinity among molecules and molecular segments should provide ample evidence that molecular guidance in nanotechnology endeavors can be found in the HSP concept.

THEORETICAL PROBLEMS AWAITING FUTURE RESOLUTION POLYMER SOLUBILITY The Flory chi parameter has been used to describe polymer–solvent interactions for many years.35,36 If this single parameter is to be complete in this function, it must include both the atomic/dispersion interactions as well as the specific interactions reflected by the δP and δH parameters. Attempts to calculate chi using HSP are reported in Chapter 2. More understanding is required before chi can be calculated with reasonable accuracy, but intensified efforts seem warranted. Zellers and coworkers have recently made an attempt to use this theory in conjunction with HSP studies.37–40 A major problem is the reliability of the chi parameter (and also HSP) values in the literature (see Chapter 2 for more details). Chapter 4 is a lengthy discussion of the use of HSP in thermodynamic models for polymer solutions. The author’s current view as expressed in Chapter 2 is that the Flory approach is a special case of the more general Prigogine corresponding states theory. This is in agreement with the view of Patterson discussed in Chapter 2. Furthermore, the very general applications of the HSP approach demonstrated in this book and elsewhere, and the apparent agreement in the treatment of specific interactions by both the HSP and Prigogine treatments, leads to confidence in the HSP approach. The geometric mean must be used in the Prigogine theory to arrive at this similarity of treatment. The statistical thermodynamic approach of Panayiotou and coworkers thoroughly discussed in Chapter 3 provides convincing evidence that the division of the cohesion energy density into three parts (at least) is the correct procedure to understand affinities among many types of materials. Finally, Chapter 4 deals with theories of vapor–liquid equilibrium in particular and how these relate to HSP.

7248_C018.fm Page 342 Wednesday, May 23, 2007 11:46 AM

342

Hansen Solubility Parameters: A User’s Handbook

SURFACE PHENOMENA Surfaces can clearly be characterized by HSP as demonstrated in Chapter 6 and Chapter 7. The work of Beerbower contained in Reference 17 and Reference 23 has also shown applications of HSP to such varied phenomena as the work of adhesion of liquids on mercury, friction of polyethylene untreated and treated with sulfuric acid, the Rehbinder effect — the crushing of aluminum oxide (Al2O3) under various liquids, and the Joffé effect — effect of liquid immersion on the fracture strength of soda-lime glass. Here again, the successful use of HSP to such applications might not have been anticipated had it been considered as a parameter for use in bulk systems only. A formalized unifying theory linking HSP to both bulk and surface phenomena is still lacking. Presently, the best that can be said is that the generality “like dissolves like” can be quantified in many cases. The extension of this, “like seeks like,” also seems to have been demonstrated. It is the surfaces of molecules which interact with each other (also in bulk and solution phenomena), so it is not surprising that cohesion parameters can be applied with success to both solubility and surface phenomena. Much more research needs to be done with these relations. A good starting point is the Handbook of Surface and Colloid Chemistry, edited by K.S. Birdi.41 If we consider chromatographic techniques as depending primarily on surface phenomena, mention should also be made of the extension of the three-parameter HSP approach to a five-parameter approach by Karger and coworkers.42 HSP characterizations of surfactants are also badly needed.

CONCLUSION HSP have been shown useful in solvent selection; predicting polymer–polymer miscibility; characterizing the surfaces of polymers, fillers, and fibers; correlating permeation phenomena; characterizing organic salts and inorganic salts; gas solubility; etc. No other parameter can be assigned to such a range of materials spanning from gases and liquids, over surfaces, to inorganic salts. These results and the close relation with the Prigogine corresponding states theory of polymer solutions, and more recently to the work of Panayiotou summarized in Chapter 3, indicate that a still more general theory exists. This theory should quantify the adage “like seeks like,” i.e., include surface phenomena as well as bulk phenomena. Specific areas needing more theoretical work related to HSP in the near future include better understanding of usage for predictions of behavior for water, gases, organic salts, inorganic salts, and organometallic compounds. Water remains special because of its low molecular volume and high δH. Most materials having HSP in the range of the customary test liquids can be studied using HSP with reasonable success. This is not fully the case for gases, many of which have much lower HSP than the well-studied liquids, and salts, many of which apparently have HSP very much higher than any of the liquids. Extensions of practical applications related to chemical resistance and the uptake of potentially dangerous materials in polymers are also required. Finally there is a great deal to be done in the areas of controlled drug release, improved understanding of some biological processes, and last but not least, the systematic use of HSP in nanotechnology to control the assembly and orientation of molecules or segments of molecules.

REFERENCES 1. Hansen, C.M., The universality of the solubility parameter, Ind. Eng. Chem. Prod. Res. Dev., 8(1), 2–11, 1969. 2. Barton, A.F.M., Handbook of Solubility Parameters and Other Cohesion Parameters, CRC Press, Boca Raton, FL, 1983; 2nd ed., 1991. 3. Barton, A.F.M., Handbook of Polymer-Liquid Interaction Parameters and Solubility Parameters, CRC Press, Boca Raton, FL, 1990.

7248_C018.fm Page 343 Wednesday, May 23, 2007 11:46 AM

The Future

343

4. Hansen, C.M., Solubility parameters, in Paint Testing Manual, Manual 17, Koleske, J.V., Ed., American Society for Testing and Materials, Philadelphia, PA, 1995, pp. 383–404. 5. Hansen, C.M., Cohesion energy parameters applied to surface phenomena, CRC Handbook on Surface Science, Birdi, K.S., Ed., CRC Press, Boca Raton, FL, 1997, chap. 10. 6. Anonymous, Brochure: Co-Act — A Dynamic Program for Solvent Selection, Exxon Chemical International Inc., 1989. 7. Dante, M.F., Bittar, A.D., and Caillault, J.J., Program calculates solvent properties and solubility parameters, Mod. Paint Coat., 79(9), 46–51, 1989. 8. Hansen, C.M. and Beerbower, A., Solubility parameters, in Kirk-Othmer Encyclopedia of Chemical Technology, Suppl. Vol., 2nd ed., Standen, A., Ed., Interscience, New York, 1971, pp. 889–910. 9. Hansen, C.M. and Andersen, B.H., The affinities of organic solvents in biological systems, J. Am. Ind. Hyg. Assoc., 49(6), 301–308, 1988. 10. Wessling, R.A., The solubility of poly(vinylidene chloride), J. Appl. Polym. Sci., 14, 1531–1545, 1970. 11. Grulke, E.A., Solubility parameter values, in Polymer Handbook, 3rd ed., Brandrup, J. and Immergut, E.H., Eds., John Wiley-Interscience, New York, 1989, pp. VII/519–559. 12. Burrell, H., Solubility of polymers, in Kirk Othmer Encycopedia of Polymer Science and Technology, Vol. 12, 2nd ed., John Wiley & Sons, New York, 1970, pp. 618–626. 13. Hoy, K.L., New values of the solubility parameters from vapor pressure data, J. Paint Technol., 42(541), 76–118, 1970. 14. Hoy, K.L., Tables of Solubility Parameters, Union Carbide Corp., Research and Development Dept., South Charleston, WV, 1985; 1st ed. 1969. 15. van Krevelen, D.W. and Hoftyzer, P.J., Properties of Polymers: Their Estimation and Correlation with Chemical Structure, 2nd ed., Elsevier, Amsterdam, 1976. 16. Riedel, G., The solubility of colorants in organic solvents (Löslichkeit von Farbstoffen in organischen Lösungsmitteln, in German), Farbe und Lack, 82(4), 281–287, 1976. 17. Beerbower, A., Environmental capability of liquids, in Interdisciplinary Approach to Liquid Lubricant Technology, NASA Publication SP-318, 1973, 365–431. 18. Fedors, R.F., A method for estimating both the solubility parameters and molar volumes of liquids, Polym. Eng. Sci., 14(2), 147–154, 472, 1974. 19. Koenhen, D.N. and Smolders, C.A., The determination of solubility parameters of solvents and polymers by means of correlation with other physical quantities, J. Appl. Polym. Sci., 19, 1163–1179, 1975. 20. Utracki, L., Personal communication, 1995. 21. Kähler, T. and Knudsen, S.L., Student Report, Technical University of Denmark, 1967. 22. Hansen, C.M., Characterization of surfaces by spreading liquids, J. Paint Technol., 42(550), 660–664, 1970. 23. Beerbower, A., Boundary Lubrication — Scientific and Technical Applications Forecast, AD747336, Office of the Chief of Research and Development, Department of the Army, Washington, D.C., 1972. 24. Hansen, C.M. and Björkman, A., The ultrastructure of wood from a solubility parameter point of view, Holzforschung, 52(4), 335–344, 1998. 25. Hansen, C.M. and Knudtson, J., Spreading and wetting of an electrodeposition coating (Ausbreitung und Benetzung bei einem Elektrotauchlack, in German), Farbe und Lack, 2, 115–118, 1973. 26. Progress in Organic Coatings, Special issue devoted to Self-Stratifying Coatings, 28(3), July 1996. 27. Thulstrup, D., Characterization of surfaces by static contact angle measurements (lecture in Danish: Karakterisering af Overflader ved Statisk Kontaktvinkelmåling), Surface Properties and Modification of Plastics (Conference Notes), Ingeniørforening i Danmark, November 13, 1997, Copenhagen. See also DSM Materialenyt 2:97, Dansk Selskab for Materialeproevning og -forskning, Ingeniørforening in Danmark, Copenhagen, 1997, 63–69. 28. König, U. and Schuch, H., Molecular composition and permeability of plastics, (Konstitution und Permeabilität von Kunststoffen, in German), Kunststoffe, 67(1), 27–31, 1977. 29. Hansen, C.M., 25 years with solubility parameters (25 År med Opløselighedsparametrene, in Danish), Dan. Kemi, 73(8), 18–22, 1992. 30. Hansen, C.M., Some aspects of acid/base interactions (Einige Aspekte der Säure/Base-Wechselwirkung, in German), Farbe und Lack, 83(7), 595–598, 1977.

7248_C018.fm Page 344 Wednesday, May 23, 2007 11:46 AM

344

Hansen Solubility Parameters: A User’s Handbook

31. Beerbower, A., Personal communication, Predicting Wetting in the Oil-Brine-Rock System, Unedited manuscript for presentation at the National AIChE Meeting, Tulsa, OK, March 10–13, 1974. 32. Panzer, J., Components of solid surface free energy from wetting measurements, J. Colloid Interface Sci., 44(1), 142–161, July 1973. 33. Hildebrand, J. and Scott, R.L., The Solubility of Nonelectrolytes, 3rd ed., Reinhold, New York, 1950. 34. Hansen, C.M., Solubility parameters for aromas and scents (Aromastoffers Opløselighedsparametre, in Danish), Plus Process, 11(9), 16–17, 1997. 35. Flory, P.J., Principles of Polymer Chemistry, Cornell University Press, New York, 1953. 36. Eichinger, B.E. and Flory, P.J., Thermodynamics of polymer solutions, Trans. Faraday Soc., 64(1), 2035–2052, 1968; 64(2), 2053–2060, 1968; 64(3), 2061–2065, 1968; 64(4), 2066–2072, 1968. 37. Zellers, E.T., Three-dimensional solubility parameters and chemical protective clothing permeation. I. Modelling the solubility of organic solvents in Viton® gloves, J. Appl. Polym. Sci., 50, 513–530, 1993. 38. Zellers, E.T., Three-dimensional solubility parameters and chemical protective clothing. II. Modelling diffusion coefficients, breakthrough times, and steady-state permeation rates of organic solvents in Viton® gloves, J. Appl. Polym. Sci., 50, 531–540, 1993. 39. Zellers, E.T., Anna, D.H., Sulewski, R., and Wei, X., Critical analysis of the graphical determination of Hansen’s solubility parameters for lightly crosslinked polymers, J. Appl. Polym. Sci., 62, 2069–2080, 1996. 40. Zellers, E.T., Anna, D.H., Sulewski, R., and Wei, X., Improved methods for the determination of Hansen’s solubility parameters and the estimation of solvent uptake for lightly crosslinked polymers, J. Appl. Polym. Sci., 62, 2081–2096, 1996. 41. Birdi, K.S., Handbook of Surface and Colloid Chemistry, CRC Press, Boca Raton, FL, 1997. 42. Karger, B.L., Snyder, L.R., and Eon, C., Expanded solubility parameter treatment for classification and use of chromatographic solvents and adsorbents, Anal. Chem., 50(14), 2126–2136, 1978.

7248_A001.fm Page 345 Wednesday, May 23, 2007 12:31 PM

Appendix A: Table A.1 Charles M. Hansen (with the help of Hanno Priebe) COMMENTS TO TABLE A.1 Table A-1 is greatly expanded compared with the first edition of this handbook. There are more chemicals, and the information for each of them is much more comprehensive. There are two lists for chemical names. The first list follows the general nomenclature that is used in the first edition. This list follows usage that is commonly found in industrial contexts. The second list of names and also the structures have been compiled by Dr. Hanno Priebe, GE Healthcare. Most of these chemical names originate from AutoNom 2000 (Copyright © 1988-2006, Beilstein-Institut zur Förderung der Chemischen Wissenschaften licensed to Beilstein GmbH and MDL Information Systems GmbH. All rights reserved). These are obviously more correct in many cases. In Autonom the Beilstein ring system naming convention was chosen. In cases where Autonom failed, the chemical names originate from ACD/Name freeware, version 8.0 (Advanced Chemistry Development, Inc., Toronto ON, Canada, www.acdlabs.com, 2005). In cases where the ACD/Name freeware failed, the name field was left empty. Most of the HSP entries in Table A-1 are calculated values. Those entries having some experimental confirmation are the original 90 from Reference 1 to Reference 3, together with some from industrial experience at PPG Industries. These are indicated in dark type. The usual procedure for the calculations was to use the procedure given in Chapter 1. The figures were used to find δD from boiling point data. δP was estimated either with a dipole moment, where this could be found, or by group contributions. There are special problems where the dipole moment is zero or close to zero for symmetrical molecules. In some cases two sets of data are given. δH was almost always found by group contributions or by similarity to related compounds. It must be said that none of the parameters are precise. Even use of experimental data does not necessarily confirm the accuracy of the values. This is particularly true where a given HSP parameter is relatively high. Whether or not given solutes dissolve is not sufficient in most cases to narrow the potential error(s) in such cases. Where latent heats are known, Equations 1.6–1.8 do limit potential errors, and experience has shown that for all practical purposes, the values assigned are most useful. Notwithstanding, several parameters have been changed since the first edition appeared. These are indicated with a “*” in the first column of names. However, no changes in the older, published figures have been made for this reason, as this was judged unnecessary. Many of the solvents for which experimental data were used to help assign the HSP were commercial in origin. This means that purity of the samples was that specified in the given case, but was usually very high, and that a mixture of isomers was present if isomers were possible. This is particularly true of the glycol ethers based on propylene oxide. For these chemicals and others where isomers are possible, the chemical names and structures indicated in Table A-1 are indicative of the type of compound present. It would be impractical to present all the possible isomers. The HSP for the organic salts formed from equimolar mixtures of organic acids and amines are those reported in Reference 4. The assignments were made on the basis of which polymers or other solutes in a test series dissolved in the given salt. This was compared with how these same solutes behaved in a series of liquids with known HSP. When the same pattern of dissolving or not was found for both a salt and a known liquid, it was assumed that their HSP were the same (or at least very close). The units used are MPa1/2 for the HSP and cc/mole for the molar volume.

345

7248_A001.fm Page 346 Wednesday, May 23, 2007 12:31 PM

346

Hansen Solubility Parameters: A User’s Handbook

REFERENCES 1. Hansen, C.M., The Three Dimensional Solubility Parameter and Solvent Diffusion Coefficient, Doctoral Dissertation, Danish Technical Press, Copenhagen, 1967. 2. Hansen, C.M., The Three Dimensional Solubility Parameter — Key to Paint Component Affinities I. — Solvents, Plasticizers, Polymers, and Resins, J. Paint Techn., 39(505), 104-117, 1967. 3. Hansen, C.M., and Skaarup, K., The Three Dimensional Solubility Parameter — Key to Paint Component Affinities III. — Independent Calculation of the Parameter Components, J. Paint Techn. 39(511), 511-514, 1967. 4. Hansen, C.M., Some Aspects of Acid/Base Interactions (in German) (Einige Aspekte der Säure/BaseWechselwirkung), Farbe und Lack, 83(7), 595-598, 1977.

7248_A001.fm Page 347 Wednesday, May 23, 2007 12:31 PM

Appendix A: Table A.1

347

TABLE A.1 No. 1

Solvent Name Acetaldehyde*

Autonom/ ACD Name

Dispersion Polarity

Hydrogen Molar Bonding Volume

Acetaldehyde

14.7

12.5

7.9

56.6

Acetaldehyde oxime

16.3

4.0

20.2

61.2

Acetamide

17.3

18.7

22.4

60.8

N-Phenyl-acetamide

20.6

13.3

12.4

110.9

Acetic acid

14.5

8.0

13.5

57.1

Acetic anhydride

16.0

11.7

10.2

94.5

Propan-2-one

15.5

10.4

7.0

74.0

2-Hydroxy-2-methylpropionitrile

16.6

12.2

15.5

94.0

O H3C 2

Acetaldoxime

H3C N 3

OH

Acetamide

O H3C NH2 4

Acetanilide

O CH3

N 5

Acetic Acid

O H3C OH 6

Acetic Anhydride

O

O

O

H3C 7

Acetone

CH3

O H3C 8

CH3

Acetonecyanhydrin N HO H3C

CH3

7248_A001.fm Page 348 Wednesday, May 23, 2007 12:31 PM

348

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No. 9

Autonom/ ACD Name

Solvent Name Acetonemethyloxime

Propan-2-one O-methyl- 14.7 oxime

CH3 N

4.6

4.6

96.7

O

Acetonitrile

Acetonitrile

15.3

18.0

6.1

52.6

1-Phenyl-ethanone

19.6

8.6

3.7

117.4

Propan-2-one oxime

16.3

3.7

10.9

80.2

Acetic acid buta-1,3dienyl ester

16.1

4.4

8.3

118.4

1-Acetyl-azepan-2-one

18.9

8.7

4.8

155.0

1-Morpholin-4-ylethanone

18.3

5.3

7.8

115.6

1-Piperazin-1-ylethanone

18.5

10.0

6.5

125.8

N

H3C 11

Hydrogen Molar Bonding Volume

CH3

H3C

10

Dispersion Polarity

Acetophenone O CH3

12

Acetoxime

OH

N

CH3

H3C 765

1-Acetoxy-1,3-butadiene

O

H3C

CH2

O 13

N-Acetyl Caprolactam O

O N

14

CH3

N-Acetyl Morpholine CH3

O

N

O

15

N-Acetyl Piperidine CH3

O N

N

7248_A001.fm Page 349 Wednesday, May 23, 2007 12:31 PM

Appendix A: Table A.1

349

TABLE A.1 (CONTINUED) No. 16

Solvent Name N-Acetyl Pyrrolidone O

Autonom/ ACD Name

Dispersion Polarity

Hydrogen Molar Bonding Volume

1-Acetyl-pyrrolidin-2one

17.8

13.1

8.3

127.0

2-Acetoxy-benzoic acid

19.0

6.6

9.3

133.5

1-Thiophen-2-ylethanone

19.1

12.2

9.3

108.0

3-Acetoxy-3ethoxycarbonylpentanedioic acid diethyl ester

16.6

3.5

8.6

279.9

Pentane-2,4-dione

16.1

11.2

6.2

103.0

Acetyl bromide

16.7

10.6

5.2

74.0

Acetyl chloride

16.2

11.2

5.8

71.4

Ethyne

14.4

4.2

11.9

42.1

O N

CH3

1116 Acetyl Salicylic Acid O

OH O

O

CH3

785

2-Acetyl Thiophene O S CH3

1219 Acetyl Triethyl Citrate O O H3C

O

O O

CH3 O

CH3

O O

CH3

17

Acetylacetone

O

O

CH3

H3C 18

Acetylbromide

O Br

H3C 19

Acetylchloride

O Cl

H3C 20

Acetylene (Ethyne)

HC

CH

7248_A001.fm Page 350 Wednesday, May 23, 2007 12:31 PM

350

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No. 21

Solvent Name Acetylfluoride

Autonom/ ACD Name

Dispersion Polarity

Hydrogen Molar Bonding Volume

Acetyl fluoride

14.7

14.0

5.7

62.0

Acridine

21.7

5.9

2.0

163.0

Propenal

15.0

7.2

7.8

66.7

Acrylamide

15.8

12.1

12.8

63.4

Acrylic acid

17.7

6.4

14.9

68.5

Acrylonitrile

16.0

12.8

6.8

67.1

Acryloyl chloride

16.2

11.6

5.4

81.3

Hexanedioic acid

17.1

9.0

14.6

107.5

4-((R)-1-Hydroxy-2methylamino-ethyl)benzene-1,2-diol

20.5

8.7

19.9

154.5

O H3C 860

F

Acridine

N 22

Acrolein

O

H2C 23

Acrylamide

O

H2C

NH2 24

Acrylic Acid

O

H2C OH 25

Acrylonitrile

H2C 26

N

Acrylylchloride

O

H2C Cl

1152 Adipic Acid (1,6-Hexanedioic) O OH

HO O

1195 Adrenaline HO

HO

N OH

CH3

7248_A001.fm Page 351 Wednesday, May 23, 2007 12:31 PM

Appendix A: Table A.1

351

TABLE A.1 (CONTINUED) No.

Autonom/ ACD Name

Solvent Name

1088 alfa-Chloro Acetophenone O

Dispersion Polarity

Hydrogen Molar Bonding Volume

2-Chloro-1-phenylethanone

20.0

9.6

5.7

116.8

Acetic acid allyl ester

15.7

4.5

8.0

108.5

Pent-4-enoic acid

16.7

4.7

11.3

102.1

3-Oxo-butyric acid allyl ester

15.9

6.9

8.6

137.8

Pent-4-enenitrile

16.3

11.2

5.0

98.5

Prop-2-en-1-ol

16.2

10.8

16.8

68.4

Allylamine

15.5

5.7

10.6

74.9

3-Bromo-propene

16.5

7.3

4.9

86.5

3-Chloro-propene

17.0

6.2

2.3

82.3

But-3-enenitrile

16.0

14.3

5.6

80.5

Cl

27

Allyl Acetate

CH3

O

H2C

O 28

Allyl Acetic Acid

O

OH

H2C 769

Allyl Acetoacetate

O

O

H3C 29

O

Allyl Acetonitrile (4-Pentenenitrile)

CH2

N 30

Allyl Alcohol

H2C 31

Br

Allyl Chloride

H2C 34

NH2

Allyl Bromide (3-Bromoprene)

H2C 33

OH

Allyl Amine

H2C 32

CH2

Cl

Allyl Cyanide

N H2C

7248_A001.fm Page 352 Wednesday, May 23, 2007 12:31 PM

352

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No. 35

Solvent Name Allyl Ethylether

O

H2C 36

Dispersion Polarity

Hydrogen Molar Bonding Volume

3-Ethoxy-propene

15.0

4.8

5.1

112.6

3-Fluoro-propene

14.9

6.4

1.0

71.5

Formic acid allyl ester

15.7

5.4

8.8

91.0

3-Iodo-propene

17.4

6.4

3.3

90.8

3-Isocyano-propene

16.1

13.0

5.4

84.5

3-Isopropoxy-propene

15.0

4.1

7.1

129.0

3-Isothiocyanatopropene

17.0

11.3

8.5

98.0

Prop-2-ene-1-thiol

16.4

6.2

7.9

80.2

2-Methyl-acrylic acid allyl ester

15.2

4.1

7.5

134.8

3-Methoxy-propene

15.0

4.3

5.9

93.7

CH3

Allyl Fluoride

F

H2C 37

Autonom/ ACD Name

Allyl Formate

O

CH2

O 38

Allyl Iodide

I

H2C 39

Allyl Isocyanide

H2C

+

N 967

C

Allyl Isopropyl Ether

O

H2C

CH3

CH3 40

Allyl Isothiocyanate

N H2C 42

S

Allyl Mercaptan

SH

H2C

1018 Allyl Methacrylate CH3 O

H2C

CH2

O

43

Allyl Methylether

H2C

O

CH3

7248_A001.fm Page 353 Wednesday, May 23, 2007 12:31 PM

Appendix A: Table A.1

353

TABLE A.1 (CONTINUED) No.

Autonom/ ACD Name

Solvent Name

1181 2-Amino-2-Methyl-1-Propanol/Acetic Acid O O

Dispersion Polarity

Hydrogen Molar Bonding Volume

Acetate 2-hydroxy-1,1dimethyl-ethylammonium

17.2

22.5

23.3

1-Amino-cyclopropanecarboxylic acid

17.0

6.3

13.6

93.7

Pyridin-2-ylamine

20.4

8.1

12.2

94.1

Pyridin-4-ylamine

20.4

16.1

12.9

87.1

Ammonia

13.7

15.7

17.8

20.8

1-Methyl-2-phenylethylamine

17.5

4.3

6.3

148.1

Acetic acid pentyl ester

15.8

3.3

6.1

148.0

1-Methoxy-4-((E)propenyl)-benzene

19.0

4.3

8.7

150.0

+

NH3

OH

1206 1-Aminocyclopropane Carboxylic Acid

O H2N

OH

1090 2-Amino Pyridine

NH2

N

1205 4-Amino Pyridine NH2

N

44

Ammonia

NH3 1179 Amphetamine

CH3 NH2 45

Amyl Acetate

H3C

O

CH3

O 699

Anethole (Trans)

H3C

CH3 O

7248_A001.fm Page 354 Wednesday, May 23, 2007 12:31 PM

354

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No. 46

Solvent Name Aniline

Autonom/ ACD Name

Dispersion Polarity

Hydrogen Molar Bonding Volume

Phenylamine

19.4

5.1

10.2

91.5

2-Methoxybenzaldehyde

19.4

11.9

8.3

120.8

2-Methoxy-phenylamine 19.5

5.4

11.4

112.2

4-Methoxy-phenylamine 19.9

6.5

11.3

113.3

Methoxy-benzene

17.8

4.1

6.7

119.1

Anthraquinone

20.3

7.6

4.8

145.6

(R)-5-((S)-1,218.0 Dihydroxy-ethyl)-3,4dihydroxy-5H-furan-2one

12.6

27.6

106.7

NH2 1113 Anisaldehyde (2-Methoxy Benzaldehyde) O O

949

CH3

Anisidine (2-Methoxyaniline) NH2 O

47

CH3

p-Anisidine (Methoxy Aniline) CH3 O

NH2

48

Anisole

H3C

O

1099 9,10-Anthraquinone O

O

1163 Ascorbic Acid OH O

O

HO HO

OH

7248_A001.fm Page 355 Wednesday, May 23, 2007 12:31 PM

Appendix A: Table A.1

355

TABLE A.1 (CONTINUED) No. 49

Solvent Name Azidoethane

H3C 50

N

N

N

807

Benzal Chloride

Hydrogen Molar Bonding Volume

Azido-ethane

15.9

8.9

12.9

79.0

3-Azido-propene

16.8

7.7

13.4

83.0

N

Dichloromethyl-benzene 19.9

6.6

2.4

134.2

Benzaldehyde

19.4

7.4

5.3

101.5

Benzamide

21.2

14.7

11.2

90.3

2.0

89.4

N

Cl

Cl

51

Dispersion Polarity

N

3-Azidopropene

H2C

Autonom/ ACD Name

Benzaldehyde O

1055 Benzamide O NH2

52

Benzene

Benzene

18.4

0

53

1,3-Benzenediol

Benzene-1,3-diol

18.0

8.4

21.0

87.5

1H-Benzoimidazole

20.6

14.9

11.0

102.7

OH

OH

894

Benzimidazole

N N

7248_A001.fm Page 356 Wednesday, May 23, 2007 12:31 PM

356

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No. 893

Solvent Name Benzisoxazole

Autonom/ ACD Name

Dispersion Polarity

Hydrogen Molar Bonding Volume

Benzo[d]isoxazole

20.6

11.5

8.8

100.7

Benzofuran

18.7

5.1

5.7

110.2

Benzoic acid

18.2

6.9

9.8

113.1

2-Hydroxy-1,2diphenyl-ethanone

19.9

9.7

10.7

187.7

Benzonitrile

17.4

9.0

3.3

102.6

Diphenyl-methanone

19.6

8.6

5.7

164.2

Benzothiazole

20.6

5.2

8.4

108.5

1H-Benzotriazole

18.7

15.6

12.4

96.2

N O 892

2,3-Benzofuran (Cumaron)

O 54

Benzoic Acid

OH O

891

Benzoin O

OH

55

Benzonitrile N

56

Benzophenone O

962

Benzothiazole

N S 1053 1,2,3-Benzotriazole

N N N

7248_A001.fm Page 357 Wednesday, May 23, 2007 12:31 PM

Appendix A: Table A.1

357

TABLE A.1 (CONTINUED) No. 866

Benzotrichloride Cl

57

Autonom/ ACD Name

Solvent Name

Cl

Dispersion Polarity

Hydrogen Molar Bonding Volume

Trichloromethylbenzene

20.2

6.6

3.2

142.1

Benzoyl chloride

20.7

8.2

4.5

116.0

Acetic acid benzyl ester 18.3

5.7

6.0

142.8

Phenyl-methanol

18.4

6.3

13.7

103.6

Benzylamine

19.2

4.6

11.7

109.2

Benzoic acid benzyl ester

20.0

5.1

5.2

191.2

Terephthalic acid 1benzyl ester 4-butyl ester

19.0

11.2

3.1

306.0

Chloromethyl-benzene

18.8

7.1

2.6

115.0

Cl

Benzoyl Chloride

Cl O

903

Benzyl Acetate O O

58

CH3

Benzyl Alcohol

OH

971

Benzyl Amine

NH2

1149 Benzyl Benzoate O O

59

60

Benzyl Butyl Phthalate O

O

O

O

Benzyl Chloride

Cl

CH3

7248_A001.fm Page 358 Wednesday, May 23, 2007 12:31 PM

358

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No. 61

Solvent Name Benzyl Methacrylate

Autonom/ ACD Name

Dispersion Polarity

Hydrogen Molar Bonding Volume

2-Methyl-acrylic acid benzyl ester

16.8

4.1

4.1

167.8

1-Benzyl-pyrrolidin-2one

18.2

6.1

5.6

160.0

Ethoxymethyl-benzene

18.4

3.8

3.8

144.2

CH3

O H2C

O

62

N-Benzyl Pyrrolidone O N

887

Benzylethyl Ether

CH3

O

867

Bicyclohexyl

Bicyclohexyl

18.6

0

0

188.5

868

Biphenyl

Biphenyl

19.7

1.0

2.0

155.1

Dicarbonimidic diamide 20.0

14.6

18.8

70.3

17.0

10.2

6.9

41.8

16.6

2.5

0

85.3

1066 Biuret

O

O

64

NH2

N

H2N

Borine Carbonyl +

H3B

O

1172 Boron Trichloride

Cl

B

Cl

Cl 1170 Bromine (P From Group Cont.)

Bromine

18.2

14.9

0

51.5

1213 Bromine (P From Dipole Moment)

Bromine

18.2

2.1

0

51.5

7248_A001.fm Page 359 Wednesday, May 23, 2007 12:31 PM

Appendix A: Table A.1

359

TABLE A.1 (CONTINUED) No. 65

Autonom/ ACD Name

Solvent Name 2-Bromo Allyl Alcohol

Dispersion Polarity

Hydrogen Molar Bonding Volume

2-Bromo-prop-2-en-1-ol 17.1

9.9

16.2

84.5

2-Bromo-propene

16.1

6.0

4.9

88.8

(Z)-1-Bromo-propene

16.3

6.4

5.0

84.7

4-Bromo-but-1-ene

16.5

6.0

4.5

102.0

4-Bromo-buta-1,2-diene

17.0

6.5

4.7

93.3

4-Bromo-1-nitro-2trifluoromethylbenzene

20.0

6.0

4.9

150.1

1-Bromo-4-ethoxybenzene

19.5

7.7

5.3

140.0

Bromo-ethyne

15.7

9.9

5.6

67.7

1-Bromo-2-methoxybenzene

19.8

8.4

6.7

124.5

Br OH

H2C 66

2-Bromo Propene

Br CH2

H3C 67

1-Bromo Propene (CIS)

H3C Br 68

4-Bromo-1-Butene

H2C 69

Br

4-Bromo-1,2-Butadiene

Br

H2C 854

5-Bromo-2-Nitrobenzotrifluoride F

F

F

O +

N

O

Br

815

1-Bromo-4-Ethoxy Benzene

H3C

O

Br 70

Bromoacetylene

Br

HC 800

o-Bromoanisole O Br

CH3

7248_A001.fm Page 360 Wednesday, May 23, 2007 12:31 PM

360

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No. 71

Solvent Name Bromobenzene

Autonom/ ACD Name

Dispersion Polarity

Hydrogen Molar Bonding Volume

Bromo-benzene

20.5

5.5

4.1

105.3

4-Bromo-benzonitrile

20.4

9.3

5.8

113.8

4-Bromo-benzoyl chloride

20.2

6.5

5.5

137.2

2-Bromo-butane

16.3

7.7

4.4

109.5

1-Bromo-2-chlorobenzene

20.3

7.7

4.6

116.9

Bromo-chloro-methane

17.3

5.7

3.5

65.0

Bromo-ethene

16.1

6.3

2.3

71.6

Tribromo-methane

21.4

4.1

6.1

87.5

Bromo-methoxymethane

16.9

8.5

7.0

81.6

Br 806

p-Bromobenzonitrile N

Br

805

p-Bromobenzoyl Chloride Br Cl O

729

2-Bromobutane

Br CH3

H3C 802

o-Bromochlorobenzene Cl Br

72

Bromochloromethane

Br 73

Cl

Bromoethylene

Br

H2C 74

Bromoform

Br

Br Br 748

Bromomethyl Methyl Ether

Br

O

CH3

7248_A001.fm Page 361 Wednesday, May 23, 2007 12:31 PM

Appendix A: Table A.1

361

TABLE A.1 (CONTINUED) No. 75

Solvent Name 1-Bromonaphtalene

Autonom/ ACD Name

Dispersion Polarity

Hydrogen Molar Bonding Volume

1-Bromo-naphthalene

20.3

3.1

4.1

140.0

1-Bromo-4-nitrobenzene

20.9

9.9

5.9

103.7

2-Bromo-buta-1,3-diene

16.9

6.4

4.7

95.2

1-Bromo-propane

16.4

7.9

4.8

90.9

2-Bromo-propane

16.1

8.3

4.7

93.6

3-Bromo-propyne

18.1

6.5

5.3

75.3

1-Bromo-2-vinylbenzene

19.5

5.2

5.3

130.0

2-Bromo-thiophene

20.1

5.2

4.6

96.8

1-Bromo-2-methylbenzene

19.3

5.0

4.2

119.0

Br

803

p-Bromonitrobenzene Br O

+

N

O

76

Bromoprene

Br CH2

H2C 1124 1-Bromopropane

Br

H3C 1125 2-Bromopropane

Br CH3

H3C 77

3-Bromopropyne

HC Br 710

o-Bromostyrene Br CH2

78

2-Bromothiophene

Br

S 713

o-Bromotoluene CH3 Br

7248_A001.fm Page 362 Wednesday, May 23, 2007 12:31 PM

362

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No. 714

Solvent Name p-Bromotoluene

Autonom/ ACD Name 1-Bromo-4-methylbenzene

Dispersion Polarity

Hydrogen Molar Bonding Volume

19.3

6.8

4.1

18.3

0.8

0

99.2

Bromo-trichloromethane

18.3

8.1

6.0

99.2

Bromotrichloro Methane (P from Group Bromo-trichlorocont.) methane

18.3

8.1

0

99.2

Bromo-trifluoromethane

9.6

2.4

0

97.0

Buta-1,2-diene

14.7

1.7

6.2

82.3

Buta-1,3-diene

14.8

2.8

5.6

83.2

1-Chloro-buta-1,3-diene

15.6

8.5

2.0

92.2

Buta-2,3-dien-1-ol

16.2

6.6

16.8

76.5

CH3

122.9

Br

743

Bromotrichloro Methane (P from Dipole Bromo-trichloroMoment) methane

Cl

Cl Cl 744

Bromotrichloro Methane (P and H from Group cont.)

Cl

Cl Cl 745

Cl

Br

Bromotrifluoromethane (Freon 1381)

F

F F

80

Br

Cl

Cl

79

Br

Br

1,2-Butadiene

H3C 81

1,3-Butadiene

H2C 82

CH2

CH2

1,3-Butadiene-1-Chloro

Cl 83

CH2

2,3-Butadiene-1-ol

OH H2C

7248_A001.fm Page 363 Wednesday, May 23, 2007 12:31 PM

Appendix A: Table A.1

363

TABLE A.1 (CONTINUED) No. 84

Autonom/ ACD Name

Solvent Name 1,3-Butadiene-1,2-Di-Chloro

Cl

Butadiene-4-Cyano

H2C 86

Hydrogen Molar Bonding Volume

1,2-Dichloro-buta-1,3diene

17.0

10.7

2.8

102.5

Penta-2,4-dienenitrile

16.2

11.7

5.2

93.7

[2,2']Bioxiranyl

18.3

14.4

6.2

77.8

Butane-2,3-dione

15.7

5.1

6.8

87.8

Acrylic acid 416.8 acryloyloxy-butyl ester

9.1

4.2

194.4

Butane

14.1

0

0

101.4

Butane-1,3-diol

16.6

10.0

21.5

89.9

Butane-1,4-diol

16.6

15.3

21.7

88.9

Butane-1-thiol

16.3

5.3

4.5

107.8

Butan-1-ol

16.0

5.7

15.8

91.5

CH2

Cl 85

Dispersion Polarity

N

Butadienedioxide

O O 87

Butadione O CH3

H3C O

88

1,4-Butandiol Diacrylate O O

H2C

O

CH2

O

89

Butane

CH3

H3C 90

1,3-Butanediol

OH H3C 730

1,4-Butanediol

HO 91

OH

1-Butanethiol

H3C 92

OH

SH

1-Butanol

H3C

OH

7248_A001.fm Page 364 Wednesday, May 23, 2007 12:31 PM

364

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No. 92

Autonom/ ACD Name

Solvent Name 2-Butanol

Dispersion Polarity

Hydrogen Molar Bonding Volume

Butan-2-ol

15.8

5.7

14.5

92.0

But-1-ene

13.2

1.3

3.9

94.3

(Z)-But-2-ene

14.7

1.3

4.1

90.3

(E)-But-2-ene

14.6

0

2.9

92.9

But-3-enenitrile

16.2

14.3

5.6

80.5

4-Methoxy-but-1-ene

15.1

5.3

5.2

111.9

(Z)-1-Methoxy-but-2ene

15.2

3.4

5.0

118.8

(E)-1-Methoxy-but-2ene

15.3

4.4

4.3

110.4

1-Butoxy-3-ethoxypropan-2-ol

15.5

6.5

10.2

185.9

CH3

H3C OH 94

1-Butene

CH3

H2C 95

2-Butene (cis)

H3C CH3 96

2-Butene (trans)

CH3

H3C 97

3-Butenenitrile

N H2C 98

1-Butenyl Methyl Ether

O

H2C 99

CH3

2-Butenyl Methyl Ether (cis)

CH3 O 100

2-Butenyl Methyl Ether (trans)

O

H3C 101

CH3

CH3

Butoxy Ethoxy Propanol Commercial H3C

O

O

OH

CH3

7248_A001.fm Page 365 Wednesday, May 23, 2007 12:31 PM

Appendix A: Table A.1

365

TABLE A.1 (CONTINUED) No. 728

Autonom/ ACD Name

Solvent Name 3-Butoxybutanol

Dispersion Polarity

Hydrogen Molar Bonding Volume

3-Butoxy-butan-1-ol

15.9

5.5

10.6

166.3

Acetic acid butyl ester

15.8

3.7

6.3

132.5

Acetic acid sec-butyl ester

15.0

3.7

7.6

133.6

Acetic acid tert-butyl ester

15.4

6.2

6.2

134.1

3-Oxo-butyric acid butyl 16.6 ester

5.8

7.3

164.3

Acrylic acid butyl ester

15.6

6.2

4.9

143.8

2-Methyl-propan-2-ol

15.2

5.1

14.7

95.8

Butylamine

16.2

4.5

8.0

99.0

CH3

O OH

H3C

102

n-Butyl Acetate

O

103

CH3

O

H3C

Sec-Butyl Acetate

CH3

O

963

tert-Butyl Acetate

CH3

O

CH3 CH3

O

H3C 768

CH3

O

H3C

n-Butyl Aceto Acetate

O

O

104

CH3

O

H3C n-Butyl Acrylate

CH3

O

H2C O 611

t-Butyl Alcohol

H3C H3C 105

CH3 OH

n-Butyl Amine

H3C

NH2

7248_A001.fm Page 366 Wednesday, May 23, 2007 12:31 PM

366

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No.

Autonom/ ACD Name

Solvent Name

1182 N-Butyl Amine/Acetic Acid

Hydrogen Molar Bonding Volume

16.0

20.3

18.4

Benzoic acid butyl ester 18.3

5.6

5.5

178.0

Butyric acid butyl ester

15.6

2.9

5.6

166.7

Butyl-cyclohexane

16.2

0

0.6

176.7

Butyl-cyclopentane

16.4

0

1.0

162.0

Formic acid butyl ester

15.7

6.5

9.2

114.8

1-Isopropenyloxybutane

14.8

5.3

5.0

145.7

2-Hydroxy-propionic acid butyl ester

15.8

6.5

10.2

149.0

Acetate butylammonium;

O

+

NH3

Dispersion Polarity

O 718

Butyl Benzoate* O CH3

O

774

n-Butyl Butyrate

O

722

CH3

O

H3C

n-Butyl Cyclohexane

CH3

723

n-Butyl Cyclopentane H3C

927

Butyl Formate

O

H3C 106

O

Butyl Isopropenyl Ether

CH2 O

H3C 107

CH3

Butyl Lactate OH

O

H3C O

CH3

7248_A001.fm Page 367 Wednesday, May 23, 2007 12:31 PM

Appendix A: Table A.1

367

TABLE A.1 (CONTINUED) No. 108

Autonom/ ACD Name

Solvent Name n-Butyl Methacrylate CH3

O

H2C

Dispersion Polarity

Hydrogen Molar Bonding Volume

2-Methyl-acrylic acid butyl ester

15.6

6.4

6.6

159.4

2-Butyl-octan-1-ol

16.1

3.6

9.3

224.2

Propionic acid butyl ester

15.7

5.5

5.9

149.7

1-Butyl-pyrrolidin-2-one 17.5

9.9

5.8

148.0

2-Hydroxy-benzoic acid butyl ester

17.9

4.8

11.7

181.7

Octadecanoic acid butyl ester

14.5

3.7

3.5

382.0

1-Butyl-3-methylbenzene

17.4

0.1

1.0

173.7

6-Methyl-cyclohex-3enecarboxylic acid butyl ester

16.1

2.5

5.7

209.1

CH3

O

717

2-Butyl Octanol OH

H3C

CH3

1006 n-Butyl Propionate

CH3

O

H3C O 109

N-Butyl Pyrrolidone O

CH3

N

764

n-Butyl Salicylate O CH3

O OH

223

Butyl Stearate O O

H3C H3C

724

3-n-Butyl Toluene CH3

CH3

747

Butyl-6-Methyl-3-Cyclohexene Carbolyate O H3C

O

CH3

7248_A001.fm Page 368 Wednesday, May 23, 2007 12:31 PM

368

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No. 725

Solvent Name n-Butylbenzene

Autonom/ ACD Name

Dispersion Polarity

Hydrogen Molar Bonding Volume

Butyl-benzene

17.4

0.1

1.1

157.0

4,5-Dimethyl[1,3]dioxolan-2-one

18.0

16.8

3.1

105.5

2-Ethyl-oxirane

16.3

6.2

5.9

87.5

1-Butyl-2-methylbenzene

17.6

0.1

1.0

171.3

1-Butyl-4-methylbenzene

17.4

0.1

1.0

174.2

But-2-ynedinitrile

15.2

16.2

8.0

78.4

Butyraldehyde

15.6

10.1

6.2

90.5

Butyramide

16.9

13.7

12.3

98.4

Butyric acid

14.9

4.1

10.6

110.0

CH3

110

2,3-Butylene Carbonate O

O H3C

111

O CH3

Butyleneoxide

H3C 719

O

o-n-Butyltoluene H3C

CH3

720

p-n-Butyltoluene CH3

CH3

112

2-Butynedinitrile

N

N

1060 Butyraldehyde*

H3C

O

1072 n-Butyramide

O NH2

H3C 114

Butyric Acid

O H3C

OH

7248_A001.fm Page 369 Wednesday, May 23, 2007 12:31 PM

Appendix A: Table A.1

369

TABLE A.1 (CONTINUED) No. 767

Butyric Anhydride

Hydrogen Molar Bonding Volume

Butanoic anhydride

16.0

6.3

7.7

164.4

Dihydro-furan-2-one

19.0

16.6

7.4

76.8

Butyronitrile

15.3

12.4

5.1

87.3

Butyryl chloride

16.8

9.4

4.8

103.6

1,3,7-Trimethyl-3,7dihydro-purine-2,6dione

19.5

10.1

13.0

157.9

Oxepan-2-one

19.7

15.0

7.4

110.8

9H-Carbazole

21.7

6.4

6.2

152.0

Methanedione

15.7

6.3

5.7

38.0

CH3

O

H3C

Dispersion Polarity

O

O

115

Autonom/ ACD Name

Solvent Name

Gamma-Butyrolactone O O

116

Butyronitrile

H3C N 117

Butyrylchloride

O Cl

H3C 1200 Caffeine O H3C

CH3 N

N

N

N

O

CH3

118

Caprolactone (Epsilon) O O

869

Carbazole (Diphenylenimine)

N

119

Carbon Dioxide* Ro = 3.3

O O

7248_A001.fm Page 370 Wednesday, May 23, 2007 12:31 PM

370

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No. 120

Autonom/ ACD Name

Solvent Name Carbon Disulfide P for 0 Dipole Moment

Dispersion Polarity

Hydrogen Molar Bonding Volume

Methanedithione

20.5

0

0.6

60.0

Methanedithione

19.9

5.8

0.6

60.0

Tetrachloro-methane

17.8

0

0.6

97.1

Tetrachloro-methane

16.1

8.3

0

97.1

2-Oxo-malononitrile

15.0

6.3

8.0

71.2

Thioxo-methanone

17.4

3.7

0

51.0

15.9

4.6

12.0

436.0

15.1

3.7

8.1

298.7

S S 121

Carbon Disulfide P for Group Contribution

S S 122

Carbon tetrachloride P for 0 Dipole Moment

Cl

Cl

Cl

Cl 862

Carbon Tetrachloride P by Group Cont.

Cl

Cl

Cl

Cl 124

Carbonyl Cyanide

O

N 123

N

Carbonyl Sulfide

S O 1224 Castor Oil

976

Cetyl Alcohol (1-Hexadecanol) H3C

Hexadecan-1-ol OH

7248_A001.fm Page 371 Wednesday, May 23, 2007 12:31 PM

Appendix A: Table A.1

371

TABLE A.1 (CONTINUED) No. 125

Solvent Name Chloral

Cl Cl

Autonom/ ACD Name

Dispersion Polarity

Trichloro-acetaldehyde

17.2

7.4

Hydrogen Molar Bonding Volume 7.6

97.5

O Cl

126

Chlorine

Chlorine

17.3

10.0

0

46.0

128

Chloro Acetaldehyde

Chloro-acetaldehyde

16.2

16.1

9.0

60.4

Chloro-acetic acid

17.7

10.4

12.3

68.6

1-(4-Chloro-phenyl)ethanone

19.6

7.6

4.0

129.7

3-Chloro-prop-2-en-1-ol 17.2

10.3

16.5

78.6

2-Chloro-prop-2-en-1-ol 17.1

10.2

16.4

79.6

3-Chloro-furan-2,5dione

20.4

17.3

11.5

89.5

Acrylic acid chloromethyl ester

15.9

7.3

8.5

101.4

1-Chloro-pentane

16.0

6.9

1.9

120.8

Cl 129

O

Chloro Acetic Acid

O

Cl 797

OH

p-Chloro Acetophenone O CH3 Cl

130

3-Chloro Allyl Alcohol

OH

Cl 131

2-Chloro Allyl Alcohol

Cl OH

H2C 983

Chloro Maleic Anhydride

O

O

O

Cl 132

1-Chloro Methyl Acrylate

O

H2C

Cl

O 955

1-Chloro Pentane

H3C

Cl

7248_A001.fm Page 372 Wednesday, May 23, 2007 12:31 PM

372

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No. 133

Solvent Name 2-Chloro Propene (Isopropenyl Chloride)

Autonom/ ACD Name

Dispersion Polarity

Hydrogen Molar Bonding Volume

2-Chloro-propene

15.5

6.7

2.2

84.9

1-Chloro-1-ethoxyethene

16.8

6.5

5.7

104.5

1-Chloro-1-fluoroethene

16.0

6.9

4.0

67.6

1-Chloro-1-nitro-ethane

16.8

13.5

4.7

85.1

1-Chloro-1-nitropropane

16.8

13.0

4.5

102.2

3-Chloro-propan-1-ol

17.5

5.7

14.7

84.2

4-Chloro-buta-1,2-diene

16.6

8.0

6.7

89.5

1-Chloro-but-2-ene

16.2

7.7

2.0

97.4

1-Chloro-2-ethoxybenzene

19.2

8.1

4.4

138.7

Cl CH3

H2C 135

1-Chloro Vinyl Ethyl Ether

Cl H2C 127

O

CH3

1-Chloro-1-Fluoro Ethylene

Cl

CH2 F

136

1-Chloro-1-Nitroethane

Cl

CH3 +

N

O

O

1083 1-Chloro-1-Nitropropane

Cl

O +

N

CH3

O 137

3-Chloro-1-Propanol

OH

Cl 138

4-Chloro-1,2-Butadiene

H2C Cl 139

1-Chloro-2-Butene

H3C 817

Cl

1-Chloro-2-Ethoxy Benzene Cl O

CH3

7248_A001.fm Page 373 Wednesday, May 23, 2007 12:31 PM

Appendix A: Table A.1

373

TABLE A.1 (CONTINUED) No.

Solvent Name

1164 1-Chloro-2-Ethyl Benzene Cl

Autonom/ ACD Name

Dispersion Polarity

Hydrogen Molar Bonding Volume

1-Chloro-2-ethylbenzene

18.9

4.9

2.2

133.0

2-Chloro-2-methylpropane

15.6

7.6

2.0

110.0

3-Chloro-2-methylpropene

16.2

5.6

2.0

98.8

1-Chloro-2-methylpropene

16.1

7.1

4.2

95.6

1-Chloro-2-methyl-3nitro-benzene

20.3

9.6

3.8

132.0

4-Chloro-1-methyl-2nitro-benzene

19.9

11.8

3.8

132.0

1-Chloro-4-ethoxybenzene

19.3

6.3

4.4

139.2

2-Chloro-4-methylphenylamine

19.7

7.4

8.9

123.0

CH3

970

2-Chloro-2-Methyl Propane

CH3 H3C 140

Cl

CH3

3-Chloro-2-Methyl Propene CH3 H2C Cl

141

1-Chloro-2-Methyl Propene CH3 CH3 Cl

813

6-Chloro-2-Nitrotoluene CH3

819

O +

N

Cl

O

4-Chloro-2-Nitrotoluene CH3

O +

N

O

Cl

816

1-Chloro-4-Ethoxy Benzene Cl

O CH3

950

2-Chloro-4-Methylaniline NH2 Cl

CH3

7248_A001.fm Page 374 Wednesday, May 23, 2007 12:31 PM

374

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No. 977

Solvent Name 2-Chloro-5-Methyl Phenol OH

Autonom/ ACD Name

Dispersion Polarity

Hydrogen Molar Bonding Volume

2-Chloro-5-methylphenol

19.0

7.7

13.1

117.4

2-Chloro-acetamide

17.6

12.4

12.8

93.5

1-Chloro-propan-2-one

16.8

9.6

5.5

80.5

Chloro-acetonitrile

17.4

13.6

2.0

63.3

Chloro-acetyl chloride

17.5

9.2

5.5

79.5

Chloro-ethyne

16.2

2.1

2.5

63.7

Acetic acid 1-acetoxy-2- 16.7 chloro-allyl ester

7.3

8.8

160.2

3-Chloro-phenylamine

20.6

9.9

9.8

105.0

1-Chloro-4-methoxybenzene

19.6

7.8

6.7

122.5

Cl

H3C

1067 2-Chloroacetamide

O Cl 142

NH2

Chloroacetone O H3C Cl

143

Chloroacetonitrile

N Cl 144

Chloroacetylchloride O

Cl Cl

145

Chloroacetylene

HC 766

Cl

2-Chloroallylidene Diacetate O

H3C Cl O

H2C

H3C

O

O

1012 m-Chloroaniline NH2

Cl

146

4-Chloroanisole

O

Cl

CH3

7248_A001.fm Page 375 Wednesday, May 23, 2007 12:31 PM

Appendix A: Table A.1

375

TABLE A.1 (CONTINUED) No. 147

Solvent Name 4-Chlorobenzaldehyde

Autonom/ ACD Name

Dispersion Polarity

Hydrogen Molar Bonding Volume

4-Chloro-benzaldehyde

19.9

7.2

5.6

113.4

Chloro-benzene

19.0

4.3

2.0

102.1

4-Chloro-benzonitrile

19.5

8.0

4.1

125.1

1-Chloro-4trichloromethylbenzene

20.3

5.5

3.5

153.8

4-Chloro-benzoyl chloride

19.9

6.7

5.1

127.1

(4-Chloro-phenyl)methanol

19.6

7.5

13.0

117.7

1-Chloro-3chloromethyl-benzene

19.9

9.3

2.6

117.7

1-Bromo-2-chloroethene

17.2

6.6

2.3

78.7

O

Cl 148

Chlorobenzene

Cl

811

4-Chlorobenzonitrile N

Cl

853

4-Chlorobenzotrichloride Cl

Cl

Cl

Cl

812

p-Chlorobenzoyl Chloride O Cl Cl

149

4-Chlorobenzyl Alcohol

OH Cl 150

3-Chlorobenzylchloride

Cl

151

Cl

1,2-Chlorobromoethylene

Cl

Br

7248_A001.fm Page 376 Wednesday, May 23, 2007 12:31 PM

376

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No. 152

Solvent Name 1-Chlorobutane

H3C 750

Autonom/ ACD Name

Dispersion Polarity

Hydrogen Molar Bonding Volume

1-Chloro-butane

16.2

5.5

2.0

104.5

2-Chloro-butane

15.8

7.6

2.0

106.8

2-Chloro-cyclohexanone 18.5

13.0

5.1

113.9

17.6

7.2

2.2

84.9

Chloro-difluoro-methane 12.3

6.3

5.7

72.9

[Chloro(methyl)amino] methane

16.0

7.8

7.9

87.4

Acetic acid 2-chloroethyl ester

16.7

9.6

8.8

107.5

(2-Chloro-ethyl)benzene

19.3

6.3

2.2

131.5

1-Chloro-2-ethoxyethane

16.3

7.9

4.6

109.5

Cl

2-Chlorobutane

CH3

H3C

Cl 780

2-Chlorocyclohexanone O Cl

153

Chlorocyclopropane

Chloro-cyclopropane

Cl

154

Chlorodifluoromethane (Freon 22)

F

Cl F

155

N-Chlorodimethylamine

Cl H3C 772

N

CH3

2-Chloroethyl Acetate

O H3C

Cl

O

1165 2-Chloroethyl Benzene

Cl

756

2-Chloroethyl Ethyl Ether

H3C

O

Cl

7248_A001.fm Page 377 Wednesday, May 23, 2007 12:31 PM

Appendix A: Table A.1

377

TABLE A.1 (CONTINUED) No. 834

2-Chloroethyl Ethyl Sulfide

Cl

S

H3C 808

Autonom/ ACD Name

Solvent Name

o-Chlorofluorobenzene F

Dispersion Polarity

Hydrogen Molar Bonding Volume

1-Chloro-2ethylsulfanyl-ethane

17.2

5.0

6.1

116.9

1-Chloro-2-fluorobenzene

19.4

8.7

2.0

104.9

Trichloro-methane

17.8

3.1

5.7

80.7

1-Chloro-hexane

16.1

6.2

1.7

138.1

Chloro-chloromethoxymethane

17.2

4.9

6.6

86.6

Chlorochloromethylsulfanylmethane

16.6

6.4

2.0

95.0

1-Chloro-naphthalene

19.9

4.9

2.5

136.2

1-Chloro-4-nitrobenzene

20.4

9.6

4.2

103.7

Chloro-nitro-methane

17.4

13.5

5.5

65.1

Cl

156

Chloroform

Cl

Cl Cl 959

1-Chlorohexane

Cl

H3C 157

Bis(Chloromethyl) Ether

O

Cl 158

Chloromethylsulfide

S

Cl 870

Cl

Cl

1-Chloronaphthalene Cl

809

p-Chloronitrobenzene O

+

N

O

Cl

159

Chloronitromethane

O

Cl

+

N

O

7248_A001.fm Page 378 Wednesday, May 23, 2007 12:31 PM

378

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No. 160

Solvent Name 2-Chlorophenol

Autonom/ ACD Name

Dispersion Polarity

Hydrogen Molar Bonding Volume

2-Chloro-phenol

20.3

5.5

13.9

102.3

Trichloro-nitro-methane

17.6

6.8

7.0

101.7

2-Chloro-buta-1,3-diene

16.1

5.4

2.1

93.2

2-Chloro-propenal

17.1

12.9

8.1

75.5

1-Chloro-propene

15.3

6.9

2.2

82.3

2-Chloro-acrylic acid

19.1

9.4

12.4

86.6

3-Chloropropionaldehyde

17.0

13.3

8.2

73.0

3-Chloro-propionitrile

17.3

15.9

6.1

77.4

3-Chloro-propyne

16.7

7.4

2.3

72.4

OH Cl

1081 Chloropicrin (Trichloronitromethane)

Cl

O +

Cl

N

Cl

O 161

Chlorprene

Cl CH2

H2C 162

2-Chloropropenal

Cl O

H2C 163

1-Chloropropene

Cl

H3C 164

2-Chloropropenoic Acid Cl OH

H2C O

165

3-Chloropropionaldehyde

Cl 166

O

3-Chloropropionitrile

Cl N 167

3-Chloropropyne

HC Cl

7248_A001.fm Page 379 Wednesday, May 23, 2007 12:31 PM

Appendix A: Table A.1

379

TABLE A.1 (CONTINUED) No. 711

Solvent Name p-Chlorostyrene CH2

Autonom/ ACD Name

Dispersion Polarity

Hydrogen Molar Bonding Volume

1-Chloro-4-vinylbenzene

18.7

4.3

3.9

128.3

1-Chloro-2-vinylbenzene

18.7

4.7

3.9

126.8

2-Chloro-thiophene

19.2

6.2

8.0

92.2

4-Chloro-benzenethiol

20.8

8.6

10.6

100.0

2-Chloro-benzenethiol

20.2

7.0

10.0

113.4

1-Chloro-4-methylbenzene

19.1

6.2

2.6

118.3

1-Chloro-1,2,2-trifluoro- 15.3 ethene

6.3

0

Cl

712

o-Chlorostyrene CH2 Cl

989

2-Chlorothiophene

Cl

S 168

4-Chlorothiophenol SH

Cl

787

o-Chlorothiophenol SH Cl

814

p-Chlorotoluene CH3

Cl

134

Chlorotrifluoroethylene (CTFE)

Cl

F

F

F

1111 trans-Cinnamaldehyde O

(E)-3-Phenyl-propenal

19.4

12.4

6.2

75.6

125.9

7248_A001.fm Page 380 Wednesday, May 23, 2007 12:31 PM

380

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No.

Solvent Name

1110 cis-Cinnamic Acid

Autonom/ ACD Name

Dispersion Polarity

Hydrogen Molar Bonding Volume

19.1

3.9

10.6

115.4

(E)-3-Phenyl-prop-2-en- 19.1 1-ol

6.0

13.0

129.1

1,3,3-Trimethyl-2-oxabicyclo[2.2.2]octane

16.7

4.6

3.4

167.5

3-Methyl-furan-2,5dione

19.2

17.0

11.2

89.9

4-((E)-3-Hydroxypropenyl)-2-methoxyphenol

19.0

7.0

16.3

171.6

Chromen-2-one

20.0

12.5

6.7

156.3

4-((E)-3-Hydroxypropenyl)-phenol

19.1

7.0

17.3

136.5

3-Methyl-phenol

18.0

5.1

12.9

104.7

(Z)-3-Phenyl-acrylic acid

OH O

1107 Cinnamyl Alcohol OH

1135 Cineol (Eucalyptol) H3C O H3C

CH3

H

982

Citraconic Anhydride O

O

O

CH3

1102 Coniferyl Alcohol OH

H3C

O

HO

974

Coumarin

O 1103 p-Coumaryl Alcohol OH

O

HO

169

m-Cresol CH3

OH

7248_A001.fm Page 381 Wednesday, May 23, 2007 12:31 PM

Appendix A: Table A.1

381

TABLE A.1 (CONTINUED) No. 170

Solvent Name Crotonaldehyde

Dispersion Polarity

Hydrogen Molar Bonding Volume

(E)-But-2-enal

16.2

14.9

7.4

82.5

(E)-But-2-enoic acid

16.8

8.7

12.0

84.6

5H-Furan-2-one

19.0

19.8

9.6

76.4

(E)-But-2-enenitrile

16.4

18.8

5.5

81.4

Cyanamide

15.5

27.6

16.8

32.8

Oxalonitrile

15.1

11.8

0

54.6

Cyanic bromide

18.3

15.2

0

52.6

Cyanic chloride

15.6

14.5

0

51.8

Cyclobutanone

18.3

11.4

5.2

73.4

Cyclodecanone

16.8

8.0

4.1

161.0

O

H3C 171

Autonom/ ACD Name

Crotonic Acid

O OH

H3C 172

Crotonlactone O O

173

Trans-Crotononitrile

N H3C 174

Cyanamid (Carbamonitrile)

N 175

NH2

Cyanogen

N 176

Cyanogen Bromide

N 177

Br

Cyanogen Chloride

N 178

N

Cl

Cyclobutanone

O

179

Cyclodecanone

O

7248_A001.fm Page 382 Wednesday, May 23, 2007 12:31 PM

382

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No. 180

Autonom/ ACD Name

Solvent Name Cycloheptanone

Dispersion Polarity

Hydrogen Molar Bonding Volume

Cycloheptanone

17.2

10.6

4.8

118.2

Cyclohexane

16.8

0

0.2

108.7

Cyclohexane-1,2dicarboxylic acid bis(7-methyl-octyl) ester

16.4

2.2

5.0

422.4

Cyclohexane-1,2-diol

17.4

9.8

18.3

112.8

Cyclohexane-1,2-dione

18.6

10.3

8.0

103.8

Cyclohexanol

17.4

4.1

13.5

106.0

Cyclohexanone

17.8

6.3

5.1

104.0

Cyclohexene

17.2

1.0

5.0

101.9

O

181

Cyclohexane

1214 Cyclohexane-1,2-Dicarboxylic Acid Di-(Isononyl) Ether H3C

CH3

CH3

H3C

O O O

O

1093 1,2-Cyclohexanediol

OH

OH 1094 1,2-Cyclohexanedione

O

O 182

Cyclohexanol

OH

183

Cyclohexanone

O

871

Cyclohexene

7248_A001.fm Page 383 Wednesday, May 23, 2007 12:31 PM

Appendix A: Table A.1

383

TABLE A.1 (CONTINUED) No.

Solvent Name

1199 Cycloheximide H3C O

CH3 H OH H

O

N

Autonom/ ACD Name

Dispersion Polarity

Hydrogen Molar Bonding Volume

4-[(R)-2-((1S,3S,5S)3,5-Dimethyl-2-oxocyclohexyl)-2hydroxy-ethyl]piperidine-2,6-dione

18.3

11.0

13.8

171.0

O

872

Cyclohexyl Benzene

Cyclohexyl-benzene

18.7

0

1.0

169.9

969

N-Cyclohexyl-2-Pyrrolidone

1-Cyclohexylpyrrolidin-2-one

18.2

6.8

6.5

163.0

Cyclohexylamine

17.2

3.1

6.5

113.8

Chloro-cyclohexane

17.3

5.5

2.0

118.6

Cyclooctanone

17.0

9.6

4.5

131.7

N

184

O

Cyclohexylamine

NH2

185

Cyclohexylchloride

Cl

186

Cyclooctanone O

920

Cyclopentadiene

Cyclopenta-1,3-diene

17.2

1.9

6.1

82.1

187

Cyclopentane

Cyclopentane

16.4

0

1.8

94.9

7248_A001.fm Page 384 Wednesday, May 23, 2007 12:31 PM

384

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No. 188

Solvent Name Cyclopentanone

Autonom/ ACD Name

Dispersion Polarity

Hydrogen Molar Bonding Volume

Cyclopentanone

17.9

11.9

5.2

89.1

O

189

Cyclopentene

Cyclopentene

16.7

3.8

1.7

89.0

735

2-Cyclopentenyl Alcohol

Cyclopent-2-enol

18.1

7.6

15.6

86.2

OH 190

Cyclopropene

Cyclopropene

17.2

2.4

2.0

50.0

191

Cyclopropane

Cyclopropane

17.6

0

0

58.3

192

Cyclopropylmethylketone

1-Cyclopropyl-ethanone 17.0

11.1

4.6

93.6

Cyclopropanecarbonitrile

18.6

16.2

5.7

75.4

(1R,4R)-1,7,717.8 Trimethylbicyclo[2.2.1]heptan-2one

9.4

4.7

153.7

CH3 O 193

Cyclopropylnitrile N

1223 d-Camphor H3C

CH3 H

CH3

O

7248_A001.fm Page 385 Wednesday, May 23, 2007 12:31 PM

Appendix A: Table A.1

385

TABLE A.1 (CONTINUED) No. 873

Autonom/ ACD Name

Solvent Name DDT Cl Cl

Cl

Cl

Dispersion Polarity

Hydrogen Molar Bonding Volume

1-Chloro-4-[2,2,2trichloro-1-(4chlorophenyl) ethyl]benzene

20.0

5.5

3.1

268.8

Cl

194

Cis-Decahydronaphthalene

Decahydro-naphthalene

18.8

0

0

156.9

195

Trans-Decahydronaphthalene

Decahydro-naphthalene

18.0

0

0

156.9

196

Decane

Decane

15.7

0

0

195.9

Decan-1-ol

16.0

4.7

10.0

191.8

Decan-2-ol

15.8

3.9

10.0

192.8

Dec-1-ene

15.8

1.0

2.2

190.6

1-Chloro-2-(2-chloro17.1 ethoxymethoxy)-ethane

10.2

7.1

141.1

CH3

H3C

197

1-Decanol* OH

H3C

734

2-Decanol CH3

H3C

OH

726

1-Decene CH2

H3C

759

Di-(2-Chloroethoxy) Methane

O

Cl 773

O

Cl

Dibutyl Fumarate O

16.7

3.0

6.7

232.7

1-Methyl-4-[(4methylphenyl)sulfinyl] benzene

20.3

11.4

3.1

209.0

CH3

O H3C

(E)-But-2-enedioic acid dibutyl ester

O O

199

Di-p-Tolyl Sulfoxide CH3

H3C S O

7248_A001.fm Page 386 Wednesday, May 23, 2007 12:31 PM

386

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No. 204

Di-(2-Chloro-Isopropyl) Ether

CH3

CH3

Cl 235

Autonom/ ACD Name

Solvent Name

Di-(2-Chloroethyl) Ether

Cl

O

1139 Di-(2-Ethyl Hexyl)azelate O

O H3C

O

O

CH3

O H3C

O

O

O

1-Chloro-2-(2-chloroethoxy)-ethane

18.8

9.0

5.7

117.2

Nonanedioic acid bis(2-ethyl-hexyl) ester

16.7

1.4

4.8

449.9

Decanedioic acid bis(2-ethyl-hexyl) ester

16.8

1.0

4.7

468.7

1-Methoxy-2(2-methoxy-ethoxy)ethane

15.7

6.1

6.5

142.0

Bis-(2-ethyl-hexyl)amine

15.6

0.8

3.2

301.5

3-(2-Ethylhexyloxymethyl)heptane

15.9

2.6

2.5

300.5

2,6-Dimethyl-heptan4-ol

14.9

3.1

10.8

177.8

2,6-Dimethyl-heptan4-one

16.0

3.7

4.1

177.1

2-Methyl-1-(2-methylpropane-1-sulfinyl)propane

16.3

10.5

6.1

195.0

CH3

Di-2-Ethyl Hexyl Ether CH3

O CH3

Di-Isobutyl Carbinol

CH3

OH

CH3 CH3

H3C Di-Isobutyl Ketone

CH3

O

CH3 CH3

H3C Di-Isobutyl Sulfoxide

CH3 H3C

146.0

CH3

N

H3C

845

CH3

Di-2-Ethyl Hexyl Amine

H3C

203

O

O

H3C

208

5.1

CH3

Di-(2-Methoxyethyl) Ether

H3C

736

8.2

O

H3C

205

19.0

CH3

1138 Di-(2-Ethyl Hexyl)sebacate

H3C

1-Chloro-2-(2-chloro-1methyl-ethoxy)propane

CH3

H3C

202

Hydrogen Molar Bonding Volume

Cl

O

Cl

Dispersion Polarity

O S

CH3 CH3

7248_A001.fm Page 387 Wednesday, May 23, 2007 12:31 PM

Appendix A: Table A.1

387

TABLE A.1 (CONTINUED) No. 279

Autonom/ ACD Name

Solvent Name Di-Isodecyl Phthalate O

H3C

CH3

H3C

CH3

Dispersion Polarity

Hydrogen Molar Bonding Volume

Phthalic acid bis-(8methyl-nonyl) ester

16.6

6.2

2.6

464.2

Phthalic acid bis-(5methyl-hexyl) ester

16.8

7.2

3.4

364.9

Hexanedioic acid bis-(7- 16.2 methyl-octyl) ester

1.8

4.9

433.7

Phthalic acid bis-(7methyl-octyl) ester

16.6

6.6

2.9

432.4

Methyl-phosphonic acid diisopropyl ester

16.4

10.0

5.7

184.4

Phosphorofluoridic acid diisopropyl ester

15.7

10.2

5.9

174.5

2-(Propane-2-sulfinyl)propane

17.0

11.5

7.4

159.9

1-Butoxy-butane

15.2

3.4

4.2

170.3

1-(Butane-1-sulfinyl)butane

16.4

10.5

6.1

195.0

O O O

280

Di-Isoheptyl Phthalate CH3

O O

CH3

O

CH3 CH3

O

281

Di-Isononyl Adipate O H3C

CH3

O O CH3

282

Di-Isononyl Phthalate CH3

O

CH3

O

CH3

O

CH3

O

840

CH3

O

Di-Isopropyl Methyl Phosphonate CH3 O

H3C O H3C

P

CH3

O CH3

842

Di-Isopropyl Phosphonofluoridate CH3 O

H3C

O P H3C

F

O CH3

198

Di-Isopropyl Sulfoxide CH3 H3C

CH3

S

CH3

O

219

Di-n-Butyl Ether

H3C 200

O

CH3

Di-n-Butyl Sulfoxide

H3C

S O

CH3

7248_A001.fm Page 388 Wednesday, May 23, 2007 12:31 PM

388

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No. 957

Di-n-Pentyl Ether O

H3C

956

O

Hydrogen Molar Bonding Volume

1-Pentyloxy-pentane

15.6

3.1

3.0

203.2

1-Propoxy-propane

15.1

4.2

3.7

137.7

1-(Propane-1-sulfinyl)propane

17.0

13.0

7.4

159.9

4-Hydroxy-4-methylpentan-2-one

15.8

8.2

10.8

124.2

Diallyl-amine

15.6

4.5

6.7

124.1

3-Allyloxy-propene

15.3

4.4

5.3

118.8

3-(1-Allyloxy-ethoxy)propene

15.4

5.1

4.8

163.3

Diazomethane

14.7

6.1

11.3

78.1

Pentanedioic acid dimethyl ester

16.2

4.7

8.4

159.0

CH3

Di-n-Propyl Sulfoxide

H3C

Dispersion Polarity

CH3

Di-n-Propyl Ether

H3C 201

Autonom/ ACD Name

Solvent Name

CH3

S O

209

Diacetone Alcohol

H3C H3C

CH3

OH 210

O

Diallyl Amine

N

H2C 211

Diallyl Ether

O

H2C 749

CH2

CH2

1,1-Diallyloxyethane

O

H2C

O

CH2

CH3 212

Diazomethane

H2C

+

N

N

1009 Dibasic Esters (dupont) Mix of Dimethyl Esters of Succinic, Glutaric, and Adipic Acids

O H3C

O

O O

CH3

7248_A001.fm Page 389 Wednesday, May 23, 2007 12:31 PM

Appendix A: Table A.1

389

TABLE A.1 (CONTINUED) No. 213

Autonom/ ACD Name

Solvent Name Dibenzyl Ether*

Dispersion Polarity

Hydrogen Molar Bonding Volume

[(Benzyloxy)methyl]benzene

19.6

3.4

5.2

197.4

Decanedioic acid dibenzyl ester

17.8

2.2

5.5

362.1

1,1-Dibromo-ethene

15.1

4.8

7.0

85.3

Dibromo-methane

17.8

6.4

7.0

69.8

1,2-Dibromo-benzene

20.7

6.5

5.3

120.0

1,1-Dibromo-ethane

18.5

8.4

8.8

91.4

1,2-Dibromo-ethene

18.0

4.9

3.0

82.7

2,3-Dibromo-buta-1,3diene

17.7

11.8

6.4

103.4

Dibutyl-amine

15.0

3.0

4.3

170.0

O

1148 Dibenzyl Sebacate

O O

O O

214

1,1-Dibromo Ethylene

Br

Br

CH2 215

Dibromo Methane

Br 804

Br

o-Dibromobenzene Br Br

216

1,1-Dibromoethane

Br

Br

CH3 217

1,2-Dibromoethylene

Br 218

Br

2,3-Dibromoprene Br

CH2

H2C Br

1015 Dibutyl Amine

H3C

N

CH3

7248_A001.fm Page 390 Wednesday, May 23, 2007 12:31 PM

390

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No. 220

Autonom/ ACD Name

Solvent Name N,N-Dibutyl Formamide

H3C

N

Dispersion Polarity

Hydrogen Molar Bonding Volume

N,N-Dibutyl-formamide 15.5

8.9

6.2

182.0

Nonan-5-one

16.0

7.7

4.4

173.4

(Z)-But-2-enedioic acid dibutyl ester

16.5

6.1

7.2

230.4

Phthalic acid dibutyl ester

17.8

8.6

4.1

266.0

Decanedioic acid dibutyl 16.7 ester

4.5

4.1

339.0

1,2-Dichloro-4trifluoromethylbenzene

20.0

4.7

2.4

145.5

1,3-Dichloro-propane

18.6

8.2

3.0

95.1

1,1-Dichloro-1-nitroethane

16.8

12.1

4.3

125.2

CH3

O 1068 Dibutyl Ketone

CH3

H3C O 1146 Dibutyl Maleate O O

CH3

O

CH3

O

221

Dibutyl Phthalate O O

CH3

O

CH3

O

222

Dibutyl Sebacate* O O

O

H3C

CH3

O

700

3,4-Dichloro α,α,α-Trifluorotoluene F

F

F

Cl Cl

874

1,3-Dichloropropane

Cl

Cl

1082 1,1-Dichloro-1-Nitroethane

Cl

O +

N O

Cl CH3

7248_A001.fm Page 391 Wednesday, May 23, 2007 12:31 PM

Appendix A: Table A.1

391

TABLE A.1 (CONTINUED) No. 229

Solvent Name 2,3-Dichloro-1,3-Butadiene Cl

Autonom/ ACD Name

Dispersion Polarity

Hydrogen Molar Bonding Volume

2,3-Dichloro-buta-1,3diene

17.1

2.3

2.8

104.0

1,3-Dichloro-but-2-ene

16.9

7.8

2.7

107.7

1,4-Dichloro-but-2-ene

17.8

7.6

2.0

97.8

1,3-Dichloro-2-fluorobenzene

19.4

9.1

2.7

117.1

1,3-Dichloro-propan-2ol

17.5

9.9

14.6

95.5

1,3-Dichloro-2-fluoro-4- 19.9 nitro-benzene

7.2

4.2

140.0

1,5-Dichloro-2-nitro-4trifluoromethylbenzene

19.9

7.6

3.7

162.0

Dichloro-acetaldehyde

16.7

9.1

7.5

94.0

CH2

H2C Cl

206

1,3-Dichloro-2-butene

Cl Cl

H3C 207

1,4-Dichloro-2-Butene

Cl

Cl 825

1,3-Dichloro-2-Fluorobenzene Cl F

Cl

739

1,3-Dichloro-2-Propanol

OH Cl 824

Cl

2,4-Dichloro-3-Fluoronitrobenzene O

+

N

O Cl

F Cl

828

2,4-Dichloro-5-Nitrobenzotrifluoride Cl

F F

Cl

F

+

N

O

O

230

Dichloroacetaldehyde

Cl

O Cl

7248_A001.fm Page 392 Wednesday, May 23, 2007 12:31 PM

392

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No. 945

Solvent Name Dichloroacetic Acid

Autonom/ ACD Name

Dispersion Polarity

Hydrogen Molar Bonding Volume

Dichloro-acetic acid

18.2

8.1

12.2

82.5

1,1-Dichloro-propan-2one

17.1

7.6

5.4

97.3

Dichloro-acetonitrile

17.4

9.4

6.4

80.3

2,4-Dichlorophenylamine

20.9

6.2

10.0

103.4

1,3-Dichloro-2methoxy-benzene

19.8

8.4

6.5

137.1

2,4-Dichlorobenzaldehyde

19.7

8.8

5.4

135.7

1,2-Dichloro-benzene

19.2

6.3

3.3

112.8

1,4-Dichloro-benzene

19.7

5.6

2.7

118.6

O Cl

OH Cl

231

1,1-Dichloroacetone O Cl

CH3 Cl

232

Dichloroacetonitrile

Cl N Cl 1000 2,4-Dichloroaniline NH2 Cl

Cl

233

2,6-Dichloroanisole O

CH3 Cl

Cl

846

2,4-Dichlorobenzaldehyde O Cl

Cl

234

o-Dichlorobenzene Cl Cl

715

p-Dichlorobenzene Cl

Cl

7248_A001.fm Page 393 Wednesday, May 23, 2007 12:31 PM

Appendix A: Table A.1

393

TABLE A.1 (CONTINUED) No. 716

Autonom/ ACD Name

Solvent Name m-Dichlorobenzene

Dispersion Polarity

Hydrogen Molar Bonding Volume

1,3-Dichloro-benzene

19.7

5.1

2.7

114.8

1,4-Dichloro-2trifluoromethylbenzene

20.0

4.7

2.4

145.5

1,4-Dichloro-butane

18.3

7.7

2.8

109.5

Dichloro-difluoromethane

12.3

2.0

0

92.3

1,1-Dichloro-ethane

16.5

7.8

3.0

84.2

Bis-chloromethyl-amine 16.8

7.6

8.0

90.8

1,1-Dichloro-ethene

16.4

5.2

2.4

79.9

(Z)-1,2-Dichloro-ethene

17.0

8.0

3.2

75.5

Bis-(2-chloro-ethyl)amine

16.8

7.6

7.7

98.3

Cl

Cl

826

2,5-Dichlorobenzotrifluoride F

F

F Cl

Cl

1019 1,4-Dichlorobutane

Cl

Cl 236

Dichlorodifluoromethane (Freon 12)

Cl

F

Cl

F 237

1,1-Dichloroethane

Cl

Cl CH3

225

N,N-Dichloromethyl Amine

Cl 239

N

Cl

1,1-Dichloroethylene

Cl CH2 Cl 224

1,2-Dichloroethylene (cis)

Cl 238

Cl

N,N-Dichloroethyl Amine

Cl

N

Cl

7248_A001.fm Page 394 Wednesday, May 23, 2007 12:31 PM

394

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No. 241

Autonom/ ACD Name

Solvent Name Dichloromethyl Methyl Ether

Hydrogen Molar Bonding Volume

17.1

12.9

6.5

90.5

Dichloromonofluoromethane (Freon 21) Dichloro-fluoro-methane 15.8

3.1

5.7

75.4

Cl

O

H3C

Dichloro-methoxymethane

Dispersion Polarity

Cl 242

Cl

Cl F 243

2,3-Dichloronitrobenzene O

+

N

O

1,2-Dichloro-3-nitrobenzene

19.7

12.6

4.4

132.5

1,2-Dichloro-4-nitrobenzene

20.1

7.2

4.1

130.6

2,4-Dichloro-1-nitrobenzene

20.4

8.7

4.2

133.4

1,5-Dichloro-pentane

19.0

7.8

1.5

127.5

2,5-Dichloro-phenol

20.0

6.3

12.1

119.0

2,6-Dichloro-phenol

20.1

7.5

10.9

119.0

Cl

Cl

801

3,4-Dichloronitrobenzene O

+

N

O

Cl Cl

822

2,4-Dichloronitrobenzene O

+

N

O Cl

Cl

833

1,5-Dichloropentane

Cl

Cl 244

2,5-Dichlorophenol OH Cl

Cl

852

2,6-Dichlorophenol OH Cl

Cl

7248_A001.fm Page 395 Wednesday, May 23, 2007 12:31 PM

Appendix A: Table A.1

395

TABLE A.1 (CONTINUED) No.

Solvent Name

1159 2,4-Dichlorophenol

Autonom/ ACD Name

Dispersion Polarity

Hydrogen Molar Bonding Volume

2,4-Dichloro-phenol

20.0

7.2

13.1

119.0

(3,4-Dichloro-phenyl)acetonitrile

20.5

10.8

4.4

148.8

1,1-Dichloro-propane

16.1

7.8

3.5

98.3

2,3-Dichloro-propan1-ol

17.5

9.2

14.6

95.2

2,3-Dichloro-propene

16.2

7.8

3.0

91.6

1,1-Dichloro-propene

16.9

6.7

2.9

93.5

(Z)-1,2-Dichloropropene

17.0

8.5

2.9

93.9

1,2-Dichloro-1,1,2,2tetrafluoro-ethane

12.6

1.8

0

OH Cl

Cl

848

3,4-Dichlorophenyl Acetonitrile N

Cl Cl

245

1,1-Dichloropropane

Cl

H3C Cl 738

2,3-Dichloropropanol

Cl OH

Cl 226

2,3-Dichloropropene

Cl Cl 246

CH2

1,1-Dichloropropene

Cl

H3C

Cl 227

1,2-Dichloropropene (cis)

Cl Cl 247

CH3

1,2-Dichlorotetrafluoroethane (Freon 114)

Cl

Cl

F

F F

F

117.6

7248_A001.fm Page 396 Wednesday, May 23, 2007 12:31 PM

396

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No. 248

Autonom/ ACD Name

Solvent Name 3,4-Dichlorotoluene CH3

Dispersion Polarity

Hydrogen Molar Bonding Volume

1,2-Dichloro-4-methylbenzene

19.8

9.8

2.5

128.7

1,2-Dichloro-1-ethoxyethene

16.9

10.5

6.0

117.8

2-(2-Hydroxyethylamino)-ethanol

17.2

10.8

21.2

95.9

1,1-Diethoxy-butane

15.4

4.9

4.6

177.6

[(Ethoxydithio)oxy]etha ne

15.1

8.3

7.4

141.5

1,1-Diethoxy-ethane

15.0

3.4

4.0

143.9

1,1-Diethoxy-ethane

15.2

5.4

5.3

142.2

2,5-Diethoxytetrahydro-furan

16.6

6.4

7.3

165.7

N,N-Diethyl-acetamide

16.4

11.3

7.5

124.5

Cl Cl

228

1,2-Dichlorovinyl Ethyl Ether

Cl

CH3

O

Cl 249

Diethanolamine

N

HO 777

OH

1,1-Diethoxy Butane

CH3

O

H3C 792

O

Diethoxy Disulfide

H3C 250

CH3

S

O

S

O

CH3

1,1-Diethoxy Ethane

CH3

791

CH3

O

O

H3C

1,1-Diethoxy Ethane (Acetal)

CH3 CH3

O

O

H3C

1045 2,5-Diethoxy Tetrahydrofuran

251

O

O

H3C

N,N-Diethyl Acetamide O H3C

N

CH3 CH3

O

CH3

7248_A001.fm Page 397 Wednesday, May 23, 2007 12:31 PM

Appendix A: Table A.1

397

TABLE A.1 (CONTINUED) No. 252

Autonom/ ACD Name

Solvent Name Diethyl Amine

H3C

Dispersion Polarity

Hydrogen Molar Bonding Volume

Diethyl-amine

14.9

2.3

6.1

103.2

Acetate diethylammonium;

16.0

20.3

18.4

1,4-Diethyl-benzene

18.0

0

0.6

156.9

1,2-Diethyl-benzene

17.7

0.1

1.0

153.5

Carbonic acid diethyl ester

15.1

6.3

3.5

121.0

Acetate diethyl-(2hydroxy-ethyl)ammonium;

16.0

20.3

18.4

Ethoxy-ethane

14.5

2.9

5.1

104.8

N,N-Diethyl-formamide

16.4

11.4

9.2

111.4

CH3

N

1183 Diethyl Amine/Acetic Acid

O

+

NH2

O

253

p-Diethyl Benzene CH3

H3C

721

1,2-Diethyl Benzene CH3 CH3

254

Diethyl Carbonate*

O O

H3C

CH3

O

1184 Diethyl Ethanolamine/Acetic Acid H + N

O

OH

255

Diethyl Ether

H3C 256

O

O

CH3

N,N-Diethyl Formamide O N

CH3

CH3

7248_A001.fm Page 398 Wednesday, May 23, 2007 12:31 PM

398

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No. 257

Autonom/ ACD Name

Solvent Name Diethyl Ketone

H3C

Dispersion Polarity

Hydrogen Molar Bonding Volume

Pentan-3-one

15.8

7.6

4.7

106.4

Malonic acid diethyl ester

16.1

7.7

8.3

152.5

Oxalic acid diethyl ester 16.2

8.0

8.8

136.6

Phthalic acid diethyl ester

17.6

9.6

4.5

198.0

Sulfuric acid diethyl ester

15.7

14.7

7.1

131.5

Ethylsulfanyl-ethane

16.8

3.1

2.0

107.4

2-Diethylamino-ethanol

14.9

5.8

12.0

133.2

Ethyldisulfanyl-ethane

16.7

6.7

5.7

123.1

2-(2-Hydroxy-ethoxy)ethanol

16.6

12.0

20.7

94.9

CH3 O

979

Diethyl Malonate

O

O

CH3

O

O

H3C

1143 Diethyl Oxalate O O

H3C

O

CH3

O

258

Diethyl Phthalate O O

CH3

O

CH3

O

259

Diethyl Sulfate

O

H3C 260

O

CH3

O

Diethyl Sulfide

H3C 261

S

O

CH3

S

2-(Diethylamino) Ethanol

H3C

OH

N

H3C 262

Diethyldisulfide

H3C 263

S

S

CH3

Diethylene Glycol Commercial

HO

O

OH

7248_A001.fm Page 399 Wednesday, May 23, 2007 12:31 PM

Appendix A: Table A.1

399

TABLE A.1 (CONTINUED) No. 264

Autonom/ ACD Name

Solvent Name

Dispersion Polarity

Diethylene Glycol Butyl Ether Acetate Acetic acid 2-(2-butoxy- 16.0 Commercial ethoxy)-ethyl ester H3C

O

O

O

Hydrogen Molar Bonding Volume

4.1

8.2

208.2

1-[2-(2-Butoxy-ethoxy)- 15.8 ethoxy]-butane

4.7

4.4

248.1

1-Ethoxy-2-(2-ethoxyethoxy)-ethane

15.8

5.9

5.6

179.8

[2-(2-Vinyloxy-ethoxy)ethoxy]-ethene

16.0

7.3

7.9

164.1

2-(2-Hexyloxy-ethoxy)ethanol

16.0

6.0

10.0

204.3

2-[2-(2-Methoxyethoxy)-ethoxy]-2methyl-propane

16.0

7.2

7.2

193.9

2-(2-Butoxy-ethoxy)ethanol

16.0

7.0

10.6

170.6

2-(2-Ethoxy-ethoxy)ethanol

16.1

9.2

12.2

130.9

Acetic acid 2-(2-ethoxy- 16.2 ethoxy)-ethyl ester

5.1

9.2

175.5

2-(2-Methoxy-ethoxy)ethanol

16.2

7.8

12.6

118.0

2-(2-Propoxy-ethoxy)ethanol

16.0

7.2

11.3

153.9

CH3 O

757

Diethylene Glycol Dibutyl Ether H3C

762

O

O

O

O

O

CH2

O

Diethylene Glycol Hexyl Ether H3C

266

CH3

O

Diethylene Glycol Divinyl Ether H2C

265

O

CH3

Diethylene Glycol Diethyl Ether H3C

755

O

O

OH

O

Diethylene Glycol Methyl t-Butyl Ether commerical H3C

O

O

O

CH3 CH3

CH3

267

Diethylene Glycol Monobutyl Ether Commercial H3C

268

O

Diethylene Glycol Monoethyl Ether Commerical

H3C 269

OH

O

O

OH

O

Diethylene Glycol Monoethyl Ether Acetate Commerical H3C

O

O

O

CH3 O

270

Diethylene Glycol Monomethyl Ether Commerical

H3C 271

O

OH

O

Diethylene Glycol Monopropyl Ether H3C

O

O

OH

7248_A001.fm Page 400 Wednesday, May 23, 2007 12:31 PM

400

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No. 272

Diethylenetriamine Commercial

H2N 810

Autonom/ ACD Name

Solvent Name

Dispersion Polarity

Hydrogen Molar Bonding Volume

N*1*-(2-Amino-ethyl)ethane-1,2-diamine

16.7

13.3

14.3

108.0

1,2-Difluoro-benzene

18.0

9.0

1.0

98.5

2,6-Difluorobenzonitrile

18.8

11.2

3.2

111.6

3,5-Difluorobenzonitrile

18.8

11.2

3.2

111.6

1,1-Difluoro-ethane

14.9

10.2

3.0

69.5

1,1-Difluoro-ethene

15.0

6.8

3.6

58.2

2,4-Difluoro-1-nitrobenzene

19.4

11.0

3.7

109.6

1-Hexyloxy-hexane

16.0

3.0

2.8

235.8

Phthalic acid dihexyl ester

17.0

7.6

3.6

332.3

NH2

N

o-Difluorobenzene F F

829

2,6-Difluorobenzonitrile N

F

F

830

3,5-Difluorobenzonitrile N

F

273

F

1,1-Difluoroethane

F

F

CH3 274

1,1-Difluoroethylene

F

F

CH2 823

2,4-Difluoronitrobenzene O

+

N

O F

F

1069 Dihexyl Ether

275

CH3

O

H3C

Dihexyl Phthalate O

O

O

CH3

O

CH3

7248_A001.fm Page 401 Wednesday, May 23, 2007 12:31 PM

Appendix A: Table A.1

401

TABLE A.1 (CONTINUED) No. 276

Dispersion Polarity

Hydrogen Molar Bonding Volume

17.3

6.3

10.7

49.6

3,4-Dihydro-2H-pyran

17.5

5.5

5.7

91.2

Benzene-1,4-diol

21.0

10.2

27.2

82.7

Benzene-1,2-diol

20.0

11.3

21.8

81.9

Hexanedioic acid diisobutyl ester

16.7

2.5

6.2

269.6

Diisopropyl-amine

14.8

1.7

3.5

141.1

4-Methylene-oxetan-2one

16.2

15.1

8.1

75.9

1,3-Dimethoxy-butane

15.6

5.5

5.2

140.0

Dihydrogen Disulfide

HS 277

Autonom/ ACD Name

Solvent Name

SH

Dihydropyran

O 917

1,4-Dihydroxybenzene (1,4Benzenediol) OH

OH

904

1,2-Dihydroxybenzene (Catechol)

OH

OH 1140 Diisobutyl Adipate O H3C

CH3 O

O

CH3

CH3

O

1084 Diisopropylamine

H3C

CH3 283

CH3

N

CH3

Diketene O O H2C

763

1,3-Dimethoxy Butane

O

H3C

CH3

O

CH3

7248_A001.fm Page 402 Wednesday, May 23, 2007 12:31 PM

402

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No. 284

Autonom/ ACD Name

Solvent Name 1,1-Dimethoxy Ethane

O

H3C

O

Dispersion Polarity

Hydrogen Molar Bonding Volume

1,1-Dimethoxy-ethane

15.1

4.9

4.9

106.7

2,6-Dimethoxy-phenol

19.3

7.6

13.7

136.4

1,2-Dimethoxy-benzene

19.2

4.4

9.4

127.7

2,5-Dimethoxytetrahydro-furan

16.8

7.2

8.2

129.6

N,N-Dimethylacetamide

16.8

11.5

10.2

92.5

But-2-yne

15.1

3.4

7.6

78.9

Dimethyl-amine

15.3

4.8

11.2

66.2

Dimethyl-amine; compound with dimethyl-amine

15.3

4.8

7.9

132.4

N,N-Dimethylbutyramide

16.4

10.6

7.4

127.8

CH3

CH3 1108 2,6-Dimethoxy Phenol OH H3C

889

O

O

1,2-Dimethoxybenzene

O

CH3 CH3

O 988

CH3

2,5-Dimethoxytetrahydrofuran

CH3

H3C O 285

O

O

N,N-Dimethyl Acetamide O N

H3C

CH3

CH3

286

Dimethyl Acetylene

H3C 287

Dimethyl Amine

H3C 288

CH3

N

CH3

Dimethyl Amine-Dimer

H N

H N 289

N,N-Dimethyl Butyramide O

H3C

N

CH3

CH3

7248_A001.fm Page 403 Wednesday, May 23, 2007 12:31 PM

Appendix A: Table A.1

403

TABLE A.1 (CONTINUED) No. 290

Autonom/ ACD Name

Solvent Name Dimethyl Carbonate

O

H3C

O

Dispersion Polarity

Hydrogen Molar Bonding Volume

Carbonic acid dimethyl ester

15.5

3.9

9.7

84.2

1,2-Dimethoxy-ethane

15.4

6.0

6.0

104.5

1-Methoxy-2-(2methoxy-ethoxy)ethane

15.8

6.1

9.2

142.0

Butane-2,3-dione

15.7

5.3

11.7

88.2

Methyldisulfanylmethane

17.3

7.8

6.5

88.6

2-Dimethylaminoethanol

16.1

9.2

15.3

101.1

Acetate (2-hydroxyethyl)-dimethylammonium;

16.8

19.8

19.8

Formate (2-hydroxyethyl)-dimethylammonium;

17.2

21.5

22.5

2-Methyl-acrylate (2hydroxy-ethyl)dimethyl-ammonium;

17.2

18.8

17.6

CH3

O 291

Dimethyl Cellosolve

O

H3C 292

CH3

Dimethyl Diethylene Glycol

O

H3C 293

O

O

O

CH3

Dimethyl Diketone O CH3

H3C O

294

Dimethyl Disulfide

H3C 295

S

S

CH3

Dimethyl Ethanolamine

CH3 N

HO

CH3

1189 Dimethyl Ethanolamine/Acetic Acid H + N

O O

OH

1188 Dimethyl Ethanolamine/Formic Acid H + N

O O

OH

1185 Dimethyl Ethanolamine/Methacrylic Acid O

H + N OH

O

7248_A001.fm Page 404 Wednesday, May 23, 2007 12:31 PM

404

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No.

Autonom/ ACD Name

Solvent Name

1186 Dimethyl Ethanolamine/p-Toluene Sulfonic Acid

H + N

O OH

S

Toluene-4-sulfonate (2hydroxy-ethyl)dimethyl-ammonium;

17.2

21.5

22.5

Mercapto-acetate (2hydroxy-ethyl)dimethyl-ammonium;

17.2

21.5

22.5

Methoxymethane

15.2

6.1

5.7

63.2

N,N-Dimethylformamide

17.4

13.7

11.3

77.0

N,N-Dimethylhydrazine

15.3

5.9

11.0

76.0

Acetate (2-hydroxypropyl)-dimethylammonium;

16.6

19.4

18.0

2-Methyl-propen-1-one

15.2

7.4

4.8

87.6

(Z)-But-2-enedioic acid dimethyl ester

16.3

8.3

9.8

125.8

O

H + N

O HS

O

OH

Dimethyl Ether

O

H3C 297

Hydrogen Molar Bonding Volume

O

1187 Dimethyl Ethanolamine/Thioglycolic Acid

296

Dispersion Polarity

CH3

Dimethyl Formamide O CH3

N

CH3

298

1,1-Dimethyl Hydrazine

NH2 N

H3C

CH3

1190 Dimethyl Isopropanol Amine/Acetic Acid H + N

O OH

299

Dimethyl Ketene

O

O

H3C

CH3

1145 Dimethyl Maleate O O O O

CH3 CH3

7248_A001.fm Page 405 Wednesday, May 23, 2007 12:31 PM

Appendix A: Table A.1

405

TABLE A.1 (CONTINUED) No. 839

Dimethyl Methyl Phosphonate H3C

Dispersion Polarity

Hydrogen Molar Bonding Volume

Methyl-phosphonic acid dimethyl ester

16.7

13.1

7.5

106.9

2,6-Dimethyl-phenol

19.1

4.9

12.9

116.3

3,4-Dimethyl-phenol

19.2

6.0

13.4

121.0

Phthalic acid dimethyl ester

18.6

10.8

4.9

163.0

2,5-Dimethyl-1Hpyrrole

18.3

7.6

6.8

101.8

Decanedioic acid dimethyl ester

16.6

2.9

6.7

233.3

Sulfuric acid dimethyl ester

17.7

17.0

9.7

94.7

Methylsulfanyl-methane 16.1

6.4

7.4

73.2

O

O

P

H3C

849

Autonom/ ACD Name

Solvent Name

CH3

O

2,6-Dimethyl Phenol OH CH3

H3C

850

3,4-Dimethyl Phenol OH

CH3 CH3

300

Dimethyl Phthalate O CH3

O O

CH3

O

782

2,5-Dimethyl Pyrrole

H3C

CH3

N

1144 Dimethyl Sebacate O H3C

O O

CH3

O

1080 Dimethyl Sulfate

O

H3C O

S

CH3

O

O 301

Dimethyl Sulfide

H3C

S

CH3

7248_A001.fm Page 406 Wednesday, May 23, 2007 12:31 PM

406

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No. 302

Solvent Name Dimethyl Sulfone

Autonom/ ACD Name

Dispersion Polarity

Hydrogen Molar Bonding Volume

Methylsulfonyl-methane 19.0

19.4

12.3

75.0

Methylsulfinyl-methane

18.4

16.4

10.2

71.3

2,3-Dimethyl-but-1-ene

14.9

1.2

2.8

125.2

2,2-Dimethyl-propan1-ol

15.6

6.5

13.5

107.5

2,4-Dimethylphenylamine

19.2

5.2

8.7

123.7

3,5-Dinitro-phenol

19.5

12.9

14.4

108.2

1,2-Dinitro-benzene

20.6

22.7

5.4

107.4

3,4-Dinitro-phenol

19.5

12.8

14.3

110.1

O S

H3C 303

CH3

O

Dimethyl Sulfoxide

O S

H3C 304

CH3

2,3-Dimethyl-1-Butene CH3 CH3

H2C

CH3

1027 2,2-Dimethyl-1-Propanol (Neopentyl alc,)

CH3 OH

H3C 952

CH3

2,4-Dimethylaniline NH2 CH3

CH3

942

3,5-Dinitrophenol O +

N

HO

O

+

O

940

N

O

1,2-Dinitrobenzene O +

N

+

N

O O

O

941

3,4-Dinitrophenol O

HO

+

N

+

N

O

O

O

7248_A001.fm Page 407 Wednesday, May 23, 2007 12:31 PM

Appendix A: Table A.1

407

TABLE A.1 (CONTINUED) Dispersion Polarity

Hydrogen Molar Bonding Volume

20.0

13.1

4.9

137.9

Hexanedioic acid dioctyl 16.7 ester

2.0

5.1

400.0

Phthalic acid dioctyl ester

16.6

7.0

3.1

377.0

[1,4]Dioxane

19.0

1.8

7.4

85.7

[1,3]Dioxane

18.1

6.6

9.3

69.9

4-Isopropenyl-1-methyl- 17.2 cyclohexene

1.8

4.3

162.9

(Phenylethynyl)benzene

20.1

0

3.2

184.3

1167 Diphenyl Acetylene P from Group Cont. (Phenylethynyl)benzene

20.1

2.0

3.2

184.3

No. 938

Autonom/ ACD Name

Solvent Name 2,4-Dinitrotoluene CH3

1-Methyl-2,4-dinitrobenzene

O +

N

O

+

O

N

O

1142 Dioctyl Adipate Commercial O H3C

O

O

CH3

O

305

Dioctyl Phthalate Commercial O O

CH3

O

CH3

O

306

1,4-Dioxane

O

O 307

1,3-Dioxolane

O O 1197 Dipentene CH3

H3C

CH2

1166 Diphenyl Acetylene P from 0 Dipole Moment

7248_A001.fm Page 408 Wednesday, May 23, 2007 12:31 PM

408

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No. 778

Autonom/ ACD Name

Solvent Name Diphenyl Ether*

Diphenyl ether

Dispersion Polarity

Hydrogen Molar Bonding Volume

19.5

3.4

5.8

160.4

(Phenylsulfonyl)benzene 21.1

14.4

3.4

174.3

O 779

Diphenyl Sulfone

O S O

875

Diphenylamine

Diphenyl-amine

20.0

3.3

5.9

145.9

N 876

Diphenylmethane

Benzylbenzene

19.5

1.0

1.0

168.2

308

Dipropyl Amine

Dipropyl-amine

15.3

1.4

4.1

136.9

Heptan-4-one

15.8

5.7

4.9

140.8

3-(3-Hydroxy-propoxy)- 16.5 propan-1-ol

10.6

17.7

130.9

3-(3-Methoxy-propoxy)- 15.5 propan-1-ol

5.7

11.2

157.4

15.7

6.5

10.0

211.2

N

H3C 737

CH3

Dipropyl Ketone*

O CH3

H3C 309

Dipropylene Glycol Commercial

310

OH

O

HO

Dipropylene Glycol Methyl Ether Commercial H3C

O

O

OH

1007 Dipropylene Glycol Mono n-Butyl Ether 1-(2-Butoxy-1-methylethoxy)-propan-2-ol CH 3

H3C

O OH

O

CH3

7248_A001.fm Page 409 Wednesday, May 23, 2007 12:31 PM

Appendix A: Table A.1

409

TABLE A.1 (CONTINUED) No.

Autonom/ ACD Name

Solvent Name

1008 Dipropylene Glycol Mono n-Propyl Ether

Dispersion Polarity

Hydrogen Molar Bonding Volume

1-(1-Methyl-2-propoxyethoxy)-propan-2-ol

15.6

6.1

11.0

185.6

Acetic acid 3-(3methoxy-propoxy)propyl ester

16.3

4.9

8.0

195.7

Methyldisulfanylmethane

17.3

7.8

6.5

88.6

Phthalic acid ditridecyl ester

16.6

5.4

1.9

558.3

1,4-Divinyl-benzene

18.6

1.0

7.0

142.8

Vinylsulfanyl-ethene

16.5

4.6

5.6

93.6

Dodecane

16.0

0

0

228.6

Dodecan-1-ol

16.0

4.0

9.3

224.5

4-(2-Amino-ethyl)benzene-1,2-diol

18.2

10.3

19.5

183.0

CH3 H3C

O

O

CH3

OH

311

Dipropylene Glycol Monomethyl Ether Acetate O H3C

312

CH3

2,3-Dithiabutane

S

H3C 313

O

O

O

S

CH3

Ditridecyl Phthalate O CH3

O O

CH3 O

314

p-Divinyl Benzene CH2

H2C

315

Divinyl Sulfide

H2C 316

S

CH2

Dodecane CH3

H3C

1075 Dodecanol H3C

OH

1180 Dopamine

HO

HO

NH2

7248_A001.fm Page 410 Wednesday, May 23, 2007 12:31 PM

410

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No.

Autonom/ ACD Name

Solvent Name

1177 Ecstasy

CH3

O O 317

H3C

Hydrogen Molar Bonding Volume

(2-Benzo[1,3]dioxol-5yl-1-methyl-ethyl)methyl-amine

18.0

5.1

6.1

202.9

Icosane

16.5

0

0

359.8

2-Chloromethyl-oxirane

18.9

7.6

6.6

78.4

2-Ethyl-oxirane

15.6

8.0

4.6

87.9

2-Vinyl-oxirane

16.6

7.7

7.4

80.7

2-Methylene-oxirane

16.5

8.6

6.7

70.0

Azepan-2-one

19.4

13.8

3.9

110.7

Ethane-1,2-dithiol

17.9

7.2

8.7

83.9

Ethanesulfonyl chloride

17.7

14.9

6.8

94.7

Ethanethiol

15.7

6.5

7.1

74.3

N

Eicosane CH3

H3C

318

Dispersion Polarity

Epichlorohydrin

O Cl 987

1,2-Epoxy Butane

O

320

CH3

3,4-Epoxy-1-Butene

H2C 319

O

1,2-Epoxy-2-Propene

O

321

CH2

Epsilon-Caprolactam O N

322

1,2-Ethane Dithiol

HS 323

SH

Ethanesulfonylchloride

O S H3C 324

Cl

O

Ethanethiol (Ethyl Mercaptan)

H3C

SH

7248_A001.fm Page 411 Wednesday, May 23, 2007 12:31 PM

Appendix A: Table A.1

411

TABLE A.1 (CONTINUED) No. 325

Ethanol

Dispersion Polarity

Hydrogen Molar Bonding Volume

Ethanol

15.8

8.8

19.4

58.5

2-Amino-ethanol

17.0

15.5

21.2

59.8

Acetate 2-hydroxyethyl-ammonium;

17.2

20.3

18.4

1-(4-Ethoxy-phenyl)ethanone

18.8

10.3

6.4

162.6

1,3-Diethoxy-propan-2ol

15.9

5.7

11.7

156.0

3-Ethoxypropionaldehyde

16.0

8.8

7.4

112.1

Propionic acid 2-ethoxy- 16.2 ethyl ester

3.3

8.8

155.5

Acetic acid ethyl ester

15.8

5.3

7.2

98.5

3-Oxo-butyric acid ethyl 16.5 ester

10.8

8.3

125.6

OH

H3C 326

Autonom/ ACD Name

Solvent Name

Ethanolamine

OH NH2 1191 Ethanolamine/Acetic Acid

O

+

NH3

O

OH 794

4-Ethoxy Acetophenone O

CH3

O CH3

776

1-Ethoxy Ethoxy-2-Propanol

OH H3C 761

3-Ethoxy Propionaldehyde

O

O

H3C 327

CH3

O

O

Ethoxyethyl Propionate

O

H3C

O

CH3

O 328

Ethyl Acetate

O

H3C

CH3

O

1016 Ethyl Aceto Acetate (Keto)

O H3C

O O

CH3

7248_A001.fm Page 412 Wednesday, May 23, 2007 12:31 PM

412

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No. 329

Autonom/ ACD Name

Solvent Name Ethyl Acetylene

Dispersion Polarity

Hydrogen Molar Bonding Volume

But-1-yne

15.1

3.4

5.0

81.5

Acrylic acid ethyl ester

15.5

7.1

5.5

108.8

Ethylamine

15.0

5.6

10.7

65.6

Octan-3-one

16.2

4.5

4.1

156.0

Ethyl-benzene

17.8

0.6

1.4

123.1

Benzoic acid ethyl ester 17.9

6.2

6.0

144.3

Bromo-ethane

16.5

8.4

2.3

74.6

Heptan-3-one

16.2

5.0

4.1

139.0

Butyric acid ethyl ester

15.5

5.6

5.0

132.9

HC CH3 330

Ethyl Acrylate

CH3

O

H2C O 331

Ethyl Amine

NH2

H3C 332

Ethyl Amyl Ketone

CH3

H3C O 333

Ethyl Benzene CH3

954

Ethyl Benzoate O CH3

O

334

Ethyl Bromide

H3C 335

Br

Ethyl Butyl Ketone

CH3

H3C

O 928

Ethyl Butyrate

O H3C

O

CH3

7248_A001.fm Page 413 Wednesday, May 23, 2007 12:31 PM

Appendix A: Table A.1

413

TABLE A.1 (CONTINUED) No. 929

Autonom/ ACD Name

Solvent Name Ethyl Caproate

H3C

O

Dispersion Polarity

Hydrogen Molar Bonding Volume

Hexanoic acid ethyl ester 15.5

3.2

5.9

149.6

Carbamic acid ethyl ester 16.8

10.1

13.0

91.2

Isocyano-ethane

15.6

15.2

5.8

74.4

Chloro-ethane

15.7

6.1

2.9

70.0

Ethyl chloridocarbonate

15.5

10.0

6.7

95.6

(E)-3-Phenyl-acrylic acid ethyl ester

18.4

8.2

4.1

166.8

2-Ethyl-but-2-enal

16.1

8.0

5.5

115.2

Cyano-acetic acid ethyl ester

16.7

7.9

8.3

107.1

2-Cyano-acrylic acid ethyl ester

15.2

10.3

9.0

117.1

CH3 O

336

Ethyl Carbamate

CH3

O

O 337

NH2

Ethyl Carbylamine +

H3C 338

N

Ethyl Chloride

H3C 339

C

Cl

Ethyl Chloroformate

O Cl

340

O

CH3

Ethyl Cinnamate O O

341

CH3

2-Ethyl Croton Aldehyde

CH3

O

CH3 1042 Ethyl Cyanoacetate

CH3

O N 342

O

Ethyl Cyanoacrylate N

O O CH2

CH3

7248_A001.fm Page 414 Wednesday, May 23, 2007 12:31 PM

414

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No. 343

Autonom/ ACD Name

Solvent Name Ethyl Ethynylether

Dispersion Polarity

Hydrogen Molar Bonding Volume

Ethoxy-ethyne

15.4

7.9

5.9

87.6

N-Ethyl-formamide

17.2

10.0

14.0

76.5

Formic acid ethyl ester

15.5

8.4

8.4

80.2

Acetic acid 2-ethylhexyl ester

15.8

2.9

5.1

196.0

Acrylic acid 2-ethylhexyl ester

14.8

4.7

3.4

208.2

17.3

6.2

6.8

265.0

Ethyl hypochlorite

15.7

8.6

6.5

79.5

Iodo-ethane

17.3

7.9

7.2

81.2

Isocyanato-ethane

15.4

12.0

2.5

78.7

O

HC

CH3 966

N-Ethyl Formamide

O CH3

N 344

Ethyl Formate

O CH3

O 346

2-Ethyl Hexyl Acetate O H3C

CH3

O H3C

347

2-Ethyl Hexyl Acrylate O

H2C

CH3

O H3C

1153 Mono 2-Ethyl-Hexyl Phthalate (MEHP) Phthalic acid mono-(2ethyl-hexyl) ester OH O O O CH3 CH3

348

Ethyl Hypochlorite

H3C 349

Cl

Ethyl Iodide

H3C 350

O

I

Ethyl Isocyanate

O H3C

N

7248_A001.fm Page 415 Wednesday, May 23, 2007 12:31 PM

Appendix A: Table A.1

415

TABLE A.1 (CONTINUED) No. 351

Solvent Name Ethyl Isopropenyl Ether

Autonom/ ACD Name

Dispersion Polarity

Hydrogen Molar Bonding Volume

2-Ethoxy-propene

14.8

3.5

5.1

113.3

Isothiocyanato-ethane

17.2

14.7

9.0

87.1

2-Hydroxy-propionic acid ethyl ester

16.0

7.6

12.5

115.0

2-Methyl-acrylic acid ethyl ester

15.8

7.2

7.5

125.8

Methylsulfanyl-ethane

17.1

4.8

2.5

91.2

4-Ethyl-morpholine

17.7

5.0

6.6

116.5

(Z)-Octadec-9-enoic acid ethyl ester

14.5

3.8

3.7

357.3

4-Ethyl-phenol

19.2

5.3

12.8

118.6

CH3

352

CH2

O

H3C

Ethyl Isothiocyanate

S H3C 353

N

Ethyl Lactate O HO

CH3

O CH3

354

Ethyl Methacrylate O H2C

O

CH3

CH3

355

Ethyl Methyl Sulfide

H3C

S

CH3

1085 4-Ethyl Morpholine O

N H3C

1117 Ethyl Oleate CH3

H3C

O O

1071 4-Ethyl Phenol OH

H3C

7248_A001.fm Page 416 Wednesday, May 23, 2007 12:31 PM

416

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No.

Solvent Name

1010 Ethyl Propionate

Dispersion Polarity

Propionic acid ethyl ester 15.5

Hydrogen Molar Bonding Volume

6.1

4.9

115.5

CH3

O

H3C

Autonom/ ACD Name

O 356

Ethyl Thiocyanate

Thiocyanato-ethane

15.4

13.4

5.4

87.1

2-Ethoxy-but-1-ene

15.3

4.2

4.6

125.6

Ethoxy-ethene

14.9

4.9

5.6

94.9

Pent-1-en-3-one

15.8

11.3

4.5

99.3

2-Ethyl-butan-1-ol

15.8

4.3

13.5

123.2

3-Methylene-pentane

14.9

1.7

3.5

123.1

1-Ethoxy-propyne

15.9

6.4

5.4

101.6

3-Methylene-pent-1-ene

15.3

1.6

3.8

115.2

N H3C 357

S

1-Ethyl Vinyl Ethyl Ether

CH3 H3C 358

Ethyl Vinylether

H3C 359

CH2

O

CH2

O

Ethyl Vinylketone

CH2

H3C O 360

2-Ethyl-1-Butanol

OH

H3C H3C 727

2-Ethyl-1-Butene

CH2

H3C H3C 361

Ethyl-1-Propynylether

CH3

H3C 362

O

2-Ethyl-1,3-Butadiene

CH3 H2C

CH2

7248_A001.fm Page 417 Wednesday, May 23, 2007 12:31 PM

Appendix A: Table A.1

417

TABLE A.1 (CONTINUED) No. 961

Solvent Name N-Ethyl-2-Pyrrolidone

Autonom/ ACD Name

Dispersion Polarity

1-Ethyl-pyrrolidin-2-one 18.0

Hydrogen Molar Bonding Volume

12.0

7.0

113.9

O

N H3C

345

2-Ethyl-hexanol

2-Ethyl-hexan-1-ol

15.9

3.3

11.8

156.6

Ethene

15.0

2.0

3.8

63.0

[1,3]Dioxolan-2-one

19.4

21.7

5.1

66.0

2-Chloro-ethanol

16.9

8.8

17.2

67.3

3-Hydroxy-propionitrile

17.2

18.8

17.6

68.3

1,2-Dibromo-ethane

19.2

3.5

8.6

87.0

1,2-Dichloro-ethane

19.0

7.4

4.1

79.4

OH

H3C H3C 1058 Ethylene

H2C 363

CH2

Ethylene Carbonate O O

364

O

Ethylene Chlorohydrin

OH Cl 365

Ethylene Cyanohydrin N

OH

366

Ethylene Dibromide

Br Br 367

Ethylene Dichloride

Cl Cl

7248_A001.fm Page 418 Wednesday, May 23, 2007 12:31 PM

418

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No. 937

Autonom/ ACD Name

Solvent Name Ethylene Dinitrate

Dispersion Polarity

Hydrogen Molar Bonding Volume

1,2-Bis-nitrooxy-ethane

16.2

13.0

4.5

101.7

Ethane-1,2-diol

17.0

11.0

26.0

55.8

Acetic acid 2-butoxyethyl ester

15.3

4.5

8.8

171.2

1-(2-Ethoxy-ethoxy)butane

15.3

4.9

4.6

175.5

1-(2-Methoxy-ethoxy)butane

15.5

5.2

4.9

157.2

2-(2-tert-Butoxyethoxy)-2-methylpropane

14.7

4.1

8.2

210.0

Acetic acid 2-acetoxyethyl ester

16.2

4.7

9.8

132.8

1-(2-Butoxy-ethoxy)butane

15.7

4.5

4.2

209.5

1,2-Diethoxy-ethane

15.4

5.4

5.2

141.6

1,2-Dimethoxy-ethane

15.4

6.3

6.0

103.9

O +

O

N

O

O

+

N

O

O

368

Ethylene Glycol

OH OH 369

Ethylene Glycol Butyl Ether Acetate O

O

H3C

752

Ethylene Glycol Butyl Ethyl Ether H3C

753

O

O

O

CH3

Ethylene Glycol Di-t-Butyl Ether CH3

HC H3C3

O

O

CH3

371

CH3

O

Ethylene Glycol Butyl Methyl Ether

H3C 370

CH3

O

CH CH3 3

Ethylene Glycol Diacetate O H3C

O

CH3

O

O

758

Ethylene Glycol Dibutyl Ether O

H3C

760

CH3

Ethylene Glycol Diethyl Ether

H3C 775

O

O

O

Ethylene Glycol Dimethyl Ether

H3C

O

O

CH3

CH3

7248_A001.fm Page 419 Wednesday, May 23, 2007 12:31 PM

Appendix A: Table A.1

419

TABLE A.1 (CONTINUED) No. 372

Autonom/ ACD Name

Solvent Name Ethylene Glycol Methyl t-Butyl Ether*

Dispersion Polarity

Hydrogen Molar Bonding Volume

2-(2-Methoxy-ethoxy)2-methyl-propane

15.3

5.1

8.2

157.4

2-(2-Ethyl-hexyloxy)ethanol

16.0

4.1

10.5

194.7

2-Benzyloxy-ethanol

17.8

5.9

12.2

143.0

Acrylic acid 2-ethoxyethyl ester

15.9

5.1

2-Hexyloxy-ethanol

16.0

5.0

11.4

164.5

2-Propoxy-ethanol

16.1

8.7

13.5

114.1

2-tert-Butoxy-ethanol

15.3

6.1

10.8

131.0

2-Butoxy-ethanol

16.0

5.1

12.3

131.6

CH3

H3C 373

O

O

CH3 CH3

Ethylene Glycol Mono-2-Ethyl Hexyl Ether* HO

CH3

O

CH3

731

Ethylene Glycol Monobenzyl Ether O

771

OH

Ethylene Glycol Monoethyl Ether Acrylate

9.3

147.7

O H3C

O

CH2

O

1056 Ethylene Glycol Mono n-Hexyl Ether CH3

O HO

1004 Ethylene Glycol Mono n-Propyl Ether

HO 374

O

CH3

Ethylene Glycol Mono-t-Butyl Ether Commercial

CH3 HO

375

O

CH CH3 3

Ethylene Glycol Monobutyl Ether Commerical

HO

O

CH3

7248_A001.fm Page 420 Wednesday, May 23, 2007 12:31 PM

420

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No. 376

Ethylene Glycol Monoethyl Ether Commercial

HO 377

Autonom/ ACD Name

Solvent Name

O

Hydrogen Molar Bonding Volume

2-Ethoxy-ethanol

16.2

9.2

14.3

97.8

Acetic acid 2-ethoxyethyl ester

15.9

4.7

10.6

136.1

2-Isobutoxy-ethanol

15.2

4.9

9.6

132.5

2-Isopropoxy-ethanol

16.0

8.2

13.1

115.8

2-Methoxy-ethanol

16.2

9.2

16.4

79.1

Acetic acid 2-methoxyethyl ester

15.9

5.5

11.6

121.6

2-Phenoxy-ethanol

17.0

7.2

12.3

125.3

[1,3,2]Dioxathiolane 2oxide

20.0

15.9

5.1

75.1

CH3

Ethylene Glycol Monoethyl Ether Acetate Commercial

O

H3C

Dispersion Polarity

O

CH3

O 378

Ethylene Glycol Monoisobutyl Ether Commercial

HO

CH3

O

CH3 379

Ethylene Glycol Monoisopropyl Ether

CH3 HO 380

O

Ethylene Glycol Monomethyl Ether Commercial

HO 381

CH3

O

CH3

Ethylene Glycol Monomethyl Ether Acetate

O

H3C

O

CH3

O 924

Ethylene Glycol Monophenyl Ether

O

HO

789

Ethylene Glycol Sulfite O S O

O

7248_A001.fm Page 421 Wednesday, May 23, 2007 12:31 PM

Appendix A: Table A.1

421

TABLE A.1 (CONTINUED) No. 382

Ethylene Methyl Sulfonate

S

O

O S

O

Dispersion Polarity

Methanesulfonic acid 2- 16.9 methanesulfonyloxyethyl ester

O

O

Hydrogen Molar Bonding Volume

9.3

9.6

97.9

O

CH3

CH3 383

Autonom/ ACD Name

Solvent Name

Ethylene Oxide

Oxirane

15.6

10.0

11.0

49.9

Thiirane

19.3

9.0

6.5

59.5

Ethane-1,2-diamine

16.6

8.8

17.0

67.3

Aziridine

18.6

9.8

7.7

51.8

Ethyl-hexyl-amine

15.7

4.2

6.1

163.3

Pent-3-en-2-one

16.2

12.1

4.5

99.0

Methoxy-ethyne

15.8

8.1

6.5

70.1

1-Ethynyloxy-propane

15.4

3.8

5.4

104.1

O 384

Ethylene Sulfide

S 385

Ethylenediamine

NH2 NH2 386

Ethyleneimine

N 946

2-Ethylhexylamine

387

CH3

N

H3C

Ethylidene Acetone

O CH3

H3C 389

Ethynyl Methyl Ether

HC

O CH3

388

Ethynyl Propyl Ether

HC

O CH3

7248_A001.fm Page 422 Wednesday, May 23, 2007 12:31 PM

422

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No.

Solvent Name

1109 Eugenol OH

Autonom/ ACD Name

Dispersion Polarity

Hydrogen Molar Bonding Volume

4-Allyl-2-methoxyphenol

19.0

7.5

13.0

154.0

7,7-Dimethyl-2methylenebicyclo[2.2.1]heptane

16.9

1.5

3.1

157.3

(E)-3-(4-Hydroxy-3methoxy-phenyl)acrylic acid

19.0

6.6

15.1

155.5

CH3

O H2C

898

Fenchene(Alfa) CH3

H3C

CH2

1104 Ferulic Acid OH O O

H3C

HO

877

Fluorene

9H-Fluorene

20.0

1.7

1.7

138.2

390

1-Fluoro Acrylic Acid

2-Fluoro-acrylic acid

16.0

8.7

13.0

90.0

2-Fluoro-acrylonitrile

14.1

15.4

5.7

88.8

1,4-Difluoro-2-nitrobenzene

19.4

7.2

3.7

108.4

1-Fluoro-4-methoxybenzene

18.7

7.3

6.7

113.2

F O

H2C OH

391

1-Fluoro Acrylonitrile

F

H2C 827

N

4-Fluoro-3-Nitrobenzofluoride F

+

N

F

799

O

O

p-Fluoroanisole O

F

CH3

7248_A001A.fm Page 423 Wednesday, May 23, 2007 12:38 PM

Appendix A: Table A.1

423

TABLE A.1 (CONTINUED) No. 392

Solvent Name Fluorobenzene

Autonom/ ACD Name

Dispersion Polarity

Hydrogen Molar Bonding Volume

Fluoro-benzene

18.7

6.1

2.0

94.7

Fluoro-ethene

15.2

7.0

1.0

57.5

Fluoromethane

13.4

10.6

9.5

40.7

2-Fluoro-buta-1,3-diene

14.2

5.8

1.0

85.5

1-(4-Fluoro-phenyl)butan-1-one

19.6

7.1

3.5

138.8

Formaldehyde

12.8

14.4

15.4

36.8

Formamide

17.2

26.2

19.0

39.8

Formic acid

14.3

11.9

16.6

37.8

Formyl fluoride

15.0

10.1

8.6

56.5

Azepane-1-carbaldehyde 18.5

10.4

7.6

127.0

F

393

Fluoroethylene

H2C 394

Fluoromethane

H3C 395

F

F

Fluoroprene

F CH2

H2C 820

4-Fluoropropiophenone CH3 O

F

396

Formaldehyde

H2C 397

O

Formamide

O 398

NH2

Formic Acid

OH

O

741

Formyl Fluoride

O 400

F

N-Formyl Hexamethylene Imine O N

7248_A001A.fm Page 424 Wednesday, May 23, 2007 12:38 PM

424

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No. 401

Autonom/ ACD Name

Solvent Name N-Formyl Piperidine O

Dispersion Polarity

Hydrogen Molar Bonding Volume

Piperidine-1carbaldehyde

18.7

10.6

7.8

111.5

(E)-But-2-enedinitrile

16.7

13.6

7.8

83.0

Furan

17.8

1.8

5.3

72.5

Furan-2-carbaldehyde

18.6

14.9

5.1

83.2

Furan-2-yl-methanol

17.4

7.6

15.1

86.5

Furan-2-carbonitrile

18.4

15.0

8.2

87.5

Propane-1,2,3-triol

17.4

12.1

29.3

73.3

Acetic acid 3-acetoxy-2- 16.4 hydroxy-propyl ester

8.9

14.2

149.4

N

402

Fumaronitrile

N N 403

Furan

O 404

Furfural

O O 405

Furfuryl Alcohol

O OH 994

2-Furonitrile

N

O 406

Glycerol OH

OH

OH

1222 Glycerol Diacetate (Isomer Mix) O H3C

CH3 O

O OH

O

7248_A001A.fm Page 425 Wednesday, May 23, 2007 12:38 PM

Appendix A: Table A.1

425

TABLE A.1 (CONTINUED) No. 990

Autonom/ ACD Name

Solvent Name Glycidaldehyde*

O

Dispersion Polarity

Hydrogen Molar Bonding Volume

Oxirane-2-carbaldehyde

17.5

13.4

9.8

63.2

Oxiranyl-methanol

18.2

9.0

17.9

66.5

2-Methyl-acrylic acid oxiranylmethyl ester

16.3

8.5

5.7

136.4

Ethanedial

15.0

17.0

13.3

50.9

Heptane

15.3

0

0

147.4

Heptan-1-ol

16.0

5.3

11.7

141.4

Heptan-2-ol

15.7

5.4

11.7

142.9

Heptan-3-ol

15.9

5.4

11.7

141.2

Hept-1-ene

15.0

1.1

2.6

141.9

Acetic acid heptyl ester

15.8

2.9

5.5

181.1

O

1087 Glycidol

O

OH 407

Glycidyl Methacrylate CH3 O

H2C

O

O

408

Glyoxal (Ethandial)

O 409

O

Heptane

H3C 931

CH3

1-Heptanol

H3C

OH

1044 2-Heptanol

OH CH3

H3C 1013 3-Heptanol

CH3

H3C

OH 410

1-Heptene

CH2

H3C 411

n-Heptyl Acetate CH3

O H3C

O

7248_A001A.fm Page 426 Wednesday, May 23, 2007 12:38 PM

426

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No. 706

Autonom/ ACD Name

Solvent Name Hexachloroacetone P Based on 0 Dipole Moment

Dispersion Polarity

Hydrogen Molar Bonding Volume

1,1,1,3,3,3-Hexachloropropan-2-one

18.3

0

0

151.9

1,1,1,3,3,3-Hexachloropropan-2-one

18.3

6.6

6.4

151.9

1,2,3,4,5,6-Hexachlorobenzene

21.9

2.1

0

139.3

1,1,1,2,2,2-Hexachloroethane

22.0

4.7

0

113.2

Hexadecane

16.3

0

0

294.1

1,1,2,3,4,4-Hexafluorobuta-1,3-diene

13.8

0

0

104.3

1,1,1,3,3,3-Hexafluoropropan-2-ol

17.2

4.5

14.7

105.3

1,2,3,4,5,6-Hexafluorobenzene

16.9

0

0

115.8

O Cl

Cl

Cl Cl Cl

707

Cl

Hexachloroacetone P based on group contributions O Cl

Cl

Cl Cl Cl

Cl

1128 Hexachlorobenzene (Lindane) Cl Cl

Cl

Cl

Cl Cl

1121 Hexachloroethane

Cl

Cl

Cl

Cl

Cl

Cl

412

Hexadecane CH3

H3C

413

Hexafluoro 1,3-Butadiene F

F F

F F

F

414

Hexafluoro Isopropanol OH F

F

F F

F

F

1119 Hexafluorobenzene F F

F

F

F F

7248_A001A.fm Page 427 Wednesday, May 23, 2007 12:38 PM

Appendix A: Table A.1

427

TABLE A.1 (CONTINUED) No. 415

Autonom/ ACD Name

Solvent Name Hexafluorohexanol

1,2,3,4,5,6-Hexafluorohexan-1-ol

F

F

F F

Dispersion Polarity

Hydrogen Molar Bonding Volume

15.1

4.4

9.9

123.3

1,2,3,4,5,6-Hexamethyl- 19.2 benzene

1.6

0

152.7

19.4

11.6

13.8

105.3

18.5

8.6

11.3

175.7

Hexanal

15.8

8.5

5.4

120.2

n-Hexane

14.9

0

0

131.6

Hexanoic acid

15.0

4.1

9.4

125.9

Hexan-1-ol

15.9

5.8

12.5

124.9

Hex-1-ene

14.7

1.1

0

126.1

OH

F

F

1123 Hexamethyl Benzene CH3 CH3

H3C

CH3

H3C CH3

948

Hexamethylenetetramine

1,3,5,7-Tetraazatricyclo[3.3.1.1*3,7*] decane

N N

N N

416

Hexamethylphosphoramide H3C O

H3C

CH3 N CH3 P

N

N

CH3 CH3

1097 Hexanal

O

H3C 417

n-Hexane

H3C

CH3

1022 Hexanoic Acid

O

H3C 930

1-Hexanol

H3C 418

OH

OH

1-Hexene*

H3C

CH2

7248_A001A.fm Page 428 Wednesday, May 23, 2007 12:38 PM

428

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No. 419

Autonom/ ACD Name

Solvent Name Hexyl Acetate

Dispersion Polarity

Hydrogen Molar Bonding Volume

Acetic acid hexyl ester

15.8

2.9

5.9

165.0

Hexane-1,6-diol

15.7

8.4

17.8

123.0

Acetic acid 6-acetoxyhexyl ester

15.3

4.5

7.2

204.3

Hydrazine

14.2

8.3

8.9

32.1

Nitrilomethane

12.3

17.6

9.0

39.3

Hydrogen peroxide

15.5

12.2

42.7

23.2

17.9

6.0

10.2

36.1

4-Hydroxybenzaldehyde

19.4

15.2

14.5

108.2

(E)-3-(4-Hydroxyphenyl)-acrylic acid

19.1

6.7

15.9

128.3

Tetrahydro-furan-3-ol

18.9

9.4

16.3

80.0

O

H3C 420

O

CH3

Hexylene Glycol

OH

HO 421

Hexylene Glycol Diacetate O CH3

O

O

H3C

O

422

Hydrazine

H2N 423

NH2

Hydrogen Cyanide

HC

N

1173 Hydrogen Peroxide

HO 424

OH

Hydrogen Sulfide

SH2 905

4-Hydroxy Benzaldehyde O

OH

1105 4-Hydroxy Cinnamic Acid OH O

HO

991

3-Hydroxy Tetrahydrofuran

O

OH

7248_A001A.fm Page 429 Wednesday, May 23, 2007 12:38 PM

Appendix A: Table A.1

429

TABLE A.1 (CONTINUED) No. 425

Autonom/ ACD Name

Solvent Name Hydroxyethyl Acrylate

O

H2C

Dispersion Polarity

Acrylic acid 2-hydroxy- 16.0 ethyl ester

Hydrogen Molar Bonding Volume

13.2

13.4

114.9

OH

O 960

N-(2-Hydroxyethyl)-2-Pyrrolidone

1-(2-Hydroxy-ethyl)pyrrolidin-2-one

18.0

9.2

15.7

113.4

1078 Indene

1H-Indene

18.7

2.6

9.0

116.5

831

1H-Indole

19.8

7.5

6.5

110.1

4-Iodo-buta-1,2-diene

17.4

6.3

6.2

105.1

Iodo-benzene

19.5

6.0

6.1

114.4

Triiodo-methane

20.2

3.6

10.6

98.2

2-Iodo-buta-1,3-diene

17.2

2.5

6.2

104.2

O

N

OH

Indole

N 426

4-Iodo-1,2-Butadiene

I H2C 427

Iodobenzene* I

1126 Iodoform

I

I I

428

Iodoprene

I H2C

CH2

7248_A001A.fm Page 430 Wednesday, May 23, 2007 12:38 PM

430

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No. 429

Isoamyl Acetate

CH3

O

H3C 754

Autonom/ ACD Name

Solvent Name

O

Dispersion Polarity

Hydrogen Molar Bonding Volume

Acetic acid 3-methylbutyl ester

15.3

3.1

7.0

148.8

3-Methyl-butan-1-ol

15.8

5.2

13.3

109.4

Propionic acid 3-methyl- 15.7 butyl ester

5.2

5.6

165.7

Acetic acid isobutyl ester 15.1

3.7

6.3

133.5

Acrylic acid isobutyl ester

15.5

6.2

5.0

145.0

2-Methyl-propan-1-ol

15.1

5.7

15.9

92.8

Formic acid isobutyl ester

15.5

6.5

6.7

117.4

Isobutyric acid isobutyl ester

15.1

2.9

5.9

163.0

CH3

Isoamyl Alcohol (3-Methyl-1-Butanol)

CH3

H3C 925

OH

Isoamyl Propionate

H3C

O

CH3 O

CH3 430

Isobutyl Acetate CH3 CH3

O

H3C

O

770

Isobutyl Acrylate CH3 O

H3C

CH2 O

431

Isobutyl Alcohol CH3 H3C OH

1035 Isobutyl Formate O CH3

O

CH3

432

Isobutyl Isobutyrate CH3

CH3 H3C

O

CH3

O

7248_A001A.fm Page 431 Wednesday, May 23, 2007 12:38 PM

Appendix A: Table A.1

431

TABLE A.1 (CONTINUED) No. 786

Isobutyl Sulfoxide

O

CH3

S

H3C 433

Autonom/ ACD Name

Solvent Name

CH3

Dispersion Polarity

Hydrogen Molar Bonding Volume

2-Methyl-1-(2-methylpropane-1-sulfinyl)propane

16.3

10.5

6.1

195.0

Isobutene

14.5

2.0

1.5

89.4

2,2-Dimethyl-oxirane

16.1

4.8

5.8

90.0

Isobutyric acid

16.5

5.4

11.1

93.4

Imino-methanone

15.8

10.5

13.6

37.7

6-Methyl-heptan-1-ol

14.4

7.3

12.9

156.6

2-Methyl-butane

13.7

0

0

117.4

3,5,5-Trimethylcyclohex-2-enone

16.6

8.2

7.4

150.5

CH3

Isobutylene

CH2 CH3

H3C 434

Isobutyleneoxide

O CH3

H3C

1039 Isobutyric Acid O H3C

OH CH3

435

Isocyanic Acid

O HN 436

Isooctyl Alcohol

CH3 OH

H3C 437

Isopentane

CH3 CH3

H3C 438

Isophorone O CH3 CH3 H3C

7248_A001A.fm Page 432 Wednesday, May 23, 2007 12:38 PM

432

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No. 439

Autonom/ ACD Name

Solvent Name Isoprene (2-Methyl-1,3-Butadiene)

Dispersion Polarity

Hydrogen Molar Bonding Volume

2-Methyl-buta-1,3-diene 14.7

1.4

4.1

100.9

Acetic acid isopropyl ester

14.9

4.5

8.2

117.1

Isopropylamine

14.8

4.4

6.6

86.8

Isopropyl-benzene

18.1

1.2

1.2

139.1

2-Chloro-propane

15.9

8.3

2.1

91.7

2-Isopropoxy-propane

13.7

3.9

2.3

140.9

Hexadecanoic acid isopropyl ester

14.3

3.9

3.7

330.0

3-Methyl-butyraldehyde 14.7

9.5

5.0

106.0

3-Methyl-butyric acid

4.1

10.7

111.0

CH3 CH2

H2C 440

Isopropyl Acetate

CH3

H3C 441

O CH3

O

Isopropyl Amine (2-Propan Amine)

CH3 NH2

H3C 878

Isopropyl Benzene (Cumene) CH3

H3C

442

Isopropyl Chloride (2-Chloro Propane)

Cl CH3

H3C 443

Isopropyl Ether

CH3

Isopropyl Palmitate H3C

CH3

O CH3

445

CH3

O

H3C 444

CH3

O

Isovaleraldehyde

CH3

O

H3C 1040 Isovaleric Acid

CH3 H3C

O OH

16.4

7248_A001A.fm Page 433 Wednesday, May 23, 2007 12:38 PM

Appendix A: Table A.1

433

TABLE A.1 (CONTINUED) No. 446

Autonom/ ACD Name

Solvent Name Isoxazole

Dispersion Polarity

Hydrogen Molar Bonding Volume

Isoxazole

18.8

13.4

11.2

64.1

Ethenone

15.4

7.3

5.8

53.0

2-Hydroxy-propionic acid

17.0

8.3

28.4

72.1

2-Methyl-acrylic acid dodecyl ester

14.4

2.2

5.1

293.1

Furan-2,5-dione

20.2

18.1

12.6

66.3

Malononitrile

17.7

18.4

6.7

55.5

(4-Chloro-phenoxy)acetic acid 2dimethylamino-ethyl ester

16.0

6.2

9.0

198.3

3,6-Dimethyl-4,5,6,7tetrahydro-benzofuran

17.2

4.6

5.6

158.0

(1R,2S,5R)-2-Isopropyl- 16.6 5-methyl-cyclohexanol

4.7

10.6

175.6

N O 447

Ketene

H2C 709

O

Lactic Acid (DL) O CH3

HO OH

448

Lauryl Methacrylate CH3 O

H2C

CH3

O

980

Maleic Anhydride

O

449

O

O

Malononitrile

N

N

1208 Meclofenoxate (Base Only) O

CH3

O O

N

CH3

Cl

1129 Menthofuran CH3 O

CH3

1132 L-Menthol H3C CH3 H3C HO

7248_A001A.fm Page 434 Wednesday, May 23, 2007 12:38 PM

434

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No.

Solvent Name

1130 L-Menthone

Autonom/ ACD Name (2S,5R)-2-Isopropyl-5methyl-cyclohexanone

O

Dispersion Polarity

Hydrogen Molar Bonding Volume

17.0

8.1

4.4

172.3

Acetic acid (1R,2S,5R)- 16.8 2-isopropyl-5-methylcyclohexyl ester

4.7

4.9

215.8

4-Methyl-pent-3-en-2one

16.4

6.1

6.1

115.6

Mesitylene

18.0

0

0.6

139.8

2-Methyl-propenal

15.7

11.1

7.4

83.3

2-Methyl-acrylamide

15.8

11.0

11.6

76.0

2-Methyl-acrylic acid

15.8

2.8

10.2

84.8

2-Methyl-acrylonitrile

15.8

15.1

5.4

83.9

H3C CH3 H3C

1131 L-Menthyl Acetate CH3

H3C O

O CH3 CH3

450

Mesityl Oxide

O

CH3

CH3

H3C 451

Mesitylene CH3

CH3

H3C

452

Methacrylaldehyde

CH3

O

H2C 453

Methacrylamide CH3 NH2

H2C O

454

Methacrylic Acid CH3 OH

H2C O

455

Methacrylonitrile

CH3 H2C

N

7248_A001A.fm Page 435 Wednesday, May 23, 2007 12:38 PM

Appendix A: Table A.1

435

TABLE A.1 (CONTINUED) No. 456

Solvent Name Methanol

4-Methoxy Acetophenone CH3

O

H3C

781

Dispersion Polarity

Hydrogen Molar Bonding Volume

Methanol

15.1

12.3

22.3

40.7

1-(4-Methoxy-phenyl)ethanone

18.9

11.2

7.0

137.8

4-Methoxy-benzonitrile

19.4

16.7

5.4

121.0

3-Methoxy-butan-1-ol

15.3

5.4

13.6

113.2

Acetic acid 3-methoxybutyl ester

15.3

4.1

8.1

153.9

2-Methoxy-tetrahydropyran

17.2

6.6

6.0

105.1

1-Methoxy-buta-1,3diene

15.5

8.3

5.4

101.8

2-Methoxy[1,3]dioxolane

17.8

8.4

7.7

95.3

OH

H3C 793

Autonom/ ACD Name

O

4-Methoxy Benzonitrile N

H3C

O

1003 3-Methoxy Butanol

O

CH3

OH

H3C

1001 3-Methoxy Butyl Acetate

2-Methoxy Tetrahydropyrane

O 457

O

CH3

1-Methoxy-1,3-Butadiene

H3C 992

CH3

O

H3C 996

CH3

O

O

O

CH2

2-Methoxy-1,3-Dioxolane O H3C

O O

7248_A001A.fm Page 436 Wednesday, May 23, 2007 12:38 PM

436

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No.

Solvent Name

1021 1-Methoxy-2-Nitrobenzene CH3 O O

Autonom/ ACD Name

Dispersion Polarity

Hydrogen Molar Bonding Volume

1-Methoxy-2-nitrobenzene

19.6

16.3

5.5

123.5

4-Methoxy-4-methylpentan-2-one

15.3

6.0

5.9

143.5

2-Methoxy-phenol

18.0

8.2

13.3

109.5

3-Methoxy-propionitrile 16.6

14.4

7.8

91.1

Acetate 3-methoxypropyl-ammonium;

17.2

22.5

23.5

(Z)-But-2-enoic acid

16.8

5.2

12.4

83.9

1-Methoxy-propene

15.0

4.3

5.7

93.5

N-Methyl-acetamide

16.9

18.7

13.9

76.9

+

N

458

Methoxyhexanone (Pentoxone) O

CH3 H3C

H3C

459

O

CH3

O

o-Methoxyphenol (Guaiacol) OH O

460

CH3

3-Methoxypropionitrile

H3C

N

O

1192 3-Methoxypropyl Amine/Acetic Acid O

+

NH3

O

O

461

2-Methyl (cis) Acrylic Acid

OH CH3 463

Methyl 1-Propenyl Ether

H3C 958

O

O

CH3

N-Methyl Acetamide

O H3C

N

CH3

7248_A001A.fm Page 437 Wednesday, May 23, 2007 12:38 PM

Appendix A: Table A.1

437

TABLE A.1 (CONTINUED) No. 464

Autonom/ ACD Name

Solvent Name Methyl Acetate

Dispersion Polarity

Hydrogen Molar Bonding Volume

Acetic acid methyl ester 15.5

7.2

7.6

79.7

3-Oxo-butyric acid methyl ester

16.4

8.6

8.9

108.3

Propyne

15.1

3.8

9.2

59.6

Acrylic acid methyl ester 15.3

6.7

9.4

90.3

But-2-en-1-ol

16.0

6.0

15.5

84.4

2-Methyl-but-3enenitrile

16.4

11.3

5.1

97.7

Methylamine

13.0

7.3

17.3

44.4

Acetic acid 1,3dimethyl-butyl ester

15.2

3.1

6.8

167.4

Benzoic acid methyl ester

18.9

8.2

4.7

124.9

O O

H3C

CH3

1037 Methyl Aceto Acetate

O

O

O

H3C 465

Methyl Acetylene

HC 467

CH3

CH3

Methyl Acrylate

O

H2C

CH3

O 468

3-Methyl Allyl Alcohol

H3C 469

OH

Methyl Allyl Cyanide

N H2C CH3 470

Methyl Amine

H3C 471

NH2

Methyl Amyl Acetate

H3C

O CH3

CH3 472

Methyl Benzoate* O O

CH3

O CH3

7248_A001A.fm Page 438 Wednesday, May 23, 2007 12:38 PM

438

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No. 473

Solvent Name Methyl Bromide

Dispersion Polarity

Hydrogen Molar Bonding Volume

Bromomethane

17.0

8.8

2.6

56.8

Hexan-2-one

15.3

6.1

4.1

123.6

Chloromethane

15.3

6.1

3.9

55.4

Methyl chloridocarbonate

16.3

9.5

8.5

77.3

Cyano-acetic acid methyl ester

16.8

14.8

9.1

88.3

Methyl-cyclohexane

16.0

0

1.0

128.3

1-Methyl-cyclohexanol

17.1

6.4

12.5

123.4

3-Methyl-cyclohexanol

17.2

6.4

12.5

124.8

4-Methyl-cyclohexanol

17.2

6.3

12.5

125.5

Br

H3C 474

Autonom/ ACD Name

Methyl Butyl Ketone

O CH3

H3C 475

Methyl Chloride

Cl

H3C 476

Methyl Chloroformate

Cl

O

CH3

O 1041 Methyl Cyanoacetate

O N 477

CH3

O

Methyl Cyclohexane CH3

1030 1-Methyl Cyclohexanol

H3C

OH

1031 3-Methyl Cyclohexanol (mix) OH

CH3

1032 4-Methyl Cyclohexanol (mix) OH

CH3

7248_A001A.fm Page 439 Wednesday, May 23, 2007 12:38 PM

Appendix A: Table A.1

439

TABLE A.1 (CONTINUED) No.

Solvent Name

1033 2-Methyl Cyclohexanol (mix)

Autonom/ ACD Name

Dispersion Polarity

Hydrogen Molar Bonding Volume

2-Methyl-cyclohexanol

17.1

6.5

12.5

124.1

3-Methylcyclohexanone

17.7

6.3

4.7

122.5

2-Methylcyclohexanone

17.6

6.3

4.7

121.3

Methoxy-ethane

14.7

4.9

6.2

84.1

Butan-2-one

16.0

9.0

5.1

90.1

Butan-2-one oxime

14.7

4.9

7.8

94.8

N-Methyl-formamide

17.4

18.8

15.9

59.1

Formic acid methyl ester 15.3

8.4

10.2

62.2

OH CH3

478

3-Methyl Cyclohexanone O

CH3

479

2-Methyl Cyclohexanone O CH3

480

Methyl Ethyl Ether

481

CH3

O

H3C

Methyl Ethyl Ketone

O CH3

H3C 482

Methyl Ethyl Ketoxime

N

OH

CH3

H3C 965

N-Methyl Formamide

O N 483

CH3

Methyl Formate

O

O

CH3

7248_A001A.fm Page 440 Wednesday, May 23, 2007 12:38 PM

440

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No. 995

Solvent Name 2-Methylfuran

Dispersion Polarity

Hydrogen Molar Bonding Volume

2-Methyl-furan

17.3

2.8

7.4

89.7

Furan-2-carboxylic acid methyl ester

17.4

6.9

9.7

107.0

2-Oxo-propionaldehyde

15.5

16.1

9.7

68.9

Methyl-hydrazine

16.2

8.7

14.8

52.7

Methyl-hydroperoxide

15.0

15.0

30.0

24.1

1-Methyl-1H-imidazole

19.7

15.6

11.2

79.5

Iodomethane

17.5

7.7

5.3

62.3

5-Methyl-hexan-2-one

16.0

5.7

4.1

142.8

4-Methyl-pentan-2-ol

15.4

3.3

12.3

127.2

CH3

O 998

Autonom/ ACD Name

Methyl Furoate

O O

CH3

O 484

Methyl Glyoxal O CH3 O

485

Methyl Hydrazine

N

H3C 486

Methyl Hydroperoxide

O

H3C 487

NH2

OH

1-Methyl Imidazole N N CH3

488

Methyl Iodide

I

H3C 489

Methyl Isoamyl Ketone O CH3

H3C

CH3

490

Methyl Isobutyl Carbinol

HO

CH3 CH3

CH3

7248_A001A.fm Page 441 Wednesday, May 23, 2007 12:38 PM

Appendix A: Table A.1

441

TABLE A.1 (CONTINUED) No. 491

Methyl Isobutyl Ketone

O

Hydrogen Molar Bonding Volume

4-Methyl-pentan-2-one

15.3

6.1

4.1

125.8

Isocyanatomethane

15.6

7.3

2.5

61.8

3-Methyl-but-3-en-2one

15.9

12.1

4.5

99.3

Isothiocyanatomethane

17.3

16.2

10.1

68.4

3-Methyl-isoxazole

19.4

14.8

11.8

57.7

Methanethiol

16.6

7.7

8.6

54.1

2-Methyl-acrylic acid methyl ester

15.8

6.5

5.4

106.1

4-Methyl-morpholine 4- 19.0 oxide

16.1

10.2

97.6

5.7

4.1

139.8

CH3

Methyl Isocyanate

N

H3C 493

Dispersion Polarity

CH3 CH3

492

Autonom/ ACD Name

Solvent Name

O

Methyl Isopropenyl Ketone O

CH2

H3C

CH3

494

Methyl Isothiocyanate

N

H3C 495

S

3-Methyl Isoxazole

CH3 N O 496

Methyl Mercaptan

SH

H3C 497

Methyl Methacrylate CH3 O

H2C

CH3

O

1106 N-Methyl Morpholine-N-Oxide O +

N H3C

498

O

Methyl n-Amyl Ketone

Heptan-2-one

O H3C

CH3

16.2

7248_A001A.fm Page 442 Wednesday, May 23, 2007 12:38 PM

442

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No. 499

Solvent Name Methyl n-Propyl Ketone

Autonom/ ACD Name

Dispersion Polarity

Hydrogen Molar Bonding Volume

Pentan-2-one

16.0

7.6

4.7

106.7

1-Methyl-naphthalene

20.6

0.8

4.7

138.8

Nitrooxymethane

15.8

14.0

4.8

63.8

(Z)-Octadec-9-enoic acid methyl ester

14.5

3.9

3.7

340.0

Methylsulfanyl-benzene

19.6

4.8

4.7

117.4

Methanesulfonylbenzene

20.0

16.9

7.8

124.6

Methylphosphonic difluoride

14.0

14.0

8.4

73.9

Propionic acid methyl ester

15.5

6.5

7.7

96.8

O H3C

CH3

500

1-Methyl Naphthalene CH3

501

Methyl Nitrate

O

H3C

+

N

O

O 502

Methyl Oleate H3C

CH3 O O

879

Methyl Phenyl Sulfide

S

CH3

1076 Methyl Phenyl Sulfone CH3 O

841

O

S

Methyl Phosphonic Difluoride

F H3C P F 503

O

Methyl Propionate

O

H3C O

CH3

7248_A001A.fm Page 443 Wednesday, May 23, 2007 12:38 PM

Appendix A: Table A.1

443

TABLE A.1 (CONTINUED) No. 902

Solvent Name 2-Methyl Pyrazine

Autonom/ ACD Name

Dispersion Polarity

Hydrogen Molar Bonding Volume

2-Methyl-pyrazine

18.3

12.3

10.5

91.4

1-Methyl-pyrrolidine

17.0

2.8

6.9

104.0

2-Hydroxy-benzoic acid methyl ester

18.1

8.0

13.9

129.6

Methyl-silane

15.5

3.3

0

71.0

1-Methyl-2-vinylbenzene

18.5

2.4

2.4

130.0

3-Methyl-tetrahydrothiophene 1,1-dioxide

19.4

17.4

5.3

112.7

2-Methyl-tetrahydrofuran

16.9

5.0

4.3

100.2

Thiocyanatomethane

17.3

15.0

6.0

68.5

Thiocyanatomethane

17.4

15.0

8.7

68.5

N

H3C

N 986

N-Methyl Pyrrolidine CH3 N

504

Methyl Salicylate O

OH

O

740

Methyl Silane

SiH3

H3C 886

CH3

alfa-Methyl Styrene CH3 CH2

506

Methyl Sulfolane

O

S

CH3

O 997

2-Methyl Tetrahydrofuran

O

507

Methyl Thiocyanate

H3C 968

CH3

S N

Methyl Thiocyanate

N

S

CH3

7248_A001A.fm Page 444 Wednesday, May 23, 2007 12:38 PM

444

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No. 508

Solvent Name Methyl Vinyl Ether

H3C 509

Autonom/ ACD Name

Dispersion Polarity

Hydrogen Molar Bonding Volume

Methoxy-ethene

14.9

5.3

6.3

75.2

But-3-en-2-one

15.6

12.5

5.0

81.2

2-Methoxy-propene

14.8

4.2

5.6

96.1

Methylsulfanyl-ethene

16.4

4.9

6.0

82.1

Methylsulfonyl-ethene

16.8

19.6

4.8

87.6

2-Methyl-butan-1-ol

16.0

5.1

14.3

109.5

3-Methyl-but-1-ene

14.0

1.4

3.8

112.9

2-Methyl-but-1-ene

14.2

1.8

2.3

108.7

2-Chloro-but-2-enal

17.1

10.6

7.3

91.7

CH2

O

Methyl Vinyl Ketone

O CH2

H3C 510

1-Methyl Vinyl Methyl Ether

CH3 O

H2C 511

Methyl Vinyl Sulfide

H3C 512

CH3

CH2

S

Methyl Vinyl Sulfone

O H3C 462

CH2

S

O

2-Methyl-1-Butanol

H3C

OH CH3

513

3-Methyl-1-Butene

CH3 CH2

H3C 514

2-Methyl-1-Butene

CH2

H3C

CH3 515

2-Methyl-1-Chloro Acrolein

Cl

H3C

O

7248_A001A.fm Page 445 Wednesday, May 23, 2007 12:38 PM

Appendix A: Table A.1

445

TABLE A.1 (CONTINUED) No.

Solvent Name

1028 2-Methyl-1-Pentanol

Autonom/ ACD Name

Dispersion Polarity

Hydrogen Molar Bonding Volume

2-Methyl-pentan-1-ol

15.8

4.9

13.5

124.0

2-Methyl-propan-1-ol

15.1

5.7

15.9

92.8

1-Methoxy-propyne

15.7

6.3

5.9

85.4

3-Methyl-buta-1,2-diene 15.1

2.5

4.5

99.7

2-Methyl-[1,3]dioxolane 17.3

4.8

5.8

89.8

2-Methyl-butan-2-ol

15.3

6.1

13.3

109.6

3-Methyl-butan-2-ol

15.6

5.2

13.4

107.7

2-Methyl-but-2-ene

14.3

2.0

3.9

106.7

CH3 OH

H3C 516

2-Methyl-1-Propanol

CH3 OH

H3C 517

Methyl-1-Propynyl Ether

H3C

O CH3

518

3-Methyl-1,2-Butadiene

CH3 H2C CH3 519

2-Methyl-1,3-Dioxolane

O

O CH3

732

2-Methyl-2-Butanol

OH H3C

CH3

CH3

1029 3-Methyl-2-Butanol CH3

CH3

H3C OH

520

2-Methyl-2-Butene CH3 H3C CH3

7248_A001A.fm Page 446 Wednesday, May 23, 2007 12:38 PM

446

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No. 521

Solvent Name N-Methyl-2-Pyrrolidone O

1002 3-Methyl-3-Methoxy Butyl Acetate O

O

972

Dispersion Polarity

Hydrogen Molar Bonding Volume

1-Methyl-pyrrolidin-2one

18.0

12.3

7.2

96.5

Acetic acid 3-methoxy3-methyl-butyl ester

15.3

3.8

7.7

168.6

Toluene-4-sulfonic acid methyl ester

19.6

15.3

3.8

152.6

4-Methyl-benzoic acid methyl ester

19.0

6.5

3.8

140.4

2-Methoxy-2-methylpropane

14.8

4.3

5.0

119.8

Dimethoxy-methane

15.0

1.8

8.6

169.4

1-Isocyanato-4-(4isocyanatobenzyl) benzene

19.5

4.1

1.7

212.1

Methyl-phenyl-amine

19.5

6.0

11.5

108.4

CH3

N

CH3

CH3 CH3

O

H3C

Autonom/ ACD Name

Methyl-4-Toluenesulfonate CH3

H3C O

S

O

O

796

Methyl-p-Toluate CH3 O CH3 O

522

Methyl-t-Butyl Ether

H3C

H3C O

CH3

CH3 523

Methylal (Dimethoxymethane)

H3C

O

O

CH3

1217 4,4'-Methylenebis(Phenylisocyanate)

951

O

O

N

N

N-Methylaniline N

CH3

7248_A001A.fm Page 447 Wednesday, May 23, 2007 12:38 PM

Appendix A: Table A.1

447

TABLE A.1 (CONTINUED) No. 524

Solvent Name Methylene Dichloride

Methylene Diiodide

I 888

Dispersion Polarity

Hydrogen Molar Bonding Volume

Dichloro-methane

18.2

6.3

6.1

63.9

Diiodo-methane

17.8

3.9

5.5

80.5

Benzo[1,3]dioxole

19.0

6.7

5.9

114.8

Morpholine

18.8

4.9

9.2

87.1

Acetate morpholin-4ium;

17.2

20.3

18.4

Tetramethyl-thiourea

17.3

6.0

10.5

132.2

17.9

0.7

1.8

181.8

Cl

Cl 525

Autonom/ ACD Name

I

1,2-Methylenedioxybenzene

O O 526

Morpholine

O

N 1193 Morpholine/Acetic Acid O O +

N H2

844

O

N,N,N,N-Tetramethylthiourea S

H3C

CH3

N

N

CH3

CH3

529

Naphtha.High-Flash

530

Naphthalene

Naphthalene

19.2

2.0

5.9

111.5

896

1-Naphthol

Naphthalen-1-ol

19.7

6.3

12.3

131.7

OH

7248_A001A.fm Page 448 Wednesday, May 23, 2007 12:38 PM

448

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No.

Solvent Name

1198 Nicotine H3C N

Autonom/ ACD Name

Dispersion Polarity

Hydrogen Molar Bonding Volume

3-((S)-1-Methylpyrrolidin-2-yl)pyridine

18.5

7.8

6.5

160.7

1-Methyl-4-nitrobenzene

20.1

9.6

3.9

98.5

2-Nitro-propan-1-ol

16.1

13.7

15.4

88.8

3-Nitro-phenylamine

21.2

18.7

10.3

96.1

Nitro-benzene

20.0

8.6

4.1

102.7

1-Chloro-4-nitrobenzene

20.0

8.8

3.9

121.2

Nitro-ethane

16.0

15.5

4.5

71.5

Nitro-ethene

16.3

16.6

5.0

59.9

N

784

p-Nitro Toluene CH3

+

O

N

O

1092 2-Nitro-1-Propanol O

O

+

N

OH

H3C

939

3-Nitroaniline NH2

N* O

531

O

Nitrobenzene

+

O

973

N

O

4-Nitrochlorobenzene Cl

+

O

532

N

O

Nitroethane

H3C +

O 533

N

O

Nitroethylene

H2C +

O

N

O

7248_A001A.fm Page 449 Wednesday, May 23, 2007 12:38 PM

Appendix A: Table A.1

449

TABLE A.1 (CONTINUED) No.

Autonom/ ACD Name

Solvent Name

1112 Nitroglycerin (Glyceryl Trinitrate) O

O

Dispersion Polarity

Hydrogen Molar Bonding Volume

1,2,3-Tris-nitrooxypropane

16.2

17.8

5.9

142.5

Nitromethane

15.8

18.8

5.1

54.3

4-Nitro-phenol

20.4

20.9

15.1

93.1

1-Nitro-propane

16.6

12.3

5.5

88.4

2-Nitro-propane

16.2

12.1

4.1

86.9

Nitroso-benzene

20.0

12.7

4.0

89.3

2-Nitro-thiophene

19.7

16.2

8.2

94.6

Nonane

15.7

0

Nonane-1,9-diol

15.7

7.0

+

+

N

O

O

O O

N

O

O

+

N

O

534

Nitromethane

CH3 +

N

O 936

O

4-Nitrophenol OH

+

O

535

N

O

1-Nitropropane CH3

+

N

O

536

O

2-Nitropropane

CH3

H3C +

O

N

O

1073 Nitrosobenzene

N

984

O

2-Nitrothiophene

+

S

O

N

O 537

Nonane

H3C

179.7

15.1

170.5

CH3

1095 1,9-Nonanediol HO

0

OH

7248_A001A.fm Page 450 Wednesday, May 23, 2007 12:38 PM

450

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No. 932

Autonom/ ACD Name

Solvent Name 1-Nonanol H3C

Hydrogen Molar Bonding Volume

Nonan-1-ol

16.0

4.8

10.6

174.4

Non-1-ene

15.4

1.0

2.2

170.5

4-Nonyl-phenol

16.5

4.1

9.2

231.0

2-(4-Nonyl-phenoxy)ethanol

16.7

10.2

8.4

275.0

2-Methylamino-1phenyl-propan-1-ol

18.0

10.7

24.1

141.9

Octadecane

16.4

0

0

326.9

Octane

15.5

0

0

163.5

Octanoic acid

15.1

3.3

8.2

159.0

Octan-1-ol

16.0

5.0

11.9

157.7

Octan-2-ol

16.1

4.9

11.0

159.1

Oct-1-ene

15.3

1.0

2.4

158.0

OH

1070 1-Nonene

CH2

H3C 538

Dispersion Polarity

Nonyl Phenol HO

CH3

539

Nonyl Phenoxy Ethanol

1209 Norephedrin OH CH3 H3C

923

N

Octadecane CH3

H3C

540

Octane

CH3

H3C 541

Octanoic Acid

O OH

H3C 542

1-Octanol*

OH

H3C 543

2-Octanol

CH3

H3C OH 544

1-Octene

HH33CC

CH3

CH2

7248_A001A.fm Page 451 Wednesday, May 23, 2007 12:38 PM

Appendix A: Table A.1

451

TABLE A.1 (CONTINUED) No.

Autonom/ ACD Name

Solvent Name

1065 Octyl Acetate

Dispersion Polarity

Hydrogen Molar Bonding Volume

Acetic acid octyl ester

15.8

2.9

5.1

196.0

(Z)-Octadec-9-enoic acid

16.0

2.8

6.2

317.0

(Z)-Octadec-9-en-1-ol

14.3

2.6

8.0

316.0

Oxalic acid

17.0

14.3

22.0

47.4

Oxalyl dichloride

16.1

3.8

7.5

85.8

Oxetane

18.0

9.1

5.4

65.0

2-(2-Oxo-pyrrolidin-1yl)-acetamide

17.5

15.6

11.2

116.6

19.8

4.2

0

24.0

17.8

10.5

13.9

151.2

O

545

CH3

O

H3C

Oleic Acid* O HO

CH3

546

Oleyl Alcohol HO

CH3

1156 Oxalic Acid O OH

HO O

547

Oxalylchloride O Cl

Cl O

1026 Oxetane (Trimethylene Oxide)

O 1210 2-Oxopyrrolidinacetamid

O O N

NH2

1169 Ozone +

O

O

O

1211 Paracetamol

N-(4-Hydroxy-phenyl)acetamide

HO CH3 N

O

7248_A001A.fm Page 452 Wednesday, May 23, 2007 12:38 PM

452

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No. 978

Solvent Name Paraldehyde

2,4,6-Trimethyl[1,3,5]trioxane

CH3

Dispersion Polarity

Hydrogen Molar Bonding Volume

16.6

7.5

7.3

132.4

1,1'-Dimethyl-4,4'19.5 bipyridinium dichloride

8.8

5.9

205.7

1,1,1,2,2-Pentachloroethane

18.2

3.2

2.4

120.9

1,1,2,2,3-Pentachlorocyclopropane

18.5

10.5

3.7

128.5

2,3,4,5,6-Pentachlorophenol

21.5

6.9

12.8

134.7

(E)-Penta-1,3-diene

14.7

2.5

5.0

101.7

1-Pentafluorophenylethanone

19.3

8.1

5.4

185.5

Tetrahydro-thiopyran

18.5

6.3

8.9

103.6

Pentanal

15.7

9.4

5.8

106.4

O

O

CH3

O

H3C

1114 Paraquat Cl

H3C

Autonom/ ACD Name

Cl

+

+

N

N

CH3

1161 Pentachloro Ethane

Cl

Cl

Cl

Cl

Cl

818

Pentachlorocyclopropane Cl Cl

Cl

Cl

Cl

788

Pentachlorophenol OH Cl

Cl

Cl

Cl Cl

548

1,3-Pentadiene (Trans)

CH3

H2C 821

Pentafluorobenzophenone F

O

F

CH3 F

F F

549

Pentamethylene Sulfide

S 1061 Pentanal (Valeraldehyde)

H3C

O

7248_A001A.fm Page 453 Wednesday, May 23, 2007 12:38 PM

Appendix A: Table A.1

453

TABLE A.1 (CONTINUED) No. 550

Pentane

H3C 551

Autonom/ ACD Name

Solvent Name

Dispersion Polarity

Hydrogen Molar Bonding Volume

Pentane

14.5

0

0

116.2

Pentane-2,4-dione

17.1

9.0

4.1

103.1

Pentanoic acid

15.0

4.1

10.3

109.2

Pentan-1-ol

15.9

5.9

13.9

108.6

Pentan-2-ol

15.6

6.4

13.3

109.6

Pent-4-enal

15.5

8.1

6.8

98.7

Pent-1-ene

13.9

1.4

3.8

110.4

Propionic acid pentyl ester

15.8

5.2

5.7

165.3

1,1,2,2,3,4,4,5,5,6Decafluoro-3,6-bistrifluoromethylcyclohexane

12.4

0

0

217.4

CH3

2,4-Pentanedione

O

O

CH3

H3C 1023 Pentanoic Acid

O H3C 552

OH

1-Pentanol*

OH

H3C 733

2-Pentanol

OH CH3

H3C 553

4-Pentenal

O

H2C 554

1-Pentene

CH2

H3C 1005 n-Pentyl Propionate

O

H3C

CH3

O 556

Perfluoro Dimethylcyclohexane F F

F

FF

F F

F

F F

F F F

F

F F

7248_A001A.fm Page 454 Wednesday, May 23, 2007 12:38 PM

454

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No. 555

557

Perfluoro Ethylene (Tetrafluoro Ethylene)

F

F

F

F

Perfluoroheptane F F F

558

Autonom/ ACD Name

Solvent Name

F

F

F

F

F

F

F

F

F F F

F

F

F

F F F F

F FF

975

F

9,10-Phenanthrenequinone O

Phenetole (Ethyl Phenyl Ether)

H3C

559

1,1,2,2-Tetrafluoroethene

15.1

0

0

65.8

1,1,1,2,2,3,3,4,4,5,5,6,6, 7,7,7-Hexadecafluoroheptane

12.0

0

0

227.3

1,1,2,2,3,3,4,4,5,5,6Undecafluoro-6trifluoromethylcyclohexane

12.4

0

0

196.0

Phenanthrene-9,10dione

20.3

17.1

4.8

148.2

Ethoxy-benzene

18.4

4.5

4.0

127.3

Phenol

18.0

5.9

14.9

87.5

4-[1-(4Hydroxyphenyl)-1methylethyl]phenol

19.2

5.9

13.8

207.5

4-[(4Hydroxyphenyl)sulfon yl]phenol

20.0

14.6

16.3

183.2

F F

O

880

Hydrogen Molar Bonding Volume

F

Perfluoromethylcyclohexane F F

Dispersion Polarity

O

Phenol OH

1100 Bisphenol A OH

H3C

OH CH3

1155 Bisphenol-S OH

O

S O

OH

7248_A001A.fm Page 455 Wednesday, May 23, 2007 12:38 PM

Appendix A: Table A.1

455

TABLE A.1 (CONTINUED) No. 560

2-Phenoxy Ethanol

Bis-(m-Phenoxyphenyl) Ether

Hydrogen Molar Bonding Volume

2-Phenoxy-ethanol

17.8

5.7

14.3

124.7

1-Phenoxy-3-(3phenoxyphenoxy)benz ene

19.6

3.1

5.1

373.0

Acetic acid phenyl ester 19.8

5.2

6.4

126.9

O

O

O

881

Dispersion Polarity

OH

O 561

Autonom/ ACD Name

Solvent Name

Phenyl Acetate

O H3C 851

O

Phenyl Acetonitrile

Phenyl-acetonitrile

19.5

12.3

3.8

114.9

Ethynyl-benzene

18.8

2.8

4.0

109.1

2-Phenyl-ethanol

19.0

5.8

12.8

120.0

Phenyl-hydrazine

20.4

6.5

14.0

98.5

Carbonyl dichloride

16.4

5.3

5.3

71.7

N

1168 Phenyl Acetylene CH

1074 2-Phenyl Ethanol

OH

1047 Phenylhydrazine

N

562

Phosgene

Cl

Cl O

NH2

7248_A001A.fm Page 456 Wednesday, May 23, 2007 12:38 PM

456

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No.

Solvent Name

1216 Phosphoric Acid

Autonom/ ACD Name

Dispersion Polarity

Hydrogen Molar Bonding Volume

Phosphoric acid

14.7

18.6

26.8

52.8

Phosphoric trichloride

18.1

9.3

0

93.2

Phosphorous trichloride

18.4

3.6

0

87.3

Isobenzofuran-1,3-dione 20.6

20.1

10.1

96.8

19.2

7.0

6.0

130.0

15.6

3.0

9.8

155.0

(1S,5S)-2,6,6-Trimethyl- 16.9 bicyclo[3.1.1]hept-2ene

1.8

3.1

159.5

Piperazine

5.6

8.0

97.9

O HO

P

OH

OH 1171 Phosphorous Oxychloride (Phosphoryl Trichloride) O P

Cl

Cl

Cl

563

Phosphorus Trichloride

Cl

Cl

P Cl

564

Phthalic Anhydride O O O

1089 Picric Acid (2,4,6-Trinitrophenol) O +

O

Picric acid

O

OH

+

N

N

O

+

O

N

O

565

Pine Oil

900

2-Pinene (dl) H3C CH3 H

H

CH3

1052 Piperazine

N

N

18.1

7248_A001A.fm Page 457 Wednesday, May 23, 2007 12:38 PM

Appendix A: Table A.1

457

TABLE A.1 (CONTINUED) No.

Solvent Name

1050 Piperidine

Autonom/ ACD Name

Dispersion Polarity

Hydrogen Molar Bonding Volume

Piperidine

17.6

4.5

8.9

98.9

Propa-1,2-diene

15.3

3.0

6.8

60.1

Propane

13.4

0

0

89.5

[1,2]Oxathiolane 2,2dioxide

18.4

16.0

9.0

87.7

Propane-1,3-diol

16.8

13.5

23.2

72.5

Propane-2-thiol

16.3

6.8

6.5

94.1

Propane-1-thiol

16.1

5.8

5.7

90.5

Propan-1-ol

16.0

6.8

17.4

75.2

Propan-2-ol

15.8

6.1

16.4

76.8

Acetic acid prop-2-ynyl ester

16.3

5.2

8.3

98.8

N 566

1,2-Propadiene (Allene)

CH2

H2C 1064 Propane

CH3

H3C 943

1,3-Propane Sultone

O

O S O

1038 1,3-Propanediol (Trimethyleneglycol)

OH

HO 567

2-Propanethiol

SH CH3

H3C 568

1-Propanethiol

SH

H3C 569

1-Propanol

OH

H3C 570

2-Propanol

OH H3C

CH3

1036 Propargyl Acetate

O

HC O

CH3

7248_A001A.fm Page 458 Wednesday, May 23, 2007 12:38 PM

458

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No. 571

Autonom/ ACD Name

Solvent Name Propargylaldehyde

Dispersion Polarity

Hydrogen Molar Bonding Volume

Propynal

16.2

11.9

8.7

60.0

Oxetan-2-one

19.7

18.2

10.3

65.5

Propionaldehyde

15.3

11.1

6.9

73.4

Oxirane-2-carbaldehyde

17.5

13.4

9.8

63.2

Propionamide

16.7

9.8

11.5

78.9

Propionic acid

14.7

5.3

12.4

75.0

Propanoic anhydride

16.2

10.0

8.7

129.4

Propionitrile

15.3

14.3

5.5

70.9

Propionyl chloride

16.1

10.3

5.3

86.9

HC O 572

beta-Propiolactone

O O 1062 Propionaldehyde*

O

H3C 574

Propionaldehyde-2,3-Epoxy*

O

O 575

Propionamide

NH2

H3C

O 576

Propionic Acid

OH

H3C

O 1043 Propionic Anhydride

O

H3C

O

O 577

Propionitrile

H3C 578

CH3

N

Propionylchloride

Cl

H3C O

7248_A001A.fm Page 459 Wednesday, May 23, 2007 12:38 PM

Appendix A: Table A.1

459

TABLE A.1 (CONTINUED) No. 579

Solvent Name n-Propyl Acetate

Dispersion Polarity

Hydrogen Molar Bonding Volume

Acetic acid propyl ester

15.3

4.3

7.6

115.3

Propylamine

16.9

4.9

8.6

83.0

1-Chloro-propane

16.0

7.8

2.0

88.1

Formic acid propyl ester 15.5

7.1

8.6

97.9

2-Methyl-acrylic acid propyl ester

15.5

6.3

6.6

158.8

1-Nitrooxy-propane

15.8

11.3

4.1

99.7

Propene

15.1

1.6

1.5

68.8

4-Methyl-[1,3]dioxolan- 20.0 2-one

18.0

4.1

85.0

9.8

15.3

85.4

CH3

O

H3C

Autonom/ ACD Name

O 580

Propyl Amine

NH2

H3C 581

Propyl Chloride

Cl

H3C 926

Propyl Formate

O

H3C 582

O

Propyl Methacrylate CH3 O

H2C

CH3

O

935

n-Propyl Nitrate

O +

N

O 583

Propylene

CH3

H2C 584

CH3

O

Propylene Carbonate O O

O CH3

746

Propylene Chlorohydrin

OH Cl

CH3

1-Chloro-propan-2-ol

16.8

7248_A001A.fm Page 460 Wednesday, May 23, 2007 12:38 PM

460

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No. 585

Autonom/ ACD Name

Solvent Name Propylene Glycol

Dispersion Polarity

Hydrogen Molar Bonding Volume

Propane-1,2-diol

16.8

9.4

23.3

73.6

1-tert-Butoxy-propan2-ol

15.3

6.1

10.8

151.6

1-Butoxy-propan-2-ol

15.3

4.5

9.2

132.0

1-Ethoxy-propan-2-ol

15.7

6.5

10.5

115.6

Acetic acid 2-ethoxy-1methyl-ethyl ester

15.6

4.3

9.0

155.1

1-Isobutoxy-propan-2-ol 15.1

4.7

9.8

132.2

15.5

6.1

11.0

134.6

15.6

6.3

11.6

93.8

OH H3C OH

586

Propylene Glycol Mono-t-Butyl Ether CH3

CH3

O

HO

CH3

H3C

587

Propylene Glycol Monobutyl Ether

HO

O

CH3

CH3 588

Propylene Glycol Monoethyl Ether

HO

O

CH3

CH3 589

Propylene Glycol Monoethyl Ether Acetate

H3C

O O

590

O CH3

Propylene Glycol Monoisobutyl Ether

HO

CH3

O CH3

857

CH3

CH3

Propylene Glycol Monoisopropyl Ether 1-Isopropoxy-propan-2ol CH 3

HO

O

CH3

CH3

591

Propylene Glycol Monomethyl Ether

HO

O CH3

CH3

1-Methoxy-propan-2-ol

7248_A001A.fm Page 461 Wednesday, May 23, 2007 12:38 PM

Appendix A: Table A.1

461

TABLE A.1 (CONTINUED) No. 592

Propylene Glycol Monomethyl Ether Acetate

H3C

O

O

O 593

Autonom/ ACD Name

Solvent Name

Hydrogen Molar Bonding Volume

Acetic acid 2-methoxy1-methyl-ethyl ester

15.6

5.6

9.8

137.1

1-Phenoxy-propan-2-ol

17.4

5.3

11.5

143.2

1-Propoxy-propan-2-ol

15.8

7.0

9.2

130.3

2-Methyl-aziridine

18.1

8.4

6.5

71.2

2-Methyl-oxirane

15.2

8.6

6.7

67.6

Prop-2-yn-1-ol

16.1

8.8

19.1

Propynenitrile

15.5

17.0

6.3

62.5

(R)-2-Isopropylidene-5methyl-cyclohexanone

17.5

8.9

5.5

162.9

1H-Pyrazole

20.2

10.4

12.4

68.1

CH3

CH3

Propylene Glycol Monophenyl Ether

HO

Dispersion Polarity

O CH3

594

Propylene Glycol Monopropyl Ether

HO

CH3

O CH3

1086 Propylene Imine (2-Methyl Aziridine)

N

H3C 595

Propylene Oxide

O CH3 1091 2-Propyn-1-ol

57.7

HC OH 596

Propynonitrile

N

HC 1133 Pulegone O H3C

CH3 H3C

985

Pyrazole

N N

7248_A001A.fm Page 462 Wednesday, May 23, 2007 12:38 PM

462

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No. 597

Solvent Name Pyridazine

Dispersion Polarity

Hydrogen Molar Bonding Volume

Pyridazine

20.2

17.4

11.7

72.6

Pyridine

19.0

8.8

5.9

80.9

Benzene-1,2,3-triol

20.7

10.7

21.1

87.0

1H-Pyrrole

19.2

7.4

6.7

69.2

Pyrrolidine

17.9

6.5

7.4

83.5

Pyrrolidin-2-one

19.4

17.4

11.3

76.4

2-Oxo-propionitrile

15.9

18.9

8.0

70.9

(R)-(6-Methoxyquinolin-4-yl)((2S,4S,5R)-5-vinyl-1aza-bicyclo[2.2.2]oct2-yl)-methanol

19.0

6.6

11.0

310.7

N

N 598

Autonom/ ACD Name

Pyridine

N 901

Pyrogallol (1,2,3-Trihydroxybenzene) OH OH

OH

600

Pyrrole

N 1051 Pyrrolidine

N

599

2-Pyrrolidone O N

601

Pyruvonitrile

O N H3C 1202 Quinine CH2 H N

HO H O H3C N

7248_A001A.fm Page 463 Wednesday, May 23, 2007 12:38 PM

Appendix A: Table A.1

463

TABLE A.1 (CONTINUED) No. 602

Autonom/ ACD Name

Solvent Name Quinoline

Dispersion Polarity

Hydrogen Molar Bonding Volume

Quinoline

19.8

5.6

5.7

118.0

1,1-Dioxo-1,2-dihydro1lambda*6*benzo[d]isothiazol-3one

21.0

13.9

8.8

206.8

2-Hydroxybenzaldehyde

19.4

10.7

14.7

104.6

2-Hydroxy-benzoic acid 19.4

10.1

17.4

95.7

N 1196 Saccharin O N S O

O

704

Salicylaldehyde O

OH

1204 Salicylic Acid OH

OH

O

1225 Sebacic Acid

Decanedioic acid

17.1

7.1

11.7

167.6

3-(2-Amino-ethyl)-1Hindol-5-ol

18.0

8.2

14.4

144.4

4-((E)-3-Hydroxypropenyl)-2,6dimethoxy-phenol

19.2

7.3

16.1

194.6

3-Methyl-1H-indole

20.0

7.1

6.2

122.6

O O

HO OH

1178 Serotonin NH2 HO N

1101 Sinapyl Alcohol OH

H3C

O

HO O

832

CH3

Skatole CH3

N

7248_A001A.fm Page 464 Wednesday, May 23, 2007 12:38 PM

464

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No.

Autonom/ ACD Name

Solvent Name

1207 Spermidin H2 N

603

NH2

N

Stearic Acid

Dispersion Polarity

Hydrogen Molar Bonding Volume

N*1*-(3-Aminopropyl)-butane-1,4diamine

16.7

11.2

12.0

155.6

Octadecanoic acid

16.3

3.3

5.5

326.0

Vinyl-benzene

18.6

1.0

4.1

115.6

2-Phenyl-oxirane

19.4

5.8

6.6

114.2

Succinaldehyde

16.8

9.8

10.5

81.2

Dihydro-furan-2,5-dione 18.6

19.2

16.6

66.8

Succinonitrile

17.9

16.2

7.9

81.2

4-Aminobenzenesulfonamide

20.0

19.5

10.7

159.5

Tetrahydro-thiophene 1,1-dioxide

20.3

18.2

10.9

95.7

OH

H3C O

604

Styrene

CH2

1014 Styrene Oxide (Phenyl Oxirane) O

605

Succinaldehyde (Butanedial)

O 606

O

Succinic Anhydride

O

O

607

O

Succinonitrile

N N 1077 Sulfanilamide

NH2 O

S

NH2

O 608

Sulfolane

O

S

O

7248_A001A.fm Page 465 Wednesday, May 23, 2007 12:38 PM

Appendix A: Table A.1

465

TABLE A.1 (CONTINUED) No. 609

Solvent Name Sulfur Dicyanide

Autonom/ ACD Name

Dispersion Polarity

Hydrogen Molar Bonding Volume

Cyanic thiocyanate

18.1

13.5

0

60.0

Oxosulfane oxide

15.8

8.4

10.0

44.0

Sulfuryl dichloride

17.6

7.2

0

81.0

Terephthalic acid

18.8

10.7

166.0

2-tert-Butyl-4-methylphenol

17.3

3.7

10.5

177.6

2,6-Dibromo-4-[1-(3,5dibromo-4hydroxyphenyl)-1methylethyl]phenol

20.2

9.1

13.8

249.5

1,1,2,2-Tetrabromoethane

22.6

5.1

8.2

116.8

1,2,4,5-Tetrachlorobenzene

21.2

10.7

3.4

116.2

S N

N 610

Sulfur Dioxide

S

O

O

1098 Sulfuryl Chloride

O Cl

S

Cl

O 897

Terephthalic Acid

6.1

OH

O

O

HO

1079 2-tert-Butyl-4-Methyl Phenol OH

CH3 CH3 CH3

CH3

1127 Tetrabromo Bisphenol A OH Br

Br

Br OH

H3C CH3 Br

612

708

1,1,2,2-Tetrabromoethane

Br

Br

Br

Br

1,2,4,5-Tetrachlorobenzene Cl Cl

Cl Cl

7248_A001A.fm Page 466 Wednesday, May 23, 2007 12:38 PM

466

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No. 783

2,2,6,6-Tetrachlorocyclohexanone

882

O

Cl

Cl

614

Autonom/ ACD Name

Solvent Name

Cl

Dispersion Polarity

Hydrogen Molar Bonding Volume

2,2,6,6-Tetrachlorocyclohexanone

19.5

14.0

6.3

138.8

1,1,2,2-Tetrachloroethane

18.8

5.1

5.3

105.2

1,1,1,2-Tetrachloroethane

18.0

4.4

4.2

105.0

1,1,2,2-Tetrachloroethene

18.3

5.7

0

101.2

1,1,2,2-Tetrachloropropane

17.9

6.7

3.3

123.7

Tetradecane

16.2

0

0

261.3

Tetradec-1-ene

16.1

0.5

1.9

253.4

13.9

4.3

0.6

224.0

16.8

5.7

8.0

81.7

Cl

1,1,2,2-Tetrachloroethane

Cl

Cl

Cl

Cl

1,1,1,2-Tetrachloroethane

Cl Cl

Cl

Cl 615

613

Tetrachloroethylene

Cl

Cl

Cl

Cl

1,1,2,2-Tetrachloropropane Cl Cl Cl

Cl

922

CH3

n-Tetradecane H3C

CH3

1011 1-Tetradecene CH2

H3C

616

Tetraethylorthosilicate CH3 O O H3C

CH3

Si O O CH3

617

Tetrahydrofuran

O

Tetrahydro-furan

7248_A001A.fm Page 467 Wednesday, May 23, 2007 12:38 PM

Appendix A: Table A.1

467

TABLE A.1 (CONTINUED) No. 999

Solvent Name 2,5-Tetrahydrofuran Dimethanol

HO

OH O

1096 Tetrahydrofurfuryl Alcohol

Autonom/ ACD Name

Dispersion Polarity

Hydrogen Molar Bonding Volume

17.9

8.5

18.9

114.5

(Tetrahydro-furan-2-yl)- 17.8 methanol

8.2

10.2

97.0

(5-Hydroxymethyltetrahydro-furan-2-yl)methanol

OH O 618

Tetrahydronaphthalene

1,2,3,4-Tetrahydronaphthalene

19.6

2.0

2.9

136.0

619

Tetrahydropyran

Tetrahydro-pyran

16.4

6.3

6.0

97.8

Tetrahydro-thiopyran

18.5

6.3

8.9

103.6

1,1,3,3-Tetramethoxypropane

15.0

7.1

6.8

164.7

1,2,3,4-Tetramethylbenzene

18.8

0.5

0.5

148.3

1,2,3,5-Tetramethylbenzene

18.6

0.5

0.5

150.8

O

620

Tetrahydrothiapyran

S

895

1,1,3,3-Tetramethoxypropane

H3C

O

H3C 884

O O

O

CH3

CH3

1,2,3,4-Tetramethylbenzene CH3 CH3 CH3 CH3

885

1,2,3,5-Tetramethylbenzene CH3

CH3 H3C

CH3

7248_A001A.fm Page 468 Wednesday, May 23, 2007 12:38 PM

468

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No. 622

Solvent Name Tetramethylene Sulfide

Autonom/ ACD Name

Dispersion Polarity

Hydrogen Molar Bonding Volume

Tetrahydro-thiophene

18.6

6.7

9.1

88.3

Tetrahydro-thiophene 1,1-dioxide

20.3

18.2

10.9

95.7

Tetrahydro-thiophene 1- 18.2 oxide

11.0

9.1

90.0

S

790

Tetramethylene Sulfone (Sulfolane)

O

623

O

S

Tetramethylene Sulfoxide O S

624

Tetramethylurea

H3C

CH3

CH3

N

N

Tetramethyl-urea

16.7

8.2

11.0

120.4

Tetranitro-methane

15.5

9.9

7.5

119.7

[2,2']Bi[[1,3]dithiolylide 21.0 ne]

8.2

8.4

204.0

Methylsulfanyl-ethane

16.2

5.9

5.3

90.4

Thiirane

19.3

9.1

5.0

58.0

Thiazole

20.5

18.8

10.8

70.9

CH3

O

933

Tetranitromethane O

+

N

O

+

N

N +

O

N

O

890

625

O O

+

O O

2,2',5,5'-Tetrathiafulvalen

S

S

S

S

2-Thiabutane

S

H3C 626

CH3

Thiacyclopropane

S 627

Thiazole*

N S

7248_A001A.fm Page 469 Wednesday, May 23, 2007 12:38 PM

Appendix A: Table A.1

469

TABLE A.1 (CONTINUED) No. 628

Autonom/ ACD Name

Solvent Name Thioacetamide

Dispersion Polarity

Hydrogen Molar Bonding Volume

Thioacetamide

17.5

20.6

20.2

75.0

Thioacetic acid

17.0

6.7

8.9

71.5

Dihydro-thiophen-2-one 19.0

6.9

6.2

86.6

Thiocyanic acid

16.8

8.9

10.9

51.7

2-(2-Hydroxyethylsulfanyl)-ethanol

17.3

8.8

19.8

103.6

Mercapto-acetic acid

16.0

8.6

14.8

69.5

Thionyl dichloride

16.9

6.2

5.9

79.0

Thiophene

18.9

2.4

7.8

79.0

Benzenethiol

20.0

4.5

10.3

102.4

S NH2

H3C 629

Thioacetic Acid

O SH

H3C 630

Gamma-Thiobutyrolactone O S

631

Thiocyanic Acid

N

SH

1034 Thiodiethylenglycol

HO

S

OH

1017 Thioglycolic Acid (Mercapto Acetic Acid)

O SH

HO 632

Thionyl Chloride

Cl

S

Cl

O 633

Thiophene

S 703

Thiophenol SH

7248_A001A.fm Page 470 Wednesday, May 23, 2007 12:38 PM

470

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No. 634

Solvent Name Thiourea

Autonom/ ACD Name

Dispersion Polarity

Hydrogen Molar Bonding Volume

Thiourea

20.0

21.7

14.8

72.8

[1,4]Oxathiane

19.0

6.6

7.7

93.5

2-Isopropyl-5-methylphenol

19.0

4.5

10.8

166.9

(E)-2-Methyl-but-2-enal 16.2

12.9

6.8

96.6

NH2

H2N S 635

1,4-Thioxane

S

O 1203 Thymol CH3

OH H3C

636

CH3

Tigaldehyde

H3C

O CH3

637

Toluene

Toluene

18.0

1.4

2.0

106.8

o-Tolylamine

19.4

5.8

9.4

107.8

2,4-Diisocyanato-1methyl-benzene

19.3

7.9

6.1

143.5

1147 Triacetin (1,2,3-Propanetriol Triacetate) Acetic acid 2-acetoxy-1- 16.5 acetoxymethyl-ethyl CH O ester O

4.5

9.1

188.2

CH3

953

2-Toluidine NH2 CH3

638

Tolylene Diisocyanate O CH3 N

N O

3

O O

O

H3C

O H3C

7248_A001A.fm Page 471 Wednesday, May 23, 2007 12:38 PM

Appendix A: Table A.1

471

TABLE A.1 (CONTINUED) No.

Solvent Name

1137 Tri-n-Butyl Acetyl Citrate CH3

CH3

O O

O

O

Autonom/ ACD Name

Dispersion Polarity

Hydrogen Molar Bonding Volume

3-Acetoxy-3butoxycarbonylpentanedioic acid dibutyl ester

16.7

2.5

7.4

384.3

Tri-n-butyl borate

16.7

1.8

4.6

269.7

3-Butoxycarbonyl-3hydroxy-pentanedioic acid dibutyl ester

16.6

3.8

10.1

345.5

Phosphoric acid tributyl ester

16.3

6.3

4.3

274.0

1H-[1,2,3]Triazole

20.7

8.8

15.0

58.2

2,4,6-Tribromo-phenol

20.6

10.2

14.1

129.7

1,1,2-Tribromo-ethene

18.3

9.4

8.0

97.8

3,3,3-Trichloro-propene

17.7

15.5

3.4

106.2

O CH3

O O O

H3 C

1020 Tri-n-Butyl Borate CH3

O H3C

B

O

CH3

O

1220 Tri-n-Butyl Citrate CH3

CH3

O O

O

O

O

OH O

H3C

641

Tri Butyl Phosphate* CH3

O P

O

O

O H3C

CH3

639

1,2,3-Triazole

N N N 1118 2,4,6-Tribromo Phenol OH Br

Br

Br

640

Tribromo Ethylene*

Br Br 642

Br

3,3,3-Trichloro Propene

Cl

H2C Cl

Cl

7248_A001A.fm Page 472 Wednesday, May 23, 2007 12:38 PM

472

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No. 643

Solvent Name 1,1,2-Trichloro Propene

Autonom/ ACD Name

Dispersion Polarity

Hydrogen Molar Bonding Volume

1,1,2-Trichloro-propene

17.7

15.7

3.4

104.8

1,2,3-Trichloro-propene

17.8

15.7

3.4

105.0

Trichloro-methyl-silane

16.5

6.6

3.5

117.4

Trichloro-acetic acid

18.3

5.8

11.4

100.2

Trichloro-acetonitrile

16.4

7.4

6.1

100.0

1,3,5-Trichloro-2methoxy-benzene

21.0

3.9

7.0

129.0

1,2,4-Trichloro-benzene

20.2

6.0

3.2

125.5

1,1,1-Trichloro-ethane

16.8

4.3

2.0

99.3

Cl Cl

CH3 Cl

644

1,2,3-Trichloro Propene

Cl Cl 981

Cl

Trichloro-Methyl-Silane

Cl H3C

Si Cl Cl

861

Trichloroacetic Acid O Cl

OH

Cl

645

Cl

Trichloroacetonitrile

Cl

Cl

N

Cl 1115 2,4,6-Trichloroanisole O

CH3 Cl

Cl

Cl

701

1,2,4-Trichlorobenzene* Cl Cl

Cl

647

1,1,1-Trichloroethane

Cl H3C

Cl Cl

7248_A001A.fm Page 473 Wednesday, May 23, 2007 12:38 PM

Appendix A: Table A.1

473

TABLE A.1 (CONTINUED) No. 648

Solvent Name 1,1,2-Trichloroethane

Autonom/ ACD Name

Dispersion Polarity

Hydrogen Molar Bonding Volume

1,1,2-Trichloro-ethane

18.2

5.3

6.8

92.9

1,1,2-Trichloro-ethene

18.0

3.1

5.3

90.2

Trichloro-fluoromethane

15.3

2.0

0

92.8

2,4,6-Trichloro-phenol

20.3

5.1

10.8

132.5

1,2,3-Trichloro-propane

17.8

12.3

3.4

106.1

14.2

3.6

3.8

101.1

2,4,5-Trichlorobenzenethiol

21.0

4.5

9.1

145.0

1,1,2-Trichloro-1,2,2trifluoro-ethane

14.7

1.6

0

119.2

Cl Cl

Cl 649

Trichloroethylene

Cl Cl

Cl 650

Trichlorofluoromethane (Freon 11)

F

Cl

Cl

Cl 702

2,4,6-Trichlorophenol OH Cl

Cl

Cl

883

1,2,3-Trichloropropane Cl

Cl

Cl

1174 Trichlorosilane

Cl

Si

Cl

Cl 651

2,4,5-Trichlorothiophenol SH Cl

Cl Cl

652

1,1,2-Trichlorotrifluoroethane (Freon 113)

Cl

F

F

Cl

F

Cl

7248_A001A.fm Page 474 Wednesday, May 23, 2007 12:38 PM

474

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No.

Autonom/ ACD Name

Solvent Name

1160 Triclosan OH

Cl O

Hydrogen Molar Bonding Volume

5-Chloro-2-(2,4dichloro-phenoxy)phenol

20.0

7.7

10.0

263.0

Phosphoric acid m-tolyl ester o-tolyl ester p-tolyl ester

19.0

12.3

4.5

316.0

(1S,2S,4R,6R)-1,7,7Trimethyltricyclo[2.2.1.0*2,6*] heptane

16.9

0

0

161.4

Tridecan-1-ol

16.2

3.1

9.0

242.0

2-[Bis-(2-hydroxyethyl)-amino]-ethanol

17.3

22.4

23.3

133.2

Acetate tris-(2-hydroxy- 17.2 ethyl)-ammonium;

20.3

18.4

Cl

Cl

653

Dispersion Polarity

Tricresyl Phosphate O P

O

O

CH3

O

H3C

899

CH3

Tricyclene CH3

H3C

CH3

H

H H

654

Tridecyl Alcohol* OH

H3C

655

Triethanolamine OH

N

HO

OH

1194 Triethanolamine/Acetic Acid OH

HO H + N

O O

OH

1218 Triethyl Citrate O H3C

OH O

O O

CH3

3-Ethoxycarbonyl-3hydroxy-pentanedioic acid diethyl ester

16.5

4.9

12.0

243.0

Triethyl-amine

17.8

0.4

1.0

138.6

2-[2-(2-Hydroxyethoxy)-ethoxy]ethanol

16.0

12.5

18.6

114.0

O O

CH3

656

Triethylamine

CH3 H3C 657

N

CH3

Triethylene Glycol

HO

O

O

OH

7248_A001A.fm Page 475 Wednesday, May 23, 2007 12:38 PM

Appendix A: Table A.1

475

TABLE A.1 (CONTINUED) No. 856

Triethylene Glycol Monomethyl Ether O

HO

658

Autonom/ ACD Name

Solvent Name

O

O

CH3

Triethylene Glycol Monooleyl Ether* HO

O

O

O

Dispersion Polarity

Hydrogen Molar Bonding Volume

2-[2-(2-Methoxyethoxy)-ethoxy]ethanol

16.2

7.6

12.5

160.0

2-(2-{2-[((Z)-Octadec9-enyl)oxy]-ethoxy}ethoxy)-ethanol

16.0

3.1

8.4

418.5

Phosphoric acid triethyl ester

16.7

11.4

9.2

171.0

2,2,2-Trifluoro-ethanol

15.4

8.3

16.4

72.3

1,2,3-Trifluoro-4-nitrobenzene

19.5

7.7

3.5

122.0

Trifluoromethyl-benzene 17.5

8.8

0

122.9

Trifluoro-acetic acid

15.6

9.9

11.6

74.2

1,1,1-Trifluoro-ethane

14.6

10.7

0

64.6

CH3

659

Triethylphosphate

O

H3C

O

P

H3C 944

CH3

O

O

2,2,2-Trifluoro Ethanol

F F

OH

F 795

2,3,4-Trifluoro Nitrobenzene O

+

N

O F

F F

1158 alpha,alpha,alpha Trifluoro Toluene F

660

F

F

Trifluoroacetic Acid

F

O

F

OH

F 661

1,1,1-Trifluoroethane

F H3C

F F

7248_A001A.fm Page 476 Wednesday, May 23, 2007 12:38 PM

476

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No. 662

Autonom/ ACD Name

Solvent Name Trifluoromethane (Freon 23)

F

Dispersion Polarity

Hydrogen Molar Bonding Volume

Trifluoro-methane

14.4

8.9

6.5

46.1

1-(4-Trifluoromethylphenyl)-ethanone

18.8

6.1

3.5

151.7

1,3-Bis-trifluoromethylbenzene

17.0

6.8

0

155.2

Benzene-1,2,4tricarboxylic acid tris(7-methyl-octyl) ester

16.6

5.7

2.2

602.9

Benzene-1,2,416.6 tricarboxylic acid tris(6-methyl-heptyl) ester

6.0

2.5

553.1

Trimethyl-amine

14.6

3.4

1.8

90.3

Isobutyric acid 3hydroxy-2,2,4trimethyl-pentyl ester

15.1

6.1

9.8

227.4

1,2,4-Trimethyl-benzene 18.0

1.0

1.0

137.3

F

F 847

4-(Trifluoromethyl) Acetophenone CH3

O

F

F

F

1157 1,3-Bis(Trifluoromethyl)Benzene F

F

F

F F

663

F

Triisononyl Trimellilate CH3 H3C

O O

CH3

O

O

CH3

O

O

H3C CH3

664

Triisooctyl Trimellitate H3C

O

CH3

O

O

H3C

665

CH3 O

O

CH3

CH3

O

Trimethyl Amine

CH3 H3C 666

N

CH3

2,2,4-Trimethyl-1,3-Pentanediol Monoisobutyrate CH3

CH3

CH3 O

H3C OH

667

CH3

Trimethylbenzene* CH3 CH3

CH3

CH3 O

7248_A001A.fm Page 477 Wednesday, May 23, 2007 12:38 PM

Appendix A: Table A.1

477

TABLE A.1 (CONTINUED) No. 669

Autonom/ ACD Name

Solvent Name Trimethylenesulfide

Dispersion Polarity

Hydrogen Molar Bonding Volume

Thietane

18.8

7.8

9.4

72.8

2,2,4-Trimethyl-pentane

14.1

0

0

166.1

Phosphoric acid trimethyl ester

16.7

15.9

10.2

115.8

Trinitro-methane

15.5

10.3

7.3

94.6

2-Methyl-1,3,5-trinitrobenzene

19.5

3.7

4.5

137.6

2-Methyl-1,3,5-trinitrobenzene

19.5

10.0

4.5

137.6

Phosphoric acid trioctyl ester

16.2

5.9

4.2

469.8

[1,3,5]Trioxane

18.7

9.2

8.6

77.0

S 670

2,2,4-Trimethylpentane

CH3 H3C 671

CH3

CH3

Trimethylphosphate O

H3C O

CH3

P

O

O

H3C

934

CH3

Trinitomethane O

O +

+

N

O

N

O

+

N

O

918

O

Trinitrotoluene (TNT) P by Dipole Moment O

O

CH3

+

+

N

N

O

O

+

O

919

N

O

Trinitrotoluene (TNT) P by Group Cont. O

O

CH3

+

O

+

N

N

O

+

O

N

O

1151 Trioctylphosphate H3C

O O P

O

O

H3C

855

CH3

1,3,5-Trioxane

O O

O

7248_A001A.fm Page 478 Wednesday, May 23, 2007 12:38 PM

478

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No.

Autonom/ ACD Name

Solvent Name

1150 Triphenyl Phosphate

Dispersion Polarity

Hydrogen Molar Bonding Volume

Phosphoric acid triphenyl ester

20.1

6.4

6.8

271.9

1-[2-(2-Methoxy-1methyl-ethoxy)-1methyl-ethoxy]propan-2-ol

15.3

5.5

10.4

214.0

Undecane

16.0

0

0

212.7

Urea

20.9

18.7

26.4

45.8

Pentanenitrile

15.3

11.0

4.8

103.8

4-Hydroxy-3-methoxybenzaldehyde

19.4

9.8

11.2

144.1

(2-Chloro-ethoxy)ethene

16.3

6.7

5.8

101.7

3-Vinyloxymethylheptane

15.6

3.4

4.2

194.2

(2-Methoxy-ethoxy)ethene

15.9

6.7

6.8

96.2

O O P

O

O

672

Tripropylene Glycol Monomethyl Ether* CH3 HO

O

O

CH3

O

CH3

CH3

1059 Undecane H3C

860

CH3

Urea (Ro = 19.4)

H2N

NH2

O 673

Valeronitrile H3C N

1162 Vanillin (4-Hydroxy-3-Methoxy Benzaldehyde) OH

CH3 O

O

674

Vinyl 2-Chloro Ethyl Ether

H2C 908

O

Cl

Vinyl 2-Ethyl-Hexyl Ether

H2C

CH3

O

H3C 675

Vinyl 2-Methoxy Ethyl Ether

H2C

O

O

CH3

7248_A001A.fm Page 479 Wednesday, May 23, 2007 12:38 PM

Appendix A: Table A.1

479

TABLE A.1 (CONTINUED) No. 676

Autonom/ ACD Name

Solvent Name Vinyl Acetate

Dispersion Polarity

Hydrogen Molar Bonding Volume

Acetic acid vinyl ester

16.0

7.2

5.9

92.6

But-3-enoic acid

16.8

5.2

12.3

85.3

But-1-en-3-yne

15.1

1.7

12.0

74.3

3-Vinyloxy-propene

14.9

6.5

5.3

105.4

Vinylamine

15.7

7.2

11.8

51.8

Bromo-ethene

15.9

6.3

5.4

71.6

1-[2-(2-Vinyloxy16.0 ethoxy)-ethoxy]-butane

5.0

6.0

206.2

1-Vinyloxy-butane

15.2

4.1

5.1

129.4

1-Vinylsulfanyl-butane

16.0

5.0

5.4

136.7

Butyric acid vinyl ester

15.6

3.9

6.9

126.5

O

H3C 677

CH2

O

Vinyl Acetic Acid

CH2

O OH

678

Vinyl Acetylene

HC CH2 679

Vinyl Allyl Ether

H2C 742

Vinyl Amine

H2C 680

O

O

CH3

S

Vinyl Butyrate

O H3C

CH2

CH3

O

Vinyl Butyl Sulfide

H2C 682

O

Vinyl Butyl Ether

H2C 681

Br

Vinyl Butyl Carbitol H3C

964

NH2

Vinyl Bromide

H2C 914

CH2

O

O

CH2

7248_A001A.fm Page 480 Wednesday, May 23, 2007 12:38 PM

480

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No. 906

Vinyl Carbitol

O

HO 683

Vinyl Crotonate

O

H3C

O

O

O

1024 Vinyl Ethyl Ether

H2C 686

687

O

7.5

13.8

129.2

Chloro-ethene

16.0

6.5

2.4

68.7

(E)-But-2-enoic acid vinyl ester

15.9

5.0

9.0

118.8

Vinyloxy-ethene

14.8

4.2

5.8

90.7

[2-(2-Ethoxy-ethoxy)ethoxy]-ethene

15.9

6.0

6.6

171.9

Ethoxy-ethene

14.5

4.9

6.0

95.0

Ethylsulfanyl-ethene

16.4

5.8

6.3

101.3

S

Formic acid vinyl ester

15.3

6.5

9.7

74.7

Iodo-ethene

17.1

5.5

7.3

75.6

2-Methyl-1-vinyloxypropane

15.1

4.1

5.1

131.3

CH2

CH3

Vinyl Ethyl Sulfide

H2C

16.3

CH2

O

Vinyl Ethyl Carbitol H3C

2-(2-Vinyloxy-ethoxy)ethanol

CH2

Vinyl Ether

H2C 907

Hydrogen Molar Bonding Volume

Cl

O

685

Dispersion Polarity

CH2

O

Vinyl Chloride

H2C 684

Autonom/ ACD Name

Solvent Name

CH3

Vinyl Formate

O

CH2

O 688

Vinyl Iodide (Iodoethene)

H2C 909

I

Vinyl Isobutyl Ether

H2C

O

CH3 CH3

7248_A001A.fm Page 481 Wednesday, May 23, 2007 12:38 PM

Appendix A: Table A.1

481

TABLE A.1 (CONTINUED) No. 910

Autonom/ ACD Name

Solvent Name Vinyl Isopropyl Ether

Dispersion Polarity

Hydrogen Molar Bonding Volume

2-Vinyloxy-propane

14.7

3.5

5.2

115.3

(2-Methoxy-ethoxy)ethene

15.6

6.2

6.8

114.7

Propionic acid vinyl ester

15.6

8.0

4.7

110.1

1-Vinyloxy-propane

14.9

3.5

5.2

113.0

4-Vinyl-pyridine

18.1

7.2

6.8

107.3

1-Vinyl-pyrrolidin-2-one 16.4

9.3

5.9

106.9

1-Ethoxy-4vinylsulfanyl-butane

16.5

5.0

5.8

174.0

(2-Ethoxyethylsulfanyl)-ethene

16.4

7.0

6.0

139.7

Vinyl-silane

15.5

2.6

4.0

89.4

CH3 H2C 911

CH3

Vinyl Methyl Cellosolve

H3C 689

O

O

O

CH2

Vinyl Propionate

CH2

O

H3C O 690

Vinyl Propyl Ether

H2C 843

CH3

O

4-Vinyl Pyridine CH2

N

691

Vinyl Pyrrolidone O N

913

Vinyl S-Butyl Mercapto Butyl Ether H2C

912

S O

CH3

Vinyl S-Ethyl Mercapto Ethyl Ether H2C

692

CH2

S O

Vinyl Silane

H2C

SiH3

CH3

7248_A001A.fm Page 482 Wednesday, May 23, 2007 12:38 PM

482

Hansen Solubility Parameters: A User’s Handbook

TABLE A.1 (CONTINUED) No. 916

Autonom/ ACD Name

Solvent Name 2-Vinyl Toluene CH3

Dispersion Polarity

Hydrogen Molar Bonding Volume

1-Methyl-2-vinylbenzene

18.6

1.0

3.8

131.8

Trifluoro-acetic acid vinyl ester

13.9

4.3

7.6

116.4

Trimethyl-vinyl-silane

14.5

1.0

2.5

145.3

2-Ethyl-hexanoic acid vinyl ester

15.6

2.7

5.8

196.0

[1,3]Dioxol-2-one

17.3

18.1

9.6

86.0

Water

15.5

16.0

42.3

18.0

Water

15.1

20.4

16.5

18.0

Water

18.1

17.1

16.9

18.0

p-Xylene

17.6

1.0

3.1

123.3

CH2

693

Vinyl Trifluoro Acetate O F

694

CH2

O

F

F

Vinyl Trimethyl Silane

CH3 Si CH3 H2C 915

CH3

Vinyl-2-Ethyl Hexanoate O H3C

O

CH2

H3C

695

Vinylenecarbonate

O O O 696

Water

OH2 859

Water - 1% in Ro = 18.1

OH2 858

Water - Complete Miscibility Ro = 13.0

OH2 697

Xylene CH3

CH3

7248_A001A.fm Page 483 Wednesday, May 23, 2007 12:38 PM

Appendix A: Table A.1

483

TABLE A.1 (CONTINUED) No. 698

Solvent Name o-Xylene

Autonom/ ACD Name o-Xylene

CH3 CH3

Dispersion Polarity

Hydrogen Molar Bonding Volume

17.8

3.1

1.0

121.2

7248_A001A.fm Page 484 Wednesday, May 23, 2007 12:38 PM

7248_A002.fm Page 485 Wednesday, May 23, 2007 12:53 PM

Appendix A: Table A.2 COMMENTS TO TABLE A.2 Documentation for the sources of data used for the HSP correlations is given in the following. The quality of the entries is sometimes less than desired because of the data being too few, too limited in scope and range of HSP, or for other reasons discussed in the text, such as the influence of molecular weight (molecular volume) of the test solvents in the given study. All entries have been included (with some apologies) as they have some value in terms of estimating, however.

POLYMERS 1–109 These polymers are listed in Reference 1 with suppliers. This report from the Scandinavian Paint and Printing Ink Research Institute updates an earlier one from 1982. The institute no longer exists. See also Reference 2.

POLYMER 110 This is an intermediate value for the permeation of chemicals through Challenge® materials [3]. See also Table 13.1 and Figure 13.2. Improved values are found below in 141 and 142. This correlation was based on few data to help locate additional solvents for testing. Results from tests with these then resulted in the correlations below.

POLYMERS 111–112 These are correlations of true solubilities for the DOW epoxy Novolacs 438 and 444.

POLYMERS 113–114 These are correlations of the chemical resistance of coatings based on inorganic zinc silicate and a two component epoxy produced by Hempel’s Marine Paints. Data taken from resistance tables.

POLYMER 115 The data are solubilities determined for PVDF with the correlation being previously published in [4].

POLYMER 116 Data for coal tar pitch generated for the solubility of the solids not dissolved in some cases where the solution was darkened with only partial solution.

POLYMERS 117–140 Permeation correlations for chemical protective clothing described in detail in Reference 5. See also Chapter 13, Table 13.1.

485

7248_A002.fm Page 486 Wednesday, May 23, 2007 12:53 PM

486

Hansen Solubility Parameters: A User’s Handbook

POLYMERS 141–142 Final permeation correlations for Challenge® 5100 and 5200 materials. Data from Reference 3 where there is considerable discussion. See also Chapter 13, Table 13.1, and Figure 13.2.

POLYMERS 143–144 These correlations are based on which solvents dissolve PVDC at elevated temperatures and use data from Wessling [6]. These were additionally used to check new calculations for solubility parameters of the solvents where these were lacking.

POLYMERS 145–148 These chemical resistance data for PES (ICI-Victrex®) and PPS (Philips-Ryton®) were based on supplier data sheets and are reported in Reference 7.

Polymers 149–160 These correlations for many common plastics types are based on the resistance tables reported in the PLASTGUIDE (1989) published by the Danish company Dukadan, which no longer exists. A single correlation for the solubility of PA 6,6 is based on its solubility only with data from Reference 8.

POLYMER 161 Beerbower treated several sets of data and made correlations of swelling and solubility (and other phenomena). This one is for polyvinyl silane.

POLYMERS 162–163 These correlations for swelling of cellophane and solubility of ethylene vinyl alcohol copolymer are based on data generated at NIF (Scandinavian Paint and Printing Ink Research Institute).

POLYMERS 164–167 These are supplementary breakthrough time correlations for Saranex®, Safety 4® 4H, and polyvinylalcohol protective gloves. See also Reference 5 and Chapter 13. Elimination of plasticizer data for the 4H gloves improved predictability for lower molecular weight materials.

POLYMERS 168–181 These correlations for common polymer types are based on data in resistance tables in the Modern Plastics Encyclopedia in the 1984/1985 issue [9]. Such data are not always sufficiently encompassing to allow good correlations.

POLYMER 182 Correlation based on high temperature solvents for ECTFE.

POLYMER 183 Data for this correlation of solubility of polyacrylonitrile were taken from the Polymer Handbook [10], Table of solvents and nonsolvents, p. VII/385-VII/386. See also Chapter 5, Table 5.3.

7248_A002.fm Page 487 Wednesday, May 23, 2007 12:53 PM

Appendix A: Table A.2

487

POLYMERS 184–186 Data for this correlation are the tendency of Polyethylene imide (PEI) (GE Ultem®) to environmental stress crack (ESC) at different stress/strain levels. These data were generated by General Electric as published in the Modern Plastics Encyclopedia 1984/1985 [9].

POLYMERS 187–224 The Handbook of Solubility Parameters and Other Cohesion Parameters [11] as well as the Polymer Handbook [12] included so-called “solvent range” data. Solvents were divided into groups of poor, moderate, and strong hydrogen bonding, and many experiments were run. The correlations show that not all the data were well taken, but a reasonable indication is possible. The full Hansen solubility parameter system is not covered very well by this limited solubility data. These polymers are included in Reference 11, Table 1, on page 280. Heating samples to speed up the solution process was also done. This can easily lead to errors.

POLYMERS 225–346 These entries have the same problem as those in 187–224 in that the data are sometimes questionable and not sufficient enough to do what has been done, i.e. convert solvent range data to Hansen solubility parameter spheres. These entries cover the acrylics, polyesters, polystyrenes, vinyls, and miscellaneous categories. Some categories are not yet included. Data on page 281-289 (Table 2) in Reference 11.

Polymer 347 These values for VYHH® (Union Carbide) were taken from Reference 1.

POLYMER 348 This questionable correlation for PVF includes only one solvent as being good [13].

POLYMER 349 Data on PES true solubility taken by author. See Chapter 5 and Table 5.4.

POLYMERS 350–358 These entries are not all polymers but mostly biological materials with the source of data being [14].

POLYMER 359 The solubility of cholesterol, data collected by the author. See Chapter 15.

POLYMER 360 Solubility data generated by high school students as part of project. Included in Reference 4. Source of chlorophyll was crushed leaves.

Polymer 361 Correlation on strength of paper immersed in different solvents reported in Reference 4. Data was taken from Reference 15.

7248_A002.fm Page 488 Wednesday, May 23, 2007 12:53 PM

488

Hansen Solubility Parameters: A User’s Handbook

POLYMER 362 Solubility of ULTRASON® PES has been reported by BASF in their product data. These data were combined with supplementary solubility data for this correlation. Also reported in Reference 16. See Chapter 5.

POLYMERS 363–364 Chemical resistance of BAREX® 210 from data in BP Chemicals datasheet. Styrene is an outlier in the first, whereas its removal from consideration gives a perfect fit and presumably a more useful correlation.

POLYMERS 365–367 These data were generated in connection with a lecture to the Nordic Conservation Congress in Copenhagen [17]. All give perfect fits, partly because of too few data, but the correlations can be useful. Paraloid B72 and Dammar are used as protective lacquers.

POLYMERS 368–369 These correlations divide the permeation coefficients given in Reference 18 into >80 and >0.8, respectively. The units are (g x mm)/(m2 x d). The fits are good. See Chapter 13.

POLYMERS 370–371 These are correlations of experimental solubility data for the Rhône-Poulenc reactive isocyanates Tolonate® HDT (which gave the same result as Tolonate® HDT-LV) and Tolonate® HDB (which gave the same results as Tolonate® HDB-LV). The fits were perfect and the numbers reasonable. The data could not include alcohol or amine solvents because of reactions.

POLYMERS 372–389 The data correlated for these 18 rubbers are from a RAPRA database [19]. The information used was satisfactory or unsatisfactory, all other information such as limited suitability was neglected. No precise weight gain or other information is available, just the general suitability or not. The values in parentheses are (data fit/number of solvents). ACM ECO CSM E EPM EPDM FQ FKM NR NBR FFKM CR AU EU T Q

acrylate rubbers (.981/55) epichlorohydrin rubbers (.988/37) chlorosulphonated polyethylene rubber (.906/53) ebonite (.722/41) ethylene-propylene copolymer (.987/47) ethylene-propylene terpolymer (.968/51) fluorosilicone rubber (.844/40) hexafluoroprop.-vinylidine fluoride copol. (Viton) (.769/50) natural rubber (1.000/59) nitrile rubber (.990/65) Kalrez® (Du Pont) (too resistant to correlate) polychloroprene (.877/54) polyester polyurethane (.959/63) polyether polyurethane (.959/63) polysulphide rubber (.799/48) silicone (.748/53)

7248_A002.fm Page 489 Wednesday, May 23, 2007 12:53 PM

Appendix A: Table A.2

SBR TFP

489

styrene butadiene rubber (.942/54) tetrafluoroethylene-propylene copolymer (.744/26)

POLYMERS 390–412 These correlations use data from the RAPRA collection of data on chemical resistance for plastics [20]. Approach same as for RAPRA rubber data just above.

POLYMERS 413–450 These data are from the collected report of the EC project on self-stratifying coatings reported in a full issue of Progress in Organic Coatings. The specific reference is Reference 21. The evaluations were made at different concentrations in many cases. Some alkyds were omitted here.

POLYMERS 451–452 These data are for strong swelling of two different film samples of brominated butyl rubber.

POLYMER 453 The correlation is based on strong swelling of a film of polyisoprene.

POLYMERS 454–458 These correlations are based on chemical resistance data from Reference 22.

POLYMER 459 Correlation based on solubility of Ethylene Vinylacetate adhesive EVA 4055.

POLYMER 460 Correlation based on solubility of Topas® 6013 from Ticona GmbH (Hoechst AG).

POLYMER 461 Correlation based on solubility of CZ® Resin from the West Company.

POLYMER 462 An older correlation for the solubility of Kauri Gum, used in the Kauri-Butanol test, was made with a data fit of 0.95 for the standard solvents.

POLYMER 463 The data for the solubility of polyvinylpyrrolidone used in this correlation are found in Reference 23. The data fit was 0.992, but as with many water soluble polymers, there is a considerable extrapolation into the “unknown” where there are no liquids.

ENTRY 464 The data fit for the correlation of solubility of palm oil with the standard set of solvents was 0.992.

7248_A002.fm Page 490 Wednesday, May 23, 2007 12:53 PM

490

Hansen Solubility Parameters: A User’s Handbook

ENTRY 465 This is a correlation of the solubility of a fungicide and algaecide called Bethoxazin using solubility data in 19 liquids from Reference 24. The data fit was 0.976.

ENTRY 466 This is a correlation for the solubility of carbon-60 at a given small level as reported in Reference 25; 15 of the 87 liquids were considered as “good” giving a data fit of 0.972.

REFERENCES 1. Saarnak, A., Hansen C.M., and Wallström E., Solubility Parameters — Characterization of Paints and Polymers, Report from Scandinavian Paint and Printing Ink Research Institute, January 1990, Hoersholm, Denmark 2. Hansen, C.M., Solubility Parameters, in Paint Testing Manual, Manual 17, J.V. Koleske, Ed., American Society for Testing and Materials, Philadelphia, 1995, pp. 38–404. 3. Hansen, C.M., Billing, C.B., and Bentz, A.P., Selection and Use of Molecular Parameters to Predict Permeation Through Fluoropolymer-Based Protective Clothing Materials, The Performance of Protective Clothing; Fourth Volume, ASTM STP 1133, J.P. McBriarty and N.W. Henry, Eds., American Society for Testing and Materials, Philadelphia, 1992, pp. 894–907. 4. Hansen, C.M., 25 Years with Solubility Parameters (in Danish: 25 År med Opløselighedsparametrene), Dansk Kemi, 73(8), 18–22, 1992. 5. Hansen, C.M. and Hansen, K.M., Solubility Parameter Prediction of the Barrier Properties of Chemical Protective Clothing, Performance of Protective Clothing: Second Symposium. ASTM STP 989, S.Z. Mansdorf, R. Sager, and A.P. Nielsen, Eds., American Society for Testing and Materials, Philadelphia, 1988, pp. 197–208. 6. Wessling, R.A., The Solubility of Poly(vinylidene Chloride), Journal of Applied Polymer Science, 14, 1531–1545, 1970. 7. Hansen, C.M., Solubility Parameters for Polyphenylene Sulfide (PPS) and Polyether Sulphone (PES), Centre for Polymer Composites (Denmark), Danish Technological Institute, Taastrup, 1991, 89 pages. ISBN 87-7756-139-2 8. Wyzgoski, M.G., The Role of Solubility in Stress Cracking of Nylon 6,6, in Macromolecular Solutions — Solvent Property Relationships in Polymers, R.B.Seymour and G.A.Stahl, Eds. Pergamon, New York, 1982, pp. 41–60. 9 Anonymous, Modern Plastics Encyclopedia 1984/1985, McGraw-Hill, New York, pp. 482–455. 10. Fuchs, O., Tables of Solvents and Non-solvents, Polymer Handbook, 3rd Ed., J. Branderup and E.H. Immergut, Eds., Wiley, New York, 1989, pp. VII/379-VII/407. 11. Barton, A.F.M., Handbook of Solubility Parameters and Other Cohesion Parameters, CRC Press Inc., Boca Raton, FL. 1983, pp. 280-289. 12. Grulke, E.A., Table 3.4, Solubility Parameter Ranges of Commercial Polymers, Polymer Handbook, 3rd Ed., J. Branderup and E.H. Immergut, Eds., Wiley, New York, 1989, pp. VII/544–VII/550. 13. Fuchs, O., Tables of Solvents and Non-solvents, Polymer Handbook, 3rd. Ed., J. Branderup and E.H. Immergut, Eds., Wiley, New York, 1989, p. VII/385. 14. Hansen, C.M. and Andersen, B.H., The Affinities of Organic Solvents in Biological Systems, American Industrial Hygiene Association Journal, 49(6), 301–308, 1988. 15. Robertson, A.A., Cellulose-Liquid Interactions, Pulp and Paper Magazine of Canada, 65(4), T-171T-178, 1964. 16. Hansen, C.M., Solvent Resistance of Polymer Composites — Glass Fibre Reinforced Polyether Sulfone (PES), Centre for Polymer Composites (Denmark), Danish Technological Institute, Taastrup, 1994. 17. Hansen, C.M., Conservation and Solubility Parameters, Nordic Conservation Congress Preprints, Copenhagen, 1994, pp. 1–13. 18. Pauly, S., Permeability and Diffusion Data, Polymer Handbook, 3rd. Ed., J. Branderup and E.H. Immergut, Eds., Wiley, New York, 1989, pp. VI/435–VI/449.

7248_A002.fm Page 491 Wednesday, May 23, 2007 12:53 PM

Appendix A: Table A.2

491

19. Anonymous, Chemical Resistance Data Sheets, Volume 2. Rubbers, New Edition — 1993, Rapra Technology, Shawbury, Shrewsbury, Shropshire, 1993. 20. Anonymous, Chemical Resistance Data Sheets, Volume 1. Plastics, New Edition — 1993, Rapra Technology, Shawbury, Shrewsbury, Shropshire, 1993. 21. Benjamin, S., Carr, C., and Walbridge, D.J., Self-stratifying Coatings for Metallic Substrates, Progress in Organic Coatings, 28, 197-207, 1996. 22. Anonymous, Engineering Guide to Du Pont Elastomers, The Du Pont Company, Switzerland, 1987. 23. Hansen, C.M., The Universality of the Solubility Parameter, Ind. Eng. Chem. Prod. Res. Dev., 8(1), 2–11, 1969. 24. Bosselaers, J., Blancquaert, P., Gors, J., Heylen, I., Lauwaerts, A., Nys, J., Van der Flaas, M., and Valcke Janssen, A., A New Fungicide and Algaecide, Färg och Lack Scandinavia, 49(1), 5–11 2003. 25. Hansen, C.M., and Smith, A.L., Using Hansen Solubility Parameters to Correlate Solubility of C60 Fullerene in Organic Solvents and in Polymers, Carbon, 42(8-9), 1591–1597, 2004.

LIST OF TRADE NAMES AND SUPPLIERS PAINTS AND BINDERS: Bayer (D): Cellit, Desmophen, Desmolac, Pergut, Cellidora, Desmodur, Baysilon, Alkydal Hercules (US): Piccolyte, Cellolyn, Pentalyn, Ester Gum, Parlon Ciba-Geigy (CH): Araldite Shell (D): Epikote, Cariflex Union Carbide (US): Vinylite, Phenoxy Hoechst (D): Macrynal, Phenodur, Alpex, Mowithal, Alfthalat, Mowilith Reichhold (CH): Super Beckasite, Uformite Polymer Corp. (CAN): Polysar Goodrich (US): Hycar Hüls (D): Vilit, Vesturit, Buna Hüls, Lutonal, Laroflex, Plastopal, Polystren Monsanto (US): Modaflow, Multiflow, Butvar Montecatini Edison (I): Vipla ICI (GB): Cereclor, Allopren, Suprasec Du Pont (US): Lucite Hagedorn (D): 1/2 sec. Nitrocellulose H 23 Röhm (D): Plexigum Rohm and Haas (U.S.): Paraloid Dynamit Nobel (D): Dynapol SOAB (S): Soamin BIP Chemicals (GB): Beetle Dyno Cyanamid (N): Dynomin DSM Resins (NL): Uracron Wacker (D): Wacker Dow Chemical (CH): Ethocel Cray Valley (GB): Versamid W. Biesterfeld (D): Chlorparaffin Synres (NL): Synresin American Cyanamide (US): Cymel Polyplex (DK): Plexal Pennsylvania Industrial Chemical Corp. (US): Piccopale, Piccoumarone OTHERS: Chemical Fabrics Corporation: Challenge

7248_A002.fm Page 492 Wednesday, May 23, 2007 12:53 PM

492

Hansen Solubility Parameters: A User’s Handbook

Chevron Phillips: Ryton ICI (Victrex plc): Victrex Saranex: Dow Safety 4, 4H: North General Electric: Ultem BASF: Ultrason BP Chemicals: Barex Rhône-Poulenc: Tolonate Ticona (Celanese): Topas West Company (DAIKYO): CZ Resin The capital letters in parenthesis are the international symbols for the respective countries: D US CH CAN I GB S N NL DK

Germany United States of America Switzerland Canada Italy Great Britian Sweden Norway Netherlands Denmark

7248_A002.fm Page 493 Wednesday, May 23, 2007 12:53 PM

Appendix A: Table A.2

493

TABLE A.2 Hansen Solubility Parameters for Selected Correlations Number

Polar

Hydrogen Bonding

Interaction Radius

12.00

6.70

10.20

18.20

12.40

10.80

7.40

17.90 20.10

4.30 6.90

3.90 5.90

5.90 9.90

14.00 23.10 20.00 17.40 21.00 19.30 23.40

7.40 14.60 10.32 10.50 11.10 9.37 7.20

9.40 5.00 10.11 9.00 13.40 10.95 14.80

13.70 20.50 10.02 7.90 11.70 8.26 14.90

Epoxy Curing Agents 23.80 20.30 24.90 26.90

5.30 6.60 3.10 2.40

16.20 14.10 18.70 18.50

16.10 9.60 20.30 24.00

17.70 19.10 21.54 16.00 20.60 19.40 17.90 18.70 19.90

10.60 12.20 14.94 13.10 7.80 7.40 9.60 9.60 8.10

11.60 9.90 12.28 9.20 11.60 6.00 5.90 9.90 6.00

9.50 8.00 16.78 11.40 13.10 9.80 8.20 8.20 9.80

23.26 19.74

6.55 11.62

8.35 14.59

19.85 12.69

Hydrocarbon Resins 16.47 17.55 19.42

0.37 1.19 5.48

2.84 3.60 5.77

8.59 6.55 9.62

Polymer

Dispersion Cellulose Acetobutyrate 16.60

1

CELLIT BP-300

2

CELLIDORA A

3 4

ETHOCEL HE10 ETHOCEL STD 20

5 6 7 8 9 10 11

ARALDITE DY O25 EPIKOTE 828 EPIKOTE 1001 EPIKOTE 1004 EPIKOTE 1007 EPIKOTE 1009 PKHH

12 13 14 15

VERSAMID VERSAMID VERSAMID VERSAMID

16 17 18 19 20 21 22 23 24

DESMOPHEN 651 DESMOPHEN 800 DESMOPHEN 850 DESMOPHEN 1100 DESMOPHEN 1150 DESMOPHEN 1200 DESMOPHEN 1700 DESMOLAC 4200 MACRYNAL SM 510N

25 26

SUPER BECKACITE 1001 PHENODUR 373U

27 28 29

PLIOLYTE S-100 PICCOPALE 110 PICCORONE 450L

30

POLYSAR 5630

Styrene-Butadiene (SBR) 17.55

3.35

2.70

6.55

31

HYCAR 1052

Acrylonitrile-Butadiene 18.62

8.78

4.17

9.62

Cellulose Acetate

Ethyl Cellulose

Epoxy

100 115 125 140

Polyurethane

Phenolic Resins

7248_A002.fm Page 494 Wednesday, May 23, 2007 12:53 PM

494

Hansen Solubility Parameters: A User’s Handbook

TABLE A.2 (CONTINUED) Hansen Solubility Parameters for Selected Correlations Number

Polymer

Polar

Hydrogen Bonding

Interaction Radius

17.53

2.25

3.42

6.55

16.57

1.41

–0.82

9.62

14.20 16.90 17.40

2.50 2.50 4.30

4.60 4.00 8.40

12.40 7.20 7.40

20.17 16.10

14.61 3.70

15.04 7.90

11.66 8.90

18.40

6.60

8.00

3.00

Dispersion Polybutadiene

32

BUNA HULS B10

33

CARIFLEX IR 305

34 35 36

LUTONAL IC/1203 LUTANAL I60 POLYVINYLBUTYL ETHER

37 38

LIGNIN MODAFLOW

39

VIPLA KR (PVC)

40 41

CERECLOR 70 CHLOROPAR 40

20.00 17.00

8.30 7.60

6.80 7.90

9.80 11.90

42 43

PERGUT S 5 ALLOPREN R10

Chlorinated Rubber 17.40 17.40

9.50 4.30

3.80 3.90

10.00 6.10

44

PARLON P 10

Chlorinated Polypropylene 20.26

6.32

5.40

10.64

45 46

HYPALON 20 HYPALON 30

Chlorosulfonated PE 18.10 18.20

3.40 4.70

4.90 2.00

3.60 5.00

47

ALPEX

19.90

0.00

0.00

9.40

48

1/2-sec.-NITRO CELLULOSE H 23

15.41

14.73

8.84

11.46

49 50 51 52

CELLOLYN 102 PENTALYN 255 PENTALYN 830 ESTER GUM BL

21.73 17.55 20.03 19.64

0.94 9.37 5.81 4.73

8.53 14.32 10.93 7.77

15.75 10.64 11.66 10.64

53 54 55

VERSAMID 930 VERSAMID 961 VERSAMID 965

17.43 18.90 20.15

–1.92 9.60 6.04

14.89 11.10 12.90

9.62 6.20 9.20

Polyisoprene

Polyisobutylene

Special

Polyvinylchloride

Chloroparaffin

Cyclized Rubber

Nitrocellulose

Rosin Derivatives

Polyamide

7248_A002.fm Page 495 Wednesday, May 23, 2007 12:53 PM

Appendix A: Table A.2

495

TABLE A.2 (CONTINUED) Hansen Solubility Parameters for Selected Correlations Number

Dispersion

Polar

Hydrogen Bonding

Interaction Radius

17.50 17.60 19.70

11.30 10.00 12.90

5.90 3.70 12.80

8.50 9.30 11.40

18.60 20.20 18.60

12.90 11.20 4.36

10.30 13.30 13.03

8.30 11.20 10.64

17.60 16.20 18.60 18.40 18.64

9.66 6.80 10.80 9.40 10.52

3.97 5.70 4.10 6.50 7.51

10.64 9.10 11.50 10.70 8.59

20.93

11.27

9.66

13.71

22.28

5.75

4.30

12.68

Vinyl Chloride Copolymers 18.40 20.00 20.00 18.40 17.10 16.50 17.70 17.60 17.60 17.40 18.10

8.40 8.30 8.30 7.60 10.40 10.90 11.10 11.10 11.10 10.20 10.30

5.80 6.70 6.70 6.70 6.50 6.40 6.90 6.80 6.40 5.90 4.20

9.00 9.40 9.40 6.80 7.50 7.70 8.70 8.80 8.60 7.80 8.30

Binders in Solution: Alkyds and Polyesters 18.60 10.00 23.00 2.20 22.90 15.20 22.60 13.80 20.50 9.30 19.20 5.30 23.60 1.00 20.60 4.60 17.30 4.20 22.60 13.10

5.00 4.20 7.60 8.10 9.10 6.30 7.60 5.50 7.90 5.80

10.40 16.90 18.10 17.10 12.40 11.90 19.00 12.60 9.30 16.80

Polymer Isocyanate

56 57 58

DESMODUR L DESMODUR N SUPRASEC F-5100

59 60 61

MOWITAL B 30 H MOWITAL B 60 H BUTVAR B 76

62 63 64 65 66

LUCITE 2042 PEMA LUCITE 2044 PMMA PLEXIGUM MB319 PLEXIGUM M527 PMMA

67

MOWILITH 50 PVAC

68

POLYSTYRENE LG

69 70 71 72 73 74 75 76 77 78 79

LAROFLEX MP 45 VILIT MB 30 VILIT MC 31 VILIT MC 39 VINYLITE VAGD VINYLITE VAGH VINYLITE VMCA VINYLITE VMCC VINYLITE VMCH VINYLITE VYHH VINYLITE VYLF

80 81 82 83 84 85 86 87 88 89

ALFTALAT AC 366 ALFTALAT AM 756 ALFTALAT AN 896 ALFTALAT AN 950 ALFTALAT AT 316 ALFTALAT AT 576 ALKYDAL F 261 HS ALKYDAL F 41 DUROFTAL T 354 DYNAPOL L 812

Polyvinylbutyral

Polyacrylate

Polyvinylacetate

Polystyrene

7248_A002.fm Page 496 Wednesday, May 23, 2007 12:53 PM

496

Hansen Solubility Parameters: A User’s Handbook

TABLE A.2 (CONTINUED) Hansen Solubility Parameters for Selected Correlations Number

Polymer

Dispersion

Polar

Hydrogen Bonding

Interaction Radius

20.00 18.50 18.00 18.80 17.70

6.20 9.21 11.60 12.00 13.00

7.00 4.91 8.50 6.00 7.60

9.50 10.64 9.00 11.50 11.50

20.70 22.20 19.35 25.50 18.80 19.90 15.90 22.10 20.81 22.70

6.10 –0.40 12.83 15.20 14.00 15.80 8.10 5.00 8.29 2.80

12.70 10.10 12.87 9.50 12.30 13.40 6.50 11.30 14.96 5.40

14.80 18.40 9.82 22.20 10.50 11.70 10.60 15.50 12.69 16.20

19.20 19.20 19.60 18.40

7.70 9.60 9.10 9.80

5.70 9.30 6.80 10.00

10.60 12.20 12.20 12.40

90 91 92 93 94

DYNAPOL L 850 PLEXAL C-34 SOALKYD 1935-EGAX VESTURIT BL 908 VESTURIT BL 915

95 96 97 98 99 100 101 102 103 104

BE 370 BEETLE 681 CYMEL 300 CYMEL 325 DYNOMIN MM 9 DYNOMIN UM 15 SOAMIN M 60 SYNRESIN A 560 PLASTOPAL H UFORMITE MX-61

105 106 107 108

URACRON 15 PARALOID P 400 PARALOID P 410 PARALOID EXPER. RES. QR 954

109 110

BAYSILON UD 125 TEFLON (SL2-)

19.40 17.10

9.90 8.10

10.10 1.30

6.90 4.70

111 112 113 114 115 116

Special Data DOW EPOXY NOVOLAC 438 DOW EPOXY NOVOLAC 444 ZINK SILICATE - CHEMICAL RES. 2-COMP EPOXY CHEMICAL RES. POLYVINYLIDINE FLUORIDE SOL. COAL TAR PITCH SOL.

20.30 19.50 23.50 18.40 17.00 18.70

15.40 11.60 17.50 9.40 12.10 7.50

5.30 9.30 16.80 10.10 10.20 8.90

15.10 10.00 15.60 7.00 4.10 5.80

Amino Resins

Acrylate Resins

Silicone Resins

117 118 119 120 121 122 123 124 125 126 127 128

Breakthrough Time (Bt) Correlations for Common Types of Chemical Protective Films at Practical Film Thickness NITRILE 20 MIN 17.50 7.30 6.50 5.10 NITRILE 1 HR 16.60 9.10 4.40 10.00 NITRILE 4 HR 19.00 12.60 3.80 13.30 BUTYL 20 MIN 16.50 1.00 5.10 5.00 BUTYL 1 HR 15.80 –2.10 4.00 8.20 BUTYL 4 HR (2) 17.60 2.10 2.10 7.00 NATURAL RUBBER 20 MIN 14.50 7.30 4.50 11.00 NATURAL RUBBER 1 HR 15.60 3.40 9.10 14.00 NATURAL RUBBER 4 HR 19.40 13.20 7.70 19.00 PVC 20 MIN 16.10 7.10 5.90 9.30 PVC 1 HR 14.90 11.10 3.80 13.20 PVC 4 HR 24.40 4.90 9.90 22.70

7248_A002.fm Page 497 Wednesday, May 23, 2007 12:53 PM

Appendix A: Table A.2

497

TABLE A.2 (CONTINUED) Hansen Solubility Parameters for Selected Correlations Number

Polymer

Dispersion

Polar

Hydrogen Bonding

Interaction Radius

11.20 15.30 17.20 16.90 17.10 24.10 10.90 16.50 13.60 17.60 19.00 14.60 16.60 16.60

129 130 131 132 133 134 135 136 137 138 139 140 141 142

POLYVINYLALCOHOL 20 MIN POLYVINYLALCOHOL 1 HR POLYVINYLALCOHOL 4 HR POLYETHYLENE 20 MIN POLYETHYLENE 1 HR POLYETHYLENE 4 HR VITON 20 MIN VITON 1 HR VITON 4 HR NEOPRENE 20 MIN NEOPRENE 1 HR NEOPRENE 4 HR CH 5100 3 HR CH 5200 3 HR

12.40 13.20 13.60 3.30 3.10 14.90 14.50 8.10 15.40 2.50 8.00 13.90 5.40 6.00

13.00 13.50 15.40 4.10 5.20 0.30 3.10 8.30 8.60 5.90 0.00 2.30 4.00 4.80

12.10 8.80 10.90 8.10 8.20 24.30 14.10 6.60 14.40 6.20 13.20 15.90 3.80 3.70

143 144

High Temperature Solubility of PVDC PVDC (110C) SOLUBILITY 17.60 9.10 PVDC (130C) SOLUBILITY 20.40 10.00

7.80 10.20

3.90 7.60

145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161

Chemical Resistance of High Performance and Other Polymers PES L C=1 18.70 10.50 7.60 PES L B + C =1 17.70 9.70 6.40 PPS CR 93°C 18.80 4.80 6.80 PPS TS60%12MO 18.70 5.30 3.70 PA6 CR 17.00 3.40 10.60 PA66 SOL 17.40 9.80 14.60 PA11 CR 17.00 4.40 10.60 POMH+POMC CR 17.10 3.10 10.70 PETP CR 18.20 6.40 6.60 PTFE L80 CR 16.20 1.80 3.40 PMMA CR 16.00 5.00 12.00 PE? CR QUESTIONABLE VALUES 16.80 5.40 2.40 PPO CR 17.90 3.10 8.50 PUR CR 18.10 9.30 4.50 ABS CR 16.30 2.70 7.10 PSU CR 16.00 6.00 6.60 VINYL SILANE 16.40 3.70 4.50

9.10 9.30 2.80 6.70 5.10 5.10 5.10 5.20 5.00 3.90 13.00 4.70 8.60 9.70 7.80 9.00 10.00

162 163 164 165 166 167

Correlations for Some Barrier-Type Polymers CELLOPHAN SW 16.10 18.50 EVOH SOL 20.50 10.50 SARANEX 4HR 17.70 18.30 4H 35°C 19.40 13.40 4H 35°C no plasticizer included 20.50 11.30 POLYVINYLALCOHOL 15.00 17.20

14.50 12.30 0.70 18.00 10.30 17.80

9.30 7.30 18.40 8.60 6.70 10.20

168 169 170

Chemical Resistance Data - Modern Plastics Encylopedia ACETAL CELANESE 21.10 9.30 ACETALHOMO-DUO 19.00 5.00 CELLULOSE ACETATE 16.90 16.30

5.90 8.00 3.70

11.40 5.00 13.70

7248_A002.fm Page 498 Wednesday, May 23, 2007 12:53 PM

498

Hansen Solubility Parameters: A User’s Handbook

TABLE A.2 (CONTINUED) Hansen Solubility Parameters for Selected Correlations Number

Polymer

Dispersion

Polar

Hydrogen Bonding

Interaction Radius

17.20 9.80 14.10 19.00 19.00 19.90 16.70 21.60 18.00 18.00 18.00

13.80 13.60 2.70 4.00 6.00 3.90 7.70 5.20 3.00 0.00 0.00

2.80 11.40 5.50 3.00 8.00 5.10 –0.50 18.80 4.00 2.00 1.00

12.60 15.20 6.60 4.00 5.00 3.80 8.10 15.40 6.00 2.00 6.00

7.30

1.70

5.10

14.10

171 172 173 174 175 176 177 178 179 180 181

CELL. ACET. BUTYRATE CELL. ACET. PROPIONATE PCTFE FEP FURAN FURF ALC PFA(?) PHENOLIC PETG HDPE PP

182

Poly(Ethylene/Chlorotrifluoroethylene) PECTFE SOL AT HIGH TEMP. 19.50

183

PAN

9.10

10.90

184 185 186

PEI - Polyethylene Imide - Environmental Stress Cracking (ESC) PEI 1200PSI 17.20 6.40 5.20 PEI 2400PSI 17.40 4.60 9.00 PEI 600PSI 17.30 5.30 4.70

3.60 7.20 3.30

187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212

Based on Solvent Range Solubility Data - Not too Reliable ESTER GUM 16.90 4.50 ALKYD 45 SOYA 17.50 2.30 SILICONE DC-1107? 19.60 3.40 PVETHYLETHER? 15.10 3.10 PBUTYLACRYLATE 16.20 9.00 PBMA? 15.90 5.50 SILICONE DC 23? 16.40 0.00 PE 16.00 0.80 GILSONITE 17.10 2.10 PVINYLBUTYLETHER 17.40 3.40 NAT RUBBER 16.00 4.00 HYP 20 CHLOROSULFONATED PE 17.40 3.20 ETHCEL N22? 22.70 0.50 CHLORINATED RUBBER 17.90 6.30 DAMMAR GUM 18.40 4.20 VERSAMID 100? 18.80 3.00 PS 18.50 4.50 PVAC 17.60 2.20 PVC 17.60 7.80 PHENOLICS 19.80 7.20 BUNA N BUTADIENE/ACRYLONITRILE 17.80 3.20 PMMA 18.10 10.50 PEO 4000 ? HEATED SAMPLES 21.50 10.90 POLYETHYLENESULFIDE (GOOD) 17.80 3.80 PC 18.10 5.90 PLIOLITE P1230 18.10 4.70

9.20 10.00 9.80 12.90 10.10 8.50 5.50 3.20 4.90 8.40 1.30 4.80 20.10 7.60 8.30 7.80 5.30 4.10 8.20 12.80 3.70 9.50 15.90 4.10 5.50 3.90

Solubility of Polyacrylonitirile 21.70

6.50 7.70 10.80 11.90 3.00 5.90 7.80 2.80 3.90 7.80 6.00 4.00 16.50 5.10 7.80 9.20 2.90 4.00 3.40 10.80 3.40 5.10 13.10 2.20 6.90 3.70

7248_A002.fm Page 499 Wednesday, May 23, 2007 12:53 PM

Appendix A: Table A.2

499

TABLE A.2 (CONTINUED) Hansen Solubility Parameters for Selected Correlations Number

Polymer

Dispersion

Polar

Hydrogen Bonding

Interaction Radius

213 214 215 216 217 218 219 220 221 222 223 224

MYLAR PET VCVA COPOLY PUR SAN VINSOL ROSIN EPON 1001 SHELLAC POLYMETHACRYLONITRILE CELLULOSE ACETATE CELLULOSE NITRATE PVOH (NOT GOOD, SEE CHAP. 5) NYLON 66

18.00 17.30 17.90 16.60 17.40 17.00 19.70 17.20 18.30 16.90 17.00 16.00

6.20 8.70 6.90 9.80 10.00 9.60 10.10 14.40 16.50 13.50 9.00 11.00

6.20 6.10 3.70 7.60 13.00 7.80 15.10 7.60 11.90 10.30 18.00 24.00

5.00 7.80 2.70 4.80 10.50 7.10 10.70 3.80 8.80 9.90 4.00 3.00

225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251

Acrylics - Solvent Range ACRYLOID B-44 19.40 ACRYLOID B-66 18.00 ACRYLOID B-72 19.20 ACRYLOID B-82 19.10 R+H PBA 16.00 R+H PiMBA 20.70 R+H PNBMA 16.00 R+H PEMA 19.00 R+H PMAA 25.60 R+H PMMA 19.10 BMA/AN 80/20 17.50 ISOB MALANH/CYCLOL 75/25 16.80 MAA/EA/ST 15/38/47 17.60 MAA/MA/VA 15/27.5/57.5 28.50 MAA/MA/VA 15/17.5/67.5 25.50 MMA/CYCLOL 58/42 18.70 MMA/EA 50/50 17.50 MMA/EA 25/75 19.00 MMA/EA/AGE 40/40/20 17.60 MMA/EA/AA 15.90 MMA/EA/AN 55/30/15? 16.70 MMA/EA/AN 40/40/20 20.40 MMA/EA/BAMA 40/40/20 17.90 MMA/EA/CYCLOL 17.60 MMA/EA/MAM 40/40/20 19.00 MMA/EA/MAM 45/45/10? 19.50 MMA/EA/BVBE 40/40/20 17.80

11.20 9.00 11.20 9.10 8.00 4.10 6.20 9.00 11.20 11.30 9.90 –0.40 5.20 15.70 15.70 9.90 9.90 9.00 9.80 15.90 10.90 13.20 8.50 9.80 9.00 11.10 10.00

4.40 3.00 1.80 3.30 8.00 10.70 6.60 8.00 19.60 4.10 4.10 7.20 7.00 18.10 18.10 8.70 4.10 15.00 5.60 11.50 8.50 11.00 11.70 6.40 15.00 8.70 6.60

10.50 9.00 11.00 9.00 12.00 11.50 9.50 11.00 20.30 10.30 9.50 8.50 4.50 21.50 21.50 8.80 9.50 12.00 9.70 11.10 8.50 12.30 12.90 9.80 12.00 11.20 9.80

252 253 254 255 256 257

ACID DEGMP CARB DEG PTH CRYPLEX 1473-5 DEG ISOPH DEG PHTH DPG PHTH

Polyesters - Solvent Range 15.30 19.40 19.20 19.20 21.00 20.10

13.30 13.40 9.40 17.20 15.20 11.50

14.90 11.60 5.60 14.60 13.20 6.70

15.60 11.10 8.90 11.80 13.70 11.60

7248_A002.fm Page 500 Wednesday, May 23, 2007 12:53 PM

500

Hansen Solubility Parameters: A User’s Handbook

TABLE A.2 (CONTINUED) Hansen Solubility Parameters for Selected Correlations Number

Polymer

Dispersion

Polar

Hydrogen Bonding

Interaction Radius

17.80 17.80 14.90 21.30 17.00 18.70 19.40 19.00 19.00 18.80 18.10 17.30

258 259 260 261 262 263 264 265 266 267 268 269

DOW ADIP TEREP DOW X-2635 MALEATE VITEL PE LINEAR VITEL PE101-X HYD BIS A FUM ISPH HYD BIS A PG FUM ISPH PENTA BENZ MAL SOL MYLAR 49001 SOL MYLAR 49002 TEG EG MAL TEREP TEG MALEATE VAREZ 123

10.40 5.60 10.10 6.30 4.40 8.90 12.20 5.00 5.00 11.40 13.90 10.90

6.80 6.80 2.90 4.70 6.20 5.50 10.20 4.00 5.00 9.20 12.10 11.90

9.30 4.50 6.10 7.30 5.00 8.40 10.80 5.00 5.00 10.20 9.70 10.70

270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296

Styrene Polymers And Copolymers - Solvent Range AMOCO 18-290 19.30 3.70 BUTON 100 BUTAD-STY 17.00 4.00 BUTON 300 17.30 3.70 KOPPERS KTPL-A 19.30 3.70 RUBBER MOD PS 20.00 5.00 STY MAL ANH 23.40 13.80 LYTRON 820 21.10 13.10 MARBON 9200 19.00 4.00 PARAPOL S-50 17.90 3.90 PARAPOL S-60 17.90 3.90 PICCOFLEX 120 17.40 7.80 SHELL POLYALDEHYDE EX 39 19.60 10.00 SHELL POLYALDEHYDE EX 40 19.60 10.00 SHELL X-450 19.30 9.50 SMA 1430A 18.80 11.40 SAN 85/15 19.10 9.50 STY/BUTENOL 85/15 17.40 7.80 STY/CYCLOL 82/18 18.20 5.60 STY/2EHA/AA 81/11/8 17.70 4.90 STY/MAA 90/10 18.70 6.30 STY/MA 85/15 18.00 9.00 STY/HALF ESTER MA 60/40 18.90 10.90 STY/PROP HALF E MA 57/43 18.00 9.80 STY/VBE 85/15 17.40 7.80 STYRON 44OM-27 MOD PS 20.00 5.00 STYRON 475M-27 20.00 5.00 STYRON 480-27 20.00 6.00

7.90 3.00 7.30 7.90 1.00 15.20 14.50 4.00 4.90 4.90 3.80 3.60 3.60 11.10 16.40 3.10 3.80 7.20 5.90 7.30 3.00 10.70 8.40 3.80 1.00 1.00 4.00

7.80 7.30 7.00 7.80 7.00 16.50 14.40 6.00 3.90 3.90 7.70 9.40 9.40 11.10 14.10 8.70 7.70 5.70 5.90 6.70 9.00 9.70 10.10 7.70 7.00 7.00 5.30

297 298 299 300 301 302

ACRYLOID K120N DODA 6225 DODA 3457 ELVAX 250 ELVAX 150 ELVAX EOD 3602-1

3.80 1.00 1.00 1.00 0.70 2.70

9.50 3.00 3.00 3.00 6.00 5.40

Vinyl Resins - Solvent Range 17.60 19.00 19.00 19.00 18.70 17.70

10.00 2.00 2.00 2.00 2.30 3.30

7248_A002.fm Page 501 Wednesday, May 23, 2007 12:53 PM

Appendix A: Table A.2

501

TABLE A.2 (CONTINUED) Hansen Solubility Parameters for Selected Correlations Number

Polymer

Dispersion

Polar

Hydrogen Bonding

Interaction Radius

303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330

EXON 470 PVC EXON 471 EXON 473 GEON 121 POLYCYCLOLa PVBE PVEE FORMVAR 7/70E PVFORMAL FORMVAR 15/95E PVIBE SARAN F-120 VCL2/AN? SARAN F-220 ? SINCLAIR 3840A VA/EHA/MA 63/33/4 VA/EHA/CYC/MAA/76/12/8/4 VA/EA/CY 70/20/10 VBE/AN/MAA 46/27/27 VBE/MA/MAC46/27/27 VDC/AA 75/25 ? VINYLITE AYAA PVAC VINYLITE VAGH VINYLITE VMCH VINYLITE VXCC VINYLITE VYHH VINYLITE VYLF VINYLITE XYHL PVBUTYRAL VINYLITE XYSG PVBUTYRAL VYSET 69

17.40 17.90 17.40 19.50 19.00 16.70 16.00 22.20 22.20 16.00 28.80 28.80 18.40 17.70 21.20 20.00 18.90 19.40 20.40 22.90 17.00 18.30 18.00 19.00 18.00 19.00 19.00 17.90

7.80 8.70 7.80 6.70 9.00 3.70 4.00 12.60 12.60 1.00 16.80 16.80 4.00 6.30 12.40 12.00 11.70 13.00 11.00 18.30 7.80 9.70 9.40 11.00 9.40 9.00 9.00 3.50

3.80 2.50 3.80 11.10 15.00 8.30 12.00 14.20 14.20 8.00 0.80 0.80 9.60 7.70 13.00 11.00 11.10 13.80 0.80 7.70 6.80 7.70 4.60 5.00 4.60 15.00 15.00 7.50

7.70 9.00 7.70 8.00 12.00 8.60 14.00 14.00 14.00 10.00 23.70 23.70 7.30 5.30 12.60 15.00 9.60 12.30 11.70 20.40 7.10 8.50 8.40 10.00 8.40 12.00 12.00 5.90

331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346

Miscellaneous - Solvent Range ACRYLAMIDE MONOMER 16.90 BAKELITE SULFONE P-47 20.00 BECKOLIN 27 MODIF OIL 11.40 PEO 4000 ? SAMPLES HEATED 22.20 CHLORINATED RUBBER 18.00 CONOCO H-35 HYDROCARBON M 11.40 DAMMAR GUM DEWAXED 19.00 EPOCRYL E-11 ? 17.30 ESTANE X-7 ?? DIOXANE ONLY 19.00 HEXADECYL MONOESTER TRIME 19.00 HYDR SPERM OIL WX135 20.00 HYPALON 20 CHL SULF PE 17.80 HYPALON 30 17.80 KETONE RESIN S588 18.00 SANTOLITE MHP ARYLSULFONA 18.40 pTOLSULFONAMIDE-FORMALDEH 24.60

18.10 3.00 0.00 11.20 6.00 0.00 2.00 12.90 1.80 11.60 4.00 3.20 3.40 10.80 12.00 18.60

19.90 6.00 3.00 13.20 5.00 3.00 9.00 12.10 7.40 14.00 2.00 4.40 3.20 13.20 8.40 16.40

17.00 3.00 18.10 17.10 7.00 18.10 9.00 8.50 1.00 11.90 5.00 4.10 5.10 12.20 10.60 20.90

7248_A002.fm Page 502 Wednesday, May 23, 2007 12:53 PM

502

Hansen Solubility Parameters: A User’s Handbook

TABLE A.2 (CONTINUED) Hansen Solubility Parameters for Selected Correlations Number

Polymer

Dispersion

Polar

Hydrogen Bonding

Interaction Radius

347 348 349

Polymer Solubility Data from Various Sources VYHH-NIF REPT 17.40 9.90 PVF? (DMF ONLY GOOD SOLVENT) 17.40 13.70 PES SOL 19.60 10.80

6.70 11.30 9.20

7.50 2.00 6.20

350 351 352 353 354 355 356 357 358 359 360 361

Biologically Interesting Systems LARD 37C 15.90 LARD 23C 17.69 1%IN WATER -AMINES 15.07 1%IN WATER +AMINES 14.96 BLOOD SERUM 23.20 SUCROSE 21.67 UREA 20.90 PSORIASIS SCALES 24.64 LIGNIN 20.61 CHOLESTEROL 20.40 CHLOROPHYLL 20.20 CELLULOSE-PAPER STRENGTH 25.40

1.16 2.66 20.44 18.33 22.73 26.26 18.70 11.94 13.88 2.80 15.60 18.60

5.41 4.36 16.50 15.15 30.60 29.62 26.40 12.92 15.25 9.40 18.20 24.80

12.03 7.98 18.12 16.22 20.48 20.44 19.42 19.04 11.83 12.60 11.10 21.70

362

PSU ULTRASON S

19.70

8.30

8.30

8.00

363 364

BAREX 210 CR BAREX 210 CR-STYRENE

20.10 17.70

9.10 8.90

12.70 10.90

10.90 6.40

365 366 367

Polymers of Interest for Conservation of Paintings PARALOID B72 17.60 7.40 ESTIMATE DRIED OIL 16.00 6.00 DAMMAR DEWAXED 19.00 2.00

5.60 7.00 9.00

9.40 5.00 9.00

368 369

LDPE PERM>80 LDPE PERM<0.8

Permeation of LDPE by Organic Liquids 16.50 4.50 15.30 5.30

0.50 2.50

6.00 10.10

370 371

Tolonate Solubility TOLONATE HDT (RH-POULENC) 19.00 TOLONATE HDB (RH-POULENC) 19.00

11.00 11.00

3.00 2.00

12.00 11.30

372 373 374 375 376 377 378 379 380 381 382

Chemical Resistance of Elastomers R ACM 16.80 R BUTYL 18.00 R ECO 21.30 R CSM 28.00 R EBONITE (DATA FIT 0.722) 18.70 R ETHYLENE/PROPYLENE 16.60 R EPDM 18.60 R FQ FL/SI 15.90 R FKM (VITON) (0.76 DATA FIT) 11.60 R NR NAT RUB 20.80 R NBR 19.80

11.80 0.00 8.10 14.00 6.10 0.00 –3.40 20.10 23.00 1.80 17.80

11.60 3.00 6.10 3.40 2.70 5.20 4.40 6.90 5.00 3.60 3.20

17.00 9.00 12.00 28.30 6.60 9.10 10.70 16.80 21.60 14.00 19.00

Polysulfone PSU

Barex

7248_A002.fm Page 503 Wednesday, May 23, 2007 12:53 PM

Appendix A: Table A.2

503

TABLE A.2 (CONTINUED) Hansen Solubility Parameters for Selected Correlations Number

Polymer

Dispersion

Polar

Hydrogen Bonding

Interaction Radius

24.60 17.90 17.90 25.30 13.80 17.20 16.60

8.60 13.30 13.30 17.30 5.00 6.00 6.80

6.40 10.70 10.70 6.70 1.20 4.60 0.60

20.40 17.10 17.10 23.60 14.30 9.80 7.90

8.60 7.10 6.50 12.20 9.00 8.10 8.80 5.70 5.60 10.90 4.30 6.30 19.50 16.70 1.40 8.90 11.20 5.60 10.80 12.30 11.30 17.40 17.40

383 384 385 386 387 388 389

R CR CHLOROPRENE R AU ESTER PU R PEU ETHER PU R T SULPHIDE R Q SILICONE (0.748 DATA FIT) R SBR R TFP TETFLPROP (0.744 DATA FIT)

390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412

Chemical Resistance of Plastics R ABS 17.60 R CELLULOSE ACETATE 14.90 R CHLORINATED PVC 17.50 R DIALLYLPHTHALATE 22.20 R POM ACETAL 17.20 R PA12 18.50 R PA66 18.20 R POLYAMIDEIMIDE 18.50 R POLYBUTYLENETEREPH 18.00 R POLYCARBONATE 19.10 R HDPE/LDPE 17.50 R PET 19.10 R POLYIMIDES 24.30 R PMMA 19.30 R TPX 18.80 R POLYPHENYLENEOXIDE 16.90 R POLYSULPHONE 19.80 R POLYPROPYLENE 17.20 R EPOXY COLD CURING 16.80 R EPOXY HOT CURING 18.30 R HET RESIN 17.50 R ISOPHTHALIC 19.80 R TEREPHTALIC 19.80

6.40 11.10 5.50 8.60 9.80 9.10 10.80 8.70 8.40 5.10 8.30 9.10 22.90 4.70 6.40 2.70 6.20 –0.40 8.80 9.70 8.30 4.20 4.20

10.90 12.40 6.30 15.80 5.30 6.30 5.20 4.20 4.50 12.10 3.90 4.80 21.60 17.40 7.90 11.70 11.30 4.50 8.20 7.30 8.60 18.00 18.00

413 414 415 416 417 418 419 420 421 422 423 424

Polymers at Different Test Concentrations - (Conc) Epoxy Polymers EPIKOTE 828 (60%) 16.60 14.00 2.80 EPIKOTE 828 (30%) 16.30 16.40 1.90 EPIKOTE 1001 (60%) 15.80 16.30 3.30 EPIKOTE 1001 (40%) 16.30 13.10 6.30 EPIKOTE 1001 (20%) 19.80 13.60 8.90 EPIKOTE 1001 (10%) 18.10 11.40 9.00 EPIKOTE 1004 (60%) 17.70 10.10 7.60 EPIKOTE 1004 (30%) 18.50 9.30 8.00 EPIKOTE 1007 (30%) 18.60 10.60 8.10 EPIKOTE 1009 (60%) 17.00 9.60 8.50 EPIKOTE 1009 (30%) 19.80 10.60 10.30 EPIKOTE 1009 (10%) 19.00 9.10 10.70

14.90 16.70 16.40 10.90 12.00 9.10 9.80 9.60 8.80 7.60 9.70 8.00

425 426

PIBMA (10%) PIBMA (30%)

Acrylics 17.00 17.10

4.60 5.90

7.60 0.70

9.50 7.30

7248_A002.fm Page 504 Wednesday, May 23, 2007 12:53 PM

504

Hansen Solubility Parameters: A User’s Handbook

TABLE A.2 (CONTINUED) Hansen Solubility Parameters for Selected Correlations Number

Dispersion

Polar

Hydrogen Bonding

Interaction Radius

17.80 17.20 20.60 18.10 17.60 17.50 16.50 16.90 17.80 21.20 16.30 16.30

10.40 7.20 3.50 5.70 10.10 5.50 8.70 7.80 6.40 1.40 10.60 10.60

2.90 3.50 7.20 0.00 5.80 3.80 5.00 0.50 4.70 10.70 7.40 7.40

9.60 4.80 12.80 8.50 9.40 4.50 10.40 7.30 10.70 12.30 12.90 12.90

Fluorinated Polyethers 18.50 20.10 17.50 18.10

5.40 4.40 6.80 3.90

6.90 3.20 10.50 8.30

9.90 8.50 12.50 8.80

Acrylic Modified Alkyd 20.10 25.20 23.70 23.90

5.70 9.20 0.50 7.80

5.30 3.70 10.30 8.80

20.00 20.00 20.00 19.90

Chlorinated Rubber 19.50 17.90 19.60

9.20 5.60 6.50

6.90 6.70 5.80

7.50 5.80 9.10

Polymer

427 428 429 430 431 432 433 434 435 436 437 438

PMMA (10%) PMMA (30%) PBMA (10%) PBMA (30%) PMMA (10%) PMMA (30%) PEMA (10%) PEMA (30%) CRODA AC500 THERMOSET CRODA AC500 THERMOSET CRODA AC550 THERMOSET CRODA AC550 THERMOSET

439 440 441 442

LUMFLON LUMFLON LUMFLON LUMFLON

443 444 445 446

PLASTOKYD S27 (30%) PLASTOKYD SC140 (30%) PLASTOKYD SC400 (30%) PLASTOKYD AC4X (30%)

447 448 449

ALLOPRENE R10 (10%) ALLOPRENE R10 (30%) ALLOPRENE R10 (60%)

450

HYPALON 20 (30%)

Chlorosulfonated Polyethylene 20.30

3.20

0.70

11.30

451

POLYISOPRENE SW

Polyisoprene Swelling 17.00

4.00

4.00

7.30

452 453

BROMOBUTYL RUBBER S BROMOBUTYL RUBBER L

Bromobutyl Rubber Swelling 17.60 17.00

1.70 3.40

2.00 2.00

6.00 6.00

454 455 456 457 458

Supplemental Chemical Resistance Corrlations NEOPRENE CR 18.10 4.30 HYTREL +/- OK 24.20 14.60 HYTREL +/- NOT OK 26.40 18.80 HYPALON +/- OK 18.40 3.60 HYPALON +/- NOT OK 18.40 5.60

6.70 13.20 7.40 6.40 6.00

8.90 18.80 26.30 9.00 9.40

459

EVA 4055 SOL

3.70

4.70

LF200 LF200 LF916 LF916

(10%) (30%) (10%) (30%)

(10%) (30%) (10%) (30%)

Ethylene Vinylacetate (EVA) Solubility 17.70

3.50

7248_A002.fm Page 505 Wednesday, May 23, 2007 12:53 PM

Appendix A: Table A.2

505

TABLE A.2 (CONTINUED) Hansen Solubility Parameters for Selected Correlations Number

Polymer

Polar

Hydrogen Bonding

Interaction Radius

18.00 18.00

3.00 1.00

2.00 3.00

5.00 4.00

18.7 21.4 17.7 22.4 19.7

8.1 11.6 3.5 7.6 2.9

13.0 21.6 3.7 10.8 2.7

8.2 17.3 4.7 13.9 3.9

Dispersion COC Solubility

460 461

TOPAS 6013 SOL CZ RESIN SOL

462 463 464 465 466

KAURI GUM POLYVINYLPYRROLIDONE (PVP) PALM OIL BETHOXAZIN CARBON-60

Miscellaneous

7248_A002.fm Page 506 Wednesday, May 23, 2007 12:53 PM

7248_A003.fm Page 507 Wednesday, May 23, 2007 12:57 PM

Appendix A: Table A.3 COMMENTS TO TABLE A.3 The data in Appendix A.3 is for the solubility of the polymers listed in Table 5.2. 0.5 g of polymer was placed in a test tube with 5 ml of the given solvent (except for the Milled Wood Lignin). The evaluation of solubility was made on a scale as follows: 1. 2. 3. 4. 5. 6.

Soluble Almost soluble Strongly swollen, slight solubility Swollen Little swelling No visible effect

Note:* indicates a reaction. Data reproduced from Hansen, C.M., “The Three Dimensional Solubility Parameter and Solvent Diffusion Coefficient,” doctoral dissertation, Danish Technical Press, Copenhagen, 1967.

507

1 1 1 1 1 1 1 1 1 5 5 1 1 1 1 1 4 1 1 1 1 3 6 5 1 1 1 1 1

A

6 5 1 6 1 4 1 5 6 6 6 1 1 6 1 1 5 5 1 4 1 3 6 6 1 1 3 1 3

B

2 1 1 1 1 1 1 1 2 5 3 1 1 1 1 1 3 4 2 2 1 1 6 6 1 1 4 1 4

C

4 3 1 6 1 1 1 1 1 5 5 1 1 1 6 1 1 1 1 1 1 1 1 1 1 1 1 1 1

D

2 2 1 5 1 1 1 1 1 4 1 1 1 1 1 3 5 2 1 1 1 1 6 1 1 1 1 5 1

E

1 5 1 5 1 1 1 1 1 6 1 1 1 1 3 2 5 5 1 4 1 1 5 1 1 1 4 3 5

F

6 6 3 4 1 1 1 1 1 6 6 1 5 4 1 1 1 1 1 1 1 5 1 6 1 1 1 1 3

G

1 1 1 1 1 1 1 1 1 6 5 1 1 1 1 1 4 1 1 1 1 1 4 1 1 1 5 1 5

H

1 1 6 6 1 1 1 5 6 5 1 6 1 1 3 5 5 6 1 3 1 1 6 1 1 1 5 5 6

I

1 1 1 1 1 3 1 6 6 6 6 1 1 6 1 1 6 6 6 6 6 5 6 6 1 * 6 5 1

J

6 6 3 6 1 1 1 1 1 6 6 1 1 6 3 3 1 1 1 1 1 6 6 6 1 * 1 1 1

K

3 1 1 4 1 3 4 6 6 6 6 6 6 6 1 4 6 6 6 6 3 4 6 3 3 3 6 5 6

L

1 1 1 1 1 1 1 1 1 6 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

M

4 1 1 5 1 1 1 6 6 1 1 6 1 6 1 3 6 6 6 6 6 1 6 6 1 1 6 6 6

N

4 5 3 5 1 1 1 1 1 6 1 1 1 1 5 3 1 1 1 1 1 1 1 1 1 1 1 1 1

O

1 1 1 6 1 1 1 6 6 1 1 1 1 1 1 4 5 5 5 3 3 1 5 1 1 5 6 6 2

P

4 4 1 5 1 1 1 6 6 4 6 6 3 6 1 5 4 6 6 6 3 1 6 5 1 6 6 5 6

Q

1 1 1 2 1 1 1 1 1 6 4 1 1 1 1 1 3 1 1 1 1 1 5 2 1 1 1 1 1

R

1 1 1 1 1 1 1 2 4 1 3 1 1 1 1 1 3 5 1 3 1 1 5 1 1 * 4 1 1

S

6 6 6 6 1 6 2 1 1 5 6 1 6 4 6 6 1 1 1 1 1 5 1 5 1 1 1 5 1

T

6 2 1 5 1 1 1 1 1 5 6 1 1 5 1 1 4 5 1 1 1 1 6 5 1 1 1 1 1

U

6 6 6 6 5 6 5 1 1 6 6 1 6 4 6 5 1 1 1 1 1 5 1 5 1 1 1 6 1

V

6 6 6 6 5 6 5 1 5 6 1 1 5 1 6 6 1 1 1 1 1 1 1 1 1 1 1 6 1

X

6 6 6 6 2 6 4 1 1 6 5 1 6 4 6 6 1 1 1 1 1 5 1 5 1 1 1 5 1

Y

3 6 6 6 6 5 3 5 6 6 1 6 3 1 6 6 4 4 5 5 1 1 6 1 6 1 6 6 6

Z

4 5 1 4 1 1 1 1 1 6 3 1 1 1 5 1 1 1 1 1 1 1 2 1 1 1 1 1 1

A

1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 4 1 1 1 1 1 1

B

6 5 6 6 5 6 5 1 1 6 6 1 6 1 6 6 1 1 1 1 1 6 1 5 1 1 1 4 1

C

1 1 1 4 1 1 1 4 5 6 4 2 1 3 1 2 6 4 3 3 1 1 6 1 1 1 3 1 6

D

6 5 6 6 5 6 4 1 1 6 6 2 6 4 6 6 1 1 1 1 1 6 1 6 1 1 1 6 1

E

6 6 5 6 2 3 1 5 4 6 6 5 6 6 4 6 6 6 4 6 5 6 6 6 1 6 5 5 6

F

5 5 2 5 1 1 1 1 1 6 6 1 6 1 2 1 1 1 1 1 1 1 2 5 1 5 1 1 1

G

4 5 5 6 4 1 3 5 4 6 6 6 1 6 1 5 4 5 5 6 6 1 6 6 5 * 6 6 6

L

508

Solvent Acetic Acid Acetic Anhydride Acetone Acetonitrile Acetophenone Aniline Benzaldehyde Benzene 1-Bromonaphthalene 1,3-Butanediol 1-Butanol n-Butyl Acetate n-Butyl Lactate Butyric Acid Gamma-Butyrolactone Butyronitrile Carbon Disulfide Carbon Tetrachloride Chlorobenzene 1-Chlorobutane Chloroform m-Cresol Cyclohexane Cyclohexanol Cyclohexanone Cyclohexylamine Cyclohexylchloride Di-(2-Chloroethyl)ether Di-isobutyl Ketone

Polymer

TABLE A.3 Solubility Data for the Original 33 Polymers and 88 Solvents

7248_A003.fm Page 508 Wednesday, May 23, 2007 12:57 PM

Hansen Solubility Parameters: A User’s Handbook

Diacetone Alcohol o-Dichlorobenzene Diethyl Amine Diethyl Ether Diethylene Glycol Diethylene Glycol Monobutyl Ether Diethylene Glycol Monomethyl Ether Dimethyl Formamide 1,4-Dioxane Dipropyl Amine Dipropylene Glycol Ethanol 96% Ethanol 99.9% Ethanolamine Ethyl Acetate Ethyl Benzene Ethyl Lactate 2-Ethyl-1-Butanol 2-Ethyl-1-Hexanol 1,2-Dichloroethane Diethyl Sulfide Dimethyl Sulfoxide Ethylene Glycol Ethylene Glycol Monobutyl Ether Ethylene Glycol Monoethyl Ether Ethylene Glycol Monoethyl Ether Acetate Ethylene Glycol Monomethyl Ether Formic Acid 90% Furan Glycerol n-Hexane Isoamyl Acetate Isobutyl Isobutyrate Isophorone Mesityl Oxide Methanol

1 1 1 3 6 3 1 1 1 1 6 4 5 6 1 1 1 3 4 1 1 4 6 1 1 1 4 1 1 4 6 1 1 1 1 5

5 1 6 6 6 3 1 1 1 6 6 6 6 6 1 5 1 2 2 1 4 1 6 5 5 1 3 6 4 6 5 1 5 1 1 6

1 2 1 3 1 1 1 1 1 1 1 4 3 1 1 4 1 5 5 1 4 1 6 1 1 1 1 5 1 6 6 1 5 1 1 4

1 1 1 1 5 1 1 1 1 1 6 6 5 6 1 1 1 3 1 1 1 3 4 1 1 1 1 3 1 5 1 1 1 1 1 5

1 1 1 4 3 1 1 1 1 1 1 4 1 2 1 2 1 1 1 1 1 1 6 1 1 1 1 6 1 6 6 1 5 1 1 4

1 1 1 5 5 1 1 1 1 1 5 1 1 6 1 3 1 1 1 1 5 2 6 1 1 1 1 1 1 6 6 1 5 1 1 5

5 1 1 3 6 4 6 1 1 1 6 6 6 6 1 1 6 6 6 1 1 6 6 6 5 1 6 6 1 6 3 1 1 1 1 6

1 1 5 4 6 1 1 1 1 5 5 1 4 6 1 1 1 5 5 1 3 1 6 2 1 1 1 1 1 6 6 1 5 1 1 1

1 2 4 6 4 1 1 1 6 6 1 1 1 5 6 4 1 1 1 1 5 1 6 1 1 5 1 1 3 6 6 5 5 1 5 3

1 6 5 6 1 1 1 1 6 5 1 6 3 * 1 6 1 6 6 6 6 1 6 1 1 1 1 6 6 6 6 1 1 1 1 1

3 1 5 6 6 1 4 1 1 5 6 6 6 6 1 1 5 6 6 1 1 1 6 2 5 1 6 6 3 6 6 1 1 1 1 6

4 6 6 6 6 6 4 1 1 6 6 6 6 4 5 6 1 6 6 5 6 1 6 6 6 4 4 1 6 6 6 6 6 6 6 6

1 1 1 1 4 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 6 1 1 1 1 5 1 6 1 1 1 1 1 2

1 6 1 5 1 1 1 1 1 1 1 1 1 1 1 6 1 6 6 6 6 1 1 1 1 1 1 1 6 6 6 6 6 1 1 1

1 1 1 1 6 1 1 1 1 1 5 5 5 5 1 1 1 1 1 1 1 4 6 1 1 1 2 5 1 6 5 1 1 1 1 4

1 4 5 3 1 1 1 1 1 5 1 1 1 1 1 6 1 1 1 4 5 1 6 1 1 1 1 6 5 6 5 5 6 1 1 1

1 6 5 6 1 1 1 1 1 5 1 5 5 5 1 6 1 6 6 5 6 1 6 1 1 1 1 2 5 6 6 6 6 1 1 5

1 1 1 5 5 1 1 1 1 1 5 5 4 6 1 1 1 3 4 1 1 1 5 1 1 1 1 5 1 4 5 1 1 1 1 4

1 1 1 3 1 1 1 1 1 2 1 1 1 1 1 3 1 3 4 1 4 1 2 1 1 1 1 1 1 3 6 1 3 1 1 1

6 1 1 1 6 6 6 6 1 1 6 6 6 6 3 1 6 6 6 1 1 6 6 6 6 5 6 6 1 6 3 1 1 1 1 6

1 1 1 6 6 4 3 1 1 4 6 6 6 6 1 1 4 6 6 1 1 5 4 4 4 1 5 6 1 6 6 1 1 1 1 5

6 1 1 1 6 5 6 5 4 1 6 6 6 6 5 1 6 6 5 1 1 6 6 6 6 6 6 6 1 5 1 1 1 1 1 6

6 1 1 1 6 5 6 5 5 1 6 6 6 6 1 1 6 1 1 1 1 6 6 1 6 6 6 6 1 6 1 1 1 1 1 6

6 1 1 1 6 6 6 6 1 1 6 6 6 5 5 1 6 6 5 1 1 6 5 6 6 6 6 6 1 6 1 1 1 1 1 6

6 6 5 6 6 6 6 6 6 6 6 5 6 6 6 6 6 1 1 6 6 6 6 4 6 6 6 3 6 6 6 6 6 6 6 5

3 1 1 1 6 1 5 4 1 1 6 5 4 6 1 1 3 1 1 1 1 6 6 1 1 1 4 6 1 6 4 1 1 1 1 6

1 1 1 1 1 1 1 1 1 1 1 1 1 6 1 1 1 1 1 1 1 1 6 1 1 1 1 1 1 6 6 1 1 1 1 1

6 1 1 1 6 6 6 6 4 1 6 6 6 6 4 1 6 5 1 1 1 6 6 6 6 6 6 6 1 6 1 1 1 1 1 6

1 3 2 4 1 1 1 1 1 3 1 5 5 1 1 6 1 4 4 1 4 1 6 1 1 1 1 6 4 6 6 4 5 6 1 5

6 1 1 1 6 6 6 6 5 1 6 6 6 6 4 1 6 6 5 2 1 6 6 6 6 6 6 6 1 6 2 1 1 1 4 6

6 4 6 6 6 6 6 4 3 6 6 6 6 6 6 6 6 6 6 3 6 2 6 6 6 5 6 6 6 6 6 5 6 6 2 6

5 1 1 1 6 1 4 1 1 1 6 6 6 6 1 1 5 6 6 1 1 6 6 1 2 1 5 6 1 6 5 1 1 1 1 6

1 4 5 6 1 1 1 1 2 6 1 5 6 1 6 6 3 6 6 6 6 1 1 6 1 6 1 1 6 6 6 6 6 6 4 5

7248_A003.fm Page 509 Wednesday, May 23, 2007 12:57 PM

Appendix A: Table A.3 509

A 1 1 4 1 1 1 1 1 1 1 1 4 5 3 6 1 1 1 1 1 1 1 1

Polymer

Methyl Ethyl Ketone Methyl Isoamyl Ketone Methyl Isobutyl Carbinol Methyl Isobutyl Ketone Methylal Methylene Dichloride Morpholine Nitrobenzene Nitroethane Nitromethane 2-Nitropropane 1-Pentanol 1-Propanol Propylene Carbonate Propylene Glycol Pyridine Styrene Tetrahydrofuran Tetrahydronaphthaline Toluene 1,1,1-Trichloroethane Trichloroethylene Xylene

1 1 5 1 1 1 1 1 1 2 2 5 3 1 5 1 2 1 4 2 4 2 3

C 1 1 1 1 1 1 1 1 1 5 1 3 5 5 6 1 1 1 1 1 1 1 1

D 1 1 1 1 1 1 1 1 5 6 6 1 1 6 4 1 1 1 1 1 1 1 5

E 1 1 1 1 1 1 1 1 5 5 5 1 1 6 6 1 1 1 2 2 4 1 3

F 1 1 6 1 1 1 1 1 2 5 1 6 6 6 6 1 1 1 1 1 1 1 1

G 1 1 5 1 1 1 1 1 1 1 1 2 4 1 6 1 1 1 5 1 1 1 4

H 6 6 1 6 6 1 1 1 6 5 6 1 1 6 1 1 3 1 5 2 6 1 5

I 1 1 6 1 1 6 1 1 1 1 1 6 6 1 6 1 6 1 6 6 6 6 5

J 1 1 6 1 5 1 1 1 5 6 3 6 6 6 6 1 1 1 1 1 1 1 1

K 4 6 6 6 5 5 1 4 5 1 5 6 6 3 6 1 6 1 6 6 6 6 6

L 1 1 1 1 1 1 1 1 1 2 1 1 1 1 6 1 1 1 1 1 1 1 1

M 1 1 6 1 3 6 1 5 6 6 6 1 1 1 1 1 6 1 6 6 6 6 6

N 1 1 1 1 1 1 1 1 3 6 4 1 2 6 6 1 1 1 1 1 1 1 1

O 1 1 1 1 1 5 1 3 6 6 6 1 1 6 4 1 4 1 4 6 3 4 6

P 1 6 6 4 5 3 1 2 5 5 4 6 6 1 6 1 6 1 6 6 6 5 6

Q 1 1 4 1 1 1 1 1 1 2 1 5 4 1 5 1 1 1 1 1 1 1 1

R 1 1 4 1 1 1 1 1 1 1 1 6 2 1 1 1 1 1 2 3 3 1 3

S 4 1 6 1 5 1 4 1 6 6 6 6 6 6 6 1 1 1 1 1 1 1 1

T 1 1 5 1 1 1 1 1 1 1 1 6 6 4 6 1 1 1 1 1 1 1 1

U 4 1 6 1 4 1 5 1 6 6 6 6 6 6 6 5 1 1 1 1 1 1 1

V 1 1 1 1 1 1 4 6 6 6 6 1 1 6 6 4 1 1 1 1 1 1 1

X 4 1 6 4 3 1 5 1 6 6 5 6 5 6 6 1 1 1 1 1 1 1 1

Y 6 6 1 6 6 2 6 6 6 6 6 1 1 6 6 5 6 1 5 6 6 2 6

Z 1 1 1 1 1 1 1 1 3 4 1 1 5 6 6 1 1 1 1 1 1 1 1

A 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

B 5 1 5 1 5 1 6 5 6 6 6 6 6 6 6 5 1 1 1 1 1 1 1

C 1 4 4 4 2 1 1 1 3 4 3 4 4 1 5 1 4 1 5 5 4 4 5

D 5 2 6 4 3 1 6 6 6 6 6 6 6 6 6 5 1 1 1 1 1 1 1

E

3 5 6 4 6 2 1 2 6 6 5 6 6 6 6 2 5 1 6 5 6 5 6

F

1 1 6 1 1 1 1 1 2 5 1 6 6 6 6 1 1 1 1 1 1 1 1

G

6 6 6 6 6 6 1 4 6 6 6 6 6 6 3 1 5 3 4 5 6 5 6

L

510

1 1 5 1 5 1 1 1 1 6 1 6 6 1 6 1 2 1 3 2 5 4 4

B

TABLE A.3 (CONTINUED) Solubility Data for the Original 33 Polymers and 88 Solvents

7248_A003.fm Page 510 Wednesday, May 23, 2007 12:57 PM

Hansen Solubility Parameters: A User’s Handbook

7248_Index.fm Page 511 Wednesday, May 9, 2007 10:49 AM

Index A Absorption concentration-dependent, 298 equations, 294–296 film formation by solvent evaporation and, 305–306 plastics, 315–316, 339 side effects, 304–305 surface resistance and, 300–301 Acetophenone, 59, 262 Acid-base theories and HSP of pigments, 130–131 Acids amines and, 140 dimerization, 65 hydrophilic bonding, 133–134 ACLAR®, 251, 252–253 Acrylonitrile/butadiene/styrene (ABS) terpolymer, 120, 300 Active agents, surface, 332–333 Activity coefficient models Flory-Huggins models, 82–84 future challenges, 90–91 at infinite dilution, 85–88 mixed solvent-polymer phase equilibria, 88–90 Adenine, 275 Adhesion maximization, physical, 122, 342 Adsorption, controlled, 132–134 Affinity and Hansen solubility parameters, 4–5 Alcohol in the bloodstream, 275–276 Alpha-helices, 272 Alternative systems, solvent, 312 Amines, 140 Amino acid side chains and water, 270–271 Ammonia, 57 Amphipathic molecules, 270–271 Anomalous diffusion, 306–308 Aromas and fragrances, 338–339 Asphalt, 152 Asphaltenes, 153, 164–166 definition, 154 molecular weight of, 154 polarity, 155 Availability of HSP data, 321–322 Azeotropes, 212, 213–215, 216–226

B Barrier polymers breakthrough times, 245–247 concentration-dependent diffusion, 244–245 human skin as, 250, 316 laminated, 253–254

RED numbers, 247, 253 solubility parameter correlation of permeation coefficients for gases, 251–254 solubility parameter correlation of permeation rates, 248–250 solubility parameter correlation of polymer swelling, 250 solubility parameter correlations based on permeation phenomena, 245–250 Benzene, 204 Beta sheets, 272 Binders coating, 138, 139, 144 pigment, 128–131 Biological materials blood serum and zein, 279, 280–281 chiral rotation, hydrogen bonding and nanoengineering, 290–291 chlorophyll, 279 cholesterol, 275–277 DNA, 273–275 HSP characterization of, 270–271 human skin, 250, 277–279 hydrophobic bonding and hydrophilic bonding of, 271–273 lard, 277 lignin, 279, 282–283, 284–288, 316–317 surface mobility, 290 urea, 274, 283, 289 water, 21–22, 289 wood chemicals and polymers, 279, 281 BISOM test, 170–173 Bitumens asphaltenes in, 153, 154–155 BISOM test, 170–173 calculation and plotting of Hansen 3D pseudosphere, 161–164 components of, 164–166 Hildebrand solubility parameters, 156, 158 HSP, 158–164 hydrocarbons in, 155–156, 164–166 models of, 152–154 polymer modified, 152 polymers and, 166–169 production process, 151–152 RED numbers, 160 solubility parameters of, 155–156 solubility SPHERE, 159–164 solvents used for determination of solubility of, 161–164 temperature effect on, 168 testing solubility of, 156, 157 turbidimetric titrations, 170

511

7248_Index.fm Page 512 Wednesday, May 9, 2007 10:49 AM

512 uses, 151, 160 Blanc fixe, 131 Blistering of coatings, 141, 281 Blood serum, 279, 280–281 Boundaries of solubility, 7–8 Breakthrough times, HSP correlations of, 245–247 Butadiene block copolymer (SBS), 168 Butyl rubber, 246

C Calculations versus experimental χ12 parameters, 34–39 polymer HSP, 97–98 Carbon dioxide, 57, 252 Hildebrand parameters of, 187–190 HSP of, 336–337 ideal solubility of gases in liquids and, 199–201 pressure effects on solubility parameters of, 187, 191–196 solubility data in various solvents, 178–186, 187 solubility in liquid solvents, 185–186 solubility parameters, 141–142 temperature effects on solubility parameters of, 190–191 three-component solubility parameters, 189–190 Carbon disulfide, 16–17 Carbon fiber surface characterization, 131–132, 133 Carbon tetrachloride, 35–36, 204 Carcinogens, 312 Castor oil, 214 Cellulose, 281, 289 acetate, 303 Challenge® 5100, 246 Chemical protective clothing, 315 Chemical resistance acceptable-or-not data on, 232 effects of solvent molecular size on, 232–233 HSP characterization of, 231–232, 339 plastics, 234–237 special effects with water, 238–239 tank coatings, 233–234 tensile strength and, 237–238 Chiral rotation and HSP, 290–291 Chlorobenzene, 245, 305 Chloroform, 35–36 Chlorophyll, 279 Cholesterol, 275–277 Cleaning solvents application of HSP methodology to, 212–213 azeotropes, 212, 213–215, 216–226 calculating HSP of composites and, 205–206 HSP of multiple-component soils and, 204–205 identification of designer, 208 method for choice of suitable, 206–208 pathology of soils and, 204 performance, 210–212 reference soils, 208, 209–210 variety of chemicals used in, 204

Hansen Solubility Parameters: A User’s Handbook Coatings amines added to, 140 binders, 138, 139, 144 blistering in, 141, 281 chemical resistance and, 231–239, 240 gases as solvents in, 141–142 for hazardous materials, 312 PET film, 234, 235 polymer compatibility in, 145–146 RED numbers, 142–143, 144 self-stratifying, 120–121 solvents, 137–142 and surface phenomena in, 144–145 SPHERE and, 143 surface mobility, 333–334 tank, 233–234 water-reducible, 139–140 Cohesion energy parameters, 2, 5, 13–18, 113 Cohesive energy density, carbon dioxide, 187 Cohesive energy difference, 27 polymer segments and solvents, 30–32 Composite soils, 204–206 Compound formation, 9 Concentration-dependent diffusion, 244–245, 297–298, 304–305 Constant diffusion coefficients, 296–297 Controlled adsorption, 132–134 Controlled release of drugs, 339–340 Cooling, rapid, 19 Corresponding states theory (CST), 3, 6 HSP and, 30–32 predicting the χ12 parameter using, 28 Critical chi parameter, 32 Critical surface tensions, 116–117 Crude oils BISOM test, 170–173 HSP of molecules in, 169–170 origins and structure, 152–153 uses, 151–152 Cyclic olefinic copolymer (COC), 120, 300–301 Cyclohexanone, 185

D Danish MAL system, 313–315 Data, HSP availability, 321–322 quality, 324, 326 Dendrimers, 90–91 Designer solvents. See also Cleaning solvents analysis of capability of, 213–215 application of HSP methodology to, 212–213 azeotropes, 212, 213–215, 216–226 cleaning performance of, 210–212 defined, 203 identification of, 208 variety of chemicals used in, 204 Desirability function, 20 Dextran C, 101–102, 322, 323–324, 325–326 Dibasic esters (DBE), 315

7248_Index.fm Page 513 Wednesday, May 9, 2007 10:49 AM

Index 1,2-dibromoethane, 185 Dichloromethane, 185 Diethyl ether, 17, 262 Diffusion anomalous, 306–308 concentration-dependent, 244–245, 297–298, 304–305 constant coefficients, 296–297 equations, 294–296 film formation by solvent evaporation and, 305–306 side effects, 304–305 steady state permeation, 296 surface effects and, 302–304 surface resistance and, 298–305 time-dependent, 308 vapor, 301–302 Dilution, infinite, 85–88 Dimerization, acid, 65 Dimethyl sulfoxide, 289 1,4-dioxane, 262 Dipolar interactions, 50–52, 66 Dipole moments, 8–9 zero, 16–17 Dispersion solubility parameters, 5, 13–16 hydrogen bonding, 52–59 Distance, solubility parameter, 7 DNA, 273–275 Dynamic shear rheometer (DSR), 152

E Elastomers, 245, 333 Electron exchange parameters, 5, 29 Electrostatic repulsion, 140 Energy density, cohesive, 13–16 Gibbs, 92–93 polar cohesive, 5 relative difference, 8 surface free, 113–114 of vaporization, 13–16 Entropic group contribution model, 78–79, 83 Environmental impact of solvents, 208 Environmental stress cracking (ESC), 107, 120 incidence of, 259–260 interpreted using HSP, 260–262 mechanism for, 259–260, 264–267 with nonabsorbing stress cracking initiators, 263–264 RED number and, 260–262 EPDM rubber, 141 Epon® 1001, 233 Epoxies, 138, 141, 233–234 Equation-of-state framework, 46–50, 60–62 carbon dioxide, 188–189 Ethanol, 4–5, 276 Ethyl acetate, 262 Ethylene glycol monomethyl ether (EGMME), 245 Ethylene vinyl alcohol copolymers (EVOH), 253–254 Evaporation, solvent, 305–306

513

F Fickian diffusion, 306–308 Field ionization mass spectrometry (FIMS), 154 Fillers and fibers, 147–148 Film formation by solvent evaporation, 305–306 Films, PET, 234, 235 First-order and second-order groups in solubility parameter prediction, 66–73 Floating rates of particles in pigments, 126–127 Flocculation, 153 Flory-Huggins model, 27–28, 77, 341 versus the GC methods, 84–90 for multicomponent mixtures, 92–93 regular solution theory and, 80–82 using Hildebrand and HSP, 82–84 Fluoropolymers, 16–17 Formamide, 274 Fragrances and aromas, 338–339 Free-volume-based models for polymers, 77–79 solvent selection for paints, 85–88 Fresh Air Numbers (FAN/MAL) system, 313–315

G Gases characterization problems, 336–337 HSP of, 336–337 ideal solubility in liquids, 199–201 permeation through polymers, 243 solubility parameter correlation of permeation coefficients for, 251–254 as solvents, 141–142 Gibbs energy, 92–93 Glycol ethers, 138, 140–141 Group-contribution method applications, 76–77 entropic model, 78–79, 83 Flory-Huggins/Hansen model versus, 84–90 Flory-Huggins model and regular solution theory, 80–82 free-volume-based models for polymers, 77 reliability of calculation procedures for, 328 UNIFAC model, 76–78, 83 Guggenheim expression, 47

H Halar®, 235, 236 Hansen 3D pseudosphere, 161–164 Hansen solubility parameters (HSP) acid-base theories and, 130–131 aromas and fragrances, 338–339 biological materials, 269–291 bitumen, 158–164 blood serum and zein, 279, 280–281 carbon dioxide, 177–201 for chemical protective clothing, 315

7248_Index.fm Page 514 Wednesday, May 9, 2007 10:49 AM

514 chemical resistance characterization using, 231–239, 240, 339 chiral rotation and, 290–291 chlorophyll, 279 cholesterol, 275–277 coatings, 145–146 composite soils, 204–206 controlled release drugs, 339–340 correlation of permeation coefficients for gases, 251–254 correlation of permeation rates, 248–250 correlation of polymer swelling, 250 correlation of zeta potential for blanc fixe, 131 correlations for chemical resistance, 234–236 correlations of breakthrough times, 245–247 correlations with surface tension, 113–114 corresponding states theories and, 30–32 data availability, 321–322 quality, 324, 326 defined, 1, 4–6 Dextran C, 322, 323–324, 325–326 electron exchange energy and, 29 environmental stress cracking (ESC) and, 260–267 fillers and fibers, 147–148 first-order and second-order groups contribution to, 66–73 Flory-Huggins model and, 81–82 gases, 336–337 GC methods versus the FH/, 84–90 human skin, 277–279, 316 hydrogen bonding and, 290–291 inorganic salts, 337 lard, 277 lignin, 279, 282–283, 284–288 methodology applied to cleaning operations, 212–213 methods and problems in determination of, 6–13 mixtures, 205–206 molecules in crude oils, 169–170 multiple-component soils, 204–206 nanoengineering and, 290–291, 340–341 organic compounds, 177–178 organic salts, 337 organometallic compounds, 338 pigments, 128–131 fillers and fibers, 125–134 plastics, 315–316, 339 polymers, 29–30, 95–109 solvents, 29–30, 40–41, 137–143 special effects with water, 238–239 surface characterization, 113–122, 131–132, 330–332 liquid adhesion, 342 mobility and, 263–264, 290 tensile strength and, 237–238 three-component, 189–190 urea, 283, 289 water, 21–22, 289, 326, 334–336 wood chemicals and polymers, 279, 281 χ12 parameter and, 32–39

Hansen Solubility Parameters: A User’s Handbook Hazardous materials chemical protective clothing for, 315 Danish MAL system, 313–315 skin penetration of, 316 solvent formulation and personal protection for least risk from, 313 substitutions for, 311–312 transport phenomena, 316–317 uptake by plastic containers, 315–316 Heat of vaporization, 50 Hemicelluloses, 281, 289 Henry's law coefficient, 178 Hexane, 116, 120, 204 Hildebrand parameters, 2–4 of bitumen, 155, 156, 158 of carbon dioxide, 187–190 solvent selection for paints, 85–88 χ12 coefficient and, 28, 32 Hoy dispersion parameter, 9, 13 Human skin HSP of, 277–279, 316 penetration of hazardous materials, 316 as polymeric barrier, 250 Hydrocarbons in bitumen, 155–156, 164–166 carbon dioxide solubility parameters and, 192–194, 336–337 coatings applications, 138–139 and corresponding states theories, 29 in crude oils, 153 ether solvents partially based on silicon, 208 Hydrochlorides, 339–340 Hydrofluoroethers (HFEs), 208 Hydrogen bonding, 5, 7 acid dimerization and, 65 coatings, 138 cohesive energy differences and, 31–32 defined, 269 dipole moments and, 9 DNA, 273–275 HSP and, 290–291 lattice-fluid, 46 nonrandom, 46, 50–59 polymer surfaces, 115 self-stratifying coatings, 120–121 solubility parameters, 17, 52–59 techniques for data treatment, 142–143 temperature increases and, 140–141 water temperature and, 141 Hydrophilic bonding, 133–134, 271–273, 333–334 Hydrophobic bonding, 271–273, 332

I Inorganic salts, 337 Interaction radius, 184 Internal pressure, 187 Intrinsic viscosity of polymers, 107–109

7248_Index.fm Page 515 Wednesday, May 9, 2007 10:49 AM

Index

515

J

N

Joffé effect, 342

Nanoengineering, 290–291, 340–341 n-butanol, 138, 140 n-butyl acetate, 262 Neoprene®, 235 New Flory theory, 27–28 n-heptane, 153, 154, 165–166 n-hexane, 243, 262 Nitrates, 337 Nitrile rubber, 246 Nitromethane, 4–5 N-methyl-morpholine-N-oxide, 279 N-methyl-2-pyrrolidone, 204 Nonabsorbing stress cracking initiators, 263–264 Nonbiodegradable substances, 312 Nonpolar interactions in organic materials, 5 Nonrandom hydrogen bonding, 46, 50–59 Normal diffusion, 306–308 Nylon-6, 303

K Keratin, 277, 317 Ketones, 138, 140

L Laminates, 253–254 Lard, 277 Lattice-fluid hydrogen bonding (LFHB), 46 Lignin, 279, 282–283, 284–288, 316–317 Liquid-liquid phase separation (LLE), 75 Liquids absorption by plastic containers, 315–316 adhesion phenomena, 342 cohesion energy parameters and surface characterization of, 114–116 environmental stress cracking (ESC) and, 262 first-order and second-order groups in, 66–73 ideal solubility of gases in, 199–201 solubility parameters for, 17–21, 32–33, 52–59 solvents, CO2 solubility in, 185–186 spontaneous spreading, 115–118 surface tension, 116–117 Lithium, 338 Lorenz-Berthelot mixtures, 41 Low density polyethylene (LDPE), 249 Lower critical solution temperature, 28 Lydersen group constants, 12–13t, 15

O Octyl alcohol, 277 Oleic acid, 185 Olive oil, 277 One-component Hildebrand parameter as a function of temperature and pressure, 187–189 Organic compounds, HSP and solvents for, 177–178 Organic salts, 337 Organoclays, 340–341 Organometallic compounds, 338 Ostwald coefficient, 178 Ozone depletion potentials (ODP), 208

M

P

MAL system, Danish, 313–315 Maltenes, 153, 164–165 MATLAB platform, 163 Melting points of polymers, 106–107 Metallic bonding, 338 Methanol, 19, 141, 238–239 2-methylcyclohexanone, 185 Methylene chloride, 140 Methyl ethyl ketone, 140 Methyl iodide-cellulose acetate, 303 Methyl isobutyl ketone, 262 Methyl methacrylate, 277 Mixtures, solubility parameters of, 205–206, 283, 289 Mobility, surface, 263–264, 290, 333–334 Molar volume and solubility parameters, 7 Molecular size and solubility parameters, 19–20, 32–33, 232–233, 243 Morpholine, 108 Multicomponent mixtures, Flory-Huggins model for, 92–93 Multiple-component soils, 204–206 Multivariable analysis and solubility parameters, 8

Paints hydrophilic bonding in, 272 older, dried, 142–143 self-stratifying coatings, 120–121 solvent selection for, 85–88 Partial solubility parameters determination, 6–13 dipolar interactions and, 50–52 for liquids, 17–21, 52–59 Particle suspensions, 126–127 Pathology of soils, 204 P-dioxane, 303 Peat moss, 333 Pentachlorophenol, 316 Permeation of liquid or gas through polymers, 243 rates, 248–250 size and shape and dependence of diffusion coefficient, 255–256 solubility parameter correlation of permeation coefficients for gases, 251–254 solubility parameter correlations based on, 245–250 steady state, 296

7248_Index.fm Page 516 Wednesday, May 9, 2007 10:49 AM

516 surface resistance in, 301–302 through human skin, 277–279, 316 PET film coating, 234, 235 Phase equilibria, mixed solvent-polymer, 88–90 Phenyl acetate, 262 Physical adhesion maximization, 122 Pigments, fillers, and fibers binders, 128–131 carbon fiber, 131–132, 133 cohesion parameter characterization study, 125 controlled adsorption, 132–134 manufacturing, 128 organic versus inorganic, 130–131 sedimentation rates, 126–127, 129 solvents, 128–131, 144–145 surface characterization methods, 126–127 Plasticizers, 21, 142, 244 Plastics absorption of chemicals in, 315–316, 339 bitumen in, 167 chemical resistance, 234–237 containers uptake, 315–316 environmental stress cracking and, 107 HSP correlations acceptable or not, 234–237 SWEAT and, 141 Plastomers, 167 Points, polymers as, 328, 330 Polar bonding and cohesive energy acid groups, 132–134 differences, 31–32 Polarity of asphaltenes, 155 Polar solubility parameters, 5, 16–17 Polyacrylonitrile (PAN), 39, 100, 132, 252 Polyamides (PA), 253–254, 260 Polybutadiene, 35–36 Polycarbonate (PC), 120, 260, 300 Poly(chlorotrifluoroethylene), 252 Polydimethyl siloxane (PDMS), 77 Polydimethylsiloxanes, 264 Polyesterimide, 108–109 Polyesters (TPU), 333 Polyether ether ketone (PEEK), 260 Polyether/polyamide block copolymer (PEBA), 333 Polyether sulfone, 141 Polyethersulfone (PES), 167–168, 238 Poly(ethylene co-chlorotrifluoroethylene) (ECTFE), 235, 236 Polyethylene (PE), 118, 246, 260, 303 low density, 249 Polyethylenesulfide, 167–168, 327–328, 329–330 Polyethylene terephthalate (PET), 252, 253–254 Polyisobutylene, 36–38 Polymer modified bitumen (PMB), 152 Polymers barrier concentration-dependent diffusion and, 244–245 human skin as, 250, 316 laminated, 253–254 RED numbers, 247, 253 solubility parameter correlation of permeation coefficients for gases, 251–254

Hansen Solubility Parameters: A User’s Handbook solubility parameter correlation of permeation rates, 248–250 solubility parameter correlations based on permeation, 245–250 swelling, 250 bitumen and, 166–169 calculated versus experimental χ12 parameters, 34–39 coatings, 139–140 cohesive energy differences between solvents and, 30–32 compatibility in coatings, 145–146 concentration and χ12, 39 concentration-dependent diffusion of, 244–245 dendrimer, 90–91 environmental stress cracking, 107 experimental evaluation of behavior of, 95–97 Flory-Huggins models using Hildebrand and HSP, 82–84 group contribution applications, 76–77 entropic model, 78–79 Flory-Huggins model and regular solution theory, 80–82 free-volume-based models for polymers, 77 UNIFAC-FV model, 77–78 high temperature solvents, 140–141 HSP for, 29–30 calculation of, 97–98 solubility and, 98–106 SPHERE program and, 95–96, 99–106, 108–109 two-dimensional plot of, 96–97 intrinsic viscosity of, 107–109 melting point determinations, 106–107 permeation of liquid or gas through, 243 as points, 328, 330 RED number, 98–100 rigid, 245 solubility, 341 solution thermodynamics, 2–4, 19, 75–76 -solvent mixtures phase equilibria, 88–90 SPHERE program and, 95–96, 99–106, 108–109 swelling of, 106 wood, 279, 281 Polymethyl methacrylate (PMMA), 243, 260 Polyolefins, 252, 253–254 Polyphenylene sulfide (PPS), 20, 141, 237 Polypropylene, 59, 106, 302 Polystyrene (PS), 38, 260 Polytetrafluoroethylene (PTFE), 252, 260 Polyurethanes, 333 Polyvinylacetate, 39 Polyvinyl acetate, 243 Polyvinylacetate, 305 Polyvinyl alcohol, 252 Polyvinyl chloride (PVC), 116–117, 243 Polyvinylchloride (PVC), 260 Polyvinyl chloride (PVC), 264–267 Polyvinylidine chloride (PVDC), 106–107 Pressure effects on carbon dioxide solubility parameters, 187, 191–196 Pressure-volume-temperature (PVT) data, 52–59

7248_Index.fm Page 517 Wednesday, May 9, 2007 10:49 AM

Index Prigogine corresponding states theory (CST), 3, 6 HSP and, 30–32 predicting the χ12 parameter using, 28 Prigogine-Patterson theory, 31, 40–41 Propylene bromide, 185 Protective clothing, chemical, 315 Proteins in blood serum, 279, 280–281 Pyrimidine, 275

Q Quasi-chemical condition, 47–49

R Rapid cooling, 19 Rates, permeation, 248–250 RED (relative energy difference) numbers, 8, 21, 40 barrier polymers, 247, 253 bitumen, 160 carbon dioxide, 185 coatings, 142–143, 144 environmental stress cracking and, 260–262 polymer solubility and, 98–100 solvent quality and, 142 transport phenomena and, 316 Reference soils, 208, 209–210 Refrigerants, 208 Regular solution theory and Flory-Huggins model, 80–82 Rehbinder effect, 120, 342 Relative energy difference, 8 Resins, 153 Resistance, surface in absorption experiments, 300–301 mathematical background, 298–299 measuring diffusion coefficients with, 304–305 in permeation experiments, 301–302 skin layer effect, 302–303 Rhodamin FB, 331, 334–335 Rigid polymers, 245 RNA, 274

S Safety and health issues. See Hazardous materials Salts inorganic, 337 organic, 337 SARA analysis, 153 Sedimentation rates, 126–127, 129, 130 Self-assembly (controlled adsorption), 132–134, 341 Self-stratifying coatings, 120–121 Side effects, 304–305 Silicon, hydrocarbon ether solvent partially based on, 208 Similia similibus solvuntur, 45–46 Skin, human. See Human skin Skin layer effects, 302–303 Soils composite, 204–206, 213–215

517 HSP of multiple-component, 204–205 method for choice of suitable solvents for, 206–208 mixtures with solvents, 205–206 pathology of, 204 reference, 208, 209–210 Solubility parameters. See also Hansen solubility parameters (HSP); Hildebrand parameters of bitumen, 155–161 boundaries and, 7–8 carbon dioxide, 141–142 dipolar interactions and, 50–52 dispersion, 13–16 distance equation, 7 first-order and second-order groups contribution to, 66–73 hydrogen bonding, 17 introduction to, 1–2 for liquids, 17–21 partial, 6–13, 17–21 polar, 5, 16–17 polymers, 341 of solvent-solute combinations, 3–4 study of, 45–46 surface science, 6 temperature dependence and, 18–19 for water, 21–22 Solubility sphere, 184 Solvents absorption, 300–301 alternatives, 312 azeotropes, 212, 213–215, 216–226 biological materials as, 274–275 carbon dioxide solubility in various, 178–186, 187 cleaning analysis of capability of, 213–215 application of HSP methodology to, 212–213 calculating HSP of mixtures of soil and, 205–206 HSP of multiple-component soils and, 204–205 method for choice of suitable, 206–208 pathology of soils and, 204 performance, 210–212 reference soils, 208, 209–210 variety of chemicals used in, 204 coating applications, 137–148 cohesive energy differences between polymers and, 30–32 diffusion coefficients, 244–245 environmental impact of, 208 evaporation film formation, 305–306 formulation and personal protection for least risk, 313 gases as, 141–142 for hazardous materials, 312 high temperature, 140–141 HSP for, 29–30, 32–33, 40–41, 137–142 for human skin, 277–279 hydrocarbon base, 138–139, 208 for lard, 277 mixtures with soils, 205–206 molecular size and solubility parameters, 19–20, 232–233 permeation rates, 248–250

7248_Index.fm Page 518 Wednesday, May 9, 2007 10:49 AM

518 pigments adsorption and, 128–131, 144–145 polymer compatibility in coatings, 145–146 -polymer mixtures phase equilibria, 88–90 selection for paints, 85–88 solubility of bitumens in, 156, 157–158 solute-combination solubility parameters, 3–4 as spheres, 328, 330 and surface phenomena in coatings, 144–145 titration tests, 170–173 used as refrigerants, 208 used for determination of solubility of bitumens, 161–164 vegetable oils in, 142 viscosity, 145–146 SPHERE and SPHERE1 analysis, 20–21 bitumen solubility and, 159–164 cholesterol, 275–276 coatings and, 143 Hansen 3D pseudo-, 161–164 polymers and, 95–96, 99–106, 108–109 Spheres, solvents as, 328, 330 Spontaneous spreading of liquid droplets, 115–118, 120 Steady state permeation, 296 Stress corrosion cracking (SCC), 260 Styrene, 277 Styrene-acrylonitrile copolymer (SAN), 260 Styrene/butadiene/styrene block copolymer (SBS), 333 Substitutions, 311–312 Sulfur trioxide, 253 Super Case II behavior, 306–308 Surfaces active agents, 332–333 -active agents in coatings, 140 carbon fiber, 131–132 characterizations and comparisons using HSP, 118–120, 131–132, 330–332 cohesion energy parameters evaluation for, 114–116 film formation by solvent evaporation and, 305–306 free energy, 113–114 HSP for, 113–122 liquid adhesion phenomena, 342 mobility, 263–264, 290, 333–334 phenomena in coatings, 144–145 physical adhesion maximization, 122 pigment, fillers, and fiber, 126–131 resistance, 298–305 self-stratifying coatings, 120–121 tension correlations with HSP, 113–114 tensions, critical, 116–117 wetting tension, 115–116, 117–118 Surfactants, 332–333 SWEAT (soluble water exuded at lower temperatures), 141, 238–240 Swelling polypropylene, 106 solubility parameter correlation of polymer, 250

Hansen Solubility Parameters: A User’s Handbook

T Tank coatings, 233–234 Tar, 152 Temperature bitumen flexibility and, 168 chiral rotation and, 290 dependence and solubility parameter calcuation, 18–19 effect on chemical resistance, 235 effects on solubility parameters of carbon dioxide, 190–196 hydrocarbon solvents and, 337 hydrogen-bonding parameters in coatings and, 140–141 melting point of polymers and, 106–107 one-component Hildebrand parameter as a function of pressure and, 187–189 SWEAT phenomena, 141, 238–240 water absorption versus, 238–240 Tensile strength, 237–238 Tetrabromobisphenol A (TBBPA), 316–317 Tetrahydrofurane, 290 Tetrahydrofuran (THF), 59, 140 Thermoplastic elastomers (TPE), 333 Three-component solubility parameters, 189–190 Thymine, 275 Titration BISOM test, 170–173 turbidimetric, 170 Toluene, 245, 277, 290, 314 Toxic substances, 312 Transport phenomena, 316–317 Trichloroethylene, 253, 277 Trichloromethane, 185 Tricresyl phosphate, 214 Turbidimetric titrations, 170

U ULTEM® 1000, 237–238 UNIFAC model, 76–78, 83 Urea, 274, 283, 289

V Van der Waals volume, 77, 79 Van Krevelen dispersion parameters, 13 Vapor degreasing operations, 210–212 Vapor diffusion, 301–302 Vapor phase osmometry (VPO), 154 Vegetable oils in solvent cleaners, 142 Vinyl acetate, 243 Vinyl chloride, 243 Viscosity crude oil, 169–170 intrinsic, 107–109 solvent, 145–146 V iton®, 106, 246, 250 Volatile organic compounds (VOC), 137, 142

7248_Index.fm Page 519 Wednesday, May 9, 2007 10:49 AM

Index

W Water absorption versus temperature, 238–240 amino acid side chains and, 270–271 characterization problems with, 334–336 chemical resistance and HSP correlations with, 238–239 as a cleaning solvent, 204 HSP for, 21–22, 289, 326 hydrogen-bonding parameter and temperature of, 141 polymer softening using, 244–245 -reducible coatings, 139–140 surface mobility, 263–264, 290 urea mixtures, 283, 289 Wetting tension, 115–116, 117–118 Wood chemicals and polymers, 279, 281, 289

X χ12 parameters

519 comparison of calculated and experimental, 34–39 defined, 27–28 HSP and, 32–39 polyacrylonitrile, 39 polybutadiene, 35–36 polyisobutylene, 36–38 polymer concentration and, 39 polystyrene, 38 polyvinylacetate, 39 scatter, 39–40 X-ray photoelectron spectroscopy (XPS) analysis, 132 Xylene, 138, 140, 314

Z Zein, 279, 280–281 Zero dipole moments, 16–17 Zeta potential, 130, 131

7248_Index.fm Page 520 Wednesday, May 9, 2007 10:49 AM