Correlation Analysis in Chemistry __ RecentAdvances __ _

Contributors

R. P. Bell, Department of Chemistry, The University of Stirling, Stirling, FK9 4LA, Scotland

M. Charton, Department of Chemistry, Pratt Institute, Brooklyn, New York, 11205, U.S.A.

C. Duboc, Universite de Paris VI, F-75005 Paris, France

D. F. Ewing, Chemistry Department, The University, Hull, HU6 7RX, England

O. Exner, Institute of Organic Chemistry and Biochemistry, Czechoslovak Academy of Sciences, Prague 6, Czechoslovakia

G. P. Ford, Department of Chemistry, University of Texas at Austin, Austin, Texas 78712, U.S.A.

M. Godfrey, Department of Chemistry, The University, Southampton, S09 5NH, England

C. Hansch, Department of Chemistry, Pomona College, Claremont, California 91711, U.S.A.

A. R. Katritzky, School of Chemical Sciences, University of East Anglia, University Plain, Norwich, NR4 7TJ, England

1. Shorter, Department of Chemistry, The University, Hull, HU67RX, England

M. Sjostrom, Research Group for Chemometrics, Institute of Chemistry, Umea University, S-901 87 Umea, Sweden

R. D. Topsom, Department of Chemistry, La Trobe University, Bundoora, Melbourne, Victoria, Australia 3083

S. Wold, Research Group for Chemometrics, Institute of Chemistry, Umea University, S-901 87 Umea, Sweden

Correlation Amdysis in Chemistry __ RecentAdvances ___ _

Edited by

N. B. Chapman and

1 Shorter The University of Hull Hull, England

Plenum Press · New lbrk and London

Library of Congress Cataloging in Publication Data

Main entry under title:

Correlation analysis in chemistry.

Includes bibliographical references and index. 1. Linear free energy relationship. l. Chapman, Norman Bellamy, 1916- II.

Shorter, John, 1926-QD501.C792 541.39 78-1081

ISBN-13: 978-1-4615-8833-7 e-ISBN-13: 978-1-4615-8831-3 DOl: 10.1007/978-1-4615-8831-3

© 1978 Plenum Press, New York Softcover reprint of the hardcover 1st edition 1978

A Division of Plenum Publishing Corporation 227 West 17th Street, New York, N.Y. 10011

All rights reserved

No part of this book may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, microfilming, recording, or otherwise, without written permission from the Publisher

Preface

This book, Correlation Analysis in Chemistry: Recent Advances, is a sequel to our Advances in Linear Free Energy Relationships. t The change in the title is designed to reflect more accurately the nature of the field and the contents of the volume. The term LFER is still widely used, but it is often applied rather loosely to correlation equations that are not LFER in the restricted sense of a relationship involving logarithms of rate or equili­brium constants on each side of the equation. The term "correlation analysis" seems to us more appropriate for the whole subject. The use of this term has compelled us also to introduce "chemistry" into the title; we have preferred not to prefix this with "organic" on the grounds that several areas of interest are not "organic chemistry" as usually understood, al­though, of course, traditional applications of the basic relationships asso­ciated with the names of Hammett and of Taft continue to be of interest.

In the first volume we sought through our authors to provide a series of general articles covering the various aspects of the field as they seemed to us. Since the book was the first international research monograph in its field, each chapter, while giving prominence to recent developments, did not neglect earlier work, so that each article presented a comprehensive account of its own area. The same general idea underlies the present volume, but most of the articles are on much more specialized and restric­ted topics, and treat in greater depth and detail matters that were dealt with in only a few pages in the earlier volume. Consequently some im­portant topics such as correlation analysis in relation to solvent effects on organic reactivity and to inorganic and organometallic chemistry are treated only incidentally in the present volume. The traditional areas of LFER associated with Hammett and Taft are now covered by articles dealing with LFER as statistical models (Chapter 1), the relationship to quantum chemistry (Chapter 3), multiparameter extensions of the Ham­mett equation (Chapter 4), applications to polycyclic arenes and heterocy­clic compounds (Chapter 5) and to olefinic systems (Chapter 6), and a critical compilation of substituent constants (Chapter 10). Applications to

tN. B. Chapman and J. Shorter, editors, Plenum Press, London and New York (1972).

v

vi Preface

spectroscopy of various types are of frequent occurrence in some of these chapters, while nmr chemical shifts are the special concern of Chapter 8. The effect of the reagent on organic reactivity is covered in two chapters, one on the Bronsted equation, which is the oldest LFER (Chapter 2), and the other on nucleophilicity (Chapter 7). The subject of biochemical quan­titative structure-activity relationships (QSAR) is developing so rapidly that another general treatment (Chapter 9) seemed justified to supplement the chapters on enzymology and drug action in the first volume.

It seems appropriate here to repeat a paragraph that we as Editors wrote in our own defense in our preface to Advances in Linear Free Energy Relationships:

Authors were given only very general guidance as to the length and content of their respective chapters, and they had full freedom to develop their topics as they felt appropriate. We are aware that the ten chapters bridge the extremes of style from the "essay-review," with a relatively low density of literature citations, to the "progress report" from which no sign/kant contribution is omitted. The chapters also vary widely in the extent to which tables or figures are used. Some authors present correlations that have been carried out specially for this book, and in some cases the approach used reflects the individual views of the author. Accordingly we are prepared for the criticism that the book has the usual failings of multi author works. However, in our editorial work we have tried to impress on the book a certain unity, despite the diversity of style and approach. We have, moreover, been mindful that parts of it may be read by scientists from a wide range of disciplines. Authors were asked to remember this and to try to make themselves intelligible to a wide readership.

For the first volume we wrote an Editors' Introduction dealing mainly with Signs, Symbols, and Terminology for Substituent Effects, which is reprinted in the present volume. The only significant change in our views is that we would now wish to encourage the use of K (konjugativ) for the total delocalization effect.

As editors we are greatly indebted to many people. The authors, like their predecessors in the earlier volume, have made strenuous efforts to meet our requirements and suggestions in many directions and have been very forbearing with the idiosyncrasies of the editors. Several colleagues in Hull have helped us in various ways. The very considerable secretarial work has been excellently done by Christina Pindar, Janet Bailey, and Dorothy Wilson, all of the Chemistry Department. We are also grateful to Jane Fisher of the British Library for work on references.

Finally, as in all our work, we are most grateful to our wives for their support and encouragement.

N. B. Chapman J. Shorter

Contents

Editors' Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi

Chapter 1. Linear Free Energy Relationships as Tools for Investigating Chemical Similarity-Theory and Practice ................ 1 Svante Wold and Michael Sjostrom

Chapter 2. The Bronsted Equation-Its First Half-Century. . . . . . . . . . . . 55 R. P. Bell

Chapter 3. Theoretical Models for Interpreting Linear Correlations in Organic Chemistry . . . . . . . . . . . . . . . . . . . . . . . . ... . . . . . . . . . . . 85 Martin Godfrey

Chapter 4. Multiparameter Extensions of the Hammett Equation 119 John Shorter

Chapter 5. Applications of Linear Free Energy Relationships to Polycyclic Arenes and to Heterocyclic Compounds .................. 175 M Charton

Chapter 6. Substituent Effects in Olefinic Systf-ms ......... . . . . . . . . . . . 269 G. P. Ford, A. R. Katritzky, and R. D. Topsom

Chapter 7. The Correlation Analysis of NUcleophilicity ............... 313 Claude Duboc

Chapter 8.

Chapter 9.

Chapter 10.

Correlation of nmr Chemical Shifts with Hammett u Values and Analogous Parameters ........................ . . . . . . 357 D. F. Ewing

Recent Advances in Biochemical QSAR Corwin Hansch

A Critical Compilation of Substituent Constants Otto Exner

397

439

Index.... .................. ...... ............................. .... 541

vii

Errata for

Advances in Linear Free Energy Relationships, edited by N. B. Chapman and J. Shorter, Plenum Press, London and New

York (1972)

p. 18 p.76

p.110 p. 121 p. 184 p. 253 p.258 p.372 p.396 p.408

Footnote t: For UR, read PRo

Table 2.1: For Ph2C read Ph2CH, and E. for Cl(CHzh should be -0.90. In footnote b, insert "hydrate" after "acetaldehyde." Line 9 from bottom: For 1 : 3, read 1 : l. Equation (3.1): Insert c2 after 3000 in the denominator. Equation (4.14): The symbol in the second term on the right side should be u~. The last line should read" ... follow as an ... ". In entry 218, Ii should be 11.5 and ni,° should be 1.4659. Section 8.1.2: Insert "energy" after "free" in first line of main text. Line 7 from bottom: The compounds should be "p-nitrophenyl benzoates." Table 9.3: The values of U m should be as follows: F, 0.34; Cl, 0.37; Br, 0.39; I, 0.35.

Editors' Introduction t

Signs, Symbols, and Terminology for Substituent Effects

Since this is an area of considerable confusion we felt it necessary to give authors some guidance, broadly as follows.

There is no uncertainty as to the sign to be attached to a a value: an electron-attracting group has a positive a value, and vice versa. There are two possible conventions, however, for the sign of a substituent effect. In the Ingold convention the movement of electrons towards a substituent is signified by a negative sign; the movement of electrons away from a substituent by a positive sign. Thus N02 has a -M effect, while OMe has a + M effect. Electronegativity is thus associated with a negative sign, and electropositivity with a positive sign. In the other convention, originally associated with Robinson, the signs are reversed.

There is little doubt that the Ingold convention is the more widely used the world over. We note, however, that many authors (especially those of textbooks and monographs published in the U.S.A.) avoid using signs and symbols for electronic effects and use phrases such as "a mesomerically electron-attracting group." From an LFER standpoint the Ingold convention has the disadvantage that the signs of the electronic effects and those of the corresponding a values are opposite. For this reason some writers on LFER have adopted the Robinson sign convention. In deciding not to follow suit in this volume, we were influenced by correspondence with Sir Christopher Ingold a few months before he died, and by the appearance of the second edition of Ingold's Structure and Mechanism in Organic Chemistry, and of Hammett's Physical Organic Chemistry. Both these books will have the status of major works for many years to come; both of them adhere to the Ingold sign convention.

We therefore asked authors to use the Ingold convention, but in the event signs and symbols to describe electronic effects have not been ex­tensively used. tThis Introduction originally appeared in Advances in Linear Free Energy Relationships,

N. B. Chapman and 1. Shorter, eds., Plenum Press, London and New York (1972).

ix

x Editors' Introduction

As to symbols, the use of I for the inductive effect (possibly with a subscript for a special purpose) is universal practice. Delocalization effects are more of a problem, the possibilities being T, R, K, or C for the total effect, with M for the time-permanent and E for the time-variable effect. T was not to be used since Ingold had abandoned it in favor of K, based on the German for "conjugative"; he preferred K to the more obvious C, clearly inconvenient in organic chemistry. However, K seemed to have definite disadvantages in connection with LFER. We suggested that K or C might be used, or R, which has the advantage of fitting in withuR, which we felt we could not tamper with, although we dislike the term "resonance." Authors were asked not to use M with reference to conditions under which time-variable effects might be considered important (for separate specification of the latter E remains appropriate). Since, however, some authors probably consider that the analysis into time-permanent and time­variable effects has sometimes been overemphasized, it is likely that this advice has not been strictly adhered to.

As to terminology, we recognized that the difficulties mentioned in the previous paragraph are carried over and others are added. Having alerted the authors to the problems as we saw them, we had hoped to examine very closely the use they made of the various terms with a view to trying to ensure some degree of uniformity of usage. In the event this task has proved beyond us, and we can do no more than lay the problems before the reader.

We consider a reactant molecule R Y and an appropriate standard molecule RoY. Initially we suppose that Y is not conjugated with either R or Ro. For RY the polar effect of the group, R, comprises all the processes whereby a substituent may modify the electrostatic forces operating at the reaction center Y, relative to the standard RoY. These forces may be governed by charge separations arising from differences in the electronegativity of atoms (leading to the presence of dipoles), the presence of unipoles, or electron delocalization. Field (or direct), inductive (through-bond polarization involving U or 1T electrons), and mesomeric effects may in principle be distinguished. Because of the difficulty of distinguishing between field and through-bond effects in practice, the term inductive effect is often used to cover both. It is so used widely in this book, but in some places the more restricted meaning is implied. The term resonance effect or resonance polar effect is often used in connection with the mesomeric effect, e.g., the analysis of U values into inductive and mesomeric components is also known as the separation of inductive and resonance effects, as shown by the use of UR as the basic symbol for the delocalization component.

When Rand Ro may be conjugated with Y, the above discussion holds, but additional influences· may arise from the more extensive elec-

Editors' Introduction xi

tron-delocalization. The mesomeric part of the polar effect will be modified. There will also be the resonance effect of R which is concerned with the extent of conjugation of Rand Y, relative to the standard RoY, and is not part of the polar effect. This distinction is sometimes not clearly made: "resonance effects" in a wide sense are lumped together and treated as if they are polar in nature; cf. the distinction between er~ on the one hand and erR, er;', and erR on the other.

O~ -o,,+~+ N02 _ NH2 +---+ _O/N~NH2

(I)

MeOOCH2'C02H +---+ MeoOCHfC02H

(II)

It thus seems that the terms resonance effect and mesomeric effect may be used to cover (a) conjugation of the substituent with the functional group, mediated by the aromatic or other delocalized system, or (b) that of the substituent with the de localized system, not including the functional group. Structure (I) is an example of (a), and (II) of (b).

It also seems common practice to use the term resonance effect in connection with (c), the conjugation of the functional group with the de localized system, e.g., as in (III)

(III)

The precise meaning of resonance effect may thus only be understood in context.

Hyperconjugation or the hyperconjugative effect, as the name implies, is historically and commonly regarded as a special kind of conjugative effect involving de localization of electrons in C-H bonds adjacent to an unsaturated system. The actual nature of this substituent effect, however, must be regarded as still in doubt. At most places in this book where the effect is referred to, no particular view of hyperconjugation seems to be implied: The term simply means a special effect of a-hydrogen atoms in the same sense as mesomeric electron release. An analogous effect of C-C bonds under circumstances in which ordinary conjugation cannot occur is also sometimes invoked; this is C-C hyperconjugation, in contrast to CH hyperconjugation.

xii Editors' Introduction

A word on steric effects is appropriate, although there are no parti­cular problems. Steric effects are caused by the intense repulsive forces operating when two non bonded atoms approach each other so closely that non bonded compressions are involved. The primary steric effect of R is the direct result of compression which arises because R differs in structure from Ro in the vicinity of the reaction center. A secondary steric effect involves the moderation of a polar effect or resonance effect by non bonded compressions.

In discussing the influence of any substituent effect it is of course necessary to consider differentially interactions in initial and transition states in the case of rate processes, and in initial and final states in the case of equilibria.

References

References are collected at the end of each chapter in the order of first citation. We have provided references to Chemical Abstracts for periodicals which, we believe, will be inaccessible to many readers. Up to volume 65 (1966) such references give the column number; from volume 66 (1967) the distinctive number of the abstract is cited.

In the case of Russian journals we have also provided references to English translations, if available. These are usually indicated simply by EE (i.e., English edition), followed by the page number, but amplification is sometimes required. The main journals involved are as follows.

Russian Title

Zhurnal Obshchei Khimii

Zhurnal Organicheskoi Khimii

Zhurnal Fizicheskoi Khimii

Uspekhi Khimii

Doklady Akademii Nauk SSR

Izvestiya Akademii Nauk SSR Seriya Khimicheskaya

English Title

Journal of General Chemistry of the USSR

Journal of Organic Chemistry of the USSR

Russian Journal of Physical Chemistry

Russian Chemical Reviews

{ Doklady Chemistry Doklady Physical Chemistry

Bulletin of the Academy of Sciences of the USSR Division of Chemical Science

Editors' Introduction

Reaktsionnaya Sposobnost Organicheskikh Soedinenii

Kinetika i Kataliz

Zhurnal Strukturnoi Khimii

Optika i Spektroskopiya

Organic Reactivity

Kinetics and Catalysis

Journal of Structural Chemistry

Optics and Spectroscopy

xiii

For Angewandte Chemie, or its International Edition, references to the English edition (EE) or to the German edition (GE) as appropriate are given.

Miscellaneous Symbols

All chapters are concerned with measures of the success of cor­relations and the following symbols are used throughout:

r correlation coefficient in a simple linear regression, R correlation coefficient in a multiple regression, s standard deviation of the estimated value of the dependent variable.