Homolytic Aromatic Substitution

137 Pages·1960·3.1 MB·English

by  G. H. Williams

INTERNATIONAL SERIES OF MONOGRAPHS ON
ORGANIC CHEMISTRY
GENERAL EDITORS: D. H. R. BARTON and W. DOERING
VOLUME 4
HOMOLYTIC AROMATIC SUBSTITUTION
OTHER TITLES IN THE SERIES ON ORGANIC CHEMISTRY
Vol. 1. WATERS—Vistas in Free Radical Chemistry
Vol. 2. TOFCHIEV ET AL.—Boron Fluoride and its Compounds as Catalysts in Organic Chemistry
Vol. 3. JANSSEN—Synthetic Analgesics—Part I: Diphenylpropylamines
Vol. 5. JACKMAN—Applications of Nuclear Magnetic Resonance Spectroscopy in Organic Chemistry
11
HOMOLYTIC AROMATIC
SUBSTITUTION
by
G. H. WILLIAMS
Reader in Organic Chemistry,
Birkbeck College,
London
PERGAMON PRESS
OXFORD · LONDON · NEW YORK · PARIS
1960
PERGAMON PRESS LTD.
Headington Hill Hall, Oxford
4 & 5 Fitzroy Square, London W.l
PERGAMON PRESS INC.
122 East 55th Street, New York 22, N. Y.
P.O. Box 47715, Los Angeles, California
PERGAMON PRESS S.A.R.L.
24 Rue des Ecoles, Paris Ve
PERGAMON PRESS G.m.b.H.
Kaiserstrasse 75, Frankfurt-am-Main
Copyright
©
1960
Pergamon Press Ltd.
LIBRARY OF CONGRESS CARD NUMBER 59-14367
PRINTED IN GREAT BRITAIN BY PAGE BROS (NORWICH) LTD
PREFACE
THE branch of organic chemistry which is concerned with the reactions of free radicals,
although fairly young, has grown, and continues to grow, with remarkable vigour,
and at a prodigious rate. Such has been its growth that the production of a compre-
hensive treatise on the whole of the field would now be an extremely difficult, if not
impossible task, and the stage has now been reached when its subdivision for the
purposes of book-writing has become necessary. This book is therefore concerned with
one group of free-radical reactions, namely those in which substitution of atoms or
groups (usually hydrogen) attached to aromatic nuclei is effected by free radicals of
various kinds. While the emphasis in this book is on the theoretical aspects of the
homolytic substitution reactions discussed, their preparative use has also been con-
sidered, and ample references are included to the original literature and to review
articles when these are available. It is hoped, therefore, that this book will serve as a
useful source of information to the reader who is interested in the practical applications
of these reactions.
After a brief introduction, I have devoted Chapter 2 to a general survey of the
methods of theoretical approach to aromatic substitution in general and homolytic
substitution in particular. In the following chapters these ideas are applied in dis-
cussion of the results of a number of homolytic substitution reactions. Reactions of
arylation, both intermolecular and intramolecular (the Pschorr phenanthrene synthesis),
alkylation with simple and substituted alkyl radicals, hydroxylation, acyloxylation,
amination, amidation, halogenation and mercuration are considered in these chapters.
I have attempted to cover the literature up to December 1958. However, since one
cannot guarantee that such a survey is complete, I can only hope that authors to
whom full justice has not been done will be patient with its shortcomings. I have not
hesitated, wherever it seemed appropriate, to place my own interpretation on experi-
mental data, to point out correlations and to draw theoretical inferences. I do not, of
course, pretend that this is the final word on any subject, and some readers may
disagree with the views I have expressed concerning controversial issues. This situation,
if realised, will give me considerable pleasure, since if it serves in any small measure to
stimulate thought, or further research into the topics discussed, then the main purpose
of this book will have been served.
I must express my thanks to those of my colleagues who have helped me during
the writing of this book. Firstly, to Professor D. H. Hey, who first excited my interest
in this subject, for many hours of stimulating and fruitful discussion. I am also in-
debted to Dr. J. I. G. Cadogan and Dr. C. W. Rees, who have read sections of the
book and given me the benefit of their views and criticism. I must also thank those
good friends who have, as indicated in the text, made their results available to me in
advance of publication. Finally, I should like to thank my wife, to whom I owe a very
great debt of gratitude, not only for her encouragement and forbearance, but also in
a more material sense, for her very able assistance at all stages in the production.
Vll
CHAPTER 1
GENERAL INTRODUCTION
1. GENERAL CHARACTERISTICS OF HOMOLYTIC REACTIONS
THE term "homolysis" was first introduced by Ingold (1938), who drew a formal
distinction between the two possible modes of bond-fission:
Heterolysis: A:B -> A~ + B+ (1)
Homolysis: A:B -> A· + B· (2)
where A and B are atoms or groups of atoms. The latter reaction is characterised by
the production of neutral fragments, each possessing an odd number of electrons, and
hence, one unpaired electron. The term "free radical" is used with this connotation
and hence may include some stable, more particularly inorganic, substances, which
possess odd numbers of electrons, for example, nitric oxide. Such substances are
free radicals according to the terms of this general definition.
Since the separation of the oppositely charged particles of equation (1) obviously
requires more energy than that of the neutral fragments of equation (2) the latter
reaction might be considered to be favoured energetically. Although in the gas phase
this is probably true, for the majority of reactions in solution, particularly with solvents
of high dielectric constant, fission by reaction (1) predominates because the solvation
of the resulting ions results in the acquisition by this mode of fission of a certain
energetic advantage over its competitor. Consequently, in solution, homolytic
reaction is the exception, rather than the rule.
The first experimental demonstration of the existence of organic free radicals was
the classic work of Gomberg (1900) on the dissociation of hexa-arylethanes into
triarylmethyl radicals. The idea of an anomalous valency for carbon was not at that
time a popular one, but the accumulation of data from many fields has resulted in the
establishment of the concept. Organic free radicals may conveniently be divided into
two classes:
(1) those of long life like the triarylmethyls, which are stabilised by resonance;
and
(2) those of short life like phenyl and methyl which do not enjoy such stabilisation.
The radicals which are capable of effecting substitution in aromatic nuclei are generally
of the second type. The inference that the short-lived radicals are intrinsically unstable
is, however, erroneous; indeed, the contrary is true. The reason why they cannot be
prepared in large quantity is not that they have any tendency to suffer molecular
disruption, for a single radical completely isolated from all contact with other
molecules would doubtless remain unchanged indefinitely. The reason is instead that
these highly energetic substances are so extraordinarily reactive that each entity
undergoes some sort of chemical change in an attempt to satisfy its normal valency
requirements, before it has made a large number of collisions with other molecules.
1
2 GENERAL INTRODUCTION
Strictly, therefore, the term "short-lived" when applied in description of a free
radical, is meaningless unless the environmental conditions are also stated. However,
except in the most rigorous analysis the above classification into two clearly defined
groups is valid and may conveniently be employed.
The physical demonstration of the existence of short-lived free radicals in solution
is extremely difficult, since owing to the very small concentration in which such labile
intermediates must occur, the techniques usually applied to the detection of radicals
of the triarylmethyl type (for example, colour reactions and magnetic measurements)
completely fail. The presence of radicals of short life is therefore inferred from their
reactions, on account of the complex kinetics which they exhibit, and the nature of the
chemical products that are obtained. Free radicals have been found to react mainly in
three ways:
by dimerisation, or combination of radicals;
R· + R· -> RR (3)
by disproportionation, involving mutual hydrogenation and dehydrogenation;
2RCH CH · -> RCH CH + RCH:CH (4)
2 2 2 3 2
by radical transfer, which is expressed by the general equation;
R· + XR' -> RX + R'· (5)
In the last reaction the nature of the final products is also dependent upon the
radical R'·. It may react as in equation (5) to give another radical R"·, which may or
may not be the same as R'·. If it is the same, then a chain reaction is set up, and the
process repeats itself until the chain is somehow interrupted. If the radical R'·, or R"·
is relatively unreactive, then the transference [reaction (5) ] results effectively in the
termination of the reaction.
In many instances these reactions may occur simultaneously, and the total reaction
may appear very complex. In other cases, however, certain of these reactions are
favoured relative to the others and specific reaction paths which lead to specific
products predominate. This may be illustrated by a comparison of the reaction medium
of the gas phase with that of the liquid phase for, although the reactions may be
initiated in the same way in both media, there are great differences in the types of
reaction occurring. For example, in the photolysis of ketones, and the pyrolysis of
certain organometallic compounds, in the gas phase, where collisions between radicals
can readily occur, reactions (3) and (4) predominate, while in solution, where the
radicals are surrounded by solvent molecules, reactions between radicals are unlikely
unless the life of a radical is for some reason prolonged. With labile radicals in solution
the radical-solvent interaction (5) is the most important reaction, and it is clear from
this equation that the product is determined by both the radical and the solvent.
With aromatic solvents, alkyl radicals behave differently from aryl radicals. Thus
alkyl radicals, in general, react with aromatic solvents by reactions other than that of
direct substitution. They do, however, effect substitution in some cases, and this will
be discussed at a later stage (Chapter 6). Similar considerations are valid in the
comparison between the reactions of a given radical with aliphatic and aromatic
solvents, since only the latter are susceptible to homolytic substitution.
GENERAL INTRODUCTION 3
During recent years these homolytic aromatic substitution reactions have been
studied in considerable detail, and the results have been discussed periodically. The
most recent, and comprehensive reviews of different aspects of the subject have been
those of Dermer and Edmison (1957) and Augood and Williams (1957).
2. HOMOLYTIC AND HETEROLYTIC AROMATIC SUBSTITUTION
Although the reactions of organic compounds are generally distinguished from those
of inorganic compounds as being reactions of non-ionic substances, it is accepted
that the course of the majority of organic reactions is determined by the electrical
influences of polarisation and polarisability. Thus, owing to the experimental cir-
cumstance that homolytic reactions are encountered less frequently than their hetero-
lytic counterparts, the so-called "normal laws" of aromatic substitution, developed as
the result of experiment, are laws governing the heterolytic reactions in which the
substances take part.
Heterolytic reagents are of two kinds: those which react preferentially at centres of
high electron density, i.e. electrophilic reagents, and those which react at electron-
deficient positions, i.e. nucleophilic reagents. Hence, in the majority of aromatic
substitution reactions, such as nitration, halogenation, sulphonation and diazo-
coupling, a proton is replaced by an electrophilic reagent capable of accommodating a
pair of electrons, e.g. the nitronium ion N0 +, which is responsible for most aromatic
2
nitrations. Many nucleophilic reactions of aromatic substitution are also known, in
which atoms and groups other than hydrogen are commonly replaced, and these have
been reviewed by Bunnett and Zahler (1951). The history of the development of the
theories of heterolytic aromatic substitution is long, and has been reviewed by Badger
(1954) and Remick (1949). The so-called "normal laws" of aromatic substitution have
been established mainly as the result of studies of nitration. (Ingold, 1934; Bird and
Ingold, 1938; Ingold and Smith, 1938; Ingold, Lapworth, Rothstein and Ward, 1939;
Hughes, Ingold and Reed, 1950; Bennett, Brand and Williams, 1946; Bennett, Brand,
James, Saunders and Williams, 1947; Roberts, Sandford, Sixma, Cerfontein and
Zagt, 1954.) On both theoretical and experimental grounds a directing group (X)
attached to the aromatic nucleus in the compound C H X, may be placed in one of
e 5
two main categories. Thus, for this type of substitution, the following groups X are
ortho-para directing, and activate the nucleus for reaction with the electrophilic
substitutient, relatively to the unsubstituted nucleus of benzene:
NH > OH > OCH > O.CO.CH3 > CH > C H > CH C0 H,
2 3 3 e 5 2 2
whereas the following groups are meta-directing and deactivating:
N(CH ) > N0 > SO3H > CHO > CO.CH3 > C0 H.
3 3 2 2
The situation is a little more complex in that certain anomalies, which are accom-
modated by the general theory, occur. Thus, for example, the halogens are deactivating
groups, but are ortho-para directing. For nucleophilic substitution these rules have
been found to be approximately reversed (for reviews see Badger, 1954; Remick,
1949; Bunnett and Zahler, 1951; Ferguson, 1952).
The contributions made by Ingold and his collaborators in this field were of the
highest importance in providing quantitive data to describe the phenomena. The
work resolved itself into two parts:
4 GENERAL INTRODUCTION
(1) the determination of the quantities ^K i.e. the ratio of the total rate of
9
nitration of the monosubstituted benzene derivative, C H X, to that of
6 5
benzene, C H ; and
6 6
(2) the determination of the proportions in which the three isomers r-XC H N0
6 4 2
(where r- = o-, m- or p-) were formed in the nitration of C H X alone.
6 5
The combined results of these experiments allowed the evaluation of the partial rate
factors £k. Each of these factors expresses the change in the specific rate of sub-
r
stitution at any point r in the nucleus due to the presence of the group X, that is, it is
the ratio of the rate of reaction at the point r in the molecule C H X to that at any one
6 5
position in benzene. The quantities ^K and £k are pure numbers; they have no
r
dimension in time. A selection of the results obtained in this work is given in Table 1-1.
TABLE 1-1. PARTIAL RATE FACTORS FOR NITRATION
X xk xk xk S*
CH 40 3 51 23
3
C0 C H 00026 00079 0001 00037
2 2 5
F — — — 015
Cl 0030 0000 0139 0033
Br 0037 0000 0106 0030
1 — — — 018
OH* — — — —104
N02* 0-24 x 10-6 2-75 x 10-6 003 x 10-6 ~io-e
* Approximate values only.
It may be seen by examination of the table that for toluene the figures are greater
than unity and show that the methyl group activates all the positions r, that is, that
toluene is more readily nitrated than benzene. Moreover, the order of the values
of %k, p > o > m, implies that this is the order of reactivity of the nuclear positions.
r
The carboxyethyl group, CO.OC H , shows general deactivation, with the mal-
2 5
position least affected, and the halogens show themselves to be exceptions to the
rule linking ortho-para direction with activation, and meta direction with deactiva-
tion (the so-called Holleman rule), as mentioned above. Details concerning other
groups, e.g., ter/.-butyl, are also available (Brown and Nelson, 1951; Cohn, Hughes,
Jones and Peeling, 1952; Holleman, 1925). Data for halogenation and sulphonation
have been reported recently. Thus for chlorination at ~25° the total relative rates for
all the groups investigated fall into sequence between the extreme values of NMffA>M018
and N%Ä:~10-6 (de la Mare, Robertson and Swedlund, 1953). The partial rate
factors for toluene are cufik = 600, CH£k = 5, and c\k = 870 (Condon,
0 m
1948, 1952; de la Mare and Vernon, 1951). The differences between these figures and
those of Table 1-1 show the electron demands of the reagents to be different (cf.
Braude, 1949; Waters, 1952) and illustrate that any set of partial rate factors applies
only to the reaction for which they were determined.
Although at that time these laws were not as clearly defined as they now are, it was
the separation of apparently anomalous products from certain substitution reactions
GENERAL INTRODUCTION 5
that first indicated the possibility of another, non-ionic, type of substitution. These
reactions were those whereby the preparation of diaryls was accomplished, and it was
suggested by Hey (1934) that, in these reactions, homolytic processes were occurring.
This suggestion was elaborated by Hey and Waters (1937). In essence, the existence of
short-lived free radicals of the type of phenyl was postulated, and these entities, being
electrically neutral, and therefore not subject to electrical effects in the same way as the
heterolytic reagents, effected substitution in aromatic molecules without conforming
to the normal laws of aromatic substitution.
These reactions can be expressed generally by equation (6),
Ar'M + ArH -> Ar'Ar + N (products) (6)
where the radical Ar'· which is produced from the radical-source Ar'M reacts with the
aromatic solvent ArH to form the diaryl Ar'Ar with ejection of a hydrogen atom which
is incorporated in the product N. It was found later that in homolytic substitution in
ArX the general tendency was for all groups X, notwithstanding their directing effect
in heterolytic substitution, to favour predominantly reaction at the /?<zra-position.
The meta-position was thought to be least favoured and the ra-diaryl was rarely
isolated. More recent work, to be described in subsequent chapters, has, however,
necessitated considerable revision of these conclusions.
REFERENCES
AUGOOD, D. R. and WILLIAMS, G. H. (1957) Chem. Rev., 57, 123.
BADGER, G. M. (1954). Structure and Reactions of Aromatic Compounds, Cambridge University Press,
Cambridge.
BENNETT, G. M., BRAND, J. C. D., JAMES, M. C. D., SAUNDERS, T. G. and WILLIAMS, G. (1947)/. Chem.
Soc, 474.
BENNETT, G. M., BRAND, J. C. D. and WILLIAMS, G. (1946) /. Chem. Soc, 869.
BIRD, M. L. and INGOLD, C. K. (1938) /. Chem. Soc, 918
BRAUDE, E. A. (1949) Ann. Rep. Progr. Chem., 46, 114.
BROWN, H. C. and NELSON, K. L. (1951) /. Amer. Chem. Soc, 73, 5605.
BUNNETT, J. F. and ZAHLER, R. E. (1951) Chem. Rev., 49, 273.
COHN, H., HUGHES, E. D., JONES, M. H. and PEELING, M. G. (1952) Nature, Lond., 169, 291.
CONDON, F. E. (1948) /. Amer. Chem. Soc, 70, 1963; (1952) 74, 2528.
DERMER, O. C. and EDMISON, M. T. (1957) Chem. Rev., 57, 77.
FERGUSON, L. N. (1952) Chem. Rev., 50, 47.
GOMBERG, M., (1900) Ber., 33, 3150.
HEY, D. H. (1934) /. Chem. Soc, 1966.
HEY, D. H. and WATERS, W. A. (1937) Chem. Rev., 21, 169.
HOLLEMAN, A. F. (1925) Chem. Rev., 1, 200.
HUGHES, E. D., INGOLD, C K., and REED, R. I. (1950) /. Chem. Soc, 2400.
INGOLD, C. K. (1934) Chem. Rev. 15, 225; (1938) Trans. Faraday Soc, 34, 227.
INGOLD, C. K., LAPWORTH, A., ROTHSTEIN, E. and WARD, D. (1939) /. Chem. Soc, 1959.
INGOLD, C. K. and SMITH, M. S. (1938) /. Chem. Soc, 905.
DE LA MARE, P. B. D., ROBERTSON, P. W. and SWEDLUND, B. E. (1953) /. Chem. Soc, 782.
DE LA MARE, P. B. D. and VERNON, C. A. (1951) /. Chem. Soc, 1764.
REMICK, A. E., (1949) Electronic Interpretations of Organic Chemistry, 2nd. edition, John Wiley, New
York.
ROBERTS, J. D., SANFORD, J. K., SIXMA, F. L. ΤΛ, CERFONTEIN, Η. and ZAGT, R. (1954) /. Amer.
Chem. Soc, 76, 4525.
WATERS, W. A. (1952) Ann. Rep. Progr. Chem., 49, 110.