Organic Chemistry vol 1

ORGANIC
CHEMISTRY

VOLUME ONE
THE FUNDAMENTAL PRINCIPLES

By

I. L. FINAR

B.Sc, Ph.D.(Lond.), A.R.I.C.

Senior Lecturer in Organic Chemistry,

Northern Polytechnic, Holloway, London

LONGMANS

Longmans, Green and Co Ltd

48 grosvenor street, london wi

railway crescent, croydon, victoria, australia

auckland, kingston (jamaica), lahore, nairobi

Longmans Southern Africa (Pty) Ltd

thibault house, thibault square, cape town

johannesburg, salisbury

Longmans of Nigeria Ltd
w. r. industrial estate, ikeja

Longmans of Ghana Ltd
industrial estate, ring road south, accra

Longmans Green (Far East) Ltd
443 lockhart road, hong kong

Longmans of Malaya Ltd
44 jalan ampang, kuala lumpur

Orient Longmans Ltd

calcutta, bombay, madras
delhi, hyderabad, dacca

Longmans Canada Ltd

137 bond street, toronto 2

© I. L. Finar 1959; 1963

First published 1951

Second Edition 1954

Third impression 1955

Fourth impression 1956

Fifth impression 1957

Third Edition 1959

Seventh impression i960

Eighth impression i960

Ninth impression 1961

Fourth Edition 1963

The fourth impression of this book was retitled Organic Chemistry
Volume One: The Fundamental Principles, as a second volume,
dealing with Stereochemistry and the Chemistry of Natural Products,
is now published and is designated Volume Two.

Printed in Great Britain by Richard Clay and Company, Ltd.,
Bungay, Suffolk

PREFACE TO THE FOURTH EDITION

This book has been revised to bring it up to date. At the same time I have
rewritten many sections on mechanisms, the added material being an
elementary account of the sort of evidence that has led workers to suggest
the mechanisms that are acceptable at the present time. This treatment
should be more interesting to the reader, and will give him a better under-
standing of this branch of organic chemistry.

Expanded subjects include quantum numbers, resonance, free radicals,
1,2-shifts, allylic rearrangements, substitution at a saturated carbon atom,
Walden inversion, hyperconjugation, the Diels-Alder reaction, aromaticity,
aromatic substitution, configuration of oximes, etc. Additional matter
includes shapes of molecules, structure and reactivity, transition state
theory of reactions, correlation and specification of configuration, ion-pairs,
neighbouring group participation, molecular overcrowding, mechanisms of
hydrolysis and esterification, Wittig reaction, hydroboronation, kinetically
and thermodynamically controlled products, principle of microscopic
reversibility, methylenes, Newman projection formulae, new reagents, etc.

A great deal of the mechanistic work is given in small print. This has
been done to keep the book within a reasonable size. The various modifica-
tions of nomenclature by the Chemical Society have been described in the
appropriate places in the text, but I have not used many of them since the
cost of resetting so many names would have increased the price of the book
out of proportion to the advantage gained. Since a summary of nomen-
clature, including some of the changes, has been given in the Appendix, the
reader should not have any difficulties with this problem.

I. L. Finar

December, 1961

,"> WITHDRAW
JS LIBRARY
\^ BOOK $

PREFACE TO THE THIRD EDITION

The fourth impression of this book (1956) was retitled Organic Chemistry
Volume One: The Fundamental Principles, and since then Volume Two:
Stereochemistry and the Chemistry of Natural Products (1956) has appeared.
The latter volume is, in effect, a continuation of the former, and so some
material that has been described in this volume (I) in a relatively elementary
manner, particularly Stereochemistry, has been dealt with far more fully in
Volume II.

Volume I has now been revised in order to bring it up to date. This
has meant rewriting many sections, and at the same time I have made some
additions which, I hope, will improve the value of this book. It may be
useful if I indicate briefly the more important changes I have made in this
new edition. Rewritten and expanded subjects include dipole moments,
resonance, Si and S2 mechanisms, steric effects, tautomerism, hyper-
conjugation, organolithium compounds, stereochemistry, Diene synthesis,
carbohydrates, aromatic substitution, transition state, and heterocyclic
compounds. Additions include the use of isotopes, molecular diagrams,
molecular crowding, Ei and E2 mechanisms, inclusion complexes, con-
formation, ferrocene, cycfoalkynes, paracyclophanes, ortho-para ratio in
aromatic substitution, and cine-substitution.

In addition to these changes, I have added many mechanisms for various
reactions and at the same time have used the more recent methods of writing
mechanisms. I have also rewritten aromatic systems with double bonds.

Once again I wish to thank those reviewers, correspondents and many of
my students who have pointed out errors and have made suggestions for
improving the book.

. I. L. Finar.

October, 1957.

PREFACE TO THE SECOND EDITION

The aim of this book has remained unchanged. Since I do not consider
the chemistry of natural products fundamental chemistry but rather the
.application of fundamental principles, I have excluded almost completely
the study of natural products. It is my hope, however, to write a com-
panion volume in which I shall deal with further aspects of stereochemistry
and also with the chemistry of many classes of natural products.

In this second edition, I have taken the opportunity to correct various
errors in the text. I wish to thank those reviewers, correspondents and
many of my students who brought these errors to my notice, and also made
suggestions for improving the book.

In this edition I have used the alphabetical order of naming prefixes,
and I have described the older method on p. 784. The principal additions
include a more detailed account of molecular orbital theory, some further
aspects of stereochemistry, various heterocyclic compounds, and a number
of dyestuffs.

My original intention was to deal with molecular orbital theory in the
future companion volume. I came to the conclusion, however, that the
treatment of this subject is best dealt with in this book. I have therefore
given an elementary account of molecular orbitals in Chapter II, and I have
discussed their applications throughout the text alongside the resonance
theory so that the student can gain some knowledge of both theories. In
order to keep the size of this book within reasonable limits, I have used
smaller type for much of the additional matter.

It is impossible to express my indebtedness to those authors of mono-
graphs, articles, etc., from which I have gained so much information. I
can only hope that some measure of my gratitude is expressed by the
references I have given to their works.

I. L. Finar.

1953

PREFACE TO THE FIRST EDITION

In this book my aim has been to describe the fundamental principles of
organic chemistry. Although the book has not been written with any
particular examination in view, nevertheless the subject matter covers most
of the organic chemistry required for the General Honours degree of the
London University. It also covers a large amount of the organic chemistry
of Part I of the Special Honours degree in chemistry of the London Univer-
sity, and a number of sections of this book should serve as an introduction
to Part II.

To many beginners organic chemistry may seem to consist of a large
variety of methods and reactions which appear to be isolated and, conse-
quently, only to be learned by heart. After many years' experience of
teaching organic chemistry to degree students, I have found that the best
method of instruction is by the introduction of electronic theories as early
as possible, with a constant application of their principles. These electronic
theories give to organic chemistry a certain coherence that is soon appre-
ciated by the beginner, and thus facilitate his study in this branch of
chemistry. Stress has been laid on structural formulae, properties of com-
pounds, and reaction mechanisms. Special attention has been given to the
systematic nomenclature of organic compounds. The alphabetical order of
naming prefixes was adopted by the Chemical Society in April, 1950. This
book was completed before this date and so this method has not been used.
The reader, however, should have no difficulty in locating a compound in
the index.

It is my experience that only a fairly detailed study of this subject matter
enables the student to appreciate the problems involved. Too short an
account usually leaves the impression that " everything works according to
plan ". This is undesirable for those who are expected to acquire a certain
amount of factual knowledge and at the same time learn to think for them-
selves. I have therefore included detailed discussions on developments of
a straightforward nature and also of a controversial nature, in the hope of
encouraging the student to weigh up the evidence for himself. This will
also give him an idea of some of the problems being investigated at the
present time, and will show him that many " facts " are subject to change.
Controversial material and the more advanced sections have generally been
printed in small type.

Only by reading original papers in which are described the " whys and
wherefores" can the student hope to gain a more mature outlook. A
selected number of reading references have therefore been given at the end
of each chapter. Since summaries of various topics by workers in special
fields are of great value in extending the student's knowledge, references of
this type have also been included. An account of the literature of organic
chemistry has been given in an appendix.

In describing methods of preparation of various compounds, I have given,
wherever possible, actual percentage yields (taken mainly from Organic
Syntheses). The student will thus be enabled to assess the value of a par-
ticular method. Where general methods of preparation have been described,
the yields have been indicated according to the following (arbitrary) scheme:

0-15% very poor (v. p.) ; 16-30% poor (p.) ; 31-45% fair (/.) ; 46-
60% fairly good (f.g.); 61-75% good (g.); 76-90% very good (v.g.);
91-100% excellent (ex.).

PREFACE TO THE FIRST EDITION ix

At the ^ end of each chapter there are questions designed to test the
student's " book knowledge " and to test the application of his book know-
ledge. At the end of the book there are also fifty questions chosen from
various examinations— B.Sc. General and Special Honours of the University
of London, and the Associateship and Fellowship of The Royal Institute of
Chemistry. I should here like to thank these Examining Boards for per-
mission to reproduce these questions.

It is hoped that the method of presentation in this book will stimulate
the reader's interest in organic chemistry and enable him to read with
understanding original papers and monographs covering specialised fields.
_ I should like to acknowledge the valuable help given me by Mr. K. Merton
in reading the manuscript and by Miss A. B. Simmonds, B.Sc., Ph.C,
A.R.I.C., in reading the proofs.

T , I. L. Finar.

July, 1950.

LIST OF JOURNAL ABBREVIATIONS

Abbreviations.
Angew. Chem.

Ann. Reports (Chem. Soc.)

Ber.

Chem. Eng. News

Chem. Reviews

Chem. and Ind.

Helv. Chim. Acta

Ind. Eng. Chem.

Ind. Eng. Chem. (Anal. Ed.

Ind. Eng. Chem. (News Ed.)

J. Amer. Chem. Soc.

J. Chem. Educ.

J.C.S.

J. Org. Chem.

J. Roy. Inst. Chem.

J.S.C.I.

J. Soc. Dyers and Col.

Nature

Proc. Chem. Soc.

Quart. Reviews (Chem. Soc.)

Rec. trav. chim.

Research

Science

Tetrahedron

Trans. Faraday Soc.

Journals.

Angewandte Chemie (the name Die Chemie was used for
vol. 55, 1942, to vol. 58, 1945).

Annual Reports of the Progress of Chemistry (The
Chemical Society, London) .

Berichte der Deutschen Chemischen Gesellschaft (name
now changed to Chemische Berichte).

Chemical and Engineering News (American Chemical
Society) .

Chemical Reviews.

Chemistry and Industry.

Helvetica Chimica Acta.

Industrial and Engineering Chemistry.

) Industrial and Engineering Chemistry (Analytical Edi-
tion) [name now changed to Analytical Chemistry] .

Industrial and Engineering Chemistry (News Edition).

Journal of the American Chemical Society.

Journal of Chemical Education.

Journal of the Chemical Society.

Journal of Organic Chemistry.

Journal of the Royal Institute of Chemistry.

Journal of the Society of Chemical Industry.

Journal of the Society of Dyers and Colourists.

Nature.

Proceedings of the Chemical Society.

Quarterly Reviews of the Chemical Society (London) .

Recueil des Travaux Chimiques des Pays-Bas.

Research.

Science.

Tetrahedron.

Transactions of the Faraday Society.

PAGE

CONTENTS
LIST OF JOURNAL ABBREVIATIONS . . . . . x

CHAPTER

I. INTRODUCTION ....... i

Historical Introduction, i. Analysis of Organic Compounds, 2.
Structural Formulae and Isomerism, 7. Saturated and Un-
saturated Compounds, 8. Classification of Organic Com-
pounds, 8.

II. STRUCTURE OF THE ATOM . . . . .10

Structure of the Atom, 10. Electronic Theory of Valency, n.
Chelate Compounds, 14. Dipole Moments, 14. Electron Dis-
placements in a Molecule, 15. Effect of Structure on Reactivity,
21. Hydrogen Bond, 21. Atomic and Molecular Orbitals, 22.
Hybridisation of Bond Orbitals, 25. General Nature of Organic
Reactions, 31. Principle of Microscopic Reversibility, 32. Ki-
netically and Therrnodynamically Controlled Products, 32.
Transition State Theory of Reactions, 33. Use of Isotopes in
Organic Chemistry, 35.

ALIPHATIC COMPOUNDS

III. PARAFFINS . .38

Some Individual Paraffins, 38. Nomenclature of the Paraffins,
44. Homologous Series, 47. General Methods of Preparation,
48. General Properties, 52. General Reactions, 54. Petro-
leum and Natural Gas, 56. Cracking, 58. Synthetic Fuels,
60.

IV. UNSATURATED HYDROCARBONS .... 62

Olefins: Nomenclature, 62. Preparation, 63. Reactions, 65.
Polymerisation and Plastics, 77. Some Individual Olefins, 80.
Unsaturated Compounds with Two or More Double Bonds :
Nomenclature, 83. Isolated Double Bonds, 83. Cumulated
Double Bonds, 83. Conjugated Double Bonds, 84. Thiele's
Theory of Partial Valencies, 85. M.O. Treatment of Butadiene,
88. Molecular Diagrams, 89. Acetylenes: Nomenclature,
90. Acetylene: Preparation, 90. Reactions, 91. Effect of Hy-
bridisation on Electronegativity of a Carbon Atom, 96. Homo-
logies of Acetylene, 97.

V. HALOGEN DERIVATIVES OF THE PARAFFINS . 100

Nomenclature, 100. Alkyl Halides: Preparation, 100.
1,2-Shifts, 101. Reactions, 105. S„i and S N 2 Mechanisms, 106.
Polar Effects, 107. Steric Effects, 107. Solvent Effects, 109.
Ion pairs, 109. Elimination Reactions, in. Some Individual
Members, 114. Dihalogen Derivatives: Preparation, 115.
Reactions, 115. Some Individual Members, 116. Trihalogen
Derivatives: Chloroform, 116. Bromoform, 118. Iodoform,
118. Polyhalogen Derivatives: Carbon Tetrachloride, 118.
Tetrachloroethane, 119. Hexachloroethane, 119. Fluorine De-
rivatives, 120.

VI. MONOHYDRIC ALCOHOLS . . . .124

Nomenclature, 124. Preparation, 124. Reactions, 128.
Methods of Distinguishing Between the Three Classes of
Alcohols, 132. Some Individual Members, 133. Determination
of Structure, 136.

Xll CONTENTS

CHAPTER

VII. ETHERS ....

Nomenclature, 140. Preparation, 140. Reactions, 141. Some
Individual Members, 143.

140

VIII. ALDEHYDES AND KETONES 145

Nomenclature, 145. Preparation of Aldehydes and Ketones,
146. Reactions common to Aldehydes and Ketones, 149. Re-
actions Given by Aldehydes Only, 160. Reactions Given by
Ketones Only, 162. Some Individual Members, 164. Chloral,
167. Pinacol-Pinacone Rearrangement, 171. Bridged-ions, 172.

IX. FATTY ACIDS 174

Fatty Acids: Nomenclature, 174. Preparation, 175. Re-
actions, 176. Some Individual Members, 179. Esters:
Nomenclature, 186. Preparation, 187. Reactions, 191. Ortho-
Esters, 194. Esters of Inorganic Acids, 195. Acid Chlorides:
Nomenclature, 197. Preparation, 197. Reactions, 198. Acetyl
Chloride, 200. Acid Anhydrides: Nomenclature, 201.
Acetic Anhydride, 201. Acid Amides: Nomenclature, 203.
Preparation, 203. Reactions, 204. Hydroxamic Acids, 207.
Imidic Esters and Amidines, 208. Acid Hydrazides and
Acid Azides, 209. Halogeno- Acids : Nomenclature, 210.
Preparation, 210. Reactions, 212. Some Individual Members,
213.

X. TAUTOMERISM 216

Tautomerism, 216. Estimation of Enol Content, 218. Enolisa-
tion, 220. Modern Theories of Tautomerism, 221. Preparation
of Acetoacetic Ester, 224. Reactions, 225. Acetoacetic
Ester Syntheses, 227. Malonic Ester Syntheses, 232.
Hydroxyaldehydes and Hydroxyketones : Principal Func-
tion, 235. Names of Prefixes and Suffixes of Functions, 235.
Some Individual Hydroxyaldehydes and Hydroxyketones, 236.
Dialdehydes : Some Individual Members, 238. Diketones:
Some Individual Members, 240. Aldehydic Acids: Some
Individual Members, 242. Ketonic Acids: Some Individual
Members, 243.

XI. POLYHYDRIC ALCOHOLS . . . . .247

Nomenclature, 247. Glycols: Preparation of Glycol, 247.
Reactions, 248. Ethylene Oxide, 250. Dioxan, 252. General
Methods of Preparing Glycols, 253. Pinacol, 253. Poly-
methylene Glycols: Some Individual Members, 254. Gly-
cerol: Preparation, 254. Reactions, 255. Polyhydric
Alcohols, 258. Oils, Fats and Waxes, 259.

XII. UNSATURATED ALCOHOLS, ETHERS, CARBONYL

COMPOUNDS AND ACIDS 265

Unsaturated Alcohols: Some Individual Members, 265.
Hyperconjugation, 269. Allylic rearrangement, 272. Un-
saturated Ethers: Some Individual Members, 274. Un-
saturated Aldehydes: Some Individual Members, 275.
Unsaturated Ketones: Some Individual Members, 277.
Unsaturated Monocarboxylic Acids: Nomenclature, 279.
Preparation, 280. Reactions, 282. Some Individual Members,
284. Principle of Vinylogy, 287. Ketens: Keten, 287.
Diketen, 289.

XIII. NITROGEN COMPOUNDS 292

Hydrogen Cyanide, 292. Alkyl Cyanides: Nomenclature,
293. Preparation, 294. Reactions, 295. Alkyl j'soCyanides :
Preparation, 297. Reactions, 297. Cyanogen and its Re-
lated Compounds, 298. Nitroparaffins : Nomenclature, 301.

CONTENTS

Structure, 301. Preparation, 302. Reactions, 303. Nitroso-
Paraffins, 306. Monoamines, 308. Nomenclature, 308.
Preparation, 308. Preparation of Quaternary Compounds, 313.
Reactions of the Amines, 313. Quaternary Ammonium Com-
pounds, 318. Diamines: Some Individual Members, 320.
Unsaturated Amines: Some Individual Members, 321.
Aminoaixohols : Some Individual Members, 322. Amino-
acids, 322. Preparation and Reactions of Glycine, 322. Be-
taines, 325. Aliphatic Diazo-Compounds : Diazomethane,
325. Diazoacetic ester, 329.

Xlll

XIV. ALIPHATIC COMPOUNDS OF SULPHUR, PHOS-
PHORUS, ARSENIC AND SILICON .

Thioaixohols : Nomenclature, 332. Preparation, 332. Re-
actions, 333. Thioethers: Preparation, 334. Reactions, 335.
Thiocyanic Acid, isoThiocyanic Acid, and Their Deriva-
tives, 336. AtKYL SULPHOXIDES, 338. ALKYL SULPHONES,

338. Sulphonic Acids: Nomenclature, 338. Preparation,

339. Reactions, 339. Thioaldehydes and Thioketones,

340. Thioactds, 340. Dithioacids, 341. Alkyl-Phosphines :
Preparation, 341. Reactions, 342. Alkyl-Arsines : Prepara-
tion 343- Reactions, 343. Cacodyl Oxide and its Related
Compounds, 344. Antimony Compounds, 345. Bismuth
Compounds, 345. Alkyl-Silanes : Preparation, 345. Re-
actions, 346. Silanols and Silicones, 346.

332

XV. ORGANO-METALLIC COMPOUNDS . . . .348
Grignard Reagents: Preparation, 348. Structure, 348.
Reactions, 349. Synthetic Uses of the-Grignard Reagents, 351.
Abnormal Behaviour of Grignard Reagents, 359. Alkyl-
Metaixic Compounds: Nomenclature, 360. Preparation, 361.
Alkali Group, 361. Copper Group, 362. Organo-Zinc Com-
pounds, 363. Organo-Cadmium Compounds, 364. Organo-
Mercury Compounds, 365. Organo-Lead Compounds, 365.
Preparation and Reactions of Free Alkyl Radicals, 365.

XVI. SATURATED DICARBOXYLIC ACIDS . . .368

Nomenclature, 368. Preparation, 368. Some Individual Mem-
bers, 370. iV-Bromosuccinimide, 375. Blanc's Rule, 380. Car-
bonic Acid and its Derivatives, 381. Urea, 383. Inclusion Com-
plexes, 387. Compounds Related to Urea, 388.

XVII. HYDROXYACIDS. STEREOCHEMISTRY. UN-
SATURATED DICARBOXYLIC ACIDS

Monobasic Hydroxy acids : Nomenclature, 393. Preparation,
393. Reactions, 394. Some Individual Members, 397. Stereo-
chemistry, 399. Elements of Symmetry, 409. Resolution of
Racemic Modifications, 411. Racemisation, 412. Walden In-
version, 413. Correlation of Configuration, 413. Neighbouring
Group Participation, 415. Asymmetric Synthesis, 416. Hy-
droxy-Dibasic and Polybasic Acids, 417. Unsaturated
DicArboxylic Acids, 423. Geometrical Isomerism, 425.
Mechanism of Addition to Double and Triple Bonds, 428. ' Some
Individual Unsaturated Dicarboxylic Acids, 430. Allenes and
Spirans, 432. Optical Isomerism of Elements other than Carbon
432.

393

XVIII. CARBOHYDRATES

Nomenclature, 436. Monosaccharides, 436. Configuration of
the Monosaccharides, 456. Determination of the Size of Sugar-
Rings, 458. Disaccharides, 460. Polysaccharides, 463.

436

XIV

CONTENTS

ALICYCLIC COMPOUNDS

XIX. ALICYCLIC COMPOUNDS

Nomenclature, 467. Preparation, 468. Diels-Alder Reaction,
472. Reactions, 474. Some Individual Members, 475. Ferro-
cene, 480. Cyc/oalkynes, 486. Baeyer's Strain Theory, 486.
Theory of Strainless Rings, 488. Conformational Analysis, 488.
Large Ring Compounds, 492.

467

AROMATIC COMPOUNDS

XX. SIMPLE AROMATIC HYDROCARBONS .

Benzene, 499. Properties, 501. Structure, 503. Aromaticity,
507. Methods of Orientation, 511. Substitution in the Benzene
Ring, 514. Electrophilic Substitution, 516. Ortho-para Ratio,
523. Nucleophilic Substitution, 524. Free-radical Substitution,
526. Preparation of Benzene Homologues, 529. Synthesis of
Aromatics from Aliphatics, and vice-versa, 541.

499

XXI. AROMATIC HALOGEN COMPOUNDS ...

Addition Compounds, 544. Nuclear Substitution Products:
Preparation, 544. Properties, 546. Benzyne mechanism, 547.
Side-Chain Substituted Compounds : Preparation, 549. Some
Individual Members, 550. Polyvalent Iodine Compounds, 551.

544

XXII. AROMATIC NITRO-COMPOUNDS ....

Methods of Nitration, 553. Properties, 554. Some Individual
Members, 554. Charge-transfer Complexes, 557. Aromatic
Nitroso-Compounds : Nitrosobenzene, 560. Phenylhydroxyl-
amine, 561. Reduction Products of Nitro-Compounds, 562.

553

XXIII. AROMATIC AMINO-COMPOUNDS .

Preparation, 564. Some Individual Members, 566.
578.

Diamines,

564

XXIV. DIAZONIUM SALTS AND THEIR RELATED COM-
POUNDS

Diazonium Salts: Diazotisation, 582. Nomenclature, 583.
Replacement Reactions, 583. Reactions in which Nitrogen is
Retained, 588. Structure of Diazonium Salts, 591. Phenyl-
hydrazine, 594. DlAZOAMINO- AND AMINOAZO-COMPOUNDS,

595. Azoxybenzene, 599. Azobenzene, 600. Hydroxyazo-
Compounds, 601. Hydrazobenzene, 602. Benzidine Rearrange-
ment, 602.

582

XXV. SULPHONIC ACIDS

Methods of Sulphonation, 606. Some Individual Sulphonic
Acids, 608. Sulphinic Acids, 614. Thiophenol, 615. Amino-
benzenesulphonic Acids, 615.

606

XXVI. PHENOLS .

Monohydric Phenols : Phenol and its Derivatives, 61 8. Homo-
logues of Phenol, 631. Dihydric Phenols, 632. Trihydric
Phenols, 635. Aromatic Ethers, 637. Claisen Rearrange-
ment, 640.

618

CONTENTS

CHAPTER

XXVII. ALCOHOLS, ALDEHYDES, KETONES AND
QUINONES

Aromatic Alcohols, 644. Aromatic Aldehydes: Benzaldehyde
and its Derivatives, 645. Phenolic Aldehydes: Preparation,
655. Some Individual Members, 658. Aromatic Ketones:
Some Individual Members, 660. Stereochemistry of Ald-
oximes and Ketoximes, 666. Quinones : Benzoquinones, 668.
Quinols, 673.

XXVIII. AROMATIC ACIDS

Nuclear-Carboxylic Acids: Preparation, 675. Some Indi-
vidual Members, 675. The tw/Ao-Effect, 686. Acids with
Carboxyl Group in Side-Chain: Some Individual Members,
689. Dibasic Aromatic Acids: Some Individual Members, 692!

POLYNUCLEAR HYDROCARBONS AND THEIR
DERIVATIVES

Diphenyl, 697- Diphenylmethane, 700. Triphenylmethane,
701. Tetraphenylmethane, 703. Hexaphenylethane, 703. Di-
benzyl, 705. Stilbene, 705. Benzoin, 706. Naphthalene, 708.
Acenaphthene, 723. Indene, 724. Fluorene, 725. Anthracene,
726. Phenanthrene, 734. Molecular Overcrowding, 739.

XXIX.

xv

PAGE
644

675

697

HETEROCYCLIC COMPOUNDS
XXX. HETEROCYCLIC COMPOUNDS ....

Furan and its Derivatives, 741. Thiophen and its Derivatives,
745- Pyrrole and its Derivatives, 746. Indole and Related
Compounds, 750. Nomenclature of Heterocyclic Compounds,
753- Pyrazoles, Glyoxalines, Oxazoles, iso-Oxazoles, Thiazoles,'
Sydnones, 754. Pyridine and its Derivatives, 756. Methods oi
Ring Fission, 761. Quinoline, 763. isoQuinoline, 765. Acri-
dine, 767. Pyrones, 767. Diazines, 769.

XXXI. DYES

Relation Between Colour and Constitution, 771. Nomenclature
of Dyes, 778. Classification, 779. Nitro-Dyes, 781. Nitroso-
Dyes, 781. Azo-Dyes, 782. Diphenylmethane Dyes, 785. Tri-
phenylmethane Dyes, 786. Xanthen Dyes, 790. Diphenyl-
amine Dyes, 795. Heterocyclic Dyes, 795. Vat Dyes, 800.
Anthraquinoid Dyes, 806. Sulphur Dyes, 808. Phthalo-
cyanine Dyes, 808.

EXAMINATION QUESTIONS

APPENDIX

Nomenclature: Order of Naming Radicals, 814. Greek
Alphabet, 815. Writing of Names, 815. Organic Chemistry
Publications: Dictionaries, 816. Physical Constants, 816. Re-
ference Books, 816. Chemical Journals, 817. Searching the
Literature, 818. Use of Beilstein, 819.

741

771

810

814

INDEX

823

CHAPTER I

INTRODUCTION

HISTORICAL INTRODUCTION

Although organic substances such as sugar, starch, alcohol, resins, oils,
indigo, etc., had been known from earliest times, very little progress in their
chemistry was made until about the beginning of the eighteenth century.
In 1675, Lemery published his famous Cours de Chymie, in which he divided
compounds from natural sources into three classes: mineral, vegetable and
animal. This classification was accepted very quickly, but it was Lavoisier
who first showed, in 1784, that all compounds obtained from vegetable
and animal sources always contained at least carbon and hydrogen, and
frequently, nitrogen and phosphorus. Lavoisier, in spite of showing this
close relationship between vegetable and animal products, still retained
Lemery's classification. Lavoisier's analytical work, however, stimulated
further research in this direction, and resulted in much-improved technique,
due to which Lemery's classification had to be modified. Lemery had based
his classification on the origin of the compound, but it was now found
(undoubtedly due to the improved analytical methods) that in a number of
cases the same compound could be obtained from both vegetable and animal
sources. Thus no difference existed between these two classes of compounds,
and hence it was no longer justifiable to consider them under separate
headings. This led to the reclassification of substances into two groups:
all those which could be obtained from vegetables or animals, i.e., substances
that were produced by the living organism, were classified as organic; and
all those substances which were not prepared from the living organism were
classified as inorganic.

At this stage of the investigation of organic compounds it appeared that
there were definite differences between organic and inorganic compounds,
e.g., complexity of composition and the combustibility of the former.
Berzelius (18 15) thought that organic compounds were produced from their
elements by laws different from those governing the formation of inorganic
compounds. This then led him to believe that organic compounds were
produced under the influence of a vital force, and that they could not be
prepared artificially.

In 1828, Wohler converted ammonium cyanate into urea, a substance
hitherto obtained only from animal sources. This synthesis weakened the
distinction between organic and inorganic compounds, and this distinction
was completely ended with the synthesis of acetic acid from its elements
by Kolbe in 1845, and the synthesis of methane by Berthelot in 1856. A
common belief appears to be that Wohler's synthesis had little effect on the
vital-force theory because it did not start with the elements. Wohler had
prepared his ammonium cyanate from ammonia and cyanic acid, both of
which were of animal origin. Partington (i960), however, has pointed out
that Priestley (178 1) had obtained ammonia by reduction of nitric acid,
which was later synthesised from its elements by Cavendish (1785). Also,
potassium cyanide was obtained by Scheele (1783) by passing nitrogen over
a strongly heated mixture of potassium carbonate and carbon, and since one
form of carbon used was graphite, this reaction was therefore carried out
with inorganic materials. Since potassium cyanide is readily converted
into potassium cyanate, Wohler's synthesis is one which starts from the
elements.

2 ORGANIC CHEMISTRY

Since the supposed differences between the two classes of compounds have
been disproved, the terms organic and inorganic would appear to be no
longer necessary. Nevertheless, they have been retained, but it should be
appreciated that they have lost their original meaning. The retention of
the terms organic and inorganic may be ascribed to several reasons : (i) all
so-called organic compounds contain carbon; (ii) the compounds of carbon
are far more numerous (over 750,000) than the known compounds of all the
other elements put together; (iii) carbon has the power to combine with
other carbon atoms to form long chains. This property, known as catenation,
is not shown to such an extent by any other element.

Hence organic chemistry is the chemistry of the carbon compounds.

This definition includes compounds such as carbon monoxide, carbon
dioxide, carbonates, carbon disulphide, etc. Since these occur chiefly in
the inorganic kingdom (original meaning), they are usually described in
text-books of inorganic chemistry.

ANALYSIS OF ORGANIC COMPOUNDS

The following is an outline of the methods used in the study of organic
compounds.

(1) Purification. Before the properties and structure of an organic
compound can be completely investigated, the compound must be prepared
in the pure state. Common methods of purification are :

(i) Recrystallisation from suitable solvents.

(ii) Distillation: (a) at atmospheric pressure; (b) under reduced pressure
or in vacuo ; (c) under increased pressure.

(iii) Steam distillation.

(iv) Sublimation.

(v) Chromatography. This method is based on the differential adsorption
of the different components of a mixture on various adsorbents. Chromato-
graphy offers a means of concentrating a product that occurs naturally in
great dilution, and is an extremely valuable method for the separation,
isolation, purification and identification of the constituents of a mixture.

It is surprising how much information has often been obtained about the
properties and structure of a substance that has not been isolated in a pure
state. Even so, purification should always be attempted, since it is much
simpler to investigate a pure substance than an impure one.

(2) Qualitative analysis. The elements commonly found in organic
substances are: carbon (always: by definition), hydrogen, oxygen, nitrogen,
halogens, sulphur, phosphorus and metals.

(i) Carbon and hydrogen. The compound is intimately mixed with dry
cupric oxide and the mixture then heated in a tube. Carbon is oxidised to
carbon dioxide (detected by means of calcium hydroxide solution), and
hydrogen is oxidised to water (detected by condensation on the cooler parts
of the tube).

(ii) Nitrogen, halogens and sulphur. These are all detected by the
Lassaigne method. The compound is fused with metallic sodium, whereby
nitrogen is converted into sodium cyanide, halogen into sodium halide, and
sulphur into sodium sulphide. The presence of these sodium salts is then
detected by inorganic qualitative methods.

(iii) Phosphorus. The compound is heated with fusion mixture, whereby
the phosphorus is converted into metallic phosphate, which is then detected
by the usual inorganic tests.

INTRODUCTION 3

(iv) Metals. When organic compounds containing metals are strongly
heated, the organic part usually burns away, leaving behind a residue. This
residue is usually the oxide of the metal, but in certain cases it may be the
free metal, e.g., silver, or the carbonate, e.g., sodium carbonate.

As a rule, no attempt is made to carry out any test for oxygen: its
presence is usually inferred from the chemical properties of the compound.

The non-metallic elements which occur in natural organic compounds, in order
of decreasing occurrence, are hydrogen, oxygen, nitrogen, sulphur, phosphorus,
iodine, bromine and chlorine. Halogen compounds are essentially synthetic
compounds, and are not found to any extent naturally. Some important
exceptions are chloramphenicol (chlorine), Tyrian Purple (bromine) and
thyroxine (iodine). In addition to these non-metallic elements, various metallic
elements occur in combination with natural organic compounds, e.g., sodium,
potassium, calcium, iron, magnesium, copper.

(3) Quantitative analysis. The methods used in the determination of the
composition by weight of an organic compound are based on simple principles.

(i) Carbon and hydrogen are estimated by burning a known weight of the
substance in a current of dry oxygen, and weighing the carbon dioxide and
water formed. If elements (non-metallic) other than carbon, hydrogen and
oxygen are present, special precautions must be taken to prevent their inter-
fering with the estimation of the carbon and hydrogen.

(ii) Nitrogen may be estimated in several ways, but only two are commonly
used.

(a) Dumas' method. This consists in oxidising the compound with copper
oxide, and measuring the volume of nitrogen formed. This method is
applicable to all organic compounds containing nitrogen.

(b) Kjeldahl's method. This depends on the fact that when organic com-
pounds containing nitrogen are heated with concentrated sulphuric acid, the
organic nitrogen is converted into ammonium sulphate. This method,
however, has certain limitations.

(iii) Halogens may be estimated in several ways. One is the classical
method of Carius. The substance is heated in a sealed tube with fuming
nitric acid in the presence of silver nitrate. Silver halide is formed, and this
is estimated gravimetrically.

A simpler method for non-volatile compounds is to fuse the substance
with sodium peroxide in a nickel crucible, whereupon the halogen is con-
verted into sodium halide, which is then estimated as before.

(iv) Sulphur may be estimated by the methods used for the halogens. In
the Carius method for sulphur, no silver nitrate is used. Organic sulphur is
converted into sulphuric acid (Carius method) or sodium sulphate (sodium
peroxide fusion). In both cases the sulphate is precipitated as barium
sulphate and weighed.

(v) Phosphorus may be estimated by heating the compound with fusion
mixture and weighing the phosphate as magnesium pyrophosphate.

The Carius determination (no silver nitrate used, cf. sulphur) invariably
gives low results for phosphorus. Olivier (1940) found that exact results
were obtained by heating the organic compound mixed with calcium oxide
in a stream of oxygen. The phosphate was then estimated as above.

(vi) Oxygen is usually estimated by difference. All direct methods are
still not completely satisfactory, but recently Aluise and co-workers (1947)
claim to have evolved a satisfactory technique. The organic compound is
subjected to pyrolysis in a stream of nitrogen, and all the oxygen in the
pyrolysis products is converted into carbon monoxide by passage over

4 ORGANIC CHEMISTRY

carbon heated at 1120 . The carbon monoxide is then passed over iodine
pentoxide, and the iodine liberated is estimated titrimetrically.

Quantitative analysis falls into three groups according to the amount of
material used for the estimation:

(i) Macro-methods which require about o-i-o-5 g. of material (actual
amount depends on the element being estimated).

(ii) Semi-micro methods which require 20-50 mg. of material,
(hi) Micro-methods which require 3-5 mg. of material.

Nowadays the tendency is to use method (ii) or (hi). Although all the
methods are simple in theory, their successful application (particularly when
using micro- or semi-micro methods) requires a great deal of technical skill.
These methods have become standardised, and are described in detail in
many books on practical organic chemistry. Improvements and new
methods for analysis, however, are always being published; e.g., chlorine and
sulphur may be determined by wrapping the sample of the compound in
filter paper, igniting and lowering it into a flask filled with oxygen. The
acid gases are absorbed in hydrogen peroxide; the sulphuric acid formed is
titrated with standard alkali, and the chloride is determined by titrating the
neutralised solution with mercuric nitrate (Mikl et al., 1953). Fluorine,
chlorine and nitrogen may be determined by decomposition in a nickel
bomb (Brown et al., 1955).

(4) Empirical formula determination. The empirical formula indicates
the relative numbers of each kind of atom in the molecule, and is calculated
from the percentage composition of the compound.

(5) Molecular weight determination. The molecular formula — this gives
the actual number of atoms of each kind in the molecule — is obtained by
multiplying the empirical formula by some whole number which is obtained
from consideration of the molecular weight of the compound. In many
cases this whole number is one.

The methods used for the determination of molecular weights fall into two
main groups : physical and chemical. The standard physical methods are the
determination of : (i) vapour density ; (ii) elevation of boiling point ; (iii) de-
pression of freezing point. These methods are described fully in text-books of
physical chemistry. In addition to these standard methods, which are used
mainly for relatively simple molecules, there are also other physical methods
used for compounds having high molecular weights, e.g., rate of diffusion, rate
of sedimentation, viscosity of the solution, osmotic pressure, etc.

The chemical methods, since they are only useful in organic work, will be
here described in detail.

(i) Molecular weights of organic acids (method of silver salt). If the

basicity of the acid is known, then the molecular weight of that acid may be
determined from the analysis of its silver salt. The silver salt is chosen
because: (a) Most silver salts are insoluble in water, and hence they are
readily prepared, (b) Most silver salts are anhydrous; this is a definite
advantage, since it does not introduce a possible source of error (i.e., the
determination of water of crystallisation), (c) All silver salts are readily
decomposed on ignition, leaving a residue of metallic silver.

The method of calculation is shown in the following example: 0-701 g. of the
silver salt of a dibasic acid on ignition yielded 0-497 g. of metallic silver.
Calculate the M.Wt. of the acid, given that the A.Wt. of silver is 108.

Since the acid is dibasic, its molecule can be represented by the formula H 2 A,
where A is that part of the molecule other than replaceable hydrogen atoms.
Hence the silver salt will be Ag 2 A, i.e., one gram molecule of it contains 216 g.
of silver.

INTRODUCTION 5

There is 0-497 g. silver in 0701 g. of AgjA.

.-. there is 216 g. silver in °' 7 °* X 2I - - g. of Ag 2 A = 3047 g.
i.e., the M.Wt. of Ag 2 A is 3047.

.-. the M.Wt. of acid H 2 Ais (Ag 2 A - 2Ag + 2H) = (3047 - 216 + 2) = 907.

(ii) Molecular weights of organic bases (method of chloroplatinate).

Organic bases combine with chloroplatinic acid, H 2 PtCl 6 , to form insoluble,
anhydrous chloroplatinates (platinichlorides) which, on ignition, leave a
residue of metallic platinum. Let B represent one molecule of the base. If
it is a " monoacid " base, the formula of its chloroplatinate will be
B 2 H 2 PtCl 6 ; if a " diacid " base, BH 2 PtCl„.

Example. 0800 g. of the chloroplatinate of a " monoacid " base on ignition
gave 0-262 g. of platinum. Calculate the M.Wt. of the base, given that the
A.Wt. of platinum is 195.

Since the base is "monoacid", the formula of its chloroplatinate will be
B 2 H 2 PtCl„ i.e., one gram molecule of the chloroplatinate contains 195 e. of
platinum.

There is 0-262 g. of platinum in o-8oo g. of B 2 H 2 PtCl„.

.-. there is 195 g- of platinum in °' 80 ° 2 g 2 195 = 595"4 g- of B 2 H 2 PtCl 8 .

i.e., the M.Wt. of B 2 H a PtCL is 595-4.
/. the M.Wt. of B is

B 2 H 2 PtCl, - H 2 PtCl 8 595-4 - (2 + 195 + 213)

- = _ =g2 . 7 _

(iii) The molecular formula of any gaseous hydrocarbon (compound con-
taining carbon and hydrogen only) may be determined by exploding a
measured volume of the gas with a measured excess of oxygen, in a eudio-
meter tube.

Example. 10 ml. of a gaseous hydrocarbon was mixed with 80 ml. of
oxygen and the mixture exploded. 70 ml. of gas remained (after cooling to
room temperature), and this was reduced to 50 ml. (of oxygen) after treatment
with potassium hydroxide solution. What is the formula of the hydrocarbon ?

There are two ways of solving this problem :

(«) C + O a >- CO a .

Thus one atom of carbon requires one molecule of oxygen.

2H 2 + O a > 2 H 2 0.

Thus one atom of hydrogen requires J molecule of oxygen. Let the formula of
the hydrocarbon be C.H,. Then x molecules of oxygen will be required to

burn the carbon to carbon dioxide, and y - molecules of oxygen to burn the
hydrogen to water. Thus we have

C,H, + (*+?)o 2 — ^*CO a +?H 2 . . . . (i)

From Avogadro's law, it follows that

1 vol. of CJHt, + [x + =n vol. of O a ^ x vol. CO a + ^ vol. H a O (as steam) (ii)

Since measurements of volume are carried out at room temperature, the
water will be present as liquid, the volume of which may be ignored. There-
fore, contraction after sparking = (1 -f- x + -) — x = (1 + -) vol.

After treatment with potassium hydroxide solution, the contraction will be
vol. {i.e., vol. of CO s ).

ORGANIC CHEMISTRY

From the figures of the experiment, we have

First contraction = 90 — 70 = 20 ml.
Second contraction = 70 — 50 = 20 ml.

y.

.". since 10 ml. of C X HL is to be taken as 1 vol. (from equation ii), then 1 + - =2.

4
.'. y = 4 and x = 2.
Hence the hydrocarbon is C 2 H 4 .

(b) Total amount of oxygen used = 80 — 50 = 30 ml.

Of this, 20 ml. was required for burning the carbon (vol. of C0 2 is equal to
vol. of O a used). Hence 10 ml. was required for the hydrogen which gives
20 ml. of steam.

.•. 10 ml. QHj, + 30 ml. O a >- 20 ml. C0 2 + 20 ml. H 2 (steam).

.'. from Avogadro's law

C X H, + 3 O a > 2CO a + 2H 2 0.

.'. x = 2, y = 4; and the hydrocarbon is C 2 H 4 .

(6) Determination of structure, i.e., the manner in which the atoms are
arranged in the molecule. The usual procedure for elucidating the structure
of an unknown compound is to make a detailed study of its chemical re-
actions. This procedure is known as the analytical method, and includes
breaking down {degrading the compound into smaller molecules of known
structure.

In addition to the purely chemical means, there are also various physical
properties which are used to elucidate structure, e.g. :

(i) Dipole moment. This gives information on the spatial arrangement of
atoms in a molecule, and so offers a means of distinguishing between alternative
arrangements.

(ii) Refractive index. This may be used to distinguish between two types of
structure, e.g., between a keto and an enol form.

(iii) Parachor. This has been used to distinguish between alternative structures.

(iv) X-Ray analysis. This offers a means of studying the arrangement of
atoms in crystalline solids, but it may also be used for liquids and gases. Since
most organic compounds are complex from the point of view of structure, X-ray
analysis has mainly been used to " round off " information obtained by purely
chemical means. Bond lengths may be measured by X-ray analysis, and devia-
tions from " normal " values give information on structure.

(v) Electron diffraction. This has been used in the same way as X-ray analysis,
and is applicable to gases, liquid and solids. It is, however, usually confined to
gases or compounds in the vapour state.

(vi) Absorption spectra. All organic compounds absorb light, which may be
in one or more of the following regions : infra-red, visible or ultra-violet. Many
bands are associated with particular groups, and it is therefore possible to
ascertain the presence of these various groups in a new compound. In general,
compounds possessing similar structures show similar absorption spectra. Hence
the structure of a new compound may be elucidated by comparing its absorption
spectrum with known spectra.

The Raman effect also is characteristic of a particular group, and has been
widely used to ascertain the nature of the groups present in a compound.

When sufficient evidence has been accumulated, a tentative structure
which best fits the facts is accepted. Sometimes two (or even more) struc-
tures fit the facts almost equally well, and it has been shown in certain cases
that the compound exists in both forms which are in equilibrium. This
phenomenon is known as tautomerism. Where tautomerism has not been
shown to be present, one must accept (with reserve) the structure that has
been chosen (see also next section).

(7) Synthesis of the compound. The term synthesis means the building
up of a compound, step by step, from a simpler substance of known structure.

INTRODUCTION 7

The term complete synthesis means the building up of a compound, step by
step, starting from its elements (and any others that may be necessary). In
either case (synthesis or complete synthesis), the structure of each inter-
mediate compound is taken as proved by its synthesis from the compound that
preceded it.

The synthesis of a compound is necessary to establish its structure be-
yond doubt. There is always the possibility of one or more steps not pro-
ceeding " according to plan ". Hence the larger the number of syntheses
of a compound by different routes, the more reliable will be the structure
assigned to that compound.

STRUCTURAL FORMULA AND ISOMERISM

In 1857, Kekule postulated the constant quadrivalency (tetravalency) of
carbon. From 1900 onwards, however, compounds containing tervalent
carbon have been prepared, and their number is increasing rapidly. These
compounds usually require special methods of preparation, and many have
a very short life (see text). Since their properties are different from those
compounds containing quadrivalent carbon, they are fairly easily recognised.
More recently, compounds containing bivalent carbon (carbenes) are believed
to be formed as intermediates during certain reactions. Hence, unless there
is definite evidence to the contrary, carbon is always assumed to be quadri-
valent.

If " valency units " or " valency bonds " (see Ch. II) are represented by
lines, then the number of lines drawn from the symbol shows the valency of
that atom, e.g.,

I I

— C— ; — O— ; — N— ; H—

I

The molecular formula shows the number of each kind of atom present
in the molecule, but does not indicate their arrangement. In organic
chemistry there are many cases where a given formula represents two or
more compounds that differ in physical and chemical properties, e.g., there
are at least seven compounds having the same molecular formula C 4 H 10 O.
Such compounds, having the same molecular formula, but differing in
physical and chemical properties, are known as isomers or isomerides, the
phenomenon itself being known as isomerism. The existence of isomerism
may be explained by assuming that the atoms are arranged in a definite
manner in a molecule, and that there is a different arrangement in each
isomer, i.e., the isomers differ in structure or constitution. This type of iso-
merism is known as structural isomerism.

Obviously, then, from what has been said above, it is always desirable to
show the arrangement (if known) of the atoms in the molecule, and this is
done by means of structural formula or bond-diagrams; e.g., the molecular

formula of ethanol is C 2 H 8 0; its structural formula is H-

H H

-C— C— 0— H.

H H

So far nothing has been said about the spatial disposition of the four
valencies of carbon. Later (Ch. II) it will be shown that when carbon is
joined to four univalent atoms or groups, the four valencies are directed
towards the four corners of a tetrahedron. Thus the above plane-structural

8 ORGANIC CHEMISTRY

formula does not show the disposition of the atoms in space; a three-
dimensional formula is necessary for this. Usually the plane-formula is
satisfactory. Since the actual spatial arrangements of a given structure may
differ, this gives rise to isomerism of the type known as stereoisomerism.
Stereoisomers have different configurations, i.e., the spatial arrangements are
different but not their structures (see p. 399).

A structural formula is really a short-hand description of the properties
of the compound. Hence the study of organic chemistry is facilitated by
mastering the structural formula of every compound the reader meets. An
organic molecule, however, is only completely described when the following
facts are known: structure or constitution (this includes a knowledge of the
electron distribution; see resonance, p. 17), configuration (p. 399), and
conformation (p. 488).

SATURATED AND UNSATURATED COMPOUNDS

If, in an organic compound containing two or more carbon atoms, there
are only single bonds Unking any two adjacent carbon atoms, then that
compound is said to be saturated, e.g., ethane, C 2 H 6 (I), normal propanol,
C 3 H 8 (II), acetaldehyde, C 2 H 4 (III).

H H H H H H

II III I //>

H— C— C— H H— C— C— C— O— H H— C— c(

II III I X H

H H (I) H H H (II) H (ill)

On the other hand, if the compound contains at least one pair of adjacent
carbon atoms linked by a multiple bond, then that compound is said to be
unsaturated, e.g., ethylene, C 2 H 4 (IV); this compound contains a double
bond. Acetylene, C 2 H 2 (V); this contains a triple bond. Acraldehyde,
C 3 H 4 (IV); this contains a double bond. The double bond between the
carbon and oxygen atoms is not a sign of unsaturation (cf. acetaldehyde above).

H

H \ / H H \ i ^°

/C=Cf H— C=C— H /C=C— Cf

(IV) (V) (VI)

CLASSIFICATION OF ORGANIC COMPOUNDS
Organic compounds are classified into three major groups:

(1) (a) Aliphatic, open-chain, or acyclic compounds.

(6) Alicyclic compounds. These are carbocyclic or ring compounds
which resemble aliphatic compounds in many ways.

(2) Aromatic compounds. These are carbocyclic or ring compounds
containing at least one benzene ring (see also p. 509) .

(3) Heterocyclic compounds. These are cyclic (ring) compounds con-
taining other elements besides carbon in the ring. In a few cases no carbon
atom is in the ring.

READING REFERENCES

Partington, A Short History of Chemistry, Macmillan (1957, 3 r d e &-)- Ch. X. The

Beginnings of Organic Chemistry.
Schorlemmer, The Rise and Development of Organic Chemistry, Macmillan (1894).

INTRODUCTION g

Japp, Kekule Memorial Lecture, J.C.S., 1898, 73, 97.

Mann and Saunders, Practical Organic Chemistry, Longmans, Green (1960, 4th ed.)
Part IV. Quantitative Analysis.

Belcher and Godbert, Semi-Micro Quantitative Organic Analysis, Longmans, Green
(1954, 2nd ed.).

Pre g'. Quantitative Organic Microanalysis, Churchill (1937).

Vogel, Practical Organic Chemistry, Longmans, Green (1956, 3rd ed.), Ch. 12. Semi-
micro Technique.

Ann. Reports (Chem. Soc), 1955, 52, 353. Classical Organic Analysis.

Ingram, The Rapid Micro-combustion Procedure, Chem. and Ind., 1956, 103.

CHAPTER II

STRUCTURE OF THE ATOM

According to modern theory, an atom consists of a nucleus which contains
■protons and neutrons, and which is surrounded by electrons. The mass of a
proton is almost the same as that of a neutron, but whereas the proton
carries a unit of positive charge, the neutron is electrically neutral. The
electron has about TsVoth or the mass of a proton, and carries a unit of
negative charge. The electrons are arranged in shells around the nucleus,
each shell being able to contain up to a maximum number of electrons, this-
maximum depending on the number of the shell, n. n is known as the
principal quantum number, and indicates the main energy level of the
electrons in that shell, n has whole number values, i, 2, 3, 4 . . ., the
shells corresponding to which are also denoted by the letters K, L, M, N
. . . respectively. In every principal quantum shell there are n energy
sublevels, and these are indicated by /, the orbital quantum number (also
known as the azimuthal or serial quantum number). Just as the principal
quantum number n can have values 1, 2, 3 . . ., so can I have values o, 1,
2, 3 . . ., n — 1. The energy state corresponding to / = o is called the s
state; I = 1, the p state; I = 2, the d state; etc. As we shall see later
(p. 23), these s, p and d states are subdivided into a number of orbitals. The
total number of orbitals that a principal quantum shell can contain is given
by n 2 . Thus, when the principal quantum number is 1 (i.e., the first or K
shell), then I = 0, i.e., there is a single orbital in this K skell and is of the
s type and is known as the is orbital. When n = 2 (i.e., the second or L
shell), then I — or 1. This means there are two energy sublevels in the L
shell. As pointed out above, the total number of orbitals in a given quantum
shell is equal to n 2 . Thus, when n = 2, there are four possible orbitals.
When l = o, this corresponds to the zs orbital. When I = 1, then there are
three equivalent orbitals; these are the three 2p orbitals. When n = 3
(i.e., the third or M shell), then the total number of possible orbitals is 9 (3 2 ).
These correspond to one 3s orbital (I = 0), three $p orbitals (/ = 1) and five
3d orbitals (I = 2). The existence of one s level, three p, five d and seven/
levels of energy was used to explain the existence of spectral lines observed
in the spectra of atoms and molecules. It should also be noted that the
farther an electron is from the nucleus, the greater is its potential energy.
Owing to the phenomenon of penetration of orbitals, i.e., outer electrons can
penetrate into the shell of inner electrons, the energy level of an electron is
thus not completely determined by its principal quantum number, but also
depends on the orbital quantum number (i.e., on the shape of the orbital).
Thus there is the following order of increasing energy: is, 2s, zp, 3s, $p,
4s, 3d, 4p, 5s, 4d, 5p . . . (see also p. 24).

In addition to the energy levels of an electron described by quantum
numbers n and I, electrons have spin about their axis, some spinning in one
direction and others in the opposite direction. This is indicated by the spin
quantum number (s), and can have values of +£ and —J. Finally, an electron
also has a magnetic quantum number (m), and this gives the allowed orienta-
tions of the orbitals in an external magnetic field. Thus an electron is
described by four quantum numbers, n, I, s and m.

By the fundamental Pauli Exclusion Principle (1925), no two electrons, in
any system, can be assigned the same set of four quantum numbers. Hence
there can be only two electrons in any one orbital, and these must be differen-
tiated from each other by their spins, which must be antiparallel, i.e., in the

STRUCTURE OF THE ATOM II

opposite sense. Such electrons are said to be paired, and a pair of electrons
with antiparallel spins in the same orbital is represented by the symbol ^j- .
Since a moving charge is accompanied by a magnetic field, a spinning electron
behaves as a small bar-magnet, and consequently two paired electrons will
give a zero resultant magnetic field.

The hydrogen atom consists of one proton and one electron. When the
hydrogen atom is in the " ground " state, i.e., the state of lowest energy, its
electron will be in the lowest energy level, i.e., the is level, and is represented
by (is). When hydrogen is in an " excited " state, its electron will occupy
a higher energy level, the actual level depending on the amount of
" excitation ".

Helium has two electrons; hence its electron configuration in the ground
state is represented as (is) 2 . Lithium has three electrons. Since the
maximum number of electrons in the K shell (n = i, I = o) is two, the third
electron must start the L shell (n = 2, I = o, 1). Electrons occupy lowest
energy levels first. Thus this third electron occupies the 2s orbital, and
not the 2p, because the 2p is a higher energy level than the 2s. Hence the
electron configuration of lithium is (is) 2 (2s). Thus the K shell is filled first.
Then the electrons enter the L shell until that is filled. In this shell the s
level is filled before the p. In fitting electrons into shells containing orbitals
of equivalent energy, Hund's rules are used to assign the electrons to their
orbitals. These rules are: (i) electrons tend to avoid being in the same
orbital as far as possible; (ii) two electrons, each singly occupying a given
pair of equivalent orbitals, tend to have their spins parallel when the atom is
in the ground state. Thus carbon, with six electrons, may be represented
as (is) 2 (2s) 2 (2^>) 2 . The K shell is filled first; the L shell is filled next, the
2s orbital being doubly filled before a higher level is used; then singly two
of the 2p orbitals. Nitrogen, with seven electrons, is (is) 2 (2s) 2 (2^>) 3 : all
three 2p orbitals each contain one electron. Oxygen, with eight electrons,
is (is) 2 (2s) 2 (2p) i : here one of the 2p orbitals is doubly filled.

THE ELECTRONIC THEORY OF VALENCY

The electronic theory of valency starts with the assumption that valency
involves the electrons in the outer shells: in some cases only those in the
highest sublevel in the outermost shell; in other cases those in the highest
and penultimate sublevels, even though the penultimate sublevel may be
in a lower quantum shell. Lewis (1916) assumed that the electron con-
figuration in the rare gases was particularly stable (since these gases are
chemically inert), and that chemical combination between atoms took place
by achieving this configuration. The outermost shell of the rare gases
always contains an octet of two s and six p electrons. Since both the s and
p sublevels are completely filled, the octet will be a stable configuratien,
In the case of helium, however, an octet is impossible; here the staWe
arrangement is the duplet, the two is electrons of which completely fill the
first quantum shell.

The octet rule applies only to atoms with 2s and 2p electrons, i.e., to
elements in the second period (Li — F). With the other elements, d orbitals
may also be used in bond formation, and hence higher covalencies {i.e., expan-
sions beyond an octet) are possible, e.g., PBr 8 (10 electrons), SF 6 (12 elec-
trons) and IF, (14 electrons). Since elements in period 2 have only 2s and
2p orbitals, the maximum covalency they can exhibit is 4.

Lewis also suggested that there was a definite tendency for electrons in
a molecule to form pairs. This rule of 2, as we have seen, became established
by the developments of quantum mechanics. There are few molecules
that contain an odd number of valencv electrons : where such odd electron

12 ORGANIC CHEMISTRY

molecules do exist, unusual properties are found to be associated with them
(see free radicals).

There are three general extreme types of chemical bonds: electro valent,
covalent and metallic bonds. In addition to these extreme types, there are
also bonds of intermediate types.

1. Electrovalency is manifested by the transfer of electrons, and gives rise to
the ionic bond. Consider sodium chloride. Sodium is (is) 2 (2s) 2 (2^>)*(3s) :
chlorine is (is) 2 (2s) 2 (2^>) 6 (3s) 2 (3^>) 5 . Sodium has completed K and L shells,
and is starting the M shell with one electron. This electron (the 3s electron)
is the valency electron of sodium. Chlorine has completed K and L shells,
and has seven electrons in the M shell. These M electrons are the valency
electrons of chlorine. If the sodium completely transfers its valency electron
to the chlorine atom, then each atom will have eight electrons in its outer-
most shell, and this, as we have seen, is a stable arrangement. Since both
atoms were originally electrically neutral, the sodium atom, in losing one
electron, will now have a single positive charge, i.e., the neutral atom has
become a positive ion. Similarly, the neutral chlorine atom, in gaining one
electron, has become a negative ion. In the sodium chloride crystal these
ions are held together by electrostatic forces. If the symbol of an element
is used to represent the nucleus of an atom and all the electrons other than
the valency electrons, and dots are used to represent the valency electrons,
then the combination of the sodium and chlorine atoms to form sodium
chloride may be represented as follows :

Na' + :ci-

2. Co valency. This type of bonding involves a sharing of electrons
in pairs, each atom contributing one electron to form a shared pair, each
pair of electrons having their spins antiparallel. This method of completing
an octet (or any of the other possible values) gives rise to the covalent bond.

Hydrogen is usually unicovalent: occasionally it is unielectrovalent, e.g.,
in sodium hydride, hydrogen exists as the hydride anion, formed by accepting
an electron from the sodium :

Na* + H* — > Na + Hi

Carbon almost invariably forms covalent compounds. The electron con-
figuration of carbon is (is) 2 (2s) 2 (2^>) 2 . Since the two 2s electrons are paired,
it would appear that carbon is bivalent, only the two single 2p electrons
being involved in compound formation. As pointed out previously, carbon
is almost always quadrivalent ; thus the 2s and 2,p electrons must be involved,
just how these four electrons readjust themselves to give quadrivalent
carbon will be described later; at this stage we shall assume it done, and

write quadrivalent carbon as • C \

In methane the four hydrogen atoms each contribute one electron and the
carbon atom four electrons towards the formation of four shared pairs:

H

4H' + -c- — > h:c:h

H

Each hydrogen atom has its duplet (as in helium), and the carbon atom has
an octet.

STRUCTURE OF THE ATOM 13

Each pair of shared electrons is equivalent to the ordinary " valency-
bond ", and so electronic formulae are readily transformed into the usual
structural formulae, each bond representing a shared pair, e.g.,

H I O I H or H— O— H; H : C j C : H or H— C^C— H;
HINIHorH— N— H

- i

From these examples it can be seen that there is a very important difference
between an electronic formula and its equivalent structural formula. In
the former, all valency electrons are shown whether they are used to form
covalent bonds or not ; in the latter, only those electrons which are actually
used to form covalent bonds are indicated. This is a limitation of the usual
structural formula. A widely used scheme is to represent structures with
ordinary valency bonds and to indicate lone pairs by pairs of dots (see
below).

3. Co-ordinate valency is a special type of covalency. Its distinguishing
feature is that both of the shared electrons forming the bond are supplied
by only one of the two atoms linked together, e.g., when ammonia combines
with boron trifluoride to form a " molecular compound ", it is the lone pair
of the nitrogen atom that is involved in the formation of the new bond. In
boron trifluoride, the boron has only six electrons in its valency shell; hence
it can accommodate two more to complete its octet. Thus, if the nitrogen
atom uses its lone pair, the combination of ammonia with boron trifluoride
may be shown as :

h :f: h:f":

h:n: + b:f: — ► h:n:b:f:
h :f:" h:f:"

In the usual structural formula notation, a co-ordinate bond may be repre-
sented by an arrow pointing away from the atom supplying the lone pair
(Sidgwick, 1927) ; thus :

H F

1 1

H F
-> H— N->B— F

u

h— n: + B— F —

1 1

H F

The atom that supplies the lone pair is known as the donor, and the atom
that receives a share is the acceptor. Since it is one atom that donates the
lone pair, the co-ordinate link is also known as the dative link (Sidgwick,
1927).

Before combination, both donor and acceptor are electrically neutral:
after combination, the donor has lost a share in the lone pair, and the
acceptor has gained a share. Therefore the donor acquires a positive
charge and the acceptor a negative charge, and the presence of these charges

may be indicated by writing the formula H 3 N — BF 3 .

Hence we have a covalent bond holding together two charged portions,
and because of this the co-ordinate link is also known as the semi-polar or
semi-ionic bond (Noyes, 1933). The co-ordinate link has also been named
the " mixed double bond " (Lowry, 1923), and the " semi-polar double bond "
(Sugden, 1925).

14 ORGANIC CHEMISTRY

Once the co-ordinate bond has been formed, there may be no way of
distinguishing it from any other covalent bond, but since one atom has
supplied the pair of shared electrons, charges are produced in the molecule.
When a covalent bond is formed, charges may also be produced in the
molecule, giving rise to a dipole (q.v.). Hence the co-ordinate bond is
effectively a covalent bond. The extent of the charge on each atom in a
dative (or covalent) bond may be found as follows. Add the number of
unshared electrons to one half of the shared electrons, and compare the
result with the number of valency electrons of the neutral atom, e.g.,
(i) methane, CH 4 . Here there are 8 shared electrons; |x8 = 4 = number
of electrons in the neutral carbon atom; therefore methane is uncharged,
(ii) H 3 N-> BF 3 . For the nitrogen atom we have |x 8 = 4; but since
the neutral nitrogen atom has 5 valency electrons, in the compound
H 3 N->BF 3 the nitrogen has a charge of +1. For boron we have
Jx8=4; but since the neutral boron atom has 3 electrons, in this mole-
cular compound the boron has a charge of — 1.

Electrovalent compounds are good electrical conductors in the fused
state or in solution. They are generally non-volatile, and are usually
insoluble in hydrocarbons and allied solvents. Covalent compounds are
non-electrical conductors, are generally volatile, and are usually soluble in
hydrocarbons and allied solvents. Since the covalent bond is directional,
stereoisomerism (space-isomerism) is possible (see p. 399). Co-ordinated
compounds behave very much like covalent compounds, but they are usually
less volatile than purely covalent compounds.

CHELATE COMPOUNDS

In the co-ordinated compounds discussed above, one donor atom has
shared its lone pair with one acceptor atom. It is possible, however, for
an acceptor atom to receive a number of shares in lone pairs, e.g., cobalt-
ammine chloride, [Co(NH 3 ) 6 ] 3+ 3Cl~. In this complex, the cobalt atom
receives shares from six lone pairs, each ammonia molecule donating its
nitrogen lone pair (I). Now let us consider ethylenediamine as the donating

molecule. Its structure is NH, — CH, — CH 2 — NH 9 . In this molecule there

(I)

NH 3 ■
H 3 N^i,/NH 3

H 3 N^ + XNH 3
NH 3 .

3C1-

Co

/NH 2 -i

CH.

NH,-CH,

(")

are two lone pairs, and it has been found that each nitrogen atom can act
independently as a donor. Thus one ethylenediamine molecule can occupy
two positions in the complex, producing the cation (II). This complex will,
therefore, contain three rings. Compounds such as this are khown as
chelate compounds, chelation taking place when the donating molecule shares
two lone pairs on different atoms within the molecule with a single acceptor
atom, thereby producing a ring. Chelation may also take place intra-
molecularly , i.e., between two atoms in the same molecule; but in these
cases the chelate rings are formed, not by co-ordinate bonds, but by hydrogen
bonds (see later).

DIPOLE MOMENTS

When a covalent bond is formed between two identical atoms, e.g., H — H,
CI — CI, etc., the two electrons forming the covalent bond may be regarded
as being symmetrically disposed between the two atoms. The centres of

STRUCTURE OF THE ATOM 15

gravity of the electrons and nuclei therefore coincide. With two dissimilar
atoms the two electrons are no longer symmetrically disposed, because each
atom has a different electron-affinity {electronegativity), i.e., attraction for
electrons. Chlorine has a much greater electron-affinity than hydrogen;
so that when chlorine and hydrogen combine to form covalent hydrogen
chloride, the electrons forming the covalent bond are displaced towards the
chlorine atom without any separation of the nuclei:

H- + :ci — >- h :ci: or h— cT

• • • •

The hydrogen atom will, therefore, be slightly positively charged, and the
chlorine atom slightly negatively charged. Thus, owing to the greater
electron-attracting power of the chlorine atom, the covalent bond in hydrogen
chloride is characterised by the separation of small charges in the bond. A
covalent bond such as this, in which one atom has a larger share of the
electron-pair, is said to possess partial ionic character.

In analogy with a magnet, such a molecule is called a dipole, and the
product of the electronic charge, e, and the distance d, between the charges
(positive and negative centres) is called the dipole moment, y.; i.e., \l = e X d.
e is of the order of io -10 e.s.u. ; d, io~ 8 cm. Therefore (i is of the order io -18
e.s.u., and this unit is known as the Debye (D), in honour of Debye, who did
a large amount of work on dipole moments.

The dipole moment is a vector quantity, and its direction is often indicated
by an arrow parallel to the line joining the points of charge, and pointing

towards the negative end, e.g., H — CI. The greater the value of the dipole
moment, the greater is the polarity of the bond. The terms polar and non-
polar are used to describe bonds, molecules and groups, and the reader is
advised to make sure he appreciates how the terms are applied in each case
under consideration.

The following points are useful in organic chemistry:

(i) In the bond H — X, where X is any atom other than hydrogen or carbon,

the hydrogen atom is the positive end of the dipole, i.e., H — X.

(ii) In the bond C — X, where X is any atom other than carbon, the carbon

atom is the positive end of the dipole, i.e., C — X. Earlier work appeared to show

that in saturated compounds of carbon, the dipole for the C — H bond was C — H.
Work by Coulson (1942), however, indicates that the dipole is in the opposite

direction, i.e., C — H, and that in methane the value is o-3oD. Both the direc-
tion and value, however, are not constant, but depend on the nature of the
hybridisation (see p. 89).

(iii) When a molecule contains two or more polar bonds, the resultant dipole
moment of the molecule is obtained by the vectorial addition of the constituent
bond dipole moments (see also pp. 427, 513).

(iv) A symmetrical molecule is non-polar, although it may contain polar bonds.

ELECTRON DISPLACEMENTS IN A MOLECULE

i. Inductive effect. Consider a carbon chain in which one terminal
carbon atom is joined to a chlorine atom : — C 3 — C 2 — C x — CI. Chlorine has a
greater electron-affinity than carbon; therefore the electron pair forming
the covalent bond between the chlorine atom and C x will be displaced
towards the chlorine atom. This causes the chlorine atom to acquire a
small negative charge, and C 2 a small positive charge. Since Cj is positively
charged, it will attract towards itself the electron pair forming the covalent

l6 ORGANIC CHEMISTRY

bond between C 2 and C 2 . This will cause C 2 to acquire a small positive
charge, but the charge will be smaller than that on C t because the effect
of the chlorine atom has been transmitted through C t to C 2 . Similarly, C s
acquires a positive charge which will be smaller than that on C 2 . This
type of electron displacement along a chain is known as the inductive effect;
it is permanent, and decreases rapidly as the distance from the source in-
creases. From the practical point of view, it may be ignored after the second
carbon atom. It is important to note that the electron pairs, although per-
manently displaced, remain in the same valency shells.

This inductive effect is sometimes referred to as a transmission effect, since it
takes place by a displacement of the intervening electrons in the molecule.
There is also another effect possible, the direct or field effect, which results from
the electrostatic interaction across space or through a solvent of two charged
centres in the same molecule, i.e., the direct effect takes place independently of
the electronic system in the molecule (Ingold, 1934). Apparently it has not been
possible to separate these two modes of inductive effect in practice.

The inductive effect may be represented in several ways. The following
will be adopted in this book : — C->-C->— C->— CI.

Inductive effects may be due to atoms or groups, and the following is the
order of decreasing inductive effects :

NO a , F, CI, Br, I, OCH3, C 6 H 5 , H, CH 3 , C 2 H 5 , (CH 3 ) 2 CH, (CH 3 ) 3 C

For measurement of relative inductive effects, hydrogen is chosen as
reference in the molecule CR 3 — H as standard. If, when the H atom in this
molecule is replaced by Z (an atom or group), the electron density in the
CR 3 part of the molecule is less in this part than in CR 3 — H, then Z is said
to have a —I effect (electron-attracting or electron-withdrawing). If the
electron density in the CR 3 part is greater than in CR 3 — H, then Z is said to
have a -f-I effect (electron-repelling or electron-releasing) e.g., Br is —I;
C 2 H 6 is +1. This terminology is due to Ingold (1926) ; Robinson suggests
the opposite signs for I, i.e., Br is +1; C 2 H 5 , —I. Ingold's terminology will
be used in this book.

2. Electromeric effect. This is a temporary effect involving the complete
transfer of a shared pair of electrons to one or other atom joined by a
multiple bond, i.e., a double or triple bond. The electromeric effect is brought
into play only at the requirements of the attacking reagent, and takes

place almost instantaneously. Consider the following: C=0 or CIO.
At the moment of reaction the oxygen atom takes complete control of one
of the shared electron pairs, the electronic structure becoming C '. O '.. Since
the carbon has lost its share in the electron pair, and the oxygen gained a
share, the carbon acquires a positive charge and the oxygen a negative one.
Removal of the attacking reagent causes the charged molecule to revert
to its original electronic condition. It should be noted that the original
condition of the molecule will have small charges on both the carbon and
oxygen atoms (positive and negative, respectively), due to the inductive
effect of the oxygen, which is more strongly electron-attracting than carbon.
Another effect may also operate to give each atom a small charge (see
resonance).

The electromeric effect is represented as follows:

The curved arrow shows the displacement of the shared electron pair,
beginning at the position where the pair was originally, and ending where

STRUCTURE OF THE ATOM 17

the pair has migrated. It should be noted that the electromeric effect
might have taken place :

vH _ +

c=o — > c— o

However, this is most unlikely, since oxygen is strongly electron-attracting,
and therefore " assists " the displacement towards itself, and " opposes "
the displacement away from itself. This is an example of the electromeric
and inductive effects aiding each other. It is possible, however, for them
to oppose each other, and when they do so, the electromeric effect generally
overcomes the inductive effect, but this happens only when the chain has
conjugated double bonds (see e.g., benzene).

The electromeric effect is represented by the symbol E, and is said to
be +E when the displacement is away from the atom or group, and — E
when towards the atom or group (cf. the I effect).

The displacement of the electron pair forming a covalent bond when a unit
charge is brought up is a measure of the polarisability of that bond. It is not a
permanent polarisation since, when the charge is removed, the electron displace-
ment disappears.

3. Mesomerism or Resonance. The theory of mesomerism was developed
on chemical grounds. It was found that no structural formula could satis-
factorily explain all the properties of certain compounds, e.g. , benzene. This
led to the idea that such compounds exist in a state which is some combina-
tion of two or more electronic structures, all of which seem equally capable
of describing most of the properties of the compound, but none of describing
all the properties. Ingold (1933) called this phenomenon mesomerism
(" between the parts ", i.e., an intermediate structure). Heisenberg (1926),
from quantum mechanics, supplied a theoretical background for meso-
merism; he called it resonance, and this is the name which is widely used.

The chief conditions for resonance are:

(i) The positions of the nuclei in each structure must be the same or nearly
the same.

(ii) The number of unpaired electrons in each structure must be the same.

(in) Each structure must have about the same internal energy, i.e., the
various structures have approximately the same stability.

Let us consider carbon dioxide as an example. The electronic structure
of carbon dioxide may be represented by at least three possible electronic
arrangements which satisfy the above conditions :

bjcjo :o:clo: :6lc:o:

(I) (II) (III)

Structures (II) and (III) are identical as a whole, since both oxygen atoms
are the same.* Each structure, however, shows a given oxygen atom to be
in a different state, e.g., the oxygen atom on the left in (II) is negative,
whereas in (III) it is positive. Although two (or more) of the electronic
structures may be the same when the molecule is considered as a whole,
each one must be treated as a separate individual which makes its own
contribution to the resonance state. Structures (I), (II) and (III) are called
the resonating, unperturbed or canonical structures of carbon dioxide, and
carbon dioxide is said to be a resonance hybrid of these structures, or in the
mesomeric state.

* If the two oxygen atoms are not the same but one is isotope "O and the other
isotope "O, then clearly structures (II) and (III) are different.

l8 ORGANIC CHEMISTRY

It is hoped that the following crude analogy will help the reader to grasp
the concept of resonance. Most readers will be familiar with the rotating
disc experiment that shows the composite nature of white light. When
stationary, the disc is seen to be coloured with the seven colours of the
rainbow. When rotating quickly, the disc appears to be white. The
resonating structures of a resonance hybrid may be compared to the seven
colours, and the actual state of the resonance hybrid to the " white "; i.e.,
the resonating structures may be regarded as superimposed on one another,
the final result being one kind of molecule. In a resonance hybrid all the
molecules are the same ; a resonance hybrid cannot be expressed by any single
structure.

In a resonance hybrid the molecules have, to some extent, the properties
of each resonating structure. The greater the contribution of any one
structure, the more closely does the actual state approach to that structure.
At the same time, however, a number of properties differ from those of any
one structure. The observed heat of formation of carbon dioxide is greater
than the calculated value by 31-6 kg. cal. In other words, carbon dioxide
requires 31-6 kg. cal. more energy than expected to break it up into its
elements, i.e., carbon dioxide is more stable than anticipated on the structure
0=0=0. How can this be explained? Arguments based on quantum
mechanics show that a resonance hybrid would be more stable than any
single resonating structure, i.e., the internal energy of a resonance hybrid
is less than that calculated for any one of the resonating structures. The
difference between the heat of formation of the actual compound, i.e., the
observed value, and that of the resonating structure which has the lowest
internal energy (obtained by calculation) is called the resonance energy.
Thus the value of the resonance energy of any resonance hybrid is not an
absolute value; it is a relative value, the resonating structure containing the
least internal energy being chosen as the arbitrary standard for the resonance
hybrid. The greater the resonance energy, the greater is the stabilisation
due to resonance. The resonance energy is a maximum when the resonating
structures have equal energy content, and the more resonating structures
there are, the greater is the resonance energy. When the resonating struc-
tures are identical in energy content, and consequently the resonance energy
is a maximum, the compound is said to be completely degenerate.

The resonance energy of a molecule is a property of the molecule in the
ground state. Most measurements of resonance energies have been obtained
from heats of combustion, but a few measurements have also been obtained
from heats of hydrogenation ; the latter method is more accurate than the
former. The heat of combustion of the most stable classical structure (i.e.,
the resonating structure with the lowest internal energy) is, as mentioned
above, obtained by calculation. This presupposes that accurate values for
bond-energies are known. If these values are not accurate, then one cannot
expect to obtain accurate resonance energies. In practice, several different
sets of bond-energies have been proposed, one set using bond-energies in
isolation (i.e., the bond between two atoms is assumed to be independent of
other atoms in the molecule), and another set using bond-energies which
depend on the environment (i.e., takes into account other atoms in the mole-
cule). Thus different resonance energies are usually obtained for any given
molecule, and so, when these values are small (i.e., the resonance energy is
small), the value may be almost, if not completely, due to the use of in-
accurate bond-energies and not actually due to stabilisation of the molecule
by resonance.

Another property of the resonance hybrid which differs from that of any
of the resonating structures is that of the bond length, i.e., the distance
between atoms joined by a covalent bond. The normal length of the car-

STRUCTURE OF THE ATOM 19

bonyl double bond (C=0) in ketones is about 1-22 A; the value found in
carbon dioxide is 1-15 A. For a given pair of atoms, the length of a single
bond is greater than that of a double bond, which, in turn, is greater than
that of a triple bond. Resonance, therefore, accounts for the carbonyl bond
in carbon dioxide not being single, double or triple (see also butadiene,
p. 87, and benzene, p. 507).

In a resonance hybrid, the electronic arrangement and bond lengths will
be different from those of the resonating structures. Consequently the
observed dipole moment may differ from that calculated for any one
structure.

As we have seen above, in a resonance hybrid all the molecules have the
same structure. A difficulty that arises with the resonance theory is the
representation of a resonance hybrid. The molecules corresponding to the
structures chosen as the resonating structures do not necessarily have an
actual existence. Thus, if these resonating structures are fictitious, what
fictitious structures are we to choose? The normal way of solving this
problem is first to ascertain the structure of the molecule by the usual
methods, and then describe it by means of the classical valency-bond
formula. Let us consider again the case of carbon dioxide. The classical
structure is (I) (see above). As we have also seen, it has been found that not
all the properties of carbon dioxide are described by this classical formula.
Thus the classical structure is an approximation, and it is in this sense that
classical structures are fictitious. By postulating other electronic structures
(II) and (III), wave functions can then also be obtained for these fictitious
structures. By a linear combination of all three functions, a " structure "
is obtained which describes the properties of carbon dioxide. This " struc-
ture " is called the resonance hybrid of the classical (I) and the two postu-
lated electronic structures (II) and (III).

It is very important to note here that wave-mechanics offers a theoretical
method of studying the electron distribution in a molecule, but starts with a
knowledge of the relative positions of all the nuclei concerned, i.e., with the
" classical structure ". Theoretically, it is possible to start from a mole-
cular formula, and then solve the structure. The number of possibilities
and mathematical difficulties, however, are far too great at present, and so it
seems that the classical chemist, who arrives at the classical structures by
classical methods, will still be " in business " for a long time to come. Since,
however, by means of wave-mechanics one can calculate the density of
electronic charge at all points in a molecule (of known classical structure),
it is possible from this information to deduce charge distributions, bond
lengths and bond angles, and consequently the size and shape of a mole-
cule.

The question that now arises is : Starting with the classical structure, what
other electronic structures are we justified in postulating ? A very important
point in this connection is that resonance can occur only when all the atoms
involved lie in the same plane {or nearly in the same plane). Thus any change
in structure which prevents planarity will diminish or inhibit resonance.
This phenomenon is known as steric inhibition of resonance (p. 688).

In practice, then, the conditions described above must be considered when
choosing canonical structures. At the same time, the following observations
will be a useful guide :

(i) Elements of the first two rows never violate the octet rule (hydrogen,
of course, can never have more than a duplet).

(ii) The more stable a structure, then generally the larger will be its con-
tribution to the resonance state. The stability of a molecule can be found
from its bond energies. The bond energy is the amount of energy required

20 ORGANIC CHEMISTRY

to dissociate a compound, say AB, in the gaseous state, into the neutral
atoms A- and B>. Generally, the structure with the largest number of bonds
is the most stable.

(iii) If the different resonating structures have the same number of
bonds, but some structures are charged, then the charged molecules will be
less stable than the uncharged. The high energy content of a charged mole-
cule is due to the work put into the molecule to separate the charges, and
the greater the distance of charge separation, the less stable is that structure.
Even so, charged structures may make a considerable contribution to the
resonance state, since resonance among a number of charged structures
gives a resonance hybrid that is more stable than any one resonating
structure.

The final problem is the method of representing a resonance hybrid.
Various methods have been used, and the one used in this book is that
introduced by Bury (1935). This consists of writing down the resonating
structures with a double-headed arrow between each pair:

o=c=o <->• 6— c=6 -<-^- 6=c— 6

Inductive and resonance (mesomeric) effects are permanently operating
in the " real " molecule; collectively they are known as the polarisation
effects. On the other hand, there are also two temporary (time-variable)
effects, the electromeric effect and the inductomeric effect (which operates by
an inductive mechanism) . Both of these are brought into play by the attack-
ing reagent, and collectively they are known as the polarisability effects.
Remick (1943) has suggested the use of subscripts s and d to represent the
static (permanent) and dynamic (time-variable) effects. Thus the inductive
effect may be represented by the symbol I„ and the inductomeric effect by
I„. Since polarisability effects aTe brought into play only by the approach
of the attacking reagent, they will therefore aid and never inhibit a reaction.

Strictly speaking, the term resonance effect (R) is not the same as the
mesomeric effect (M). The mesomeric effect is a permanent polarisation, and
the mechanism of electron transfer is the same as that in the electromeric
effect, i.e., the mesomeric effect is a permanent displacement of electron pairs
which occurs in a system of the type Z — C=C; e.g., Z=R 2 N, CI:

R,N— C=C— ; XI— 0=C—

Thus the essential requirement for mesomerism is the presence of a multiple
bond in the molecule. On the other hand, the resonance effect embraces all
permanent electron displacements in the molecule in the ground state, e.g.,
the hydrogen chloride molecule is a resonance hybrid of two resonating
structures:

H— CI ^->- H+Cl-

Since there is no multiple bond in this molecule, the mesomeric effect is not
possible.

When the electronic displacement is away from the group the mesomeric
(resonance) effect is said to be +M (+R), and when towards the group,
-M(-R).

The mesomeric effect is particularly important in conjugated systems
(p. 84), and the combined mesomeric and electromeric effects are known as
the conjugative effect. This term is also used in the same sense as the reson-
ance effect. Also, since this combined effect was first recognised in connec-
tion with tautomerism, it has also been called the tautomeric effect (±T).

STRUCTURE OF THE ATOM 21

The possible polar influences of groups are shown in the following table:

Electronic mechanism

Polarisation effect
(permanent)

Polarisability effect
(temporary)

Inductive (±1)

Inductive (I or I,)

Inductomeric (I*)

Conjugative (Tautomeric,
±T; or Resonance, ±R)

Mesomeric (M) or Reson-
ance (R)

Electromeric (E)

Fields in which operative

Physical properties.
Reaction equilibria.
Reaction "rates.

Reaction rates only

As has been pointed out above, resonance describes all permanent electron
displacements in the molecule in the ground state. It therefore follows that
the I-effect can be described as due to resonance. Thus resonance is the
combination of I- and M-effects. It is more convenient, however, from the
point of view of the organic chemist, to consider a molecule with respect to
its I-effect and " resonance " (mesomeric) effect separately. Hence from
this point of view, resonance is the additional permanent electronic displace-
ments to the I-effect, and it is customary to ignore the latter effect when
discussing " resonance ". In other words, resonance, in this context, is
concerned only with the part of the molecule containing multiple bonds and
is therefore, strictly speaking, n-eledron resonance. This is the sense in
which the term resonance will be used in this book.

Effect of structure on reactivity. The type of reaction of an organic com-
pound is largely dependent on the nature of the functional group present
(p. 48). It has been found that various structural changes, e.g., the intro-
duction of a given group into different positions in a molecule containing a
given functional group, usually affect the rate of a given type of reaction
and also the equilibrium position, and may even change the type of mechan-
ism of the reaction. Much work has been done to try to correlate structure
and reactivity (i.e., rate of reaction), and as an outcome of this work, it
appears that some sort of quantitative correlation can be made on the basis
of consideration of independent contributions of inductive, resonance
(mesomeric) and steric effects. When each effect has been assessed, then
all three may be combined, and in this way there is obtained a relationship
between structure and reactivity. In the text are discussed many cases of
the effects on reaction rates and mechanism by polar (I and R) and steric
effects.

Reactions in organic chemistry may be classified into the following main
types: (i) substitution; (ii) replacement or displacement; (iii) addition;
(iv) elimination; (v) isomerisation (rearrangement).

THE HYDROGEN BOND OR HYDROGEN BRIDGE

Compounds containing OH or NH groups often exhibit unexpected
properties such as relatively high boiling points, and it was soon felt necessary
to assume that the elements oxygen or nitrogen were linked by means of
hydrogen, thereby producing the hydrogen bond. Detailed study has shown
that the unexpected properties were exhibited only when the atoms partici-
pating in the bond had high electron-affinity — fluorine, oxygen and nitrogen
(decreasing in this order), and to a less extent, chlorine and sulphur. Thus
the hydrogen bond explained, for example, the existence of the HF a _ ion, the
association of hydroxylic compounds such as water, alcohols, etc., and the
association of ammonia.

22 ORGANIC CHEMISTRY

The exact nature of the hydrogen bond has been the subject of much dis-
cussion. Possibly a number of factors contribute, but it appears that the
most important one is electrostatic. In bond Z — H, if Z has high electron-
affinity, there will be a relatively large amount of polarity, i.e., the state of

8- S+-

affairs will be Z — H, where 8 -f- is relatively large. Since the hydrogen atom

8+

has a tiny volume, the H will exert a large electrostatic force and so can
attract atoms with a relatively large 8- — charge, providing these atoms have
a small atomic radius. Fluorine, oxygen and nitrogen are of this character.
If the atom has a greater radius the electrostatic forces are weaker; thus
chlorine, although it has about the same electron-affinity as nitrogen, forms
very weak hydrogen bonds since its atomic radius is greater.

This theory of electrostatic union has much to support it. The hydrogen
bond is very weak, and has more in common with the " van der Waals
forces " (which are electrical in nature) than with anything else. Values

obtained for the energy of the hydrogen bond are H — F H, io k.cal./mole;

H— O H, 7; H— N H, 2.

Hydrogen bond formation intramolecularly , i.e., involving one molecule
only, gives rise to ring formation or chelation, and this usually when the
formation of a 5-, 6- or 7-membered ring is possible. Hydrogen bonding
intermolecularly , i.e., between two or more molecules, gives rise to association.
Many examples of hydrogen bonding will be found in the text, and this is
represented by a dotted line between the hydrogen and other atom involved
(as shown above).

Hydrogen bonding affects all physico-chemical properties such as m.p.,
b.p., solubility, spectra (infra-red and Raman shifts), etc., e.g., association
produces a higher boiling point than expected (e.g., from the molecular
weight of the compound). On the other hand, chelation usually produces
a lower boiling point than expected, e.g., a nitro-compound usually has a
higher boiling point than its parent compound, but if chelation is possible in
the nitro-compound, the boiling point is lowered (see, e.g., nitrophenols,
p. 626).

ATOMIC AND MOLECULAR ORBITALS

So far, we have discussed the structure of molecules in ^terms of valency
bonds. There is an alternative method of investigating the structure of
molecules, and to appreciate this approach — and to extend the other — it is
necessary to consider the structure of matter from the point of view of
wave-mechanics. Classical physics (i.e., the laws of mechanics, etc.) is
satisfactory when dealing with large masses. These laws are approxima-
tions, but deviations become significant only when dealing with very small
particles such as electrons and nuclei. The behaviour of these small par-
ticles, however, may be satisfactorily studied by wave (quantum) mechanics.
This uses the idea of the particle-wave duality of matter. It has already
been pointed out that the electron may be regarded as a tiny mass carrying
a negative charge. In 1923, de Broglie proposed that every moving particle
has wave properties associated with it. This was first experimentally
verified in the case of the electron (Davisson and Germer, 1927; G. P.
Thomson, 1928). Thus an electron has a dual nature, particle and wave,
but it behaves as one or the other according to the nature of the experiment ;
it cannot at the same time behave as both. According to wave-mechanics, a
moving particle is represented by a wave function 1/1 such that ip 2 dv is the
probability of finding the particle in the element of volume dv. The greater
the value of i/j 2 , the greater is the probability of finding the electron in that
volume dv. Theoretically, ip has a finite value at a large distance (compared

STRUCTURE OF THE ATOM

23

with atomic dimensions) from the nucleus, but in practice there is very little
probability of finding the electron beyond a distance of 2-3A. It is therefore
possible to map out regions or contours within which the probability of
finding the electron is high, and outside which there is very little likelihood
of finding the electron.

An alternative interpretation of the wave function tp is to regard the elec-
tron as a cloud (charge-cloud), the density of the cloud at any point being
proportional to tfi 2 . Hence once again it is possible to draw contours within
which almost all the electron charge is to be found. These regions (of
probability or of density of charge-cloud) are known as atomic orbitals
(A.O.s), and have characteristic shapes.

In 1926, Schrodinger developed the wave-equation, which connected the
wave function ip of an electron with its energy, E. This equation has an
infinite number of solutions, but very few of these solutions describe the
known behaviour of electrons. Thus only certain values for E are permis-
sible, since certain conditions must be satisfied. The permitted solutions
for ifi are called the eigenf unctions, and the corresponding values of E are
called the eigenvalues. A number of eigenfunctions exist, the simplest being
those which possess spherical symmetry (ip s function), and the next simplest
being those which possess an axis of symmetry (if/ p function) . The eigenvalues
(i.e., the energy values) for the if/ s functions are not, in general, the same as
those for the ^ functions. Since i/i s dv measures the probability of finding
the electron in the element of volume dv, we can therefore picture " proba-
bility regions " which will be generated by the expressions tp s 2 and ifi p *. Such
regions are those we have called atomic orbitals, and we can speak of the
energy of an A.O. if we mean the eigenvalue (energy value) corresponding to
that wave function tjj.

In addition to its wave function if>, an electron also has spin. Two electrons
can have the same wave function, i.e., can occupy the same orbital provided
their spins are opposite (Pauli exclusion principle). In this case the electrons
are said to be paired.

Fig. 2.i.

The various A.O.s are classified as s, p,
and p orbitals need concern us, and Fig.

d.f,

orbitals. Only the s

I „„, „„^ _ b . 1* shows their shapes. The s-

orbital is spherically symmetrical, which means that the region within
which it is reasonable to expect to find the electron is a sphere having the
nucleus as centre. The ^-orbitals are dumb-bell in shape, and the two
halves are separated by a nodal plane, over which the value of iff is zero,
i.e., there is no likelihood of finding the electron in this plane. In these
^-orbitals, the electron is confined to regions which have a marked directional
character, each orbital having an axis at right angles to those of the other

* Each diagram has been given two numbers, the first indicating the chapter, and the
second the order in that chapter. Reference to a diagram in its own chapter will be
indicated by the second number only.

24 ORGANIC CHEMISTRY

two, and hence they are known as the p x , p v , p z orbitals, respectively. These
orbitals are entirely equivalent except for their directional property.

The order of orbital energies is is < 2s < 2p < 3s < 3p < . . . . Since
an electron must occupy some particular orbital, when the electron " jumps "
from that orbital to another, it acquires the energy of the new orbital,
absorbing or emitting the difference in a " discrete energy packet " or
quantum. Not all transitions between different energy levels are allow-
able; a definite rule of selection exists, e.g., permitted transitions are
s — > p,p — >s or d, etc.; s — > s is not permitted. When an electron
absorbs a quantum of energy, it is driven into an (allowable) orbital of
higher energy. The atom is then said to be " excited " , and is more reactive.
On returning to its normal orbital, the electron emits the quantum of energy
at a definite wave-length, giving rise to a particular line in the emission
spectrum (see p. 773). When all the electrons in an atom are in their
normal orbitals, i.e., orbitals of lowest energy, the atom is said to be in the
" ground " state.

So far, we have dealt only with atoms, i.e., with electrons associated with
one nucleus. The wave-equations for molecules cannot be solved without
making some approximations. Two types of approximations have been
made, one set giving rise to the valence-bond method (V.B.) ; and the other
set to the molecular orbital method (M.O.). The V.B. method — due mainly
to the work of Heitler, London, Slater and Pauling — considers the molecule
as being made up of atoms with electrons in atomic orbitals on each atom.
Thus a molecule is treated as if it were composed of atoms which, to some
extent, retain their individual character when linked to other atoms. The
M.O. method — due mainly to the work of Hund, Lennard- Jones and Mulliken
— treats a molecule in the same way as an atom, except that in the molecule
an electron moves in the field of more than one nucleus, i.e., molecular orbitals
are pplycentric. Thus each electron in a molecule is described by a certain
wave function, the molecular orbital, for which contours can be drawn as
for A.O.s, but differing in that the former are polycentric and the latter
monocentric. In general, the greater the freedom {i.e., the larger the region
for movement) allowed to an electron, the lower will be its energy. Thus
atoms combine to form a molecule because, owing to the overlap of the A.O.s
when the atoms are brought together, the electrons acquire a greater freedom,
and the energy of the system is lowered below that of the separate atoms.
Energy would therefore have to be supplied to separate the atoms in the
molecule, and the greater the amount of energy necessary, the stronger are
the bonds formed between the various atoms.

Let us now consider the case of the hydrogen molecule. A hydrogen
atom has one is electron. When the bond is formed between two hydrogen
atoms to form the hydrogen molecule, these two is electrons become paired
to form molecular electrons, i.e., both occupy the same M.O., a state of
affairs which is possible provided their spins are antiparallel. A very
important principle for obtaining the M.O. is that the bond energy is greatest
when the component A.O.s overlap one another as much as possible. To
get the maximum amount of overlap of orbitals, the orbitals should be in the
same plane. Thus the M.O. is considered as being a linear combination of
atomic orbitals with maximum overlap (L.C.A.O.). Furthermore, according
to L.C.A.O. theory, the binding energy is greater the more nearly equal are
the energies of the component A.O.s. If these energies differ very much,
then there will be no significant combination between the two atoms con-
cerned.

Since the hydrogen molecule is composed of two identical atoms, the
probability of finding both electrons simultaneously near the same nucleus is
very small. Hence one might expect the M.O. to be symmetrical with respect

STRUCTURE OF THE ATOM 25

to the two hydrogen nuclei, i.e., the M.O. in the hydrogen molecule will be
" plum-shaped " (Fig. 2).

Fig. 2.2.

Although the probability of finding the two electrons simultaneously
near the same nucleus is very small, nevertheless this probability exists, and
gives rise to the two ionic structures H + H~ and H~H + . Thus the hydrogen
molecule will be a resonance hybrid of three resonating structures, one purely
covalent (i.e., the two electrons are equally shared), and two ionic (i.e.,
the pair of electrons are associated with one nucleus all the time) :

h:h -<-> h :h <-> ii: h

Calculation has shown that the ionic structures contribute very little to the
actual state of the hydrogen molecule, and the bond between the two hydro-
gen atoms is described as a covalent bond with partial ionic character. It
should here be noted that when the single bond is formed between the two
hydrogen atoms, the probability of finding the electrons is greatest in the
region between the two nuclei. It is this concentration of the negatively
charged electrons between the two positive hydrogen nuclei that binds the
nuclei together. Since electrons are negatively charged, they will repel
each other and so tend to keep out of the region between the two nuclei.
On the other hand, since the spins of the two electrons are antiparallel, this
produces attraction between the two electrons, thereby tending to concen-
trate them in the internuclear region. The net result is that the electron
density for -paired electrons is greatest between the two nuclei. Such a bond
is said to be a localised M.Q., and preserves the idea of a bond connecting
the two atoms. This localisation (in a covalent bond) gives rise to the
properties of bond lengths, dipole moments, polarisabilitv and force con-
stants.

HYBRIDISATION OF BOND ORBITALS

The electron configuration of carbon is (is) 2 (2s) 2 (2p x , 2p tf ). It therefore
appears to be bivalent. To be quadrivalent, the (2s) 2 and the (2p x , 2p y )
electrons must be involved. One way is to uncouple the paired 25 electrons,
and then promote one of them to the empty zp, orbital. Should this be done,
four valencies would be obtained, since each of the electrons could now be
paired with an electron of another atom. The resulting bonds, however,
would not all be equivalent, since we would now have the component
A.O.s 2s, 2p x , 2p s , 2p s . All work on saturated carbon compounds indicates
that the four valencies of carbon are equivalent (but see below). In order
to get four equivalent valencies, the four '* pure " A.O.s must be " mixed "
or hybridised. It is possible, however, to hybridise these four " pure "
A.O.s in a number of ways to give four valencies which may, or may not, be
equivalent. Three methods of hybridisation are important: (i) tetrahedral
(sp 3 bond), (ii) trigonal {sp* bond), (hi) digonal (sp bond).

(i) In tetrahedral hybridisation, the (2s) and (zp x , 2p v , 2p z ) electrons are all
hybridised, resulting in four equivalent orbitals arranged tetrahedrally, i.e.,
pointing towards the four corners of a regular tetrahedron (Fig. 3). The
orbitals are greatly concentrated along these four directions (Fig. a shows

26

ORGANIC CHEMISTRY

the shape along one of these directions). Then by linear combination with
the is orbitals of four hydrogen atoms, four equivalent M.O.s are obtained
for methane. Because of the large amount of overlapping between the
hybridised A.O.s of the carbon and the s A.O. of the hydrogen atom, there
will be strong binding between the nuclei. As in the case of the hydrogen
molecule, each M.O. is almost completely confined to the region between
the two nuclei concerned, i.e., in methane are four localised molecular
orbitals. This scheme of localised M.O.s may be satisfactorily applied to
all compounds containing single covalent bonds. Bond orbitals which are
symmetrical about the line joining the two nuclei concerned are known as
o-bonds.

Fig. 2.3.

The above state of affairs holds good only so long as four identical groups
are attached to the carbon atom, e.g., in CH 4 , CC1 4 , C(CH 3 ) 4 , etc. When
the groups are different, e.g., in CHC1 3 , the four bonds are no longer equiva-
lent. The four carbon valencies are now hybridised in a non-equivalent
fashion, pointing towards the four corners of an irregular tetrahedron. In
CHC1 3 , the three CI — C — CI angles are increased from the normal angle of
109° 28' to about 111°, and the three CI — C — H angles decreased to about
108 .

(ii) In trigonal hybridisation, the 2s, 2p x and 2p y orbitals are hybridised,

Fig. 2.4.

resulting in three equivalent coplanar orbitals pointing at angles of 120° in the
aey plane (Fig. 4a). The remaining orbital is the undisturbed o.p^ (Fig. b).
Thus there will be three equivalent valencies in one plane and a fourth
pointing at right angles to this plane (Fig. c). The three coplanar valencies
form o-bonds, and the 2p z valency forms the so-called re-bond. The 2p z
electrons are known as -re-electrons, mobile electrons, or unsaturation elec-
trons when they form the re-bond. The trigonal arrangement occurs in
compounds containing a double bond, which is regarded as being made up

STRUCTURE OF THE ATOM

27

of a strong bond (a-bond) between two trigonal hybrid A.O.s of carbon, and
a weaker bond (7r-bond) due to the relatively small overlap of the two pure
p, orbitals in a plane at right angles to the trigonal hybrids. Fig. 5(a) shows
the plan, and (6) the elevation of ethylene, CH 2 =CH 2 (see also p. 427).

The H C H angle in ethylene has been measured spectroscopically, and
it has been found to be 119 55' (Gallaway et al., 1942). This is in agreement
with the value expected for trigonal hybridisation.

It is the K-electrons which are involved in the electromeric and resonance effects.

When a compound contains two or more double bonds, the resulting

Fig. 2.5.

xy is plane of trigonal
hybridisation

M.O.s depend on the positions of these bonds with respect to one another
(see, e.g., butadiene, p. 88, and benzene, p. 509).

(iii) In digonal hybridisation, only one 2s electron and the 2p x electron
are hybridised, resulting in two equivalent collinear orbitals (Fig. 6); the
2p v and 2,pz electrons remain undisturbed. Thus we get two equivalent
valencies (forming the c-type of bond) pointing in opposite directions along
a straight line, and two other valencies (each forming a 7r-type of bond),
one concentrated along the jy-axis
(the zp v orbital), and the other
along the z-axis (the ip z orbital).
The digonal arrangement occurs in
compounds containing a triple
bond, e.g., acetylene.

When the electrons of any atom
have been placed in hybridised or-
bitals, that atom is said to be in a
" valence state ". The atom on its
own cannot exist in a valence state; Fig. 2.6.

energy is required to promote the

atom to this condition. This energy is obtained through the formation of bonds
which are stronger with the hybridised orbitals than with the '* pure " orbitals,
i.e., more energy is released with the former than with the latter.

In the foregoing account of carbon hybridisation, multiple bonds (double
and triple) have been described in terms of o-bonds and rc-bonds. There is,
however, an alternative approach to this problem of valency theory. The
application of this theory makes use of the Pauli exclusion principle, and also
takes into account the electrostatic repulsion between electrons. These prin-
ciples lead to the conclusion that electrons with the same spin avoid each other.
Thus, for example, Zimmerman et al. (1949) have shown, as a consequence of the
exclusion principle, that the most probable arrangement of the four electrons of
carbon (25, 2p x , -zp y , 2fi z ) is at the corners of a regular tetrahedron provided all
the electrons have the same spin. This tetrahedral configuration, however, is
achieved only if the four electrons are equivalent, i.e., the electrons occupy
orbitals which in carbon atoms are sp 3 hybrid orbitals.

28 ORGANIC CHEMISTRY

As we have seen, in saturated carbon compounds the four sp 3 orbitals point
towards the corners of a tetrahedron. In ethylene we have used sp 2 trigonal
hybridisation (one a- and one re-bond) to describe the double bond. It is possible,
however, to still use sp 3 hybridisation to describe ethylene. In this case two
electrons are in one orbital of " banana " shape (" bent " bond), and the other
two electrons in a second " banana " orbital, equivalent to the first but the
mirror image of it. Thus ethylene may be represented (Fig. 7; see also p. 488) :

H \ yv /H PK c^=^) /H

>rt/ Xp' >p p/

H / ^ X H h/^^^H
Fig. 2.7.

It is interesting to note that this " bent " bond method of representing ethylene
is equivalent to Baeyer's description of a double bond (p. 487).

In the same way, the triple bond in acetylene (previously described in terms of
one a- and two re-bonds) may also be regarded as made up of three equivalent
sp 3 hybrids symmetrically disposed round the C — C axis.

Quantum mechanical arguments show that both methods of representing these
multiple bonds are equal to each other, each method having certain advantages.
The a-re bond method is more convenient for describing transitions from one state
into another (e.g., in electronic spectra), whereas the " bent " bond method is
more convenient for describing electron distribution in a molecule.

Now let us consider molecules in which the central atom has lone pairs. Con-
sider nitrogen, with electron configuration (is) a (2s) 2 (2^>) 3 . This has three ip
orbitals, and if each combines with a hydrogen atom, then the molecule of
ammonia is !NH 3 . In this molecule there is a lone pair (2s) a , and the three
hydrogen atoms, bound by the overlap with ip orbitals, will therefore have a
valency angle of 90 . The nitrogen atom is tercovalent, and by virtue of its lone
pair, can act as a donor to become quadrivalent uni-electrovalent, e.g.,

->- NH 4 +C1-

At first sight it might appear that the four hydrogen atoms in the ammonium
ion are not equivalent; three bonds are formed from ip electrons, and one by the
lone 2s 3 . However, the fact is that all four hydrogen atoms in NH 4 + are equiva-
lent, and also the valency angle is about 109-5° (the tetrahedral value). More-
over, the valency angle in ammonia is about 107° and not 90° (the expected value
from ip bonding).

Now consider oxygen (\s) % (is)\ipY. One ip orbital is doubly filled, and so,
in water, one would expect that the two single ip electrons would combine with
hydrogen to form water in which the bond angle HOH is 90°. Actually the bond
angle is about 104-5°. Also, since the water molecule has two lone pairs, it can
act as a donor to form the hydroxonium ion, H s OI + , and it is found that all the
three hydrogen atoms are equivalent. The valency angles are nearly those
expected of tetrahedral configurations, and the difference from 90° is far too
large to be accounted for by repulsion between hydrogen atoms. A satisfactory
explanation is as follows. Sidgwick et al. (1940) assumed that lone pairs of
electrons and bonding pairs were of equal importance, and that they arranged
themselves symmetrically so as to minimise the repulsions between them. Thus
pairs of electrons in a valency shell, whether a bonding pair or a lone pair, are
always arranged in the same way, and this depends only on the total number of
pairs. Thus two pairs are arranged linearly (e.g., HgCl 2 ), three pairs in the form
of an equilateral triangle (e.g., BC1 3 ), four pairs tetrahedrally (e.g., CH 4 ), five
pairs in the form of trigonal bipyramid and six pairs octahedrally. In all cases
it is assumed that all the pairs of electrons occupy hybridised orbitals and only
a-bonds are present (see also p. 335).

Now let us consider ammonia and water from this point of view. In each of
the valency shells of these central atoms there are four pairs of electrons (nitrogen
with one lone pair, and oxygen with two lone pairs) . If these four pairs occupy
four tetrahedrally hybridised orbitals, then the valency angles should be about

STRUCTURE OF THE ATOM 29

I09-5 . As we have seen above, in ammonia the angle is about 107 , and in
water about 104-5 °. The problem then is to account for these deviations from
the anticipated " regular " shapes. This may be explained by assuming electro-
static repulsions between electron pairs in the valency shell (both bonding and
lone pairs) are in the following order:

lone pair — lone pair > lone pair — bond pair > bond pair — bond pair.

This order can be explained on the basis that lone pairs are more concentrated
than bonding pairs (the latter are " stretched " in the formation of covalent
bonds). Thus lone-pair electrons exert greater electrostatic repulsions on other
lone pairs than on bonding pairs. Consequently, when lone pairs occupy the
valency shell, bonding pairs are " forced together ". In CH 4 there are four
bonding pairs, and so the distribution is symmetrical with a valency angle of
109 28'. In NH 3 , there are three bonding pairs and one lone pair, and since
the latter has a bigger repulsive force, the bonding pairs are forced closer to-
gether (the bond angle is about 107 ). It should also be noted that the lone
pair is also in a hybridised orbital, and so when the ammonia molecule is con-
verted into the ammonium ion the four bonding pairs are in the same state of
hybridisation, and consequently all four hydrogen atoms are equivalent.

In the water molecule (bond angle 104-5°), there are two lone pairs, and conse-
quently the bonding pairs are " closed up " more than in ammonia.

Other factors also play a part in deciding the values of the bond angles: the
electronegativity of the central atom and the electronegativities of the attached
atoms.

Let us now consider the hydrogen chloride molecule, H '. CI '.. Here again
the four pairs of electrons of the chlorine atom (one bonding* pair and three
lone pairs) each occupy each sp s hybrid orbital. In this molecule, however,
since chlorine is more strongly electron-attracting than hydrogen, the
electrons are more likely to be simultaneously near the former atom than
the latter. Thus, although the electrons will be found with great probability
in the region between the atoms, i.e., we have a localised M.O., nevertheless
the region near the chlorine atom will tend to be occupied more than that
near the hydrogen atom. In other words, in addition to the covalent
structure H — CI, there will also be a significant contribution of the ionic
structure H + C1~, the contribution of H~C1 + being negligible. Thus we may
say that hydrogen chloride is a resonance hybrid of the two resonating
structures H — CI and H+C1 - . The actual hydrogen chloride molecule will

therefore have a dipole moment H — CI.

In general, when two dissimilar atoms are linked, the contribution of the
two ionic structures, A + B~ and A _ B + , will not be equally important. The
greater the electron-affinity of B with respect to A, the greater will be the
contribution of A+B~ to the actual state of the molecule. The problem is
then to decide what are the weights of the contributions of the resonating
structures. The importance of this problem is readily seen from a con-
sideration of carbon dioxide.

o=c=o O—O^O 6=C— 6

(I) (II) (ill)

Suppose these three structures are described by the wave functions x/i v ^ 2
and tfi 3 . Then the actual molecule will be represented by a wave function
which is a linear combination of the three structures :

The best approximation is given by values for the coefficients a x , a % and a 3
that give the lowest energy to the resonance hybrid, i.e., coefficients a v a 2

30 ORGANIC CHEMISTRY

and a 3 are a measure of the weights of the resonating structures. This may
be done by calculation. On the other hand, in qualitative resonance argu-
ments, it is usual to assume that the weight of a resonating structure is
directly related to its energy content. This, however, appears to be satis-
factory only when two structures are involved, but often leads to erroneous
results when three or more are involved.

Now let us consider the problem of ionic character of a bond, and for this
purpose let us examine the hydrogen chloride molecule. There are two
important structures for this molecule, one purely covalent and the other
purely ionic: H — CI and H + C1~. The wave function of the resonance
hybrid is :

<A = ^covalent + Clonic

When the value of a is such that minimum energy is obtained, then
(a 2 /i -f- a 2 ) X ioo is called the per cent, ionic character of the bond. This
may be calculated from a knowledge of the dipole moment of the bond.
The following values have been found for the hydrogen halides :

HF, 60; HC1, 17; HBr, n; HI, 5.

Thus HF is largely ionic, and HI is mainly covalent.

We have already discussed the problem of the ground and excited states
of an atom. Now let us consider the analogous position of molecules.
Suppose we are dealing with a diatomic molecule in which each atom has supplied
one electron to form the bond. By means of the L.C.A.O. theory, the solution
for the M.O. is found by ultimately solving a quad-
ratic equation derived from the combination of the
two wave functions (A.O.s). Two real roots are
obtained, i.e., two M.O.s of different energy levels .
are possible when two A.O.s are combined. Now J
it is possible for both electrons to occupy either g
M.O., or for one electron to be in the M.O. of the
lower energy level and the other in the higher energy
level. When both electrons occupy the lower M.O.,
the molecule is in the ground state, and when one
or both electrons occupy the higher M.O., the mole- p IG 2 g

cule is in an excited state. Suppose E t and E 2

(where E x < E 2 ) are the energies of the two contributing A.O.s and <f x and <f 2 ,
(where Si < $ 2 ) are the energies of the two resulting M.O.s. On comparing these
energies, it will be found that $ x < E t and (f 2 > E 2 , i.e., one M.O. has lower
energy than the lower of its components (this is the ground state), and the other
M.O. has a higher energy than the higher of its components (this is an excited
state). In general, if n A.O.s are combined, then there are n resultant M.O.s, and
any two consecutive M.O.s embrace one of the contributing A.O.s (see Fig. 8).

It can now be seen that when two atoms combine to form a bond, two
types of bonding are possible. In one M.O., the energy level is lower than
either of its component A.O.s. In this M.O., the electron charge is concentrated
in the region between the two nuclei, resulting in strong bonding between the
atoms. This type of M.O. is called a bonding orbital or a, bond (Fig. ga) . As we
have seen, a bonding orbital is formed by two electrons with their spins anti-
parallel. In the other M.O., the energy level is higher than either of its com-
ponent A.O.s. In this M.O., the charge is pushed away from the region between
the two nuclei, resulting in a nodal plane midway between A and B. In this
condition A and B repel each other, and this M.O. is said to be <m<i-bonding or
a a„ bond (Fig. 96) . In an anti-bonding orbital, the two electrons have the same
spin, and since negatively charged electrons repel each other, and since electrons
with the same spin tend to keep as far away from each other as possible, the
concentration of two such electrons in the internuclear region is reduced to such
a great extent that there is no bonding between the two nuclei; the result is thus
an anti-bond.

-m.0.23
■ a.0.3

-m.o.u
— a.0.2
-m.0.23
-a.o.|
-m.0.11

STRUCTURE OF THE ATOM 31

Any two A.O.s can be combined provided their energies are approximately
the same. In the above example of bonding and anti-bonding orbitals, A and B
were atoms with s electrons. Combination of two ip electrons also gives one
bonding and one anti-bonding orbital (Fig. 10a and 6). In the bonding orbital,
a 7ybond, there is one nodal plane (which contains the molecular axis), but in the
anti-bonding orbital, a Tt^-bond, there are two nodal planes (one containing the

<T e orbital
(a)

°" ^ bital v x orbital

(ft) g (i)

Fig. 2.9. Fig. 2.10.

molecular axis, and the other perpendicular to it). It should here be noted that
as the number of nodes in a bond increases, the energy level rises, and con-
sequently the bond becomes weaker. "When a molecule undergoes transitions
from one energy level to another (with emission or absorption of light), a g state
must go to a u state, or vice-versa. Transitions from one g state to another
g state, and from one u state to another u state, are forbidden.

An important difference between V.B. and M.O. theories is that when dealing
with energy levels of electrons in molecules, in the V.B. method, electrons are
dealt with in pairs, whereas in the M.O. method, electrons can be dealt with
individually (see also p. 777).

THE GENERAL NATURE OF ORGANIC REACTIONS

Much work has been done on reaction mechanisms, i.e., the actual steps
by which a reaction takes place. A chemical equation indicates the initial
and final products of a reaction; rarely does it indicate how the reaction
proceeds. Many reactions take place via intermediates which may or may
not have been isolated. When the products of a reaction are formed by a
single collision of the reactant molecules, i.e., the reaction proceeds without
any intermediates, the reaction is said to be a one-step (or elementary) re-
action. Most reactions, however, are complex, i.e., they occur via a number
of reaction steps. The rate of the overall reaction is controlled by the slowest
step; this is known as the rate-determining step.

There is, in general, no one method that is satisfactory for the deter-
mination of mechanisms of reactions, but the use of a number of methods
may lead to an acceptable answer. It should be borne in mind that mechan-
isms are, in general, theories that have been devised to explain the facts
which have been obtained experimentally. Some of the commoner methods
used for elucidating mechanisms are :

(1) Kinetics. Kinetic studies are concerned with rates of reactions and
provide the most general method for determining reaction mechanisms.

(2) The identification of all the products of a reaction.

(3) The detection, or better still (if it is possible), the isolation of inter-
mediates.

(4) The effect on reaction rates of changing the structure of the reactants.

(5) The effect on reaction rates of changing the solvent.

32 ORGANIC CHEMISTRY

(6) Stereochemical evidence. This type of approach can only be used
when dealing with optically active compounds.

(7) The use of isotopes. This method is particularly useful for tracing
the part played by a particular atom in a reaction.

Applications of these methods are discussed in the text, but before ending
this discussion, there are two other points of interest. One is the principle
of microscopic reversibility. According to this principle, the mechanism of
any reaction, under a given set of conditions, is identical in microscopic
detail to that of the reverse reaction under the same conditions, except that
it proceeds in the opposite way. With this principle it has been possible to
deduce mechanisms where the forward or backward reactions do not lend
themselves to kinetic studies (see, e.g., p. 70).

The other point is that when the various possible products of a reaction
are not interconvertible under the conditions of the reaction, then the pro-
duct formed most rapidly will be the one that predominates in the products.
This most rapidly formed product is known as the kinetically controlled
product. If, however, the possible products of the reaction are inter-
convertible under the reaction conditions, then the most stable product will
predominate in the final products. This most stable product is known as
the thermodynamically controlled product (see, e.g., p. 87).

Now let us examine in more detail what happens when molecules containing
covalent bonds undergo chemical reaction. Consider the reaction

Y + R- X — > Y-R + X

where RX and RY are both covalent molecules. It can be seen that in this
reaction, bond R — X has been broken and the new bond Y — R has been
formed. The mechanism of the reaction depends on the way in which these
bonds are broken. There are three possible ways in which this may occur,
and the result of much work has shown that the actual way in which the
break occurs depends on the nature of R, X and Y, and the experimental
conditions.

(i) Each atom (forming the X— R bond) retains one electron of the
shared pair, i.e.,

X-|-R — > X- + R-

This equation is usually written :

X— R — >X- + R-

This gives rise to free radicals, and the breaking of the bond in this manner
is known as homolytic fission (homolysis). Free radicals are odd electron
molecules, e.g., methyl radical CH 3 -, triphenylmethyl radical (C 6 H 6 ) 3 0, etc.
The majority are electrically neutral (a few free radical ions are known). All
possess addition properties, and are extremely reactive; when a free radical
is stable, its stability is believed to be due to resonance. Free radicals are
paramagnetic, i.e., possess a small permanent magnetic moment, due to
the presence of the odd (unpaired) electron. This property is used to detect
the presence of free radicals. Diradicals are also known; these have an
even number of electrons, but two are unpaired (see, e.g., methylene, an-
thracene). In general, free-radical reactions are catalysed or initiated by
compounds which generate free radicals on decomposition, or by heat or
light. Furthermore, a reaction which proceeds by a free-radical mechanism
can be inhibited by the presence of compounds that are known to combine
with free radicals. Another important characteristic is that a free-radical
mechanism leads to abnormal orientation in aromatic substitution.

STRUCTURE OF THE ATOM 33

(ii) Atom (or group) R retains the shared pair. This may be represented
as:

x|— r — > x + :r

This equation is now usually written :

X— R — >■ X+ + R"

This is known as heterolytic fission (heterolysis), and Y is said to be an electro-
philic (electron-seeking) or cationoid reagent, since it gains a share in the
two electrons retained by R. Obviously an electrophilic reagent attacks a
molecule at the point of high electron density. When an electrophilic
reagent is involved in a substitution or a replacement reaction, that reaction is
represented by S B (S referring to substitution, and E to the electrophilic

reagent). When R! is a negative group in which the carbon atom carries
the negative charge, i.e., has an unshared pair of electrons, the group is
known as a carbanion.

(iii) Atom (or group) R loses the shared pair, i.e., the shared pair remains
with X. This may be represented as:

X— |R-

-^x: + r

X— R-

-> X- + R +

or

This also is heterolytic fission (heterolysis), and Y is said to be a nucleophilic
(nucleus-seeking) or anionoid reagent, since it supplies the electron pair.
Obviously a nucleophilic reagent attacks a molecule at the point of low
electron density. When a nucleophilic reagent is involved in a substitution
or a replacement reaction, that reaction is represented by Sn. When R + is a
positive group in which the carbon atom carries the positive charge, i.e.,
lacks a pair of electrons in its valency shell, the group is known as a carbonium
ion, and is in a trigonal state of hybridisation. Furthermore, such a car-
bonium ion is said to have a " classical " structure. There are, however,
various cases where the ion is better represented as a bridged carbonium
ion, and in these cases the ions are said to have a " non-classical " structure
(see, e.g., p. 101).

Because of their positive charge, carbonium ions are very unstable, but
they may be stabilised by delocalisation (i.e., by spreading) of the charge by
means of solvation. Alternatively, the ion may be stabilised by delocalisa-
tion of the charge within the molecule by inductive and/or resonance effects.
In certain cases, delocalisation may be the result of hyperconjugation (see
p. 269).

Transition state theory of reactions. According to the collision theory of
reactions, -before molecules can enter into chemical reaction, they must collide
and they must be activated, i.e., they must attain a certain amount of energy (E)
above the average value. However, the rate of a reaction depends not only
on the frequency of collisions in which the energy of activation is exceeded
but also on whether the colliding molecules are suitably oriented with respect to
each other for effective reaction to occur. This limitation is known as the proba-
bility or steric factor, and depends, for a given type of reaction, on the geometry
of the reacting molecules. A simple example of the steric factor is that in the
reaction

2HI >- H, + I.

34 ORGANIC CHEMISTRY

If hydrogen iodide decomposes on collision, then activated molecules can collide
in one of two ways, the " right " way leading to decomposition, and the " wrong "
way leading to merely a " change in partners ".

H ->H H— H

"Right way" | I — >■ +

I -> I I— I

H~~> I H— I

" Wrong way " | | >• +

I ->H I— H

One can expect that when various paths are possible for a given reaction under
given conditions, then the path actually followed will be the one requiring the
lowest energy of activation. The problem therefore is to try to work out the
path that requires the minimum energy of activation.

The transition state theory of reactions does not use the simple idea of collision,
but considers how the potential energy of a system of atoms and/or molecules
varies as the molecules are brought together. Consider the reaction

A + BC ■>- AB + C

This is known as a three-centre reaction. London (1929), by making certain
approximations, showed that the minimum energy required in a three-centre
reaction is when the reaction proceeds by an end-on approach, i.e., in the above
reaction, the approach of A to BC requiring the minimum activation energy is
for A to approach BC along the bonding line of BC and on the side remote from C :

A •> B— C >- A— B + C

In this three-centre reaction, the value of the activation energy depends on four
factors: (i) The strength of the B — C bond. The stronger this is, the greater
will be E. (ii) The repulsion between A and BC. The greater this repulsion, the
greater will be E. (iii) The repulsion between AB and C. The greater this
repulsion, the greater will be E. (iv) The strength of the A — B bond. The
greater the strength of this bond, the lower be E.

Since most reactions are carried out in solution, another factor affecting the
value of £ is solvation of molecules and ions.

When we consider the mechanism of activation of this three-centre reaction,
we can imagine that there are two extreme cases possible: (i) A is forced up
against the repulsion of BC until it is close enough to compete with B on equal
terms with C, which is finally expelled, (ii) BC acquires so much energy that
the bond B — C is broken, and then A and C combine without any opposition.

Polanyi et al. (1931-1938) amplified London's ideas into the transition state
theory. These authors showed by mathematical treatment that the lowest
value of E is obtained when the reaction proceeds through a compromise be-
tween the two extremes (i) and (ii) mentioned above. A approaches BC along
the bonding line of BC remote from C, and is forced against the repulsion of BC,
and at the same time bond BC stretches until A and C can compete on equal
terms for B. Thus a point is reached when the distances A — B and B — C are
such that the forces between each pair are the same. This condition is the
transition state {activated complex) ; in this state neither molecule AB nor BC
exists independently. The system can now proceed in either direction to form
A and BC or AB and C. This sequence of events may be represented by the
following equation (T.S. = transition state) :

A + BC — > A -B C > AB + C

T.S.

It should be noted that the new bond is formed on the side opposite to that of
the original bond (which was broken). Thus the original molecule is turned
" inside out ", i.e., inverted when the new molecule is formed. This inversion
may be observed if B contains an asymmetric carbon atom (see the Walden
inversion, p. 413).

STRUCTURE OF THE ATOM

35

The above sequence of events may also be represented graphically by means
of an energy profile diagram (Fig. n). This diagram is obtained by plotting the
potential energy (P.E.) of the system against the reaction co-ordinates (the
various distances between the nuclei of A, B and C).

t
P.E.

t
P.E.

A+BC

AB+C

T.S.

Reactants

Products

Reaction co-ordinate •
Fig. 2. ii.

Reaction co-ordinate -
Fig. 2.12.

E is the activation energy, and AH is the heat of reaction at constant pressure.
It is assumed that the reaction rate is given by the rate at which the reactant
molecules pass through the transition state. For a given shaped " hump ",
the lower it lies (i.e., the lower the energy barrier is), the easier it is for the reactant
molecules to enter the transition state. Also, the wider the hump for a given
height, the easier it is for the reactant molecules to enter the transition state,
since there is now a wider latitude in nuclei positions for the activated complex.

The activated complex is not a true molecule; it contains partial bonds, and
the energy content of the system is a maximum. Its life is extremely short, and
hence it cannot be isolated; it is always decomposing into reactants or products.
The reaction, however, if complex, will proceed through true intermediates which
possess some measure of stability, and if this is great enough, the intermediates
may be isolated. If the reaction proceeds through a true intermediate (I)
(Fig. 12), there will be a minimum in the energy profile diagram. The greater
the dip, the more stable will be the intermediate, and conversely, the shallower
the dip, the less stable will be the intermediate. In the extreme case, the dip
may be so shallow that the intermediate is indistinguishable from the transition
state. It should be noted here that each intermediate has its own transition
state, and it appears that it is usual to assume that the structure of the transition
state closely resembles the Structure of the intermediate (i.e., the product
of that step). There is, however, evidence to show that in some cases the
transition state may be closer to the parent system than to the intermediate (or
product) .

Use of isotopes in organic chemistry. In recent years the use of isotopes
has been extremely helpful in the study of reaction mechanisms and re-
arrangements, in the elucidation of structures, and also in quantitative
analysis. The application of isotopes in biochemistry has also been particu-
larly fruitful, since they offer a means of identifying intermediates and the
" brickwork " of the final products. The common isotopes that have been
used in organic chemistry are: deuterium ( 2 H, D; stable), tritium ( 3 H, T;
radioactive), 13 C (stable), 14 C (radioactive), 18 N (stable), 18 (stable), 32 P
(radioactive), 35 S (radioactive), 37 C1 (stable), 82 Br (radioactive), 131 I (radio-
active).

Various methods of analysis are used. Radioactive isotopes are usually
analysed with the Geiger-Muller counter, and the stable isotopes by means of
the mass spectrograph. Deuterium is often determined by means of infra-
red spectroscopy; and there are also the older methods for deuterium and
18 of density, or refractive index measurements (of the water produced
after combustion of the compound). A new method is that of nuclear

36 ORGANIC CHEMISTRY

magnetic resonance, this technique being applicable only to those isotopes
having nuclear magnetic moments, e.g., D, T, 13 C and 15 N.

Isotopes are usually used as tracers, i.e., the starting material is labelled at
some particular position, and after reaction the labelled atom is then located
in the product. This does not mean that labelled compounds contain ioo
per cent, of the isotope, but that they usually contain an abnormal amount
of the isotope. Many examples of the use of isotopic indicators will be
found in the text (see Index, Isotopic indicators).

The use of isotopes, stable or radioactive, is based on the fact that the
chemical behaviour of any particular isotope is the same as that of the other
atoms isotopic with it (the chemical properties of an element depend on the
nuclear positive charge and the number of electrons surrounding the nucleus,
and not on the number of neutrons in the nucleus) . This identity in chemical
behaviour is essentially true for the heavy atoms, but in the case of the
lightest elements, reactions involving heavier isotopes are slower, but so
long as identical paths are followed, the final result is unaffected. This
difference in rates of reaction of isotopes is known as the kinetic isotope effect,
and the magnitude of such effects depends on the weight ratio of the isotopes
involved. Thus the kinetic isotope effect is greatest with H and D (and T).
The kinetic effect has been widely used to study reaction mechanisms, since
this difference in rate is significant when the bond attaching the isotopic
atom is stretched in the activated state, e.g., a protonated molecule may
react about seven times as fast as the corresponding deuterium compound.

Ion accelerators and nuclear reactors produce, as by-products, artificial

isotopes, particularly those which are radioactive. These isotopes are then

supplied in the form of some compound from which a labelled compound can

be synthesised, e.g., 14 C is usually supplied as Ba 14 C0 3 , 16 N as 16 NH 4 C1, etc.

A simple example of the synthesis of a labelled compound is that of acetic

*
acid (in the following equations, which involve a Grignard reagent, C is 14 C;

this is a common method of representing a tracer atom, provided its nature

has been specified).

BaC0 3 + 2HCI — > BaCl 2 + H 2 + C0 2
(i) C0 2 + CH 3 MgI — > CH 3 -C0 2 H

(ii) C0 2 + 3 H 2 ^> H s O + CH 3 OH -^ CH 3 I -*► CH 3 MgI

co,

CO,

CH 3 -CO,H CH s -C0 2 H

In a number of cases, an exchange reaction is a very simple means of
preparing a labelled compound, e.g., dissolving a fatty acid in water en-
riched with deuterium.

R-C0 2 H + D 2 ^ R-C0 2 D + DHO

Because of the possibility of this exchange reaction, it is often necessary to
carry out control experiments.

Isotopes are also very useful in the analysis of mixtures, particularly for
the determination of the yield of products in a chemical reaction when
isolation is difficult. A simple method is that of isotopic dilution. The
labelled compound is prepared, and a known amount is then added to the
mixture to be analysed. A portion of the substance is now taken and

STRUCTURE OF THE ATOM 37

analysed for its isotopic content. From a knowledge of the isotopic content
of the labelled compound added and recovered, and the weight of the labelled
compound added, it is thus possible to calculate the weight of the labelled
compound in the mixture. This method can only be used as long as there is
no isotopic exchange during the isolation.

READING REFERENCES*

Sidgwick, Electronic Theory of Valency, Oxford Press (1927).

Watson, Modern Theories of Organic Chemistry, Oxford Press (1941, 2nd ed.).

Remick, Electronic Interpretations of Organic Chemistry, Wiley (1943).

Pauling, The Nature of the Chemical Bond, Cornell University Press (i960, 3rd ed.).

Hammett, Physical Organic Chemistry, McGraw-Hill (1940).

Branch and Calvin, The Theory of Organic Chemistry (An Advanced Course), Prentice-
Hall (1941)-

Dewar, The Electronic Theory of Organic Chemistry, Oxford Press (1949)-

Walsh, Remarks on the Strengths of Bonds, Trans. Faraday Soc, 1947, 43, 60 (also
pp. 158, 342).

Coulson, Representation of Simple Molecules by Molecular Orbitals, Quart. Reviews
(Chem. Soc), 1947, I, 144.

Coulson, Valence, Oxford Press (1961, 2nd ed.).

Wheland, Resonance in Organic Chemistry, Wiley (1955).

Wheland, Advanced Organic Chemistry, Wiley (i960, 3rd ed.).

Ferguson, Electronic Structures of Organic Molecules, Prentice-Hall (1952) .

Hermans, Theoretical Organic Chemistry, Elsevier (1954).

Ingold, Structure and Mechanism in Organic Chemistry, Bell and Sons (1953).

Cartmell and Fowles, Valency and Molecular Structure, Butterworths (1956).

Hine, Physical Organic Chemistry, McGraw-Hill (1962, 2nd ed.).

Gould, Mechanism and Structure in Organic Chemistry, Holt and Co. (i959)-

Pimental and McClellan, The Hydrogen Bond, Freeman and Co. (i960).

Leffler, The Reactive Intermediates of Organic Chemistry, Interscience (1956).

Dickens and Linnett, Electron Correlation and Chemical Consequences, Quart. Reviews
(Chem. Soc), 1957, 11, 291.

Gillespie and Nyholm, Inorganic Stereochemistry, Quart. Reviews (Chem. Soc), 1957, 11,

339-
Bent, Distribution of Atomic s Character in Molecules and its Chemical Implications,

J. Chem. Educ, i960, 37, 616.
Sanderson, Principles of Chemical Bonding, /. Chem. Educ, 1961, 38, 382.
Orville-Thomas, Nuclear Quadruple Coupling and Chemical Bonding, Quart. Reviews

(Chem. Soc), 1957, 11, 162.
Pople, The Molecular-Orbital and Equivalent Orbital Approach to Molecular Structure,

ibid., 1957, 11, 273.
Arnstein and Bentley, Isotopic Tracer Technique, ibid., 1950, 4, 172.
Thomas and Turner, The Syntheses of Isotopically Labelled Organic Compounds,

ibid., 1953. 7. 4°7-
Gold and Satchell, The Principles of Hydrogen Isotope Exchange Reactions in Solution,

ibid., 1955, 9, 51.
Semenow and Roberts, Uses of Isotopes in Organic Chemistry, /. Chem. Educ, 1956,

33. 2-

Popjak, Chemistry, Biochemistry, and Isotopic Tracer Technique, Lectures, Mono-
graphs and Reports of the Royal Institute of Chemistry, 1955, No. 2.

* Books which deal with the electronic theories of organic chemistry will not be given
as reading references in subsequent chapters in this book. The reader should always be
prepared to refer to them on any matter dealing with mechanisms.

CHAPTER III

ALIPHATIC COMPOUNDS

PARAFFINS

Aliphatic compounds are open-chain or acyclic compounds, and the name
aliphatic arises from the fact that the first compounds of this class to be
studied were the fatty acids (Greek: aliphos, fat).

Carbon forms a large number of compounds with hydrogen only and these
are known collectively as hydrocarbons. There are two groups of hydro-
carbons: (i) saturated hydrocarbons; (ii) unsaturated hydrocarbons.

The paraffin hydrocarbons or the paraffins are the saturated hydrocarbons.
Many occur naturally, and the chief source of the paraffins is mineral oil or
petroleum, which occurs in many parts of the world.

The simplest paraffin is methane, CH 4 , which occurs in " natural gas "
(q.v.) and the gases from oil-wells. Methane is the principal product of
organic decay in swamps and marshes, the gas being set free by the action
of bacteria; this method of formation in nature has given rise to the name
" marsh-gas " for methane. Sewage sludge which has been fermented by
bacteria yields a gas containing about 70 per cent, methane, and this is used
as a liquid fuel. Methane also forms about 40 per cent, by volume of
coal-gas.

Methane may be synthesised as follows:

(i) By striking an electric arc between carbon electrodes in an atmosphere
of hydrogen. Only a very small amount of methane is obtained in this way.

(ii) A mixture of carbon and reduced nickel is heated at 475 ° in the pre-
sence of hydrogen.

(hi) The first synthesis of methane was carried out by Berthelot in 1856,
who passed a mixture of carbon disulphide and hydrogen sulphide over
heated copper:

CS 2 + 2H 2 S -f 8Cu — >■ CH 4 + 4Cu 2 S

(iv) A mixture of carbon monoxide or dioxide and hydrogen is passed
over finely divided nickel heated at about 300 :

C0 + 3 H 2 -^CH 4 + H 2
C0 2 + 4H 2 -^ CH 4 + 2H 2

This synthesis is due to Sabatier and Senderens (1897). Until recently
the nickel catalyst was usually prepared by reducing with hydrogen nickel
oxide deposited on a suitable inert, porous support, e.g., kieselguhr. The
support is impregnated with a nickel salt, treated with sodium hydroxide,
washed and dried, and the resulting nickel oxide reduced with hydrogen at
300-450°. Many organic compounds may be reduced by passing their
vapours mixed with hydrogen over nickel heated at 200-300°. Any
reduction that is carried out in this manner is referred to as the Sabatier—
Senderens reduction, in honour of the workers who first introduced this
method. It is quite a common feature in organic chemistry to name a
reaction after its discoverer or, in certain cases, after a worker who investi-
gated the reaction and extended its application. The reader should always
make himself familiar with the reaction associated with a particular name.

The most common nickel catalyst used to-day is that prepared by the
method introduced by Raney (1927). An alloy containing equal amounts
of nickel and aluminium is digested with sodium hydroxide; the aluminium

38

ALIPHATIC COMPOUNDS 39

is dissolved away, and the residual very finely divided nickel is washed
and stored under water, ethanol, or any other suitable liquid. Raney
nickel is more reactive than the supported nickel catalyst, and is usually
effective at lower temperatures, often at room temperature.

None of the syntheses described above is of any practical importance as
a method of preparing methane in quantity. The following methods are
those which may be used in the laboratory, i.e., convenient methods of
preparing methane in reasonable quantities. The degree of purity of the
product depends on the particular method used. Quite often it is the
purification of the crude product which causes an appreciable loss of material,
and the more we aim at getting the " pure " compound, the smaller is the
final yield.

1. The most convenient method is to heat a mixture of anhydrous sodium
acetate and soda-lime :

CH s -C0 2 Na + NaOH(CaO) — > CH 4 + Na 2 CO s

In chemical reactions soda-lime behaves as sodium hydroxide (or calcium
hydroxide) ; it is not deliquescent and does not attack glass, and is therefore
more convenient to use than solid sodium hydroxide.

2. By boiling aluminium carbide with water:

A1 4 C S + i 2 H 2 — ^ 3 CH 4 + 4A1(0H) 3

The methane is impure; it contains hydrogen.

3. Almost pure methane (it contains traces of hydrogen) is obtained by
the reduction of methyl iodide with nascent hydrogen:

CH3I + 2[H] — > CH 4 + HI

A common method for generating hydrogen uses zinc and acetic acid
saturated with hydrogen chloride, or hydrochloric acid alone, or with
aqueous sodium hydroxide. Another useful method is by the action of a
zinc-copper couple on ethanol.

Lithium aluminium hydride or lithium hydride may also be used for
reducing alkyl bromides.

4. By the action of water on:

(a) dimethylzinc, (CH^aZn + 2H 2 — >■ 2CH 4 + Zn(OH) 2

(b) Methylmagnesium iodide,

CH 3 - Mg - I + H 2 — > CH 4 + Mgl(OH).

Method (b) is far more convenient than (a). Methylmagnesium iodide is
a' member of a group of compounds known as the Grignard reagents (p. 348).

Methane is obtained in vast quantities from natural gas, gas from the
oil-wells, and from cracked petroleum (q.v .).

Properties of methane. Methane is a colourless, odourless, non-poisonous
gas; its b.p. is — 1647760 mm., and m.p. —184°. It is somewhat soluble
in water, 100 ml. of water dissolving about 5 ml. of methane at 20 ; but
is quite soluble in ethanol and ether. It burns with a non-luminous flame
in air or oxygen, forming carbon dioxide and water:

CH 4 + 20 2 — > C0 2 + 2H a O

It explodes violently when mixed with air (or oxygen) and ignited, and this
is believed to be the cause of explosions in coal-mines, where methane is
known as fire-damp. Methane may be catalytically oxidised to methanol
and formaldehyde.

40 ORGANIC CHEMISTRY

Substitution reactions of methane. Chlorine has no action on methane
in the dark. In bright sunlight the reaction is explosive, and hydrogen
chloride and carbon are formed:

CH 4 + 2Cl a — -> C + 4HCI

In diffused sunlight no explosion occurs, but a series of reactions takes
place whereby the four hydrogen atoms in methane are successively replaced
by chlorine atoms:

CH 4 + Cl a — > CH3CI + HC1 methyl chloride

CH S C1 + Cl a — y CH 2 C1 2 + HC1 methylene chloride

CH 2 C1 2 + Cl 2 — > CHCI3 + HC1 chloroform

CHCI3 + Cl 2 — y CCI4 + HC1 carbon tetrachloride

In methane the four carbon valencies are satisfied by combination with
four hydrogen atoms. Carbon never exhibits a valency of more than four,
and so cannot combine with more than four hydrogen atoms or four other
univalent atoms or groups. Hence in the reaction with chlorine, the
hydrogen atoms are displaced, and chlorine atoms take their place. This
type of reaction is known as substitution, and is the direct replacement of
hydrogen by some other atom or grpup. The products so formed are known
as substitution products. The atom or group that has replaced the hydrogen
atom is called the substituent, and when a substituent atom or group is re-
placed by some other atom or group, the reaction is referred to as a re-
placement (or displacement) reaction. It should be noted that in substitution
or replacement reactions there is no change in structure. The spatial arrange-
ment of the molecule, however, may have changed (see p. 413)-

In the substitution reaction between methane and chlorine, all four
substitution products are obtained, since it is impossible to stop the reaction
at any particular stage. It has been found possible, however, to control
the reaction so as to obtain mainly methyl chloride {q. v.).

Methane also undergoes substitution with bromine, but the reaction is less
vigorous than that with chlorine. With iodine the reaction is reversible :

CH 4 + I 2 ^ CH3I + HI

The equilibrium lies almost completely on the left, and consequently the
yield of methyl iodide is negligible. On the other hand, in the presence of
an oxidising agent, e.g., iodic acid, nitric acid, mercuric oxide, etc., the
reaction proceeds to the right, since the equilibrium is upset by the removal
of the hydrogen iodide which is oxidised, e.g.,

5HI + HI0 3 — > 3I2 + 3H 2

Methane reacts explosively with gaseous fluorine. The initial reaction is
possibly:

CH 4 + 2F 2 — 5-C + 4HF

Structure of methane. The molecular formula of methane is CH 4 .
Assuming the quadrivalency of carbon and the univalency of hydrogen, we
find that there is only one structure possible for methane, viz., (I). Study of
the reactions of methane shows that all four hydrogen atoms are equivalent,

h a a

H-i-H H— C-H H-C— CI

(I) H CI (ii) H (ill)

ALIPHATIC COMPOUNDS 41

e.g., methylene chloride, CH 2 C1 8 , prepared by totally different methods, is
always the same. Thus (II) and (III) are different ways of writing the same
structure. At first sight it may appear that these two structural formula
are different. They are different if the molecule is two-dimensional, but, as
we have seen (p. 25), in saturated compounds the four valencies of carbon are
arranged tetrahedrally. Examination of the two structures as tetrahedral
figures shows that they are identical.* The chief disadvantage of the plane-
structural formula is that it does not show the spatial arrangement of the
atoms. On the other hand, the three-dimensional structural formula is
cumbersome, and for many complicated molecules cannot easily be drawn
on paper. Hence we usually adopt the plane-formulae when dealing with
compounds from the point of view of their structure; only when we wish
to stress the spatial arrangement of the atoms or groups in a molecule do
we resort to solid diagrams (see, e.g., Ch. XVII). How we show the relative
positions of the various atoms or groups in the plane-structural formula
of a given compound usually depends on ourselves, but where possible the
simplest method of writing the structure should be chosen. Consider
1 : 5-dichloropentane, C B H 10 C1 2 . This is a " straight " chain compound,
and its structural formula is usually written CH 2 C1-CH 2 -CH 2 -CH 2 -CH 2 C1.
Now, 1 : 5-dichloropentane gives cycfopentane when treated with zinc.
cycfoPentane is a ring compound, and to show its formation from 1 : 5-di-
chloropentane, we write the structure of the latter as follows:

^CHa-CHaCl /CH 8V

\CH 2
/CH 2

CH 2 + Zn — >• CH 2 V"s + ZnCl 8

N CH 2 -CH 2 C1 \CH 2

Thus " straight " chains may be " bent " to stress a particular point we
may have in mind.

Uses of methane, (i) When heated at 1000 , methane is decomposed into
carbon and hydrogen :

CH 4 — > C + 2H 2

The carbon is formed in a very finely divided state, and is known as carbon
black. This is used in making printers' ink and paints; it is also used in
the rubber industry for motor tyres, etc.

(ii) Methane is used as a source of hydrogen (for synthetic ammonia and
for synthesis gas). Methane is mixed with steam and passed over nickel
supported on alumina heated at 725 °:

CH 4 + H 2 0— ^CO + 3H 2

(iii) For the technical preparation of methyl chloride and methylene
chloride (q.v.).

(iv) For the technical preparation of methanol and formaldehyde (q.v.).
(v) As a liquid fuel.

Ethane, C 2 H 8 , occurs with methane in natural gas and the gases from
the oil-wells. It is formed to a slight extent when an electric arc is struck
between carbon rods in an atmosphere of hydrogen. It may be prepared
by any of the following methods:

I. By reduction of ethyl iodide with nascent hydrogen using, for
example, the zinc-copper couple and ethanol:

C 2 H S I + 2[H] — > C 2 H„ + HI

* " Atomic models " are very useful to the organic chemist. Sets of these models
may be bought; alternatively the reader can build his models from plasticine and
matchsticks, which provide crude but usually satisfactory models.

42 ORGANIC CHEMISTRY

The ethane is contaminated with traces of hydrogen :

Zn/Cu + 2C 2 H 6 OH — > H 2 + (C 2 H 5 0)2Zn + Cu

2. By treating a dry ethereal solution of methyl iodide with sodium:

2CH3I + 2Na — > C 2 H 8 + 2NaI

This is an example of the Wurtz reaction. It is not so straightforward
as the equation indicates; other products are obtained in addition to
ethane (see p. 51).

3. By the electrolysis of a concentrated solution of sodium or potas-
sium acetate. A mixture of ethane and carbon dioxide is evolved at
the anode, and hydrogen is evolved at the cathode:

2CH 3 -C0 2 Na + 2H 2 — > C 2 H 6 + 2C0 2 + 2NaOH + H 2

This means of preparation is an example of Kolbe's electrolytic method.
The ethane may be freed from carbon dioxide by washing the mixture
with aqueous sodium hydroxide, but still contains other impurities
(see p. 52).

4. Pure ethane may be obtained by the action of water on ethyl-
magnesium iodide (a Grignard reagent) :

C 2 H 5 — Mg-I + H 2 — > C 2 H 6 + Mgl(OH)

5. The best way of preparing pure ethane in large quantity is by
the catalytic hydrogenation of ethylene :

C 2 H 4 -f- H 2 — > C 2 H 6

3°°

Properties of ethane. Ethane is a colourless gas, b.p. —89°, sparingly
soluble in water but readily soluble in ethanol. It burns in air or oxygen
to form carbon dioxide and water:

2C 2 H 6 + 70 2 — > 4C0 2 + 6H 2

It reacts with halogens in a similar manner to methane to form substitution
products, but a much greater number of products are possible due, firstly
to the presence of six hydrogen atoms in ethane compared with four in
methane, and secondly, to the fact that isomerism is possible in the sub-
stitution products of ethane, and not in those of methane. Thus, for
example, two dichloroethanes are possible: CH 3 *CHC1 2 and CH 2 ChCH 2 Cl
(see later).
Structure of ethane. The molecular formula of ethane is C 2 H 6 . Assum-
ing the quadrivalency of carbon and the univalency of
I 1 hydrogen, the only possible structure for ethane is (I). This

I 1 structure agrees with all the known properties of ethane.

"ll Writing the structural formula of ethane in the bond-diagram

I X wa y uses U P a l°t 0I space. Hence it has become customary

" " to use a " contracted " structural formula: CH 3 - — CH 3 or

W CH 3 -CH 3 or CH 3 CH 3 . The reader should make himself

familiar with these different ways of writing structural formulae as soon as

possible.

Propane. C 3 H 8 is a constituent of natural gas and gas from the oil-
wells. It may be prepared by the following methods:

I, By reduction of propyl iodide with nascent hydrogen:
C 3 H 7 I + 2[H]-^^C 3 H 8 + HI

ALIPHATIC COMPOUNDS 43

2. By the Wurtz reaction, using a mixture of methyl and ethyl
halides, e.g.,

CH 3 Br + CH 3 -CH 2 Br + 2Na ether > CH 3 -CH 2 -CH 3 + 2NaBr

The yield of propane is poor, since ethane and butane, C 4 H 10 (as well
as other compounds), are obtained as by-products:

2CH 3 Br + 2Na — > C 2 H„ + 2NaBr
2C 2 H 5 Br + 2Na — y C 4 H 10 + 2NaBr

3. By Kolbe's electrolytic method, using a mixture of sodium acetate
and sodium propionate:

CH 3 -C0 2 Na + C 2 H 5 -C0 2 Na + 2H 2 — > C 3 H 8 + 2C0 2 + 2NaOH + H 2

Apart from other products, ethane and butane are obtained in relatively
large quantities, resulting in a poor yield of propane (cf. 2 above).

4. By the action of water on propylrnagnesium iodide :

C 3 H 7 — Mg— I + H 2 — > C 3 H 8 + MgI(OH)

Properties of propane. Propane is a colourless gas, b.p. —44-5°. It
resembles methane and ethane in many of its chemical properties.

Structure of propane. The molecular formula of propane is C 3 H 8 . Assum-
ing the quadrivalency of carbon and univalency of hydrogen, the only
possible structure for propane is CH 3 -CH 2 -CH 3 . This structure agrees with
all the known properties of propane.

Butanes, C 4 H 10 . Theoretical consideration of this formula shows that
two structures are possible :

H
H H H H H H— C— H H
H— C— C— C— C— H H— C C C— H

UU k k k

CH 3

•CH"

CH 3 -CH 2 -CH 2 -CH 3 CH 3 -CH-CH 3 or (CH 3 ) 3 CH

(I) (ID

(I) has a straight chain, and (II) a branched chain. Both isomers are known,
and thus " butane " is the first paraffin to exhibit structural isomerism.
This is an example of chain or nuclear isomerism, and is characterised by
the manner of linking of the carbon chain. (I) is known as normal butane,
and (II) as /sobutane. Both occur in natural gas and in petroleum gas, and
they may be separated by fractional distillation under pressure.

normal Butane may be prepared by the Wurtz reaction using ethyl iodide :

2CH 3 -CH S I + 2Na -^» CH 3 -CH 2 -CH 2 -CH 3 + 2NaI

It is a colourless gas, b.p. — 0-5°.

woButane may be prepared by reducing tertiary butyl iodide with nascent
hydrogen:

(CH 3 ) 3 CI + 2[H] -|g|> (CH 3 ) 3 CH + HI
It is a colourless gas, b.p. —10-2°.

44 ORGANIC CHEMISTRY

Examination of the two butane structures shows that all the carbon
atoms are not equivalent, and also that all th'e hydrogen atoms are not
equivalent. A primary carbon atom is one that is joined to one other
carbon atom; a secondary carbon atom to two other carbon atoms; and a
tertiary carbon atom to three other carbon atoms. Hydrogen atoms joined
to primary, secondary and tertiary carbon atoms are known as primary,
secondary, or tertiary hydrogen atoms, respectively. Thus normal butane
contains two primary carbon atoms, two secondary carbon atoms, six
primary and four secondary hydrogen atoms. /soButane contains three
primary carbon atoms, one tertiary carbon atom, nine primary hydrogen
atoms, and one tertiary hydrogen atom. As we shall see later, the behaviour
of these various types of hydrogen atoms differs considerably.

Pentanes, C 5 H 12 . Three pentanes are possible theoretically, and all are
known.

Structure. Name. b.p.

CHa'CHj-CHj-CHyCH, normal pentane 36

CH 3 \

/CH-CHyCHj, isopentane 28

CH/

CH,

i
CH 3 — C — CH 3 »eopentane 9-4°

CH 3

All the pentanes occur in natural gas and petroleum gas. ««oPentane
contains a quaternary carbon atom, i.e., a carbon atom joined to four other
carbon atoms.

As the number of carbon atoms in the paraffin increases, the number of
possible isomers increases rapidly, e.g., the paraffin C 18 H 32 can exist in 4347
isomeric forms. The number of isomers of a given paraffin may be calculated
by means of mathematical formula;; in most cases very few have actually
been prepared.

Nomenclature of the Paraffins

Whenever a new branch of knowledge is opened up, there is always the
problem of introducing a system of nomenclature. The early chemists
usually named a compound on the basis of its history, e.g., methane. This
is the parent hydrocarbon of methyl alcohol, CH 3 OH. Methyl alcohol
was originally obtained by the destructive distillation of wood, and was
named " wood-spirit ". From this arose the word methyl, which is a
combination of two Greek words, methu (wine) and hule (wood). Other
examples of this way of naming compounds are acetic acid, which is the
chief constituent of vinegar (Latin: acetum, vinegar); malic acid, which
was first isolated from apples (Latin: malum, apple), and so on. Thus
grew up a system of common or trivial names, and in many cases the origin
of the name has been forgotten. One advantage of the trivial system is
that the names are usually short and easily remembered, but a disadvantage
is that a particular compound may have a number of names.

As the number of known organic compounds increased, it became apparent
that it was necessary to systematise the method of nomenclature. The
most satisfactory system is one which indicates the structure of the com-
pound. This task was originally begun in 1892 by an international com-
mittee of chemists at Geneva, and hence is referred to as the Geneva system
of nomenclature. The work was carried on by the International Union of
Chemists (I.U.C.) by a committee appointed in 1922, and in 1931 these

ALIPHATIC COMPOUNDS

45

drew up a report which is often referred to as the I.U.C. system. Nomen-
clature is always undergoing revision, and the latest rules are those recom-
mended in 1957 by the Commission on the Nomenclature of Organic
Chemistry of the International Union of Pure and Applied Chemistry
(I.U.P.A.C). Various changes have been made, but two that have not been
incorporated into this book are: (i) Numbers indicating the positions of
substituents are to be separated by commas (not by colons as used in this
book), (ii) Prefixes are to be italicised when they define the positions of
named substituents or which are used to define stereoisomers (see p. 46).
The reader should always consult the Chemical Society Handbook if he wishes
to publish work in the Journal, and any mistakes he makes in nomenclature
will soon be put right by the Editors!

Dyson (1946) has developed " a new notation for organic chemistry ".
This scheme does not provide a new means of naming compounds, but shows
how it is possible to portray the structure of an organic compound irrespec-
tive of its complexity, and how it may be used for indexing.

There are at least three systems in use for naming paraffins, and in all
three the class-suffix, i.e., the ending of the name which indicates the par-
ticular homologous series (see later), is -ane.

1. In the trivial system of nomenclature the straight-chain compounds
are always designated as normal compounds, and the word normal is usually
abbreviated to »-. If the compound contains the grouping (CH 3 ) 2 C — H, it
is known as the tso-compound; if it contains a quaternary carbon atom,
the compound is known as the ««o-compound. It is impossible to name
many of the more complex paraffins by the trivial system (see later for
examples of the trivial system).

The first four paraffins have special names (related to their history);
from the fifth member onwards, Latin or Greek numerals are used to indicate
the number of carbon atoms in the molecule.

Name.

Formula.

Name.

Formula.

methane .... CH 4

hexadecane

C 16 H M

ethane

CjH a

heptadecane

C],H 38

propane

CjH g

octadecane

C l8 H 38

butane

C 4 H 10

nonadecane

C 19 H 10

pentane

• C 6 H 1S

eicosane .

C 20 H 42

hexane

C e H 14

heneicosane

C 2 iH 44

heptane

C 7 H„

docosane .

C 22 H 4e

octane

C S H 18

tricosane, etc. .

v> 2 3-H 48 , etc

nonane

C 8 H 20

triacontane

C 30 rl 62

decane

C 10 H 22

hexatriacontane

• C 38 H 74

undecane \
hendecane-*

• C U H 24

tetracontane .

C 40 H 82

pentacontane .

C 50 H 10S

dodecane

C 12 H 2S

hexacontane .

C 80 H 122

tridecane

C 13 H 28

heptacontane .

C 70 H 142

tetradecane

C 14 H 30

octacontane

C 80 H 1M

pentadecane

C«H 32

Univalent radicals that are formed by the removal of one hydrogen atom
from a paraffin are known as alkyl or alphyl radicals or groups. The name
of each individual radical is obtained by changing the suffix -ane of the
parent hydrocarbon into -yl. The first five alkyl radicals are often repre-
sented by a shorthand notation.

The radical derived from pentane has, in the past, been usually named
amyl. This name is now abandoned and pentyl and isopentyl are to be
used.

The paraffins are also known as the alkanes, since an alkyl radical plus
one hydrogen atom gives a paraffin, i.e., alkyl + H = a\kane.

In chemical equations, if we are dealing with alkyl compounds as a group

4 6

y

Paraffin.
methane
ethane .

propane .

butane

pentane

{

ORGANIC CHEMISTRY

Radical.
methyl CH S —
ethyl C a H 6 —

»-propyl CH 3 *CH a - CH 2 —

isopropyl CHj-CH-CHj
f M-butyl CH 3 -CH 2 -CH 2 -CH 2 —

sec-butyl CH 3 -CH 2 -<iH-CH 3

iso-butyl (CH 3 ) 2 CH-CH 2 —
Ue^.-butyl (CH 3 ) 3 C—

amyl CH 3 -(CH 2 ) a -CH 2 —
isoamyl (CH 3 ) 2 CH-CH 2 -CH 2 —

Short-hand notation.
Me
Et

M-Pr, Pr°t, or Pr

s'soPr, Pr/3, or Pr 4
w-Bu, Bua, or Bu

sec.-Bu, Bu0, Bu", or s-Bu

isoBu or Bu'
tert.-Bu, Bu', or t-Bu

n-Atn
isoAm or Am'

and we do not wish to specify any particular member, we use the symbol
R to represent the unspecified alkyl radical, e.g., RC1 represents any alkyl
chloride.

2. In this system of nomenclature the hydrocarbon, except the %-com-
pound, is regarded as a substitution product of methane. The most highly
branched carbon atom in the compound is named as the methane nucleus,
and the alkyl groups attached to this carbon atom are named in order
of increasing molecular weight of the groups (or in alphabetical order). If
two groups have the same molecular weight, the simpler is named before
the more complex, e.g., propyl before wopropyl. Hydrogen atoms, if
joined to the carbon atom chosen as the methane nucleus, are not named.
Since April, 1950, however, the Chemical Society has adopted an alphabetical
order for prefixes denoting substituents. This order follows in general that
adopted in Chemical A bstracts except for differences in nomenclature, spelling,
italicising, or punctuation. Italicised prefixes are neglected when assembling
substituents, e.g., wobutylwill be named before ethyl. Isomeric substituents,
however, are arranged in alphabetical order of the italicised prefixes, except
that iso follows directly after n, e.g., w-butyl, isobutyl, sec-butyl, tert.-butyl.

In the I.U.P.A.C. rules, however, the prefixes n, iso, sec, tert., neo, cyclo,
epi, alio, etc., are no longer to be italicised, i.e., they are to be written n,
iso, s, t, neo, cyclo, epi, alio, etc. Also, the alphabetical order of prefixes
follows the first roman letter, e.g., ethyl, isobutyl, etc. (see appendix for
further information on nomenclature).

This system of nomenclature is fairly good, since the name indicates the
structure of the compound. It is impossible, however, to name the complex
paraffins by this system.

3. In the I.U.P.A.C. system of nomenclature the longest chain possible is
chosen, and the compound is named as a derivative of this »-hydrocarbon.
The carbon chain is numbered from one end to the other by arabic numerals,
and the positions of side-chains are indicated by numbers, the direction of
numbering being so chosen as to give the lowest numbers possible to the
side-chains. When series of locants containing the same number of terms
are compared term by term, that series is "lowest " which contains the
lowest number on occasion of the first difference, and this principle is used
irrespective of the nature of the substituents, e.g.,

CH,

9
•CH,

CH-

765
-CH-CH 2 -CH a

4
CH.

3
•CH,

CHo Clio

2 1
CH-CH 3

CH,

This is named 2 : 7 : 8-trimethyldecane and not 3 : 4 : 9-trimethyldecane;
the first set is " lower " than the second set because at the first difference
2 is less than 3. When two sets of numbers are equally possible, then the

ALIPHATIC COMPOUNDS 47

order of the prefixes in the name decides which shall be used, e.g., i-bromo-
3-chloropropane and not 3-bromo-i-chloropropane. It should also be noted
that the names of prefixes are arranged alphabetically, regardless of the
number of each, e.g., 5-ethyl-2 : 3-dimethyloctane.

The I.U.P.A.C. system of nomenclature is undoubtedly superior to the
other two, since it permits the naming of any paraffin on sight.

The following are examples of the three systems of nomenclature:

1.

2.

3-

CH3* CHj'CHj* CH a

n-butane

n-butane

«-butane

12 3 4

CH3' CH" CH 2 * CH 8

isopentane

ethyldimethyl-

tsopentane

CH 3

methane

2

(C is most highly

branched)

CH 3

1 2 1 S 4

neohexane

ethyltrimethyl-

2 : 2-dimethyl-

1

methane

butane

CH,

2

(C is most highly
branched)

8 6 7 8

OH 3* OH 2* CH* CH 2* CHg" CH 3

~~ '

~

5-ethyI-2 : 3-
dimethyloctane

*CH 2

3 CH— CH 3

8 CH— CH 3

KH,

N.B. The following names are retained for unsubstituted hydrocarbons
only: isobutane, isopentane, neopentane, isohexane.

When several chains are of equal length, that chain chosen goes in series to :
(a) the chain which has the greatest number of side-chains ; (b) the chain whose
side-chains have the lowest-numbered Iocants; (c) the chain having the greatest
number of carbon atoms in the smaller side-chains; (d) the chain having the
least-branched side-chains. Also, where there is a side-chain within a side-chain,
the latter is also numbered, and the name of the complex radical is considered to
begin with the ■ first letter of its complete name, e.g.,

6 5 4 3 21

CH.-CH— CH— CH— CH-CH 3

CH3 CH a CH 3

k

"H s

4-ethyl-2 : 3 : 5-
trimethylhexane

H,

CH,

|i 2 3
CH.-C— CH.-CH.

9 8 7 6 5 4321

CH 3 -CH a 'CH 2 'CH 2 — C— CH a -CH a -CH a -CH s

5-(i : i-dimethylpropyl)-
5-(2-methylpropyl)nonane

CH„

Homologous Series

If we examine the formulae of the various paraffins we find that the
formula of each individual differs from that of its " neighbour " by CH 2 ,
e.g., CH 4 , C 2 H 6 , C 3 H g , C 4 H 10 , C 5 H 12 , ... A set of compounds, such as

48 ORGANIC CHEMISTRY

the paraffins, in which the members differ in composition from one another
by CH Z , is known as an homologous series, the individual members being
known as homologues.

Throughout organic chemistry we find homologous series, each series
being characterised by the presence of a functional group. The functional
group is an atom or a group of atoms that causes a compound to behave
in a particular way, i.e., it is the functional group that gives rise to homo-
logous series. Some of the more important functional groups and the
classes of compounds to which they give rise are shown in Table I.

It is also possible for a compound to contain two (or more) identical or
different functional groups, and this gives rise to polyfunctional compounds
(see text).

If we examine the formulae of the various paraffins, we find that the
formula C^H^.,^ will represent any particular homologue when n is given
the appropriate value, e.g., for pentane n is 5; therefore the formula of
pentane is C 8 H 12 . The formula QiHan+a is known as the general formula of
the paraffins. The composition of any homologous series can be expressed
by means of a general formula.

When we study the methods of preparation of the different paraffins,
we find that several methods are common to all, i.e., similar methods may
be used for the preparation of all the homologues. This gives rise to the
general methods of preparation of a particular homologous series.

TABLE I

Class of Compound.

Functional Group.

Formula.

Name.*

Alcohols ....
Aldehydes and ketones

Carboxylic acids .

Cyanides ....

Nitro-compounds

Amines ....

Mercaptans

Sulphonic acids .

—OH

~^c=o

_/

— O^N
—NO,
— NH,
— SH
— SO s H

Hydroxyl group
Carbonyl group

Carboxyl group

Cyano group
Nitro group
Amino group
Mercapto group
Sulphonic acid group

* Many functional groups are known by more than one name. Nomenclature is
dealt with in each homologous series described in the text.

Examination of the properties of the paraffins shows that many pro-
perties are, more or less, common to all the paraffin homologues. This
gives rise to the general properties of an homologous series.

The occurrence of homologous series facilitates the study of organic
chemistry, since it groups together compounds having many resemblances.
If we know the properties of several of the lower homologues, we can obtain
a fair idea of the properties of higher homologues, i.e., we can forecast
(within limits) the properties of a compound that we have not yet prepared.
The reader, however, must never be too hasty in predicting the properties
of an unknown homologue. The idea of homologous series should be used
as a guide, not as a hard-and-fast rule.

In view of what has been said above, we can see that in studying organic
chemistry it is advantageous to describe first the general methods of prepara-
tion of an homologous series, and then the general properties of that series.
It is also usual to describe the more important members individually, and

ALIPHATIC COMPOUNDS 49

to indicate, at this stage, any special methods of preparation and any special
properties. This is the way (wherever possible) in which we shall deal
with organic chemistry throughout this book, and we shall start by recon-
sidering the paraffins from this point of view.
The general methods of preparation of the paraffins fall into three groups.

A. From compounds containing the same number of carbon atoms.

1. By the catalytic reduction of unsaturated hydrocarbons, e.g., reduction
of ethylene:

C 2 H 4 + H 2 Ac 2 H, (ex.)*
300

2. (a) An alcohol, ROH, is converted into its corresponding alkyl iodide
using, e.g., phosphorus triodide (see alkyl halides):

3ROH + PI 3 -^ 3RI + H 3 P0 3 (v.g.)

The alkyl iodide may then be converted into the paraffin by various means:
(i) Reduction with nascent hydrogen :

RI + 2[H] — >• RH + HI (g.-v.g.)

(ii) Catalytic reduction using palladium as catalyst (see p. 65 for the
preparation of this catalyst) :

RI + H 2 -^>RH + HI (v.g.)

(iii) Reduction by heating with concentrated hydriodic acid at 150°.
This high temperature necessitates heating under pressure, since the maxi-
mum boiling point of hydriodic acid (57 per cent. HI) is 126°. High-pressure
work is carried out in autoclaves, but where the pressure is not excessive,
sealed, thick-walled glass tubes may be used. Reductions with hydriodic
acid under pressure are carried out in sealed tubes :

RI + HI— >RH + I 2 (g.-ex.)

The reduction can be performed directly on the alcohol, using excess of
hydriodic acid:

ROH + 2HI ^-> RH + I 2 + H 2 (g.-ex.)

Reduction with concentrated hydriodic acid is usually carried out in the
presence of a small amount of red phosphorus which regenerates the hydriodic
acid from the iodine formed. The hydriodic acid-red phosphorus mixture
is one of the most powerful reducing agents used in organic chemistry.

Instead of reduction, the alkyl iodide may be converted into the corre-
sponding Grignard reagent, which is then decomposed by water to form
the paraffin :

RI + Mg -^h* R— Mg— I — °-> RH (v.g.)

(b) By the reduction of a carbonyl compound with concentrated hydriodic
acid and red phosphorus, heated under pressure at 150°, e.g., acetone is
converted into propane:

CH 3 -CO-CH 3 ^> CH 3 -CH a -CH 3 (v.g.)
* See pre/ace for the significance of these terms in parentheses.

50 ORGANIC CHEMISTRY

Alternatively, the carbonyl compound can be reduced to the corresponding
alcohol, which is then reduced by HI/P (cf. above) . On the other hand, ketones
may be converted into the corresponding paraffins by the Clemmensen (see p. 150)
and Wolff-Kishner (see p. 153) reductions.

(c) By the reduction of fatty acids, R # C0 2 H, with HI/P in a sealed tube
at 200 :

R-C0 2 H -^> R-CH 3

200°

The yields are very good for the higher paraffins, and may even be improved
by heating the fatty acids with hydrogen under pressure in the presence of
a nickel catalyst.

B. From compounds containing a larger number of carbon atoms.

1. By heating a mixture of the sodium salt of a fatty acid and soda-lime:

R-C0 2 Na + NaOH(CaO) — > RH + Na 2 C0 3

This process of eliminating carbon dioxide from a carboxylic acid is known
as decarboxylation. Soda-lime is a very useful reagent for this process, but
various other reagents may also be used.

Oakwood et al. (1950) have shown that only sodium acetate decomposes
according to the equation given above. In all of the other cases tested —
propionate, butyrate and caproate — various products were obtained, e.g.,
with sodium propionate:

C 2 HyC0 2 Na — - — > C 2 H 6 + CH 4 + H 2 + Unsaturated compounds.
(44%) (20%) (33%)

This method, therefore, is not suitable for the preparation of simple paraffins
since, apart from the low yield of the desired product, it is very difficult to
separate the mixtures obtained.

2. By heating a mixture of the disodium salt of a dicarboxylic acid and
soda-lime, e.g., sodium adipate gives ^-butane:

C0 2 Na-CH 2 -CH 2 -CH 2 -CH 2 -C0 2 Na + 2NaOH(CaO) — >

CH 3 -CH 2 CH 2 -CH 3 + 2Na 2 C0 3

The reader may well ask: why not use tricarboxylic acids, etc.? When
dealing with a preparation, there are at least three important points that
must be considered: (i) the yield of crude product; (ii) the yield of pure
product; (iii) the accessibility of the starting materials. In certain cases
the yield of crude material is high, but the nature of the impurities is such
that purification causes a large loss of material, resulting in a poor yield
of pure product. On the other hand, it often happens that the product
of one reaction is to be used as the starting material for some other com-
pound which can readily be freed (i.e., purified with very little loss) from
the original impurity. Provided, then, that this impurity does not interfere
with the second reaction, the crude material of the first step can be used as
the starting material for the second. Thus the yield alone of a particular
reaction cannot decide the usefulness of that method of preparation; the
subsequent history of the product must also be taken into consideration.
Furthermore, all things being equal, the more accessible materials, i.e.,
readily prepared or purchased, are used as the starting materials.

With respect to the decarboxylation of acids as a means of preparing
paraffins, the reader will find that tricarboxylic acids, etc., are not readily
accessible; in fact, they are less accessible than the paraffins that can be

ALIPHATIC COMPOUNDS 51

prepared from them. It would, therefore, be useless, from the practical
point of view, to use these acids as starting materials for paraffins. From
the point of view of learning the subject, however, some useful purpose is
served in carrying out " paper reactions " with inaccessible materials,
since the reader may then master reactions of practical value.

C. From compounds containing fewer carbon atoms.

i. By the Wurtz reaction (1854). An ethereal solution of an alkyl halide
(preferably the bromide or iodide) is treated with sodium, e.g.,

RX + R'X + 2Na —> R— R' + 2NaX

As previously pointed out, when we do not wish to specify a particular
alkyl radical, we use the symbol R. When we deal with two unspecified
alkyl radicals which may, or may not, be the same, we can indicate this by
R and R' : also, when dealing with compounds containing a halogen atom,
and we do not wish to specify the halogen, we can indicate the presence of
the unspecified halogen atom by means of X.

Consideration of the equation given above shows that in addition to the
desired paraffin R — R', there will also be present the paraffins R — R and
R' — R'. Unsaturated hydrocarbons are also obtained. Obviously, then,
the best yield of a paraffin will be obtained when R and R' are the same,
i.e., when the paraffin contains an even number of carbon atoms and is
symmetrical. It has been found that the Wurtz reaction gives good yields
only for " even carbon " paraffins of high molecular weight, and that the
reaction generally fails with tertiary alkyl halides {q.v.).

Sodium is used in the Wurtz reaction. Other metals, however, in a finely-
divided state, may also be used, e.g., Ag, Cu (see text).

Two mechanisms have been suggested for the Wurtz reaction, and there
is evidence in favour of both. It is even possible that both take place
simultaneously.

(i) The intermediate formation of an organo-metallic compound, e.g., the
formation of w-butane from ethyl bromide :

C 2 H— Br + aNa- > C 2 H~ 5 Na + + NaBr

C 2 H~ s Na + + C„H 5 Br — > C 2 H S — C 2 H 5 + NaBr
(ii) The intermediate formation of free radicals, e.g.,

C 2 H S — Br + Na- — >■ C 3 H 5 - + NaBr
C 2 H 5 - + C 2 H 5 - — > C 2 H 5 — C 2 H 5

One of the properties of free radicals is disproportionation, i.e., intermolecular
hydrogenation, one molecule acquiring hydrogen at the expense of the
other, e.g.,

C a H 5 - + C a H 5 - — > C 2 H 6 + C 2 H 4

This would account for the presence of ethane and ethylene in the products.
According to Morton et al. (1942), however, ethane and ethylene may be
produced as follows:

Na+ *Br
H a C <-] CH a ^ CH 3 _,_ CH 2 _j_ NaBr

8 1 "*1 1 * ^ I s 4-

CH3 H CHjj CH S

CH,

2

52 ORGANIC CHEMISTRY

This mechanism is particularly interesting in view of the fact that dispropor-
tionation is commonly accepted as a criterion for a free-radical mechanism.
Furthermore, Bryce-Smith (1956) has obtained evidence that free radicals
play only a minor part in the formation of the usual Wurtz coupling and
disproportionation products. Also, Le Goff et al. (1958) have obtained
evidence to show that the Wurtz reaction of sodium with 2-chloro-octane is
a bimolecular reaction of an alkylsodium with an alkyl halide.

2. Kolbe's electrolytic method (1849). A concentrated solution of the
sodium or potassium salt of a fatty acid or mixture of fatty acids is electro-
lysed, e.g.,

R-C0 2 K + R'-C0 2 K + 2H a O — >■ R— R' + 2C0 2 + H 2 + 2KOH

If R and R' are different, then hydrocarbons R — R and R' — R' are also
obtained (c/. Wurtz reaction) . Such mixtures can often be separated readily.
Yields of 50-90 per cent, have been obtained with straight-chain acids con-
taining 2-18 carbon atoms. Alkyl groups in the a-position decrease the
yield (usually below 10 per cent.). The by-products are olefins, alcohols
(particularly in alkaline solution), and esters. It is also interesting to note
that the yields of the alkanes are increased when dimethylformamide is
used as solvent (Finkelstein et al., i960).

The Kolbe electrolytic method now has application in the synthesis of
natural compounds, particularly lipids.

The mechanism of the reaction is still obscure ; a possibility is via free radicals,
e.g., when sodium propionate is electrolysed, «-butane, ethane, ethylene and
ethyl propionate are obtained. The propionate ion discharges at the anode to
form a free radical:

C 2 H 5 C0 2 f— > C 2 H 5 C0 2 - + e

This free propionate radical then breaks up into the free ethyl radical and
carbon dioxide:

C 2 H 5 CCy — > C 2 H B - + CO,
Then:

(i) 2C a H 5 - — > C 4 H 10

(ii) C 2 H 6 - + C 2 H 6 - — > C,H 6 + C 2 H 4

(ill) C 2 H 5 - + C 2 H 6 C0 2 > C a H 5 -CO a C 2 H 5

Reaction (i) gives w-butane; (ii) gives ethane and ethylene by disproportionation
(cf. Wurtz reaction) ; and (iii) gives ethyl propionate.

3. By the action of an alkyl halide on a Grignard reagent:

R— Mg— I + R'l ^> R— R' + Mgl 2 (g.-v.g.)

4. Frankland's method (1850). Dialkyl-zinc compounds readily react
with alkyl halides to form hydrocarbons:

R^Zn + R'l — > R— R' + R— Zn— I

Dialkyl-zinc compounds are difficult to handle and as far as hydrocarbons
are concerned, are used only for the preparation of paraffins containing a
quaternary carbon atom, e.g.,

(CH 3 ) 3 CC1 + (CH^n —+ (CH 3 ) 4 C -f- CH 3 — Zn— CI

General properties of the paraffins. The name paraffin arose through
contracting the two Latin words " parum affinis", which means " little
affinity ". This name was suggested because these hydrocarbons were
apparently very unreactive. It is difficult to define the terms " reactive "

ALIPHATIC COMPOUNDS 53

and " unreactive ", since a compound may be reactive under one set of
conditions and unreactive under another. Under " ordinary " conditions, the
paraffins are inert towards reagents such as acids, alkalis, oxidising reagents,
reducing reagents, etc. In recent years, however, it has been shown that the
paraffins are reactive if the " right " conditions are used (see below).

General physical properties of the paraffins. The normal paraffins from C x
to C 4 are colourless gases; C 5 to C 17 , colourless liquids; and from C 18 on-
wards, colourless solids. The b.ps. rise fairly regularly as the number of
carbon atoms in the compound increases. This holds good only for the
normal compounds, and the difference in b.ps. decreases as the higher
homologues are reached. Other physical properties, such as m.p., specific
gravity, viscosity, also increase in the same way as the b.ps. (of the normal
paraffins), e.g., the specific gravity of the normal paraffins increases fairly
steadily for the lower members, and eventually tends to a maximum value
of about 0*79. Straight-chain paraffins containing at least six carbon
atoms form inclusion compounds with urea (see p. 387).

At the moment comparatively little is known about the quantitative relation-
ships between physical properties and chemical constitution. It is believed that
variation in b.ps. of compounds is due to different intermolecular forces such as
hydrogen bonding, dipole moments, etc. Hydrogen bonding may produce
association, and this will cause the b.p. to be higher than anticipated (see, e.g.,
alcohols). The greater the dipole moment of the compound, the higher is the
b.p., since, owing to the charges, more work is required to separate the molecules,
e.g., nitro-compounds, R # N0 2 , which have large dipole moments, have much
higher b.ps. than the paraffins in which the dipole moment is absent or very
small.

Observation has shown that in a group of isomeric compounds (acyclic),
the normal compound always has the highest b.p. and m.p., and generally,
the greater the branching, the lower the b.p.

The paraffins are almost insoluble in water, but readily soluble in ethanol
and ether, the solubility diminishing with increase in molecular weight.

It is believed that solubility depends on the following intermolecular forces:
solvent/solute; solute/solute; solvent/solvent. A non-electrolyte dissolves
readily in water only if it can form hydrogen bonds with the water. Thus
paraffins are insoluble, or almost insoluble, in water. Methane is more soluble
than any of its homologues; hydrogen bonding with the water is unlikely, and
so other factors — possibly molecular size — must also play a part. A useful rule
in organic chemistry with respect to solubility is that " like dissolves like ",
e.g., if a compound contains a hydroxyl group, then the best solvents usually
contain hydroxyl groups. This rule is not rigid (cf. paraffins).

X-Ray analysis of solid paraffins has shown that the carbon chains are fully
extended, i.e., zigzag. In the liquid state this extended form is also one of the

W

Fig. 3.1.

stable conformations provided that the carbon chain is not very long. When
the number of carbon atoms is sixteen or more, the extended form is no longer
present in the liquid state. The presence of a number of these different
conformations (or rotational isomers) has been shown by a study of infra-red and
Raman spectra of liquid paraffins. The various conformations arise from the
fact that groups can rotate about single bonds (p. 404). Furthermore, it has
been found that not all possible conformations are present, e.g., M-heptane shows
the presence of three conformations (Sheppard et al., T948, 1949). Fig. r shows
diagrammatically three possible forms.

54 ORGANIC CHEMISTRY

Since the dipole moment of methane is zero, the dipole moment of the
methyl group is equal to that of the fourth C — H bond (in methane) and is
directed along this axis. Thus replacement of hydrogen by a methyl group
will not be expected to change the dipole moment, i.e., the dipole moment of
all paraffins, whether straight- or branched-chain, will be zero. This has
been found to be so in practice. This will always hold good whatever con-
formation is taken up by the paraffin provided that no deformation of the
normal carbon valency angle (of 109° 28') is produced in the twisting, since
all methyl groups will be balanced by a C — H bond. It therefore follows
that the electronegativity of all alkyl groups is equal to that of hydrogen,
namely zero (p. 16). As soon as one hydrogen atom is replaced by another
atom or group (other than alkyl), the resultant molecule will now be found
to possess a dipole moment (see also p. 106).

General Chemical Properties of the Paraffins

1. Halogenation (see also the alkyl halides). Chlorination has been
studied in very great detail. It may be brought about by light, heat or
catalysts, and the extent of chlorination depends largely on the amount
of , chlorine used. A mixture of all possible isomeric monochlorides is
Obtained, but the isomers are formed in unequal amounts, due to the
difference of the reactivity of primary, secondary and tertiary hydrogen
atoms. Markownikoff (1875) found experimentally that the order of ease
of substitution is tertiary hydrogen >secondary>primary. This observa-
tion is very useful for predicting the possible courses of a reaction, and
qualitatively, to what extent each course will proceed, e.g., chlorination of
wobutane at 300 gives a mixture of two isomeric monochlorides:

CH 3

CH 3

CH 3 -CH-CH 2 C1

and

CH 3 -CC1-CH 3

(I)

(n>

We should expect to find more of (II) than (I) ; quantitative experiments
show that this is so.

Bromination is similar to chlorination, but not so vigorous. Iodination
is reversible, but it may be carried out in the presence of an oxidising agent,
such as HI0 3 , HN0 3 , HgO, etc., which destroys the hydrogen iodide as it
is formed (see p. 40). Iodides are more conveniently prepared by treating
the chloro- or bromo-derivative with sodium iodide in methanol or acetone
solution, e.g.,

RC1 + Nal ^ff^V RI + NaCl

This reaction is possible because sodium iodide is soluble in methanol or
acetone, whereas sodium chloride and sodium bromide are not.

Direct fluorination is usually explosive; special conditions are necessary
for the preparation of the fluorine derivatives of the paraffins (see p. 120).

2. Nitration (see also p. 302). Under certain conditions, paraffins react
with nitric acid, a hydrogen atom being replaced by a nitro-group, N0 2 -
This process is known as nitration. Nitration of the paraffins may be carried
out in the vapour phase between 150° and 475°, whereupon a complex mix-
ture of mononitroparaffins is obtained. The mixture consists of all the
possible mononitro-derivatives and the nitro-compounds formed by every

ALIPHATIC COMPOUNDS 55

possibility of chain fission of the paraffin; e.g., propane gives a mixture of
i-nitropropane, 2-nitropropane, nitroethane and nitromethane:

CH a -CH 9 -CH 3

HNO,

NO a

er

n

jiaraffin

n with

atoms

than

is very

is

converts

CH 3 -CH 2 -CH 2 -N0 2 + CH 3 -CH-CH 3 + C 2 H 5 -N0 2 + Crf 3 -N0 2

As in the case of halogenation, the various hydrogen atoms in propane are
not replaced with equal ease.

3. Sulphonation (see also p. 606) is the process of replacing a hydroge:
atom by a sulphonic acid group, S0 3 H. Sulphonation of a normal
from hexane onwards may be carried out by treating the paraffi:
oleum (fuming sulphuric acid). The ease of replacement of hydrogen
is : tertiary very much greater than secondary, and secondary great
primary; replacement of a primary hydrogen atom in sulphonation
slow indeed. isoButane, which contains a tertiary hydrogen atjom
readily sulphonated to give ter<.-butylsulphonic acid:

(CH 3 ) 3 CH + H 2 S0 4 /S0 3 — > (CH 3 ) 3 OS0 3 H + H 2 S0 4

Sulphuryl chloride, in the presence of light and a catalyst,
hydrocarbons into sulphonyl chlorides (p. 339).

It has been shown that in concentrated sulphuric acid, hydrocarbons contain-
ing a tertiary hydrogen atom undergo hydrogen exchange (Ingold et al\ 1936).
The mechanism is believed to occur via a carbonium ion :

R 3 CH + 2H 2 S0 4 > R 3 C+ + HS0 4 " + SO a + 2H a O

R 3 C+ + R 3 CH > R 3 CH + R 3 C+ etc.

This reaction is of particular interest since optically active hydrocarbons have
been racemised in sulphuric acid (see p. 412) ; e.g., Burwell et al. (i94p) have
shown that optically active 3-methylheptane is racemised in sulphuric acid.

4. Oxidation. All paraffins readily burn in excess of air or oxygen to form
carbon dioxide and water. Incomplete oxidation, due to insufficient air,
produces carbon-black in variable yields. The mechanism of the oxidation
of paraffins in the vapour state appears to take place via the formation of a
hydrocarbon peroxide; eg.,

R-CH 3 + O a — > R-CH 2 OOH — > R-CHO + H 2

Other products are also obtained by fission of the carbon chain.

Oxidising reagents such as potassium permanganate readily oxidise
tertiary hydrogen atom to a hydroxyl group, e.g., wobutane is oxidised
tert. -hutzxiol:

(CH 3 ) 3 CH + [O] -^^-> (CH 3 ) 3 -COH

„OH

The catalytic oxidation of methane produces methanol, CH 3 <
formaldehyde, H'CHO. The catalytic oxidation of higher horrjologui
(C 16 — ) produces long-chain fatty acids and some other products

5. Isomerisation of M-paraffins into branched-chain paraffins in which
side-chain is a methyl group, may be brought about by heating
paraffin with aluminium chloride at 300 °, e.g., M-hexane
2- and 3-methylpentanes:

isomerisms

CH,-CH,-CH 2 -CH 2 -CH.-CH,

AK3,
300*

CH,

CH a

a
to

and
es

the

the n-

intb

CH 3 -CH-CH 2 -CH 2 -CH 3 + CH 3 -CH 2 -CH-CH 2 -CH 3

56 ORGANIC CHEMISTRY

According to Pines et al. (1946), this isomerisation does not occur unless a
trace of water is present (to form HC1 from the A1C1 3 ) together with a trace of
alkyl halide or an olefin. The isomerisation is believed to be an ionic chain
reaction. The olefin " impurity " is converted into a carbonium ion (by the
AICI3 and HC1), and this initiates the chain reaction, isomerisation then occurring
by 1,2-shift (p. 101).

Instead of writing out the formulas of the 2- and 3-methylpentanes as shown in
the equation, an alternative way is to write the formula in a " straight " line,
enclosing in parentheses any side-chain; thus: CH 3 'CH(CH 3 )'CH2*CH 2 'CH3
and CH 3 "CH 2 "CH(CH3) # CH 2 'CH 3 . The atoms or groups in parentheses are
joined directly to the preceding carbon atom in the chain not placed in
parentheses, e.g.,

/OH
CH 3 -C— CH( may be written as CH 3 -C(:N-OH)-CH(OH)-CH a

I! X CH 3

N-OH

In many cases where there is no ambiguity, the parentheses may be omitted,
e.g., isopropanol is CH 3 -CH>CH S ; this is often written CH g *CHOH*CH 3 ,

OH
but if the rules are strictly adhered to, it should be written CH 3 *CH(OH)-CH 3 .
6. The thermal decomposition of the paraffins (see cracking).

PETROLEUM AND NATURAL GAS

Crude petroleum (mineral oil) is the term usually applied to the gases
occurring naturally in the oilfields, the liquid from the wells, and the solids
which are dissolved in, or have separated from, the liquid. The composition
of crude petroleum varies with the locality of occurrence, but all contain
paraffins (from about C x to C 4? ), cyc/oparaffins or naphthenes, and aromatic
hydrocarbons. The low-boiling fractions of almost all petroleums are
composed of paraffins; it is the composition of the higher-boiling fractions
which differs according to the source of the petroleum. In addition to
hydrocarbons, there are also present compounds containing oxygen,
nitrogen, sulphur and metallic constituents.

Natural gas is the term applied to the large quantities of gas associated
with or unassociated with liquid petroleum. The composition of natural
gas varies with the source, and consists chiefly of the first six paraffins,
the percentage of each decreasing with increasing molecular weight. Other
gases such as water vapour, hydrogen, nitrogen, carbon dioxide and hydrogen
sulphide may be present in amounts that vary with the locality of occurrence.

The origin of petroleum and natural gas is still uncertain. Many theories
have been suggested, but not one explains all the known facts. There is now,
however, general agreement that petroleum has organic origin, and this is due
to the fact that the higher boiling fractions of petroleum contain optically active
compounds (p. 399) and that petroleum has been shown to contain both animal
and plant type of porphyrins, i.e., haemin and chlorophyll (Treibs, 1934-36}.
The problem is how organic matter is converted into petroleum. A very highly
favoured theory is that it takes place by means of bacterial decomposition, but
there is a growing belief that bacterial action is only the first stage in the con-
version and this is then followed by physical and chemical stages.

Another highly favoured theory is that petroleum is formed from organic
matter by the catalytic activity of certain natural inorganic compounds. There
is a great deal of experimental work that supports this theory which, in many
ways, appears superior to any other. It is probable that both mechanisms
are operating.

ALIPHATIC COMPOUHDS

57

Distribution and general composition of crude petroleum. If the residue of
petroleum, after removal of volatile compounds, contains a large amount of
paraffins or wax, the petroleum is classified as paraffinic or paraffin base oil. If
naphthenes predominate, the petroleum is classified as asphaltic or asphalt base
oil. The crudes from the wells in Pennsylvania, Iran, Irak and Rumania are
paraffinic; those from Baku and Venezuela are asphaltic; and those from
Oklahoma, Texas and Mexico are intermediate in composition, and may be
classified as paraffinic and asphaltic.

Distillation of petroleum. The crude oil is nearly always associated with water
and sand; hence the crude petroleum discharged from the top of the well contains
water and sand in suspension. The mixture is passed, under pressure, into
cylindrical tanks, and the gas, oil and solids are drawn off separately.

Except for the low-boiling hydrocarbons, no attempt is made to separate the
individual hydrocarbons. The crude oil is fractionated by continuous distillation
into four main fractions: petrol {gasoline), kerosene (kerosine, paraffin oil), gas oil
(heavy oil) and lubricating oiL The residue may be fractionated by means of
vacuum-distillation to give light, medium and heavy lubricating oils, paraffin
wax, and asphaltic bitumen. Each of the four main fractions may be further
split up by batch distillation into fractions of narrow boiling range. Recently, it
has been possible to isolate individuals by " superfractionation ". The final
number of fractions taken depends on the purpose in view.

Table II shows one set of fractions that may be obtained.

Refining of the various fractions. It appears that refining was originally
introduced to remove the bad colour and objectionable odour of petrol. To-day
it is realised that it is more important to remove sulphur compounds which lower
the response of petrol to added tetraethyl-lead.

An internal-combustion engine, i.e., one which burns fuel within the vjorking
cylinder, is more efficient the higher the compression ratio. Petrol engines use
" spark ignition ", and as the compression ratio increases, a point is reached
when " knocking " is observed, i.e., after passage of the firing spark, instead of

TABLE II

Name.

B.P. ° C.

Approximate
Composition.

Uses.

Light petrol .

Benzine

Ligroin

Petrol (gasoline) .

Kerosene (paraffin oil)

Gas oil (heavy oil) .

Lubricating oil (mineral oil)

Greases, vaseline, petrolatum

Paraffin wax (hard wax)

Residue (asphaltic bitumen)

20— ioo
70-90
80-120
70-200
200-300
above 300

c.-c,
c, c xl

C12 — *-*l«
C l3 - C 18

C lg — C 2 2

Solvent
Dry cleaning
Solvent
Motor fuel
Lighting
Fuel oil
Lubricants
Pharmaceutical

parations
Candles, waxed

etc.
Asphalt tar;

leum coke

pre-
paper,
petro-

all the fuel gas burning smoothly, the end portion burns with explosive violence,
giving rise to a metallic rattle. The phenomenon of knocking is still not fully
understood, but it has been found that, among other factors, the tendency to
knock depends on the nature of the petrol. ^-Paraffins tend to produce knock-
ing far more than branched-chain paraffins. Edgar (1927) introduced ■>. : 2 : 4-
trimethylpentane (incorrectly known as iso-octane), which has higher antiknock
properties, and w-heptane, which has lower antiknock properties than any
commercial petrol, as standards for rating fuels. " iso-Octane " is arbitrarily
given the value of 100 and w-heptane, o, and the octane number of any fuel is the
per cent, of " iso-octane " in a mixture of this compound and w-heptane which
will knock under the same conditions as the fuel being tested.

Olefins and aromatic compounds have high octane numbers. Tetraetiyl-lead
also raises the octane number of a given petrol, but if sulphur compounds are

58 ORGANIC CHEMISTRY

present, the response to this " dope " is lowered. Hence it is very important
to remove sulphur compounds from petrol. The method of refining depends on
the particular fraction concerned, and it is not practicable to refine before
distillation (of the petroleum).

Gasoline refining. I. Petroleum may be treated with concentrated sulphuric
acid which reduces the sulphur content and also removes unsaturated compounds
which polymerise on standing to form gums (see olefins). For straight-run
gasoline, i.e., gasoline obtained directly from crude petroleum, 98 per cent,
sulphuric acid is used; for cracked gasoline (see later), 80 per cent, acid is used.
This diluted acid removes only unstable unsaturated hydrocarbons, i.e., those
which tend to polymerise, leaving the stable unsaturated hydrocarbons, which
raise the octane number of the gasoline. The diluted acid, however, causes
some polymerisation to take place, but the gasoline is readily separated from
these high-boiling polymers by distillation.

2. Instead of sulphuric acid the adsorption process can be used to remove
thioalcohols (the chief group of sulphur compounds occurring in petroleum)
from straight-run gasoline. The gasoline vapour is passed, under pressure, over
an adsorbent such as clays, bauxite, etc., heated at about 450 C. Cracked
gasoline contains sulphur as thiophens, and these cannot be removed so easily
this way. Thiophens, however, are not so objectionable as thioalcohols.

3. Straight-run gasolines may be refined by sodium hydroxide washing, which
may remove almost all thioalcohol sulphur. Where this simple sodium
hydroxide treatment is insufficient, it is followed by " sweetening ". By
sweetening, the thioalcohols (which give gasoline an unpleasant odour) are
converted into disulphides, thereby improving the odour. Common sweetening
agents are:

(i) An alkaline solution of sodium plumbite (" doctor solution ") :

2RSH + NajPbOa + S >■ R— S— S— R + PbS + zNaOH

Free sulphur is added as required in carefully controlled amounts. The spent
doctor solution is regenerated by blowing with air.

(ii) Sodium hypochlorite :

NaOC!

2RSH + [O] l_rr-> R— S— S— R + H a O
(iii) Cupric chloride:

2RSH + 2CuCl a >- R— S— S— R + 2CuCl + 2HCI

Sweetening of " sour " gasoline does not appreciably alter the total sulphur
content, and will not improve the octane number or lead susceptibility; in fact
sweetened gasoline may have a lower octane number and lead susceptibility than
unsweetened ; only the odour is improved.

4. The solutiser process involves the use of solvents, and the more common
ones are a methanolic solution of sodium hydroxide, sodium hydroxide and
sodium butyrate, and potassium hydroxide and potassium butyrate. By this
means the thioalcohols are removed completely, and thus the octane number
and lead susceptibility are raised.

Kerosene refining, (i) The kerosene is washed first with sulphuric acid, then
with sodium hydroxide solution, and finally with water, (ii) The kerosene is
treated with liquid sulphur dioxide, which removes most of the sulphur com-
pounds and aromatic hydrocarbons. Because of the removal of the latter, this
method of refining cannot be used for gasoline.

Gas oil and lubricating oil refining is carried out by extraction with liquid
sulphur dioxide.

CRACKING

The thermal decomposition of organic compounds is known as pyrolysis;
pyrolysis, when applied to paraffins, is known as cracking.

When heated to about 500-600 , paraffins are decomposed into smaller
molecules, and the products obtained from a given paraffin depend on:
(i) the structure of the paraffin; (ii) the pressure under which cracking is

ALIPHATIC COMPOUNDS 59

carried out; and (iii) the presence or absence of catalysts such as silica-
alumina, sUica-alumina-thoria, sihca-alumina-zirconia.

The mechanism of cracking is still obscure. Many theories have been
suggested, and one that is highly favoured is a free-radical mechanism,
evidence for which has been obtained from the observation that at cracking
temperatures many hydrocarbons produce free alkyl radicals.

When petroleum is cracked, of all the compounds produced, the most
important are those containing up to four carbon atoms: methane, ethane,
ethylene, propane, propylene, butane, butylene and isobutylene. All of
these have found wide application as the materials for the preparation of a
large number of chemicals (see text).

By using suitable catalysts, paraffins containing six or more carbon atoms
may be catalytically cyclised, e.g., w-hexane, under pressure, passed over
chromic oxide carried on an alumina support and heated at 480-550 , gives
benzene (see also p. 500) :

C 6 H 14 — > C 6 H 6 -f 4H2

There are two main types of cracking: (i) liquid phase, and (ii) vapour
phase.

(i) Liquid phase cracking. Heavy oil (from the petroleum distillation)
is cracked by heating at a suitable temperature (475-530°) and under
pressure (100-1000 lb/sq. in.), by means of which the cracked material is
maintained in the liquid condition. The heavy oil is converted into gasoline
to the extent of 60-65 P er cent - 0I tne ou (by volume), and has an octane
number of 65-70. If attempts are made to increase the yield of gasoline,
the octane number decreases.

(ii) In vapour phase cracking the cracking temperature is 6oo° and the
pressure is 50-150 lb/sq. in. The cracking stock may be gasoline, kerosene,
gas oils, but not the heavy oils, since these cannot be completely vaporised
under the above conditions.

Reforming is the process whereby straight-run gasoline is cracked in
order to raise the octane number. The gasoline is heated to about 6oo°
and under pressure of 400-750 lb/sq. in., and the yield varies from 60 to
90 per cent. ; the greater the yield, the lower is the octane number. Catalysts
— the oxides of silicon and aluminium, plus small amounts of other oxides
such as magnesia, zirconia, etc. — are usually employed in the reforming
process, and their use produces a higher octane number (which is partly
due to the increased content of benzene and toluene), and also increases the
yield of gasoline.

Catalytic cracking is also increasing in use, since it has been found that
catalytically cracked gasoline contains few olefins that readily polymerise.
Gum formation in cracked gasolines is prevented by the addition of inhibitors,
which are mainly phenols or aromatic amines, e.g., catechol, p-benzyl-
aminophenol, naphthylamine.

As pointed out above, large quantities of gases up to C 4 (and small amounts
of C s , i.e., the pentanes and pentenes) are produced in cracking. These
gases may be used as the starting materials for various chemicals. Alter-
natively, by polymerising the olefins under the influence of a catalyst, e.g.,
phosphoric acid on kieselguhr, or sulphuric acid, a high-octane (80-85)
gasoline can be obtained.

Treatment of natural gas. When natural gas does not contain hydrocarbons
above ethane, it is said to be " lean " or " dry "; when it .contains the higher
hydrocarbons (up to hexane) it is said to be " rich " or " wet ". The paraffins
may be separated by fractional distillation under increased pressure, thereby
giving methane, ethane, propane, n- and isobutanes, n-, iso- and weopentanes,
and hexane. These gases are used for various purposes (see text). On the other

60 ORGANIC CHEMISTRY

hand, natural gas itself may be used in the manufacture of various compounds.
Oxidation of natural gas under carefully controlled conditions produces a
complex mixture of compounds, among which are formaldehyde, acetaldehyde,
acetic acid, acetone, methanol, ethanol, propanols and butanols. These are
separated by distillation, solvent extraction, etc.

Wet gas is also used as a source of gasoline. The vapours of the liquid hydro-
carbons (pentanes and hexane) in the wet gas are removed by various methods,
e.g., compression, and cooling. The liquid product obtained from wet gas is
known as " natural " or " casinghead " gas, and is " wild " because of the
dissolved gases in it. These gases may be removed by distillation under pressure,
and the resulting liquid is known as " stabilised natural gasoline "; this has very
high antiknock properties.

Synthetic Fuels, (i) Fischer-Tropsch Gasoline Synthesis or Synthine Process.
Synthesis gas, which is water-gas mixed with half its volume of hydrogen, at
about 200-300 and a pressure of 1-200 atm., is passed over a catalyst.

;vCO + yH z >■ mixture of hydrocarbons + water

(saturated and unsaturated)

The water-gas is made from coke. One third of the water-gas is passed with
steam at 400 over iron, and the hydrogen-enriched water-gas is then mixed with
the rest of the water-gas to produce the synthesis gas. This contains about
20 per cent, of inert gases. If the synthine process is carried out at atmospheric
pressure, the carbon dioxide (about one third of the volume of the inert gases) is
left in; if the synthesis is carried out under pressure, most of the carbon dioxide
is washed out.

Synthesis gas contains sulphur compounds, and since these poison the catalyst,
their removal is necessary. Hydrogen sulphide is removed by bog-iron ore.
Organic sulphur compounds are oxidised under carefully controlled conditions,
the sulphur being retained as sodium sulphate. If the synthesis gas contains
large amounts of hydrocarbons, the latter are almost completely removed by
passing through active carbon adsorbers before removing the sulphur com-
pounds. The purified gas is then passed to the catalyst chambers, with or
without compression. Other methods of preparing synthesis gas are also
available, e.g., from methane and steam (p. 41).

Various metals and oxides have been used as catalysts. One of the best is:
cobalt (100 parts), thoria (5 parts), magnesia (8 parts) and kieselguhr (200 parts).
When synthesis gas is passed over the catalyst at moderate pressure (9-11 atm.),
more of the high-boiling fraction is obtained than when the process is carried out
at atmospheric pressure. The liquid products are fractionally distilled, and
refined in the same way as are the petroleum fractions ; furthermore the higher-
boiling fractions are cracked.

Gasoline from the synthine process costs more than that from petroleum. The
Fischer-Tropsch oils appear to be more valuable as chemical raw materials than
as fuels, and have been used for the production of higher olefins, fatty acids,
detergents and in the ow-process (see p. 127).

(ii) Petrol from coal, (a) Distillation of coal-tar gives a fuel oil, fractionation
of which yields petrol (about one sixth of the volume). On the other hand,
hydrogenation of the fuel oil under a pressure of 200 atm. and at about 475
produces petrol in 100 per cent, yield.

(6) In the Bergius process coal dust is heated to 400-500 in hydrogen at
250 atm., preferably in the presence of a catalyst, one of the best being an
organic compound of tin. The yield of petrol may be as high as 60 per cent, (on
the coal used).

(c) In the I.C.I, process coal dust is mixed with heavy oil to form a paste
{50 per cent, of oil), which is pumped, with hydrogen, under pressure (250 atm.),
into chambers containing the catalyst (organic tin compound) heated at 450 .
The gases produced are scrubbed and condensed, and the liquid fractions are
distilled to give the petrol fraction. The higher-boiling oils may be further
hydrogenated to give more petrol.

ALIPHATIC COMPOUNDS 6l

QUESTIONS
i. Write out the structures and names (by the three methods described in the text)
of the isomeric hexanes. State how many primary, secondary, tertiary and quaternary
carbon atoms there are in each isomer.

2. By means of equations, show how you would convert methane into propane.

3. What is the percentage of carbon and hydrogen in the paraffin C 30 H, a ? Could
you distinguish this paraffin from its next homologue by determining the percentage
composition of each hydrocarbon?

4. Synthesise all the alkanes you can, by methods dealt with so far, from (a)
CH a -CH 2 -CH(OH)-CH 3 ; (6) CH s -CH,-CH,-CH 2 -CO a H.

5. Define and give examples of: — (a) isomerism, (b) substitution, (c) homologous
series, (d) cracking, (e) nitration, (/) sulphonation, (g) decarboxylation.

6. What reagents could you use to convert: — (a) a monohalogen derivative of a
paraffin and (6) carbonyl compounds, into the corresponding and higher paraffins?

7. Write notes on: — (a) the Wurtz reaction, (b) Kolbe's electrolytic method, (c)
Frankland's method, {d) Fischer-Tropsch reaction.

The reader should test himself on the general methods of preparation and general
properties of any homologous series.

READING REFERENCES

Handbook for Chemical Society Authors. Special Publication No. 14. The Chemical
Society (i960).

Rules for I.U.P.A.C. Notation for Organic Compounds, Longmans, Green (1961).

Beezer, Latin and Greek Roots in Chemical Terminology, /. Chem. Educ, 1940, 17, 63.

Hurd, The General Philosophy of Organic Nomenclature, /. Chem. Educ, 1961, 38, 43.

Gilman, Advanced Organic Chemistry, Wiley (1942, 2nd ed.), (i) Vol. I, Ch. 1. Reactions
of the Paraffins, (ii) Vol. II, Ch. 23. Constitution and Physical Properties of
Organic Compounds.

Advances in Organic Chemistry, Vol. I (i960). Weedon, The Kolbe Electrolytic Syn-
thesis, p. 1.

Goldstein, The Petroleum Chemicals Industry, Spon (1949).

Brooks, The Chemistry of the Non-benzenoid Hydrocarbons, Reinhold (1950, 2nd ed.j.

Astle, The Chemistry of Petrochemicals, Reinhold (1956).

Rossini, Hydrocarbons in Petroleum, /. Chem. Educ, i960, 37, 554.

CHAPTER IV

UNSATURATED HYDROCARBONS

Olefins, Alkylenes or Alkenes

The olefins are the unsaturated hydrocarbons that contain one double
bond. The simplest member of the series is ethylene, CH 2 =CH 2 ; hence
this homologous series is often referred to as the " ethylene series ". The
olefins have the general formula CbH^, and the double bond is also known as
the " olefinic bond " or " ethylenic bond ".

Olefins have recently become very important technically, since they are
obtained in huge quantities in the cracking of petroleum, and may be used
to prepare a large variety of organic compounds (see text).

Nomenclature. The name olefin arose from the fact that ethylene was
called " defiant gas " (oil-forming gas), since it formed oily liquids when
treated with chlorine or bromine. The original name given to this homo-
logous series was olefine; but it was later decided to reserve the suffix -ine
for basic substances only. Since the name olefine had gained wide usage,
it was decided to compromise and call the series the olefins.

One method of nomenclature is to name the olefin from the corresponding
paraffin by changing the suffix -ane of the latter into -ylene, e.g., (methylene),
ethylene, propylene, etc.

The name alkylene is obtained in a similar manner, aikane being converted
into alkylene.

Isomers differing only in the position of the double bond are prefixed by
Greek letters or numbers which indicate the position of the double bond.
The lowest number is usually given to the double bond (see below), and
the number (or Greek letter) indicates the first of the two carbon atoms that
are joined together by the double bond, e.g.,

CH 3 -CH 2 -CH=CH 2

a-Butylene

or i-Butylene

CHg* C rl === CH *CH 3

(3-Butylene

or 2-Butylene

(CH 3 ) 2 C^=CH a

isoButylene

(CH^C^CH-CHs

(J-isoAmylene

{cf. pentane and amyl)

Another method of nomenclature is to consider ethylene as the parent
substance and the higher members as derivatives of ethylene. If the com-
pound is a monosubstituted derivative of ethylene, then no difficulty is
encountered in naming it; if the compound is a disubstituted derivative
of ethylene isomerism is possible, since the alkyl groups can be attached to
the same or different carbon atoms. When the groups are attached to the
same carbon atom the olefin is named as the asymmetrical or unsymmelrical
compound (abbreviated to as- or unsym-); when attached to different
carbon atoms the olefin is named as the symmetrical (sym- or s-) compound,
e-g-,

CH 3 -CH=CH 2 Methylethylene

CH 3 -CH=CH-CH 3 s-Dimethylethylene

(CH 3 ) 2 C=CH 2 as-Dimethylethylene

(CH 3 ) a C=CH-CH 3 Trimethylethylene

According to the I.U.P.A.C. system of nomenclature, the class suffix of the
olefins is -ene, and so the series becomes the alkene series. The longest
carbon chain containing the double bond is chosen as the parent alkene,

62

UNSATURATED HYDROCARBONS 63

the name of which is obtained by changing the suffix -ane of the correspond-
ing paraffin into -ene. The positions of the double bond and side-chains are
indicated by numbers, the lowest number possible being given to the double
bond, and this is placed before the suffix, e.g.,

CH 3 -CH 2 -CH=CH 2 but-i-ene

CH 3 -CH=C(CH 3 )-CH 2 -CH 3 3-methylpent-2-ene

When there are several chains of equal length containing the double bond,
then the same principles apply as for the alkanes (p. 47), e.g.,

4

CH 3 'CH 2 -C-CH( * 2-ethyl-3-methylbut-i-ene

4, ^

A double or triple bond is regarded as a functional group. When there
are several functional groups in the molecule, then the lowest number possible
is given, in order of preference: (i) to the principal functional group of the
compound; (ii) to the double or triple bond; and (iii) to atoms or groups
designated by prefixes.

General Methods of Preparation of the Olefins. 1. By the action of con-
centrated sulphuric acid, at 160-170 °, on primary alcohols. The acid acts
as a dehydrating agent, removing one molecule of water from the alcohol
to form the olefin, e.g., ethylene from ethanol:

C 2 H 5 OH -^-> C 2 H 4 (f.-g.)

Dehydration of secondary and tertiary alcohols is best carried out using
dilute sulphuric acid, since the olefins produced from these alcohols (particu-
larly tertiary alcohols) tend to polymerise under the influence of the con-
centrated acid. The yields of olefin from secondary and tertiary alcohols
are very good.

Instead of sulphuric acid, glacial phosphoric acid (HP0 3 ), phosphorus
pentoxide, or alumina may be used. With alumina, at 350 , the yields are
v.g.-ex. Brandenberg et al. (1950) have converted all three classes of alcohols
into olefins in excellent yields by means of boric acid as catalyst (borates are
formed as intermediates) :

3R-CH a -CH 2 OH + H s BO» — > (R-CH 2 -CH 2 0) 3 B — ^

3 R-CH=CH 2 + H 3 B0 3

Methyl xanthates may also be used to prepare olefins (p. 131).
2. By the action of ethanolic potassium hydroxide on monohalogen.
derivatives of the paraffins, e.g., propylene from propyl bromide:

CH a -CH 2 -CH 2 Br + KOH ethano1 > CH 3 -CH=CH 2 + KBr"+ H 2

This is not a very important method for the preparation of the lower alkenes,
since these may be prepared directly from the corresponding alcohols, which
are readily accessible. The reaction, however, is very important, since by
means of it a double bond can be introduced into an organic compound
(see text). The yield of olefin depends on the nature of the alkyl halide
used; it is fair with primary, and very good for secondary and tertiary
alkyl halides. Ethylene cannot be prepared by this method from ethyl
halide (see alkyl halides for the mechanism of dehydrohalogenation, i.e.,
removal of halogen acid).

64 ORGANIC CHEMISTRY

3. (i) By the action of zinc dust on methanolic solutions of gew-dihalogen
derivatives of the paraffins, e.g., propylene from propylidene bromide:

CH 3 -CH 2 -CHBr 2 + Zn — > ZnBr 2 + [CH 3 -CH 2 -CH<] — ^CH,-CH=CH 2

If sodium is used instead of zinc, and the reaction is carried out preferably
in ether solution, comparatively little propylene is formed, the main product
being hex-3-ene:

2CH 3 -CH 2 -CHBr 2 + 4Na — > CH 3 -CHyCH=CH-CH 2 -CH 3 + 4NaBr

This reaction is really an extension of the Wurtz synthesis, and the im-
portant point to note is that the use of sodium tends to produce lengthening
of the carbon chain.

(ii) By the action of zinc dust on methanol solutions of w'c-dihalogen
derivatives of the paraffins, e.g., propylene from propylene bromide:

CH 3 -CHBr-CH 2 Br + Zn — y CH 3 -CH=CH 2 + ZnBr 2

Sodium can also be used, but zinc dust is usually more satisfactory.

Neither (i) nor (ii) is used very much for preparing alkenes, since the
necessary dihalogen compounds are not readily accessible. The method,
however, is very useful for purifying alkenes or for " protecting " a double
bond (see, e.g., allyl alcohol). Sodium iodide may be used instead of zinc
dust, and its use depends on the fact that the w'c-di-iodide which is formed
is unstable, and readily eliminates iodine to form a double bond:

>CBr-CB< + 2NaI — > 2NaBr + [>CI-CI<] — > >C=C< 4- I*

4. By heating a quaternary ammonium hydroxide, e.g., ethylene from
tetraethylammonium hydroxide :

(C 2 H 6 ) 4 NOH — > C 2 H 4 + (C 2 H 6 ) S N + H 2

This method of preparation is more important as a means of ascertaining
the structure of a compound containing nitrogen in a ring, and is the basis
of the Hofmann Exhaustive Methylation reaction (see p. 318).

5. Boord et al. (1930-33) have prepared olefins by conversion of an
aldehyde into its chloro-ether, treating this with bromine followed by a
Grignard reagent, and finally treating the product with zinc and »-butanol.

R-CH 2 -CHO °^ H > R-CH 2 -CHC1-0C 2 H B -^» R-CHBr-CHBr-OC 8 H 8

R-MgB - > - R-CHBr-CHR'-OC 2 H s Za > R-CH=CH-R'

* 6 C.H.OH

This method is very useful for preparing olefins of definite structure, and an
interesting point about it is the replacement of the a-chlorine atom by
bromine when the a-chloro-ether undergoes bromination in the p-position.

6. The Wittig reaction (1953, 1956). This is also a means of preparing
olefins where the position of the double bond is definite. An alkyl (or
aralkyl) triphenylphosphonium halide is treated with, e.g., sodium ethoxide
and the alkylidene- (or arylidene-) phosphorane produced is then warmed
with an aldehyde or ketone (Ph = C 6 H S ) :

Ph 3 P 4- RCH 2 Br — =* Ph 3 PCH 2 R} + Br- Et0Na >

Ph 3 P=CHR-^% R' 2 -C-CHR.pPh 8 — >

i-

R' 2 C=CHR + Ph 8 P=0

UNSATURATED HYDROCARBONS 65

7. A number of olefins are prepared by the cracking of petroleum (p. 58),
e.g., ethylene, propylene, butylenes, etc. (see also the individuals). For the
production of the lower olefins the most suitable starting material is gas oil,
whereas for the higher olefins it is best to use paraffin wax or Fischer-Tropsch
wax (p. 60). The lower olefins (C a — C 6 ) are also prepared by the catalytic
dehydrogenation of saturated hydrocarbons, the most satisfactory catalysts
being those of the chromium oxide— alumina type.

General properties of the olefins. The members containing two to four
carbon atoms are gases; five to fifteen, liquids; sixteen onwards, solids at
room temperature. All are lighter than water, in which they are insoluble,
and they burn in air with a luminous smoky flame.

Owing to the presence of a double bond, the olefins undergo a large number
of addition reactions, but under special conditions they also undergo sub-
stitution reactions. The high reactivity of the olefinic bond is due to the
presence of the two Tt-electrons. These are less firmly held between the two
nuclei than the {^electrons, and are more exposed to external influences, and
so are readily polarisable. It is the ^-electrons which undergo the electro-
meric effect at the requirements of the attacking reagent, and when addition
occurs, the trigonal arrangement in the olefin changes to the tetrahedral
arrangement in the saturated compound produced (see p. 429).

1. Olefins are readily hydrogenated under pressure in the presence of a
catalyst. Finely divided platinum and palladium are effective at room
temperature; nickel on a support (Sabatier-Senderens reduction) requires
a temperature between 200 and 300°; Raney nickel is effective at room
temperature and atmospheric pressure:

G a H 4 -)- H 2 — >- C 2 H 6

Platinum and palladium-black, i.e., the metals in a very finely divided
state, may be prepared by reducing their soluble salts with formaldehyde.
Adams' platinum-platinum oxide catalyst is prepared by reducing platinum
oxide with hydrogen before the addition of the compound being hydrogen-
ated, or it may be added to the compound, reduction of the oxide taking place
during hydrogenation.

One molecule of hydrogen is absorbed for each double bond present in
the unsaturated compound.

The olefinic bond is readily reduced catalytically, but it is not reduced by
metals and acid, or sodium and ethanol, unless the double bond is in the
afJ-position with respect to certain groups (see text).

The mechanism of catalytic hydrogenation is not yet fully understood. It
appears that adsorption occurs to give metal-carbon and metal-hydrogen bonds.
In this way the catalyst lowers the energy of activation.

2. Olefins form addition compounds with chlorine or bromine, e.g., ethylene
adds bromine to form ethylene bromide:

CH 2 =CH 2 + Br 2 — ^ CH 2 Br-CH 2 Br (85%)

Addition of halogen to olefins can take place by two types of mechanism, polar
or free-radical.

Polar mechanism. Evidence for this mechanism is as follows. Stewart et al.
(1923) showed that the reaction between ethylene and bromine (in the absence
of light) occurs only at the surface of the reaction vessel. Norrish (1923)
showed that a polar surface was necessary. When the walls of the container were
coated with paraffin wax (a non-polar substance), the rate of reaction between
ethylene and bromine was very much reduced, whereas when the walls were
coated with stearic acid (a polar substance), the reaction rate was very much
increased. Norrish et al. (1926), examining the addition of chlorine to ethylene,
showed that a small amount of water vapour catalyses the reaction when the

D

66 ORGANIC CHEMISTRY

walls of the container were " bare ", but had no effect when the walls had been
previously coated with paraffin wax. Water vapour can be adsorbed on bare
glass but not on wax-coated glass. All these experiments lead to the conclusion
that reaction occurs at the surface and takes place by a polar mechanism. That
a polar mechanism is operating is supported by the fact that the reaction is also
catalysed by inorganic halides such as aluminium chloride, etc. The addition of
halogen to olefins may also be carried out in a suitable solvent, e.g., chloroform,
carbon tetrachloride.

The next problem to be considered is whether in this addition reaction the
halogen behaves as an electrophilic or a nucleophilic reagent. All the evidence
has shown that reaction takes place in two stages, with the halogen behaving as
an electrophilic reagent, e.g., Francis (1925) showed that when ethylene reacts
with bromine in aqueous sodium chloride solution, the products are ethylene
dibromide and i-bromo-2-chloroethane; no ethylene dichloride is obtained. These
results are readily explained by the following mechanism :

y — 2I rv s+ s- +

CH 2 =CH 2 Br— Br ^=i CH^CHa Br Br — y CH 2 — CH 2 Br + Br~

T.S.

The bromine atom that adds on shares a pair of electrons from the ethylene
molecule, and in doing so releases its own bonding pair to the other bromine atom
(of the bromine molecule), the latter thus being released as a bromide ion. Since
the bromine atom that adds on first gains a share in the two electrons retained
by the ethylene molecule, bromine is thus an electrophilic reagent. The car-
bonium ion produced can now combine with any negative ion, and so, since both
bromide and chloride ions are available, both of these can add on to give the
final products :

Br-Q-CHa— CH 2 Br ^ B~r CH 2 — CH 2 Br >- CH 2 Br— CHJBr

T.S.

In the same way, the Cl~ adds on to give CH 2 Cl»CH 2 Br.

It has been shown, however, that the above mechanism for addition of halogen
to olefins satisfies many reactions but not others (Robertson et al., 1937- ).

The addition of halogen to a double bond is trans; this stereochemical aspect is
discussed on p. 428.

An interesting point about the above mechanism is why, with a symmetrical
olefin such as ethylene, the addition is not nucleophilic, i.e.,

Br— Br CH 2 =CH 2 y CH 2 Br— CH 2 + Br+ — > CH 2 Br-CH 2 Br

Several reasons have been proposed to explain electrophiUc attack, e.g., negative
ions (nucleophilic reagents) are hindered from attacking ethylenic carbon atoms
by the screen of ^-electrons. In fact, ^-electrons are very susceptible to attack
by electrophilic reagents. Thus olefins are themselves nucleophilic reagents;
e.g., many olefins form addition complexes with the silver ion as, e.g., perchlorate.
These addition compounds are known as ^-complexes (p. 70) .

(CH 3 ) 2 C=C(CH 3 ) 2 + Ag+ y (CH 3 ) 2 GjC(CH :

Ag

fC(CH 3 ) 2 |

+ >cio 4 -

Lg + J

Free-radical mechanism. Under suitable conditions, halogens may add to
olefins by a free-radical mechanism, e.g., Stewart et al. (1935) have shown that
the addition of chlorine to ethylene is accelerated by light. This suggests a
free-radical mechanism.

■2C1-

CH 2 =CH 8 + CI > CH 2 C1— CH 2 -

CH 2 C1— CH 2 - + Cl 2 — ^CH 2 C1— CH 2 C1 + CI-
CH 2 =CH 2 + CI > CH 2 C1— CH 2 -, etc.

«?*.3^

UNSATURATED HYDROCARBONS 67

It may be asked why the free radical CH 2 C1*CH 2 - does not combine with
the other chlorine atom (free radical). There is, of course, always a chance
of this occurring, but since the concentration of the chlorine molecules is
infinitely greater, the reaction will therefore proceed as shown above. On
the other hand, since the concentration of the ethylene is high, it would appear
that the free radical CH 2 C1*CH 2 ' could react with ethylene molecules. If this
were to happen, polymerisation (or at least dimerisation) would take place.
There appears to be no evidence for this, and so, if the reaction is via free radicals,
we must suppose it to take place as shown. This type of reaction is known as a
free-radical chain reaction, and once started, carries on until the reactants are
used up, or the chain broken by the destruction of the free radicals (see below,
polymerisation) .

Instead of addition reactions with the halogens, the olefins may undergo
substitution provided the right conditions are used. Thus when straight-
chain olefins are treated with chlorine at a high temperature, they form

mainly monoehlorides of the allyl type, i.e., in the chain — C — C — C=C — ,

it is the hydrogen of C that is substituted (see allyl compounds); e.g.,
propylene heated with chlorine between 400 and 600° gives allyl chloride:

CH 3 -CH==CH 2 + Cl a — > CH 2 C1-CH=CH 2 + HC1

Above a certain temperature range, substitution takes place; below this
range, addition takes place. The temperature range varies according to
the olefin used, but for most olefins lies between 300 and 600°.

On the other hand, substitution is fairly easy with branched-chain olefins,
and again occurs in the allyl position (see, e.g., wobutene).

The action of fluorine on olefins usually results in the formation of carbon
tetrafluoride, but addition to the double bond may be effected by treating the
olefin with hydrogen fluoride in the presence of lead dioxide (Henne et al., 1945) ;
the fluorinating agent is lead tetrafluoride :

PbO a + 4HF — >■ PbF 4 + 2H 2

+ PbF 4 ■ — > ^CF-CF< + PbF 2

3. Olefins form addition compounds with the halogen acids, e.g., ethylene
adds hydrogen bromide to form ethyl bromide :

C 2 H 4 -f HBr — > C 2 H 5 Br

The order of reactivity of the addition of the halogen acids is hydrogen
iodide >hydrogen bromide >hydrogen chloride >hydrogen fluoride. The
conditions for the addition are similar to those for the halogens; the addition
of hydrogen fluoride, however, is effected only under pressure.

In the case of unsymmetrical olefins it is possible for the addition of the
halogen acid to take place in two different ways, e.g., propylene might add
on hydrogen iodide to form propyl iodide :

CH 3 -CH=€H 2 + HI — y CH 3 -CH 2 -CH 2 I

or it might form isopropyl iodide :

CH 3 -CH=CH 2 + HI — > CH 3 -CHI-CH 3

Markownikoff studied many reactions of this kind, and as a result of his
work, formulated the following rule: the negative part of the addendum adds
on to the carbon atom that is joined to the least number of hydrogen atoms.

68 ORGANIC CHEMISTRY

In the case, of the halogen acids the halogen atom is the negative part,
and so isopropyl halide is obtained.

Markownikoff 's rule is empirical, but may be explained theoretically on the
basis that the addition occurs by a polar mechanism. As with halogens, the
addition of halogen acid is an electrophilic reaction, the proton adding first,
followed by the halide ion. That the polar mechanism operates is supported by
much experimental work. One piece of evidence that may be cited is that
Hennion et al. (1939, 1941) have shown that the addition of hydrogen chloride
or bromide to, e.g., cyclohexene, is faster in hydrocarbon solvents such as heptane
than in nucleophilic solvents such as ether. These rate differences may be
explained by the fact that ether but not heptane can form oxonium ions with
protons and thereby greatly reduce the proton concentration :

(C a H 6 ) a O + HC1 >■ (C 2 H 6 ) 2 OH + }Cl-

This also indicates that the addition of the proton is the rate-determining step.
Thus:

S "al rV slow n + fast

CH 2 =CH 2 H— CI << CH 2 -CH 3 + Cl~ >- CH 2 C1-CH 3

Now consider the case of propylene. Since the methyl group has a +1
effect, the electromeric effect will be away from the methyl group. Thus the
proton adds on to the carbon farthest from the methyl group, and the halide ion
then adds to the carbonium ion :

s~± CV +
CH 3 — >- CH=CH 2 H— I ^=±: CH 3 -CH-CH 3 + I~ > CH 3 -CHI-CH 3

The Peroxide Effect (Kharasch, 1933) . The presence of oxygen or peroxides
that are formed when the olefin stands exposed to the air, or added peroxides
such as benzoyl peroxide, causes the addition of hydrogen bromide to take
place in the direction opposite to that predicted by Markownikoff 's rule.
This departure from the rule is known as the " abnormal " reaction, and
was shown to be due to the " peroxide effect " (Kharasch et al., 1933).
Hydrogen chloride, hydrogen iodide and hydrogen fluoride do not exhibit
the abnormal reaction. The abnormal reaction in the presence of peroxides
can be prevented by the addition of an " inhibitor " such as diphenylamine,
catechol, etc. It has been found that the addition of hydrogen bromide is
" abnormally " effected photochemically as well as by peroxide catalysts
(Vaughan, et al., 1942).

The mechanism of the peroxide effect is believed to be a free-radical chain
reaction, the peroxide generating the free radical R- (cf. polymerisation,
below) :

(R-C0 2 ) 2 — > 2RC0 2 >■ 2R- + 2C0 2

R- + HBr — >RH + Br-

R'-CH=CH 2 + Br ► R'-CH-CH 2 Br -^> R'-CH 2 -CH 2 Br + Br-, etc.

In the photochemical addition, the bromine atom is produced by a quantum
of light :

HBr -^-> H- + Br-

At least two explanations may be offered for the fact that the bromine
atom attacks the carbon atom not joined to the least number of hydrogen
atoms :

(i) Free halogen atoms are electrophilic reagents owing to their tendency
to complete their octets, and hence will attack the olefin at its point of highest
electron density. As we have seen above, the electromeric effect in olefins

UNSATURATED HYDROCARBONS 69

of the type R-CH=CH a takes place away from the CH group, owing to the
electron-repelling effect of the R group, i.e., we have:

R-CH=CH 2 — > R-CH— C*H 2

+ •"" • • HBr

R-CH— CH 2 -f Bi > R-CH— CH 2 Br — -> R-CH 2 -CH 2 Br + Br-, etc.

(ii) Each of two carbon atoms joined by the double bond retains its
^-electron. Thus a bromine atom can attack either carbon atom equally
well, but of the two free radicals that can be produced, viz., R-CH-CHgBr and
R-CHBr-CH 2 -, it is the former which has the lower free energy, and hence
this one is more likely to be formed.

Another problem is why only hydrogen bromide exhibits the peroxide effect.
The answer is possibly as follows. In HX, the bond strength order is HC1>
HBr>HI. Thus the H — CI bond is too strong to be broken (homolytically).
On the other hand, since the H — I bond is weaker than that of H — Br, one might
have expected HI to also exhibit the peroxide effect. The reason that it does
not is believed to be due to the fact that although HI is split into hydrogen and
iodine atoms, iodine atoms are not reactive enough to add on to a double bond,
but combine with each other to form iodine molecules, thereby continuously
breaking the chain reaction.

Tri- and tetra-halogenated methanes also add on to a terminal double bond in
the presence of peroxides (Kharasch et al., 1945- ). Here again the mechanism
is believed to be a free-radical chain reaction, e.g.,

(R-CO a ) 2 y 2R- + 2C0 2

R- + CHC1 3 > RH + -CC1 3

R-CH=CH 2 + -CClg >■ R-CH-CH 2 -CC1 3

— l*^.R-CH s -CH a -CCl3 + -CCl 3 ; etc.

4. Hypohalous acids add on to olefins to form halohydrins. Usually the
reaction is carried out by treating the olefin with chlorine- or bromine-
water.

The mechanism is believed to be (cf. p. 66) :

+ H.O
CH 2 =CH a + Br 2 ^ Br" + CH 2 Br-CH 2 >

CH 2 Br-CH 2 OH 2 > H+ + CH 2 Br-CH a OH

With unsymmetrical olefins, the hydroxyl group adds on to the carbon atom
joined to the least number of hydrogen atoms :

R-CH=CH„ CI— CI > C1-+R-CH-CH 2 C1 -^— > R-CHOH-CH a Cl

Aqueous solutions of hypohalous acid also add on to olefins in the presence of
strong acids to form halohydrins, e.g., the addition of hypochlorous acid to
ethylene. The mechanism is probably via the formation of the chlorinium ion:

ClOH + H + ^ C10H 2 ^=±: C1+ + H 2

CH,=CH 2 + C1+ — > CH 8 C1-CH 2 J^.>

+

CH 2 Cl-CH 2 OH 2 + ^~ H N CH 3 Cl-CH 2 OH

Various sulphenyl halides form adducts with olefins; 2 : 4-dinitrobenzene-
sulphenyl chloride in particular has been found extremely useful for identify-
ing olefins (Kharasch et al., 1949- ).

70 ORGANIC CHEMISTRY

The mechanism of the addition is possibly as follows, the sulphur being the
positive end of the dipole :

CH 2 — CH 2 + NO,/" \- S— CI — >- C1CH 2 -CH 2 -S^ %N0 2
~N0 2 N<5 2

Two products are obtained with unsymmetrical olefins, the predominating
adduct being the one formed in accordance with Markownikoff ' s rule.

Compounds containing triple bonds also form adducts with one molecule
of the sulphenyl chloride.

5. Olefins are absorbed by concentrated sulphuric acid to form alkyl
hydrogen sulphates. Addition takes place according to Markownikoff 's
rule, e.g., propylene reacts with sulphuric acid to form wopropyl hydrogen
sulphate :

CH 3 -CH=CH 2 H— OSO a -OH — >

CH 3 -CH-CH 3 + 6-S0 2 -OH — > (CH 3 ) 2 CHO-S0 2 -OH

Paraffins are not absorbed by cold concentrated sulphuric acid, and hence
may be separated from olefins (see also ethers and alcohols).

6. Hydration of olefins. Olefins may be hydrated to alcohols by absorp-
tion in concentrated sulphuric acid followed by hydrolysis of the alkyl
sulphate (see above section and p. 134). Olefins, however, may also be cata-
lytically hydrated in dilute acid solution, e.g., Lucas et al. (1934) showed
that wobutylene forms t-butanol in dilute acid solution :

+

(CH 3 ) 2 C=CH 2 + H 2 -^-> (CH 3 ) 3 COH

The mechanism of olefin hydration has been the subject of much discussion.
The hydration reaction, as such, cannot be examined kinetically, but since the
reaction is reversible, the principle of microscopic reversibility may be applied
(p. 32). Hughes and Ingold (1941) have examined the elimination of water from
t-alcohols, and on the basis of their results have proposed the following mechan-
ism for hydration :

S — ^ f~V slow \ +
Me a C=CH 2 H— OH a + n Me 2 C— Me + H 2

-- 5=i Me 2 C— Me ^=h±j> Me 2 OMe
I vtht- I

+OH 2 OH

Thus the first step (which is the rate determining step) ie protonation to form a
carbonium ion. Taft et al. (1952— 1954), however, have obtained evidence that
the classical carbonium ion (shown in the equation) is not formed first. Accord-
ing to these authors, a it-complex is formed first, and this is then converted into
the classical carbonium ion. In the re-complex the proton is not directly bound
to either carbon atom; the re-orbital overlaps the vacant hydrogen orbital.

Me 2 C Me 2 C Me 2 C +

\\ + H 3 0+ =?=i H a O + || — y H + — > I

CH 2 CH 2 CH 3

The problem, however, cannot be regarded as settled.

7. Olefins add on nitrosyl chloride, nitrosyl bromide and oxides of nitro-
gen; e.g., ethylene forms ethylene nitrosochloride with nitrosyl chloride:

CH 2 =CH a + N0C1 — > CH 2 C1-CH 2 -N0

Since the X atom is the negative end of the dipole in NOX, it will add
on to the carbon atom joined to the least number of hydrogen atoms (Mar-

UNSATURATED HYDROCARBONS

71

kownikoff 's rule), e.g., trimethylethylene adds on nitrosyl bromide to form
the following trimethylethylene nitrosobromide:

rv rv

(CH 3 ) 2 C=CH-CH 3 + NO— Br — > (CH 3 ) 2 CBr-CH(NO)-CH 3

The reaction with nitrosyl chloride is usually carried out by treating a
solution of the olefin and ethyl or pentyl (amyl) nitrite in glacial acetic acid
with concentrated hydrochloric acid, the temperature being maintained at
about io°. The nitrosochlorides (and nitrosobromides) are usually bimole-
cular crystalline solids, e.g.,

CH 2

JJ + NOC1 -

CH»

LCH 2 -i

NO"

CI

2 molecules

CH 2 -NO"
CH 2 C1

The addition of the oxides of nitrogen to olefins is complicated, and much
of the work done is of a doubtful nature. The compound formed depends
on the structure of the olefin and the nature of the "nitrous fumes";
usually a mixture of addition products is formed. According to Levy,
Scaife et al. (1946), when ethylene, propylene, and some other olefins react
•with dinitrogen tetroxide, N 2 4 , then according to the conditions dinitro-
paraffin, nitro-alcohols and nitro-alkyl nitrates can be obtained in high yield.
The reaction is best carried out in the liquid phase at — io° to +25°. Di-
nitro-compounds or nitro-nitrites are produced, but the latter are usually
partly oxidised to nitro-nitrate. The unchanged nitro-nitrite is unstable,
tending to explode, but it may be converted into the stable nitro-alcohol
when treated in the cold with water or a lower aliphatic alcohol. The
nitrogen tetroxide behaves as (i) two N0 2 groups to give dinitro-compounds,
and (ii) one N0 2 group and one ONO group (nitrite radical) to give nitro-
nitrites (see also nitro-compounds) :

(NO,)(ONO)

ROH

> >C C< > >C C<

<NO,)(NO,)

N0 2 O-NO

(nitro-nitrite)
oxidation

^C-

['

N0 2 OH

(nitro-alcohol)

<<

N0 2 NO a

(dinitro-paraffin)

N0 2 0-N0 2

(nitro-nitrate)

According to Schechter et al. (1953), the addition of dinitrogen tetroxide
occurs via the formation of the nitronium free-radical :

•NO,

2 N— NO,

•NO,

2 -NO,

Dinitrogen trioxide in ether adds on to olefins at —70 to 5 to give mainly
dimeric nitro-nitroso compounds:

>:=c< + NO-NO.

NO NO,

72 ORGANIC CHEMISTRY

If the olefin is unsymmetrical, then the nitro group adds to the carbon
joined to the larger number of hydrogen atoms. This may be explained by
assuming that the nitro group is the positive part of the addendum (i.e., the
addition takes place according to Markownikoff 's rule) :

CH a -CH=CH 2 2 N— NO — > CH 3 -CH-CH 2 -N0 2 + NO — >

CH 3 -CH(NO)-CH 2 -N0 2

Acetyl nitrate (70 per cent, nitric acid and excess of acetic anhydride) reacts
with many alkenes to give a mixture of (3-nitro-acetate, |3-nitro compound and
P-nitro-nitrate (Bordwell et al., i960), e.g. (Ac = CH 3 -CO) :

MeC=CH 2 + 2AcON0 2 -^!-> Me 2 OCH 2 -N0 2 (64%)

Me OAc

+ CH 2 =C-CH 2 -N0 2 + Me 2 C-CH 2 'N0 2

Me ONO a

(5%) (4%)

8. Olefins are readily hydroxylated, i.e., add on hydroxyl groups, to
form dihydroxy-compounds known as glycols (q. v.). Hydroxylation may
be effected:

(i) By cold dilute alkaline permanganate solution (cis-hydroxylation) ;
e.g., ethylene is converted into ethylene glycol:

CH 2 =CH 2 + H 2 + [O] KMn °' > CH 2 OH-CH 2 OH

The mechanism of hydroxylation with permanganate is believed to proceed
via a cyclic intermediate (cf. osmium tetroxide, below).

Hi

H— C— OH

1 1 KMnO. H— C— Ov JO hydrolysis H— C~

H«i > I >M< * I

I H— C— O/ ^ H— C-

-OH

This mechanism accounts for cis-hydroxylation and is supported by the work of
Wiberg et al. (1957), w h° used potassium permanganate labelled with ls O and
showed that both glycol oxygen atoms come from the oxidising agent.

(ii) By 90 per cent, hydrogen peroxide in glacial acetic acid or better,
in formic acid; e.g., oleic acid is converted into 9 : 10-dihydroxystearic
acid (see also reaction 9) :

CH s -(CH 2 ) 7 -CH=CH-(CH 2 ) 7 -C0 2 H + H 2 2 — >

CH 3 -(CH 2 ) 7 -CH(OH)-CH(OH)-(CH 2 ) 7 -C0 2 H

The addition of hydrogen peroxide may be catalysed by various oxides,
e.g., osmium tetroxide in terf.-butanol (m-addition), selenium dioxide in
ferrf.-butanol or acetone (/raws-addition ; see p. 429).

(hi) By means of osmium tetroxide. This compound adds very
readily to an ethylenic double bond at room temperature (Criegee,
1936):

R-CH R-CH— 0\ />0

|| +0s0 4 > I ^Os^ (v.g.-ex.)

R-CH R-CH— 0/ ^O

UNSATURATED HYDROCARBONS 73

These cyclic compounds (osmic esters), on refluxing with aqueous
ethanolic sodium hydrogen sulphite, are hydrolysed to i : 2-glycols
(m-glycols).

If the addition of osmium tetroxide is carried out in the presence of pyridine,
coloured crystalline compounds are obtained, usually in theoretical yield :

RCH R-CH— (X

|| + Os0 4 + 2C 6 H 6 N — > I )Os0 2 , 2C 5 H 5 N

R-CH R-CH— CK

Berkowitz et al. (1958) have shown that ruthenium tetroxide is more convenient
to use than osmium tetroxide; it is less toxic.

m-Hydroxylation of a double bond may also be effected by treating the
olefin with iodine and silver acetate in wet acetic acid, and then hydrolysing
the mixed mono- and di-acetates with alkali (Barkley et al., 1954) :

>C=C< — -*>C C< ■ + ^c-c<-^oc C<

' ^ CH.-CO^g -^ I p~ II I I

OH 0-CO-CH 3 CH 3 -COO OCO-CH 3 OH OH

Glycols are readily oxidised to acids or ketones by means of acid per-
manganate or acid dichromate, the nature of the products being determined
by the structure of the glycol, e.g.,

(a) propylene glycol gives acetic and formic acid:

CH 3 -CH(OH)-CH 2 OH -^ CH 3 -C0 2 H + H-C0 2 H

(b) j'soButylene glycol gives acetone and formic acid:

CH 3 \ f01 CH 3 \

)C(OH)-CH 2 OH ^> /CO + H-C0 2 H + H 2

_/ CH /

CH/ CH ;

3

Sodium bismuthate, in acid solution, also effects similar oxidations
(Rigby, 1950). An advantage of this reagent is that an aldehyde is one of
the products (when possible) ; it is not further oxidised.

Oxidation of a glycol may also be effected by lead tetra-acetate,
(CH 3 -C0 2 ) 4 Pb, or by periodic acid, HI0 4 or H 5 IO„, the products being
aldehydes or ketones, according to the structure of the glycol, e.g.,

(a) ethylene glycol gives two molecules of formaldehyde:

CH 2 OH-CH 2 OH ^— > H-CHO + H-CHO

2 z (CH,CO,).Pb

(b) isoButylene glycol gives acetone and formaldehyde:

CH 3 \ r01 CH 3 \

)C(OH)-CH 2 OH -i^ )C0 + H-CHO

CH,/ ' CH 3 /

It can be seen that whatever oxidising agent is used the glycol is split
into two fragments, the rupture of the carbon chain occurring between the
two carbon atoms joined to the hydroxyl groups. Since these two carbon
atoms were linked together by the double bond in the original olefin, identi-
fication of the two fragments which may be acids, aldehydes or ketones, will
indicate the position of the double bond in the olefin, e.g.,

R-CH=CH-R' ^% R-CH(OH)-CH(OH)-R' ^VR-CHO + R'-CHO

74 ORGANIC CHEMISTRY

9. Prileschaiev's reaction (1912). By means of per-acids, the double
bond in olefins is converted into the epoxide (olefin oxide). Perbenzoic acid,
C<jH 6 -CO0 2 H, and monoperphthalic acid, C0 a H-C 6 H 4 -C0-0 2 H, have been
widely used for this reaction, e.g.,

/0\
R-CH=CH-R' + C 6 H 8 -CO0 2 Na — > R-CH-CH-R' + C 6 H 5 -C0 2 Na

Emmons et al. (1954, 1955) have found that peroxytrifluoroacetic acid
(CF 8 -C0-0 2 H) is a very good reagent for epoxidation and hydroxylation.

R-CH=CH-R CF - CO ' H > R-CH-CH-R -2£^

R-CHOH-CH(0-CO-CF 3 )-R HC ""> R-CHOH-CHOH-R (60-95%)

CH g OH

This method is particularly useful for high molecular-weight alkenes with a
terminal double bond (these are only slowly hydroxylated by other per-
acids) . Furthermore, peroxytrifluoroacetic acid may be used to hydroxylate
negatively substituted olefins, e.g., ethyl acrylate, CH 2 =CH-C0 2 C 2 H 5 .

Epoxides are readily converted into glycols (p. 248).

Many mechanisms have been proposed for epoxidation, but none is certain.
According to Pausacker et al. (1955), the mechanism is

H I/O— C-C 6 H 5 CH\ ^.-O^C-CsHj CH\ 0=C-C 6 H

+
H— O

transition state

10. Olefins add on ozone to form ozonides. These are usually explosive in
the free state, and their structure and mechanism of formation have been the
subject of a great deal of work. Staudinger (1922) suggested that the
molozonide is formed first, and this then rearranges to the ozonide, some
polymerising as well.

/°\
R-CH=CH-R' + 3 -^ R-CH— CH-R' — > R-CH CH-R' + polymer

I + - II

o — 0—0 o — o

molozonide ozonide

Criegee (1959) has isolated the molozonide of ^aws-di-t-butylethylene at
— 8o° (R = R' = (CH 3 ) 3 C — ); reduction of this compound gives the glycol,
thereby showing that one of the C — C bonds is still intact. When the tempera-
ture was allowed to rise to — 6o°, the molozonide rearranged to the ozonide.

According to Criegee et al. (1954), the formation of the ozonide takes place as
follows :

-o 3 -^[>c-^c<]_>
>c=o+ x c< — >>cC Jcc

UNSATURATED HYDROCARBONS 75

Bailey (1957), however, has proposed a mechanism in which the first step is the
formation of a 7t-complex. This then produces a zwitterion which breaks down
to produce the zwitterion and carbonyl compound in the Criegee mechanism :

+ o 3 — ± >C==C< — > >c J -<< — > >C— O— O + >c=o
In this way Bailey explains the ready fission of carbon-carbon multiple bonds.

The ozonide is prepared by dissolving the olefinic compound in a solvent
that is unaffected by ozone, e.g., chloroform, carbon tetrachloride, glacial
acetic acid, light petrol, etc., and a stream of ozonised oxygen is passed
through. Subsequent treatment may be by one of the following procedures.

(i) The solvent is evaporated under reduced pressure, and the residual
ozonide is treated with water and zinc dust in the presence of traces of silver
and hydroquinone (Whitmore, 1932). Aldehydes or ketones are obtained
according to the structure of the olefin. The function of the zinc dust is
to destroy the hydrogen peroxide which is formed in the reaction, and which
tends to oxidise the aldehyde (if this is a primary cleavage product) to the
corresponding acid, e.g.,

R 2 C— O— CH-R' HO

I — i-> R 2 CO + R'-CHO + H 2 2

-O
R'-CHO + H 2 O a — > R'-C0 2 H + H 2

2^

A-

In practice both aldehyde and acid are obtained.

(ii) A better method than the above is the reductive decomposition of the
ozonide (Fischer, 1928, 1932). A palladium catalyst carried on a calcium
carbonate support is added to the solution of the ozonide and then hydrogen
is passed in. Usually a good yield (50-90 per cent.) of aldehyde or ketone
is obtained:

R-CH— O— CH-R' „

I J — ■!> R-CHO + R'-CHO + H 2

O O

(hi) Wilms (1950) has found that peracetic acid in acetic acid oxidises
ozonides to carboxylic acids in high yields.

The resulting aldehydes (or acids) and ketones are identified, and thus
the position of the double bond in the olefinic compound is found. The
complete process of preparing the ozonide and decomposing it (and identify-
ing the products formed) is known as ozonolysis, and this is probably the
best method for determining the position of a double bond in any olefinic
compound. Recently, however, some doubt has been cast on ozonolysis as
a means of determining the positions of double bonds in unsaturated com-
pounds, e.g., according to Barnard et al. (1950), during the ozonolysis of
citral (an acyclic terpene), partial rearrangement from the t'sopropylideoe
(I) to the t'sopropenyl structure (II) occurs (this is an example of a three-,
carbon tautomeric system; see p. 220) :•

CH 3 CH 3 CH 3 CH 3

CH 3 -C=CH-CH 2 -CH 2 -C=CH-CHO ^=ft CH 2 =C-CH 2 -CH 2 -CH 2 -C=CH-CHO

(I) (ii)

Ozonides may be reduced to alcohols directly by sodium borohydride
[inter alia, Sousa et al., i960).

76 ORGANIC CHEMISTRY

ii. Olefins isomerise when heated at high temperature (500-700 ), or
at a lower temperature (200-300°) in the presence of various catalysts,
e.g., aluminium sulphate. Isomerisation may be due (i) to the change in
position of the double bond, which always tends to move towards the centre
of the chain, e.g., pent-i-ene isomerises to pent-2-ene:

CH 3 -CH 2 -CH 2 -CH=CH 2 — > CH 3 -CH 2 -CH=CH-CH 3

(ii) To the migration of a methyl group, e.g., but-i-ene isomerises to
isobutene:

CH g -CH 2 -CH=CH 2 — > (CH S ) 2 C=CH 2

(i) and (ii) may, or may not, occur together.

12. Olefins add on to isoparaffins in the presence of a catalyst, many of
which are known, but the one usually employed is concentrated sulphuric
acid, e.g., ethylene adds on to jsobutane to form a mixture of 2-methyl-
pentane and 2 : 3-dimethylbutane :

(CH 3 ) 2 CH-CH 3 -^> (CH 3 ) 2 CH.CH 2 -CH 2 .CH 3 + (CH 3 ) 2 CH.CH(CH 3 ) 2

This reaction is particularly useful for preparing " j'so-octane ", 2:2:4-
trimethylpentane (see p. 57) by treating wobutane with isobutene in the
presence of concentrated sulphuric acid:

(CH 3 ) 3 CH + CH 2 =C(CH 3 ) 2 — ■> (CH 3 ) 3 OCH 2 -CH(CH 3 ) 2

13. Diborane reacts rapidly at room temperature with olefins, giving the
trialkylborons. Terminal olefins give the primary alkylborons, which can
be oxidised by hydrogen peroxide to primary alcohols (Brown et al., 1957) :

R-CH==CH 2 -^> (R-CH 2 -CH a — ) 3 B ^^-> 3 R-CH 2 -CH 2 OH (g.)

The s- and t-alkylborons obtained from non-terminal olefins readily undergo
isomerisation to primary alkylborons when heated. The addition of di-
borane is known as hydroboronation (hydroboration).

If the trialkylboron is treated with a carboxylic acid, the corresponding alkane
is obtained by protolysis :

(R-CH 2 -CH 2 — ) 3 B -51> 3 R-CH 2 -CH 3

Thus the final product is formed by reduction of an olefin by a non-catalytic
method.

In the same way, acetylenes undergo monohydroboronation and protolysis
to give almost pure cw-olefin:

. — CH=CH—

14. Olefins condense with acetic anhydride, (CH 3 *CO) 2 0, in the presence
of a catalyst, e.g., zinc chloride, to form unsaturated ketones, e.g., ethylene
forms methyl vinyl ketone:

CH 2 =CH 2 + (CH 3 -C0) 2 ^> CH 2 =CH-CO-CH 3 + CH 3 -CO a H

Acid chlorides, alkyl chlorides and a-halogenated ethers also combine with
olefins in the presence of aluminium chloride, e.g.,

CH 3 -COCl + CH 2 =CH 2 ^%- CH 3 -C0-CH 2 -CH 2 C1

A1C1

(CH 3 ) 3 CC1 + CH 2 =CH 2 '-> (CH 3 ) 3 CH 2 -CH 2 C1

AICI

CH 3 -0-CH 2 Cl + CH a =CH 2 ?-> CHs-O-CHg-CHg-CHjCl

UNSATURATED HYDROCARBONS 77

All of these condensations are examples of the Friedel-Crafts reaction in
aliphatic compounds (see p. 529).

15. Olefins readily polymerise in the presence of suitable catalysts, e.g.,
isobutene gives a polymer in the presence of concentrated sulphuric arid:

wC 4 H 8 — MC 4 H 8 ) re

When two compounds have the same empirical formula but differ in
molecular weight, the more complicated compound is called a polymer of
the simpler one. The term polymerisation was used originally to indicate
the process that took place when a single substance — the monomer — gave
products having the same empirical formula but different molecular weights,
each of these being a multiple of that of the monomer. As the investigation
of polymerisation reactions progressed, it was found that many compounds
of high molecular weight, although they produced a large number of
monomer molecules on suitable treatment, did not always have exactly
the same empirical formula as the parent monomer. This led to a modifica-
tion of the definitions of the terms polymer and polymerisation. Accord-
ing to Carothers (1931) polymerisation is best defined as intermodular
combinations that are functionally capable of proceeding indefinitely. This
definition implies that there is no limit theoretically to the size of the polymer
molecule. In practice, however, the polymer ceases to grow, for various
reasons (see below). The terms polymer and polymerisation are now used
mainly in connection with high molecular weight compounds, which, in
addition to being called polymers, are also known as macromolecules.

There are two types of polymerisation, addition polymerisation and
condensation polymerisation.

Addition Polymerisation. Addition polymerisation occurs among mole-
cules containing double or triple bonds; but in certain cases it can also
occur between bifunctional compounds that result from the opening of
ring structures (see, e.g., ethylene oxide). There is no liberation of small
molecules during addition polymerisation.

A very important group of olefinic compounds that undergo addition
polymerisation is of the type CH a =CHY, where Y may be H, X, CO a R
CN, etc. :

H

«CH a =C

I
Y

There are three possible ways in which this polymerisation can occur:

(i) Head to tail: — CH 2 -CHY— CH 2 -CHY—
(ii) Head to head and tail to tail:

— CHY-CH 2 — CH 2 -CHY— CHY-CH 2 — CH 2 -CHY—

(hi) A random arrangement involving (i) and (ii).

Experimental work seems to indicate that (i) is favoured.

Most polymerisations are carried out in the presence of catalysts, and
polymerisation of olefins can be accelerated by ionic-type catalysts or radical-
type catalysts. Both types of reaction consist of a number of steps which
follow one another consecutively and rapidly, and appear to take place
in three principal steps :

(i) The initiation or activation,
(ii) The growth or propagation,
(iii) The termination or cessation.

78 ORGANIC CHEMISTRY

If M represents the monomer, the series of reactions may be represented
as follows:

(i) M — > M* M

(ii) M + M* — > MM* > MMM* — > >■ M„*

(iii) M K * — > M„.

The Ionic Mechanism of Catalysis. The ionic mechanism is believed to take
place in the presence of certain metallic and non-metallic halides such as A1C1 3 ,
SnCl 4 or BF 3 . In certain cases sulphuric acid also catalyses polymerisation.
Ionic catalysts are usually electrophilic reagents, and Hunter and Yohe (1933)
have suggested that the chain-initiating action of these catalysts depends on their
electrophilic nature, and consists in the catalyst acquiring a share in a pair of
electrons (the ^-electrons) from the double bond of the monomer, e.g.,

(i) A1C1. + C=C — y CI3AI— C— C

II II

(ii) C1.A1— C— C+ wC=C — > CI3AI— (— C— C— ) — C— C

II II I I I I

(iii) C1.A1— (— C— C— )„— C— C — > C1 2 A1— (— C— C— )„— C=C + HC1

I I I I I I I I

In (iii) a proton is lost, thus producing a double bond at the end of the chain so
that the molecule becomes deactivated, and hence ceases to grow (see also below).

The Free-Radical Mechanism of Catalysis. The most important cases of addi-
tion polymerisation are those which take place by chain reactions and ate brought
about by catalysts that are known to generate free radicals. The most widely
used catalysts are the organic and inorganic peroxides and the salts of the per-
acids, e.g., benzoyl peroxide, acetyl peroxide, hydrogen peroxide, potassium
perborate, etc.

Staudinger (1932) was the first to suggest a free-radical mechanism, and it
may be as follows for an organic peroxide :

(i) (R-C0 2 ) 2 — >-2RCCy
RCO a >R- +C0 2

R- + C=C — > R— C— O

I I I I

(ii) R— C— C- + wC=C — y R— {— C— C— )„—C— C-

II II I II I

(iii) The cessation reaction may take place:

(a) By the collision between two growing chains which unite to form a
deactivated molecule:

2R— M,- > R— M„— M„— R

Alternatively, disproportionation (see p. 51) may take place, and thereby
deactivate the growing molecules.

H

2R— (— c— c— )„-c-c._ > l| |f 11 I H \

ill II R— (— c— c— )„— c— c— H + R— (— C— C— ) — 6=c

* I I I I I I I I

UNSATURATED HYDROCARBONS 79

(6) By the collision between the growing chain and a catalyst radical :

R— - M„- + R- > R— M„— R

(c) By the collision between the growing chain and impurities which have
become activated jiuring the polymerisation, e.g.,

R— M„- + YZ y R— M— Y + Z-

Z- + M — > ZM- -^-> ZM„-
ZM„- -f- YZ — »- Z— M B — Y + Z-

In " uncatalyseld " polymerisation, i.e., in the absence of foreign substances,
initiation may bejjjin by dimerisation of the monomer :

11 1 1 1 1

Not only can addition polymerisation take place among molecules of one kind,
but it can also take place among molecules of two kinds, when the phenomenon
is known as copolymerisation or interpolymerisation.

Condensation Polymerisation or Polycondensation. In condensation
polymerisation, bi- or polyfunctional molecules condense with one another,
and in doing so repeatedly eliminate a small molecule such as water, am-
monia, hydrogen chloride, etc., as the reaction proceeds. This type of
polymerisation takes place by a series of steps, and is discussed in various
parts of the text (see, for example, the aldol condensation, the esterification
of glycols with dibasic acids, etc.).

Polymers may Jbe classified into three groups :

(i) natural, e\.g., rubber, proteins, cellulose; (ii) semi-synthetic, e.g. nitro-
cellulose, cellulose acetate; and (iii) synthetic, e.g., nylon, bakelite, perspex.

Plastics form a group of high polymers which have a fair range of def ormability
and mouldability, particularly at high temperatures. In plastics the polymers
formed do not alllhave the same molecular weight, and since the polymers are not
amenable to thej ordinary methods of separation, the molecular weight of a
" polymer " is trie average molecular weight. Polymerisation is carried out with
the object of building up compounds with predicted properties, and since the
properties of a plastic depend on the degree of polymerisation >t is necessary to
stop polymerisation when the desired average molecular weig* 1 * is reached. This
may be done by Various means, e.g., variation of the concentration of the catalyst.
The average molecular weight of plastics varies from about 20,000 (e.g., nylon)
to several hundrejd thousand (e.g., polyvinyl chloride, 250,000).

Plastics are generally tough, resistant to the action of acids and alkalis, and not
very much affected over a fair range of temperature. They can be moulded to
any desired shapje or form.

Plastics are cjf two main types, thermoplastic and thermosetting. Thermo-
plastics are linear polymers which are soluble in many organic solvents, and
which soften oni heating and become rigid on cooling. The process of heat-
softening, mouldjing and cooling can be repeated as often as desired, and hardly
affects the properties of the plastic. Typical thermoplastics are cellulose acetate,
nitrocellulose an(l vinyl polymers such as polythene, perspex, etc.

Thermosetting plastics are three-dimensional polymers which are insoluble in
any kind of solvent, and which can be heat-treated only once before they set, i.e.,
their formation, after which heating results in chemical decomposition, and hence
they cannot bef "reworked". Typical thermosetting plasti6s are phenol-
formaldehyde, urea-formaldehyde, melamine-formaldehyde, silicones, etc.

In thermoplastics the chains are, more or less, free chemically, but are held
together by van der Waals' forces. It is possible, however, to link together these
linear molecules (cf. the rungs of a ladder) and the cross-linking agent converts the
thermoplastic into a thermosetting plastic, e.g., in the vulcanisation of rubber the

8o

ORGANIC CHEMISTRY

sulphur cross-links the long chains. Furthermore, such thermosetting plastics
may be reconverted into thermoplastics by opening the cross-links, e.g., the
reclaiming of rubber. Most thermosetting plastics may be regarded as cross-
linked polymers.

Those plastics which do not soften very much with rise in temperature are made
soft and readily workable by the addition of certain compounds known as
plasticisers; e.g., polyvinyl chloride is extremely stiff and hard, but addition of
tricresyl phosphate makes it soft and rubber-like.

Some Individual Olefins.

Tlje first member of the olefin series is methylene (carbene), CH 2 , but it
existp only as a free diradical with a very short life period. It has been
prepared by heating diazomethane at very low pressure :

CH 2 N 2 — > -CH 2 + N 2

The j:wo electrons must be unpaired for methylene to be a diradical. It has
been shown that methylene and its derivatives behave as electrophilic
reagents. Methylene reacts with alkyl chlorides to attack both the C — CI
and <fc-C — H bonds {inter alia, Bradley et al., 1961) ; e.g. (see also p. 483) :

CH

CH S -CH 2 -CH 2 C1 ^> CH 3 -CH 2 -CH 2 -CH 2 C1 + CH 3 -CH 2 -CHC1-CH 3

Ethylene, ethene, C 2 H 4 . Ethylene may be prepared by any of the general
methods of preparation (except 2), but the most convenient laboratory
methpd is to heat ethanol with excess of concentrated sulphuric acid.

Th^ mechanism of this reaction is the reverse of hydration of olefins (p. 70) :

C 2 H 5 OH + H+ ^^ C 2 H 6 OH 2 + ^± C 2 H 5 + + H 2

The ckrbonium ion, which is unstable, stabilises itself by eliminating a proton to
form Ethylene; the proton is accepted by a water molecule (a Lewis base) :

I CH 2 — CH 2 — H OH 2 ^==± CH 2 =CH 2 + H s O +

Foil each molecule of ethanol converted into ethylene, one molecule of water
is produced, and hence, after a time the sulphuric acid becomes too dilute to
behavte as a dehydrating agent.

Ethylene may also be prepared by the electrolysis of sodium succinate
(Koltje):

CH 2 -COONa CH 2

I +2H 2 — > |J + 2 C0 8 + 2 NaOH + H.

CH.-COONa CH 2

±2

Ethylene is still prepared industrially by passing ethanol vapour over
heated alumina at about 350 :

1

It is, fie

C 2 H s OH— ^C 2 H 4 +H 2

lowever, now being obtained in huge quantities as a by-product in
the crkcking of crude oil and of ethane, propane and butane. Ethylene is
also manufactured by the partial hydrogenation of acetylene, which is mixed
with hydrogen and passed, at 200°, over a palladium catalyst carried on a
silica-gel support :

CH=CH + H 2 -+ CH a =CH 2 (95%)

Ethylene is a colourless gas, b.p. — 105 °, sparingly soluble in water. It
burns jwith a smoky luminous flame. It has been claimed that carefully

UNSATURATED HYDROCARBONS 8l

purified ethylene does not react with chlorine in the absence of light. When
ethylene is heated with chlorine at 350-450°, vinyl chloride is obtained:

CH 2 =CH 2 + Cl 3 — > CH a =CHCl + HC1

The unsaturated radical CH 2 =CH — is known as the vinyl or ethenyl radical.
Ethylene may be oxidised by atmospheric oxygen in the presence of
silver as catalyst, and at a temperature of 200-400 °, to ethylene oxide:

CH 2 =CH 2 + £0 2

Ethylene polymerises under high pressure and high temperature to form
polyethylene or polythene:

»CH 2 =CH 2 -+ — (— CH 2 — CH a — )„—

This polymerisation is catalysed by traces of oxygen (which produces the
free radicals). Polythene is very resistant to acids, bases, and most of the
usual organic solvents.

Ethylene is used for ripening fruit. Unripe fruit may be transported
easily without damage, and ripens on exposure to ethylene gas for a few
days, the product being apparently indistinguishable from the natural
ripened fruit. Ethylene is also used as an anaesthetic, in the manufacture
of mustard gas and plastics (polythene, polystyrene), and in the preparation
of various solvents such as glycol, dioxan, cellosolves, etc.

Structure of Ethylene. The molecular formula of ethylene is C 2 H 4 .
Two carbon atoms have the power to combine with six univalent atoms or
groups, as in ethane, «eopentane, etc. There are only four univalent
hydrogen atoms present in ethylene: therefore ethylene is said to be un-
saturated, and should be capable of adding on two univalent atoms or
groups, and this, as we have seen above, is observed in practice. Thus the
structure of ethylene must be such as to be capable of undergoing addition
reactions. Assuming carbon to be quadrivalent and hydrogen univalent,
three structures are possible for ethylene :

H H H H H H

h44- - 4- U

or

CH 8 — CH< -CH 2 -CH — CH a =CH 2

(I) (ID (Hi)

Two isomeric compounds of molecular formula C 2 H 4 C1 2 are possible:
CH 3 -CHCl a and CH 2 C1-CH S C1. Both isomers are known, one (ethylene
chloride) being formed by the direct combination between ethylene and
chlorine, and the other (ethylidene chloride) by the action of phosphorus
pentachloride on acetaldehyde. The structure of ethylidene chloride is
CH 8 -CHC1 2 (see p. 116) ; hence the structure of ethylene chloride is CH 2 C1-CH 2 C1.
If (I) were the structure of ethylene, then the addition of chlorine should give
ethylidene chloride, and not ethylene chloride. We may, therefore, reject
structure (I). Furthermore, since (I) is unsymmetrical it would have a fairly
large dipole moment ; actually ethylene has a zero dipole moment.

Structure (II) represents ethylene as possessing " free " bonds by means
of which addition compounds are formed. If this is the structure, then we
might expect that the " free " carbon valencies could be satisfied one at a

82

ORGANIC CHEMISTRY

timej i.e., a compound such as CH 2 C1*CH 2 - should be possible, since if two
" free " valencies can exist independently of each other, it is logical to
suppose that one can exist by itself. No such compounds have yet been
obtained, and in practice it is found that unsaturated compounds always
combine with an even number of univalent atoms or groups. Hence struc-
ture III) is accepted for ethylene, and the presence of the double bond (con-
sisting of one a-bond and one 7t-bond) is supported by other evidence (length
of the carbon-carbon bond; geometrical isomerism).

In structures (I) and (II) the unconnected bonds indicate one electron. If
the tjro electrons in (II) axe paired, they form a covalent bond, and so (II) and
(III) jare the same. On the other hand, if the two electrons in (II) are un-
paired, then (II) is a diradical. Since ethylene does not exhibit the usual
properties of a free diradical, we must reject (II).

The presence of a double (or triple bond) in an organic compound may be
found readily by means of bromine water, bromine in chloroform solution,
or dilute alkaline permanganate. If the compound under investigation
is unsaturated, then the above reagents are decolorised. Perbenzoic acid
or monoperphthalic acid can be used to detect the presence of a double
bond, and also to estimate the number of double bonds (see also iodine
value, p. 261).

Propylene, propene, C 3 H 6 , may be prepared by heating propanol or iso-
propa.nol with sulphuric acid (mechanism as for ethylene from ethanol) :

CH 3 -CH 2 -CH 2 OH -
CH 3 -CH(OH)-CH 3

-H.O

> CH,-CH=CH,

-H 2

> CH,-CH=CHo

It inay also be prepared by heating propyl iodide with ethanolic potassium
hydroxide:

CH 3 -CH 2 -CH 2 I + KOH

ethanol

> CH 3 -CH=CH 2 + KI + H a O

Propylene is obtained commercially in huge quantities as a by-product
in thf! cracking of petroleum. It is a colourless gas, b.p. — 48°, insoluble
in water but fairly soluble in ethanol. It is used industrially for the pre-
paration of wopropanol, glycerol, etc.

The unsaturated radical CH 2 =CH*CH 2 — is known as the allyl radical.

Butylenes, butenes, C 4 H 8 . There are three isomeric butylenes, and all
are gases :

CH,-CH 2 -CH=CH 2 , a-butylene,but-i-ene (b.p. -6-i°); CH 3 -CH=CH-CH 3 ,
(3-butylene, but-2-ene (b.p. 1°); (CH 3 ) 2 C=CH g , isobutylene, wobutene
(b.p. ^6-6°).

All the butylenes are obtained from cracked petroleum. The 1- and 2-
butenes are used for the preparation of sec.-butanol (q.v.), and isobutene
for tert.-huta.nol (q.v.). But-2-ene differs from its isomers in that it exhibits
geometrical isomerism (see p. 425). isoButene differs from its isomers in
that it reacts with chlorine at room temperature to give mainly substitution
products, substitution occurring in the allyl position (see p. 67). Thus
3-chloro-2-methylprop-i-ene is the main product, and is accompanied by a
small amount of the addition product 1 : 2-dichloro-2-methylpropane :

CH,

CH„

X=CH.

a. C1CH 2 n
CH/

CH 3

)C^CH 8 +

ch/

CC1-CH.C1

UNSATURATED HYDROCARBONS 83

UNSATURATED COMPOUNDS WITH TWO OR MORE
DOUBLE BONDS

When the compound contains two double bonds, it is known as a diolefin
or alkadiene, and has the general formula C M H 2n _ 2 ; when there are three
double bonds present, the compound is known as a triolefin or alkatriene,
and has the general formula C„H 2n _4,' etc.

Nomenclature. The longest carbon chain containing the maximum
number of double bonds is chosen as the parent hydrocarbon, and the
chain is so numbered as to give the lowest possible numbers to the double
bonds, e.g.,

s
CFT

\c=C(CH 3 )-C(CH 3 )=CH 2 2:3: 4-trimethylpenta-i : 3-diene

CH/

There are three different types of compounds with two double bonds.

1. Hydrocarbons with isolated double bonds contain the arrangement
^C=CH>(CH 2 )„*CH==C<^ where «>0. One of the simplest compounds
of this type is diallyl or hexa-i : 5-diene, which may be prepared by the
action of sodium on allyl iodide (Wurtz reaction) :

2CH 2 =CH-CH 2 I + 2 Na — > CH 2 =CH-CH 2 -CH 2 -CH==CH 2 + 2NaI

Diallyl is a liquid, b.p. 59-6°. It resembles the olefins chemically, but
since there are two double bonds present, it may add on two or four univalent
atoms or groups according to the relative concentration of the addendum ;
e.g., with excess bromine, diallyl forms 1:2:5: 6-tetrabromohexane.

2. Hydrocarbons with cumulated double bonds contain the arrangement
^C=C=C^. The simplest compound of this type is allene'or propadiene,
which may be prepared by heating 1:2: 3-tribromopropane with solid
potassium hydroxide, and then treating the resulting 2 : 3-dibromopropylene
with zinc dust in methanol solution :

CH 2 Br-CHBr-CH 2 Br-^-> CH 2 Br-CBr=CH 2 Zn/CH,0H ^ CH 2 t=C=CH 2

Allene is a gas, b.p. —32°. With bromine it forms 1:2:2: 3-tetrabromo-
propane; with sulphuric acid it forms acetone; and when treated with
sodium in ether, the sodium derivative of propyne (CH 3 *C=CNa) is pro-
duced. Allene compounds are very important from the view of stereo-
chemistry (see p. 432).

In recent years, allenes have been prepared from acetylenes by rearrangement,
e.g..

an K rn

•C0 2 H

A novel way of preparing allenes is by treating the cycfopropane derivative
formed from an olefin and bromoform and alkali with magnesium in ether (Doer-
ing et al., 1958). Dibromomethylene is an intermediate (p. 117).

R 2 C=CR 2 + ICBr, > R 2 C CR 2 — -> R2C=C=CR 2 .

Br„

84 ORGANIC CHEMISTRY

An extended allene type of linkage gives the cumulene system, the simplest
member of which is butatriene, and this has been prepared by debrominating
i : 4-dibromobut-2-yne with zinc (Schubert et al., 1952, 1954) :

Zn

CH 2 Br-C^OCH 2 Br > CH 2 =C=C=CH !!

3. Hydrocarbons with conjugated double bonds contain single and
double bonds arranged alternately, i.e., they contain the arrangement
— C=CH — CH=CH — CH=CH — . The simplest member of this group of
compounds is buta-i : 3-diene, which may be prepared by passing cyclo-
hexene over a heated nichrome wire (an alloy of nickel, chromium and iron) :

XH, V XH,

•-2

/r

CH 2 CH CH Cii 2

CH 2 CH CH CH

+ 0; (65-75%)

^ch/

Butadiene is prepared technically :

(i) By dehydrogenating w-butane (from natural gas or from petroleum gas)
or but-i-ene (from cracked petroleum) by passing the gas over a heated
catalyst, e.g., chromic oxide on an alumina support.

(ii) By passing a mixture of butane-i : 3-diol and steam, in proportions
4 : 1, over trisodium phosphate containing 20 per cent, free phosphoric acid,
heated at 270 :

CH 3 -CH(OH)-CH 2 -CH 2 OH-^% CH 2 =CH— CH=CH 2 (85-90%)

(iii) By passing ethanol vapour over a catalyst of alumina-zinc oxide
heated at 420-470 :

2C 2 H 6 OH — > CH 2 =CH— CH=CH 2 + 2H 2 + H 2 (20%)

The yield is low owing to the production of many by-products such as acetic
acid, ethyl acetate, ether, etc. The yield of butadiene has been improved by
passing a mixture of ethanol and acetaldehyde over a heated catalyst of
silica-gel plus 2 per cent, tantalum oxide:

C 2 H B OH + CHg-CHO — > CH 2 =CH— CH=CH 2 + 2 H 2

(iv) By passing a mixture of acetylene and formaldehyde over copper
acetylide as catalyst, whereupon butynediol CH^H-C^C'CH^H is formed.
This is hydrogenated catalytically to butane-i : 4-diol, which, on catalytic
dehydration, gives butadiene:

C 2 H 2 + 2H-CHO -^> CH 2 OH-C==C-CH 2 OH -~>

CH 2 OH-CH 2 -CH 2 -CH 2 OH ~ 2H '°> CH 2 =CH— CH=CH 2

Butadiene is a gas, b.p. — 2-6°. Under the influence of sodium as catalyst,
butadiene readily polymerises to a product which has been used as a rubber
substitute known as buna (ft^tadiene + Na). The mechanism of this
polymerisation is uncertain, but a possibility is discussed in connection with
isoprene (see below).

A very important diolefin is isoprene or 2-methylbut-i : 3-diene,
CH.

I
CH 2 =C — CH=CH 2 , which may be obtained, in poor yield, by the slow

UNSATURATED HYDROCARBONS 85

distillation of rubber. It may be prepared by heating isopentanol with
hydrogen chloride, and chlorinating the resulting woamyl chloride, three
dichlorides thereby being formed :

(CH 3 ) 2 CH-CH 2 -CH 2 OH + HC1 — >■ (CH 3 ) 2 CH-CH 2 -CH 2 C1 + H 2

(CH 3 ) 2 CH-CH 2 .CH 2 Cl-^-> CHsX

(CH 3 ) 2 CC1-CH 2 -CH 2 C1 + (CH 3 ) 2 CH-CHC1-CH 2 C1 + /CH-CH 2 -CH 2 C1

(I) (II) CH 2 CK (in)

(I) is the main product (c/. p. 54), and this, when passed over soda-lime
heated at 500 , gives isoprene:

CH 3

(CH 3 ) 2 CC1-CH 8 -CH 2 C1 _£E». CH 2 =€— CH=CH 8

Isoprene is prepared technically:

(i) By passing tsopentane or isopentene over heated chromic oxide on
an alumina support (c/. butadiene),
(ii) By treating acetone as follows :

ONa ONa

^ 3 ) C =0i^CH 3 4 ^^CH 3 -i-C^CH-^>

CH 2 CH 3

QH OH CH,

CH 8

H, . m I _ A.,0,. ^J_ CK==Q

CH 3 — C— C^CH — -i-v CH 3 — C— CH=CH 2 — ^> CH a =C— CH==CH.

1 (cat.) " • 400 *

CH a

Isoprene is a liquid, b.p. 35 °, and when heated with sodium at 6o° it
polymerises to a substance resembling natural rubber. The mechanism
of this polymerisation is not settled; one that has been suggested is as
follows :

CH a =C(CH 3 )— CH=CH 2 + Na > NaCH 2 — C(CH 3 )=CH— CH a -

[,=C(CH,)-CH=CH,

| CH ' =

NaCH 2 — C(eH 3 )=CH— CH 2 — CH 2 — C(CH 3 )=CH— CH 2 «, etc.

Compounds containing conjugated double bonds have physical and
chemical properties that are not usually shown by compounds containing
isolated or cumulated double bonds, e.g., they show optical exaltation,
undergo abnormal addition reactions, readily polymerise, and undergo the
Diels-Alder reaction (p. 472). Another typical reaction of conjugated
dienes is their combination with sulphur dioxide to form a cyclic sulphone,
e-g.,

CH— CH 2 \
CH 2 =CH— CH=CH 2 + S0 2 — > || )S0 8

CH— CH/

sulpholene

Thiele's Theory of Partial Valencies. Conjugated compounds undergo
abnormal addition reactions, e.g., when butadiene is treated with bromine

86

ORGANIC CHEMISTRY

(one molecule), two dibromo-derivatives are obtained, the " expected "
3 : 4-dibromobut-i-ene (i : 2-addition), and the "unexpected" 1:4-
dibromobut-2-ene (1 : 4-addition) :

CH 2 =CH-CH=CH 2 -^-> CH 2 Br-CHBr-CH=CH 2 + CH 2 Br-CH=CH-CH 2 Br

It has been found that 1 : 2- and 1 : 4-additions usually take place together,
and the relative amount of each generally depends on the nature of the
addendum and the conditions of the experiment, e.g., type of solvent,
temperature.

Thiele (1899) suggested his theory of partial valencies to account for
1 : 4-addition. According to Thiele, a single bond is sufficient to hold two
carbon atoms together, and the two valencies of the double bond are not
used completely to link the two carbon atoms, but only one valency and
part of the other, leaving a surplus on each carbon atom. Thiele called
this surplus valency the residual or partial valency, and if we represent
it by a broken line, the formula of butadiene (and similarly for any other
conjugated compound) may be written CH 2 — -CH— CH^CH 2 . "Thiele

■ r i r

thought that the two middle partial valencies mutually satisfied each other
rather than remain free. Thus the actual state of butadiene is

CH 2 =CH— CH— CH, or CH,— CH— CH— CH„.

The ends of this molecule are therefore the most active parts, and so addition
of, e.g., bromine will occur at these ends, first by attachment through the
partial valencies, and then by each bromine atom acquiring a full valency,
causing the two middle carbon atoms to utilise completely the two valencies
left:

CH 2 — CH— CH— CH 2 + Br 2

— > CH 2 — CH— CH— CH 2

I I

Br Br

Thiele's theory explains 1 : 4-addition so well that it does not account at
all for 1 : 2-addition !

The mechanism of addition to conjugated systems is now believed to be as
follows. The first problem to solve is which carbon atom in butadiene is attacked
initially. Assuming that the addition mechanism is the same as for mono-
olefins {i.e., a two-stage electrophilic reaction), then attack on butadiene will be
at a terminal carbon atom since this becomes electron rich through the +E effect
of the vinyl group :

^— si Cit +

CH 2 =CH— CH=CH 2 CI— CI > CH 2 =CH— CH— CH 2 C1 + Cl~

The carbonium ion produced is a resonance hybrid :

+ +

CH 2 =CH— CH— CH 2 C1 ^-^ CH 2 — CH=CH— CH 2 Cl

This resonance hybrid may be represented as

8+ 8+

CH 2 — CH^CH— CH 2 C1

Thus there are two positive centres which may be attacked by the chloride ion
in the second step.

8+ 8+

CH 2 — CH^CH— CH 2 C1 + CI" >

CH 2 =CH— CHC1-CH 2 C1 + CH a Cl'CH=CH— CH 2 C1

UNSATURATED HYDROCARBONS 87

Since both the 1 : 2- and 1 : 4-dichlorides are formed in practice, the interesting
question is whether one can predict which isomer will predominate. The answer
to this depends on whether the isomers are interconvertible or not under the
conditions of the experiment. If the conditions permit interconversion, then
the thermodynamically controlled product (i.e., the more stable one) will pre-
dominate. If the conditions do not permit interconversion, then the kinetically
controlled product {i.e., the one formed faster) will predominate (p. 32). Con-
sider the following :

rv rv

In this type of structure, halogen hyperconjugation is possible (p. 270), and this
makes the molecule more stable than one in which such hyperconjugation is not
possible. Inspection of the two butadiene dichlorides shows that the 1 : 2-
product has one chlorine-hyperconjugated system whereas the 1 : 4-product has
two. Thus, the latter is more stable and so should be the thermodynamically
controlled product if the two forms are interconvertible, and consequently will
be the predominant product under these conditions. These predictions have
been observed experimentally, e.g., Muskat et al. (1930) treated butadiene with
chlorine under conditions where the two isomers were not interconvertible, and
obtained about 60 per cent, of the 1 : 2- and 40 per cent, of the 1 : 4-product.
This implies that the 1 : 2-compound is the kinetically controlled product. Pudovic
(1949) heated each isomer at 200 and obtained the same equilibrium mixture con-
taining about 30 per cent. 1 : 2- and 70 per cent. 1 : 4-. Thus, the 1 : 4-isomer is
the thermodynamically controlled product, and predominates under conditions
of interconvertibility.

The addition of halogen acid to butadiene also produces two products, the
1 : 2- and the 1 : 4-. Furthermore, since the proton adds on first, this adds
always to the terminal carbon, the halogen then adding at position 2 or 4.

In the foregoing account of the reactions of butadiene, we have assumed
that the molecule has the structure CH 2 =CH — CH=CH 2 . There are,
however, alternative electronic structures which are charged. Hence
butadiene is a resonance hybrid of a number of resonating structures:

CH 2 =CH— CH=CH 2 «--» CH 2 =CH— CH— CH 2 <->-

CH 2 — CH=CH— CH 2 , etc.

There is still, however, another contributing structure of butadiene, viz., (I).
The two electrons have antiparallel spins, since the number of unpaired
electrons in each resonating structure must be the same (p. 17). Thus these
paired electrons would, in the ordinary way, form a covalent bond. The
distance between them, however, is too great for them to form an effective
bond. Consequently this bond is referred to as a formal bond, and may be

CH 2 — CH=CH~CH 2 CH 2 — CH=CH— CH 2 CH 2 =CH— CH^CH^

(I) (II) . (Ill)

represented by a dotted line (II). Structure (II) carries no charges and so (II)
and (III) (also uncharged) are probably the main contributing resonance
structures (p. 20) . If only these two contribute significantly to the resonance
hybrid, the resonance energy can be expected to be small. Actually,
calculation has shown it to be about 3-5 k.cal./mole. Thus in butadiene
there are no " pure " single or double bonds. The lengths of the bonds will
therefore be somewhere between the extremes (of single and double bonds),
the actual values depending on the relative contributions of the resonating
structures to the resonance hybrid. Furthermore, if we assume that the
charged structures make a very small contribution, it might at first sight

88

ORGANIC CHEMISTRY

appear that the butadiene molecule possesses a dipole moment. This,
however, is not so in practice, the reason being that the terminal carbon
atoms can be either positive or negative, since the electronic displacements
can occur equally well in either direction. For this reason butadiene is said
to exhibit balanced conjugation, and this may be represented as :

^-\ rv

CH,=CH— CH=CH 2

So far we have considered the structure of conjugated compounds from V.B.
theory. When we consider their structure from M.O. theory, we get a different
picture. Each carbon atom in butadiene has the trigonal arrangement, and
Fig. i (a) shows the p z electrons associated with each carbon atom. If the mole-
cule is planar, the p, electron of C a overlaps that of C x as much as it does that of
C 3 , etc. Therefore all four p„ orbitals can be treated as forming an M.O. covering
all four carbon atoms (b). In this condition, a pair of electrons are no longer
mainly confined to the region between two nuclei, i.e., the bond formed is no

HH f

.^

n

~i '<r^\

A A d

Q

n

(b)

G

f ■«.

(

r •
(

B

n

Q

(e)

L)

(J

Fl

')

3.4.

6-6 6-6
(g)

longer a localised bond. The bonds produced are therefore called delocalised
bonds. In all, there are four delocalised M.O.s possible from the combination of
the four ir-electrons, two bonding and two anti-bonding M.O.s (p. 30). The two
bonding M.O.s are shown in Fig. 1(6) and (c), the former having one nodal plane,
and the latter, two. The two anti-bonding M.O.s are shown in (e), with three
nodal planes, and in (/), with four nodal planes. As pointed out on p. 31, as the
number of nodes in an orbital increases, so does the energy associated with that
orbital. Furthermore, according to the Pauli exclusion principle, no more than
two electrons can occupy the same M.O. Therefore in the ground state of
butadiene, two of the rc-electrons will occupy the M.O. in (b), and the other two
the M.O. with the next higher energy level, i.e., (c) . Fig. (d) represents these two
in one diagram, i.e., (d) represents the ground state of butadiene. In any excited
state of butadiene, electrons will occupy orbitals (e) or (/) (see p. 777).

In general, in a conjugated system containing zn rc-electrons, there are n bond-
ing and n anti-bonding orbitals, and in the ground state these electrons will
occupy, in pairs, the n M.O.s of lowest energy.

In delocalised bonds, the electrons have greater freedom of movement than in
localised bonds. Thus the total energy of the system is lowered, i.e.,
delocalisation of bonds makes the molecule more stable. Hence the butadiene
molecule in state (d) is more stable than in state (g), in which the Tt-electrons are
paired as " isolated " pairs, each pair covering two carbon atoms. This energy
of stabilisation could be called the delocalisation energy (Coulson), but it is more
usual to call it the resonance energy. It should here be noted that delocalisation
of bonds is in M.O. theory what resonance is in V.B. theory.

It can be seen from the foregoing discussion that the M.O. treatment of
conjugated systems does away with the idea of " bonds " between atoms (this
applies to the rc-bonds, and not to the a-bonds). Also the term conjugation is

UNSATURATED HYDROCARBONS 89

used in M.O. theory to indicate the existence in any part of a molecule of molecular
orbitals which embrace three or more nuclei. It is important to note that a conjug-
ated system (denned as above) always contains double bonds, but that the
reverse is not necessarily the case, e.g., ethylene.

Since the electron cloud covers the whole of the butadiene molecule, an electrical
influence in one part of the system is easily propagated to another (cf. p. 517).
Calculation (Coulson and Longuet-Higgins, 1947) has shown that when, for
example, bromine attacks butadiene, the bromine molecule approaches the end of
butadiene molecule most easily, and produces an alternate polarity:

© © ©
>- C, C 2 C 3 C 4 + Br

Br

s*

Br The negative bromide ion can then attack C a or C 4 .

I The relationship between the observed bond-length and the value

18- expected on the assumption that it is a " pure " single or double

° r bond has been put on a quantitative basis. In the V.B. method the

double-bond character of a bond may be calculated from a knowledge

of the observed bond length, the values of the single bond in ethane, and the

double bond in ethylene being taken as standard lengths for " pure " single and

double bonds, respectively. Calculations by Pauling et al. (1933) have shown

that the "single" bond in butadiene has about 20 per cent, double-bond character

(see also below).

In the M.O. method, the character of a bond is denned by its fractional bond
order, where the bond orders of 1, 2 and 3 are given to the bonds in ethane,
ethylene and acetylene, respectively. Since also the method of calculation is
different from that of the V.B. method, the numerical values obtained by the
two methods are different. Even so, these values always correspond.

Coulson (194 1, 1947) has shown that the butadiene molecule may be
represented as shown in (IV).

H H

"\ 1-894 I 1-447 I 1-891 /^
0-888 0-381 0-391 0-838

(IV)

Calculation gives a bond order of 1-894 to the two outer " double " bonds, and
a bond order of 1-447 to the central " single " bond. Thus the total bond number
of either of the end carbon atoms is 2 X i-o + 1-894 = 3-894, and the total
bond number of either middle carbon is i-o + 1-447 + I- 894 = 4-341. Further-
more, since calculation has shown that the maximum bond number for a carbon
atom is 4-732, it follows that each carbon atom in butadiene has the " free
valency " shown in (IV) (Coulson has suggested that free valency be represented
by an arrow). On the other hand, if a structure containing fractional double
bonds is written with single bonds labelled with the bond order, and charges are
placed on the atoms, then the resulting diagram is known as a molecular
diagram, e.g., the molecular diagram for benzene is (V) (see also p. 528). Among
other things, a molecular diagram enables one to estimate the most likely points
of attack (see, e.g., pyrrole, p. 750).

ACETYLENES OR ALKYNES
The acetylenes are unsaturated hydrocarbons that contain one triple
bond. The simplest member of the series is acetylene CH^CH, and hence
this homologous series is often referred to as the " acetylene series ". The
acetylenes have the general formula C n H 2 „_ 2 and the triple bond is also
known as the " acetylenic bond .".

90 ORGANIC CHEMISTRY

Nomenclature. One method is to name higher homologues as derivatives
of acetylene, the first member of the series, e.g.,

CH 3 *C=CH methylacetylene

CH 3 *C=OCH 2 *CH 3 ethylmethylacetylene

In the I.U.P.A.C. system of nomenclature the class suffix is -yne, and the
rules for numbering are as for the olefins (p. 63), e.g.,

CH=CHethyne; CH 3 -C=OCH 3 but-2-yne
(CH 3 ) 2 CH , C=OCH 3 4-methylpent-2-yne

Acetylene or ethyne, C 2 H 2 , is the most important member of this series,
and it may be prepared by any of the following methods :

1. By the action of water on calcium carbide:

CaC 2 + 2H 2 — > C 2 H 2 + Ca(OH) a

This method of preparation is used industrially, since calcium carbide is
readily manufactured by heating calcium oxide with coke in an electric
furnace:

CaO + 3C — > CaC 2 + CO

Acetylene prepared from calcium carbide is not pure, but contains small
amounts of phosphine, hydrogen sulphide, arsine, ammonia, etc.; the
impurities present depend on the purity of the calcium carbide used.
Scrubbing with water is usually sufficient to reduce the amount of all the
impurities, except phosphine, below the limit necessary for the safe applica-
tion of acetylene for technical purposes, Phosphine is removed by means
of oxidising agents, e.g., acid dichromate or bleaching-powder, whereby the
phosphine is retained as phosphoric acid.

A more recent industrial preparation of acetylene is by the electric arc
cracking of methane-ethane mixtures which are derived from coal hydro-
genation (p. 60).

2. By the action of ethanolic potassium hydroxide on ethylene bromide.
The reaction proceeds in two steps, and under suitable conditions the
intermediate product vinyl bromide may be isolated:

CH 2 Br-CH 2 Br + KOH -^> CH 2 =CHBr + KBr + H 2

CH 2 =CHBr + KOH -^4- CH=CH + KBr + H 2

Ethylidene chloride may be used instead of ethylene bromide (or chloride),
and this reaction also proceeds in two steps :

CH 3 -CHC1 2 -^* CH 2 =CHC1 -^-> CH=CH

Sodamide can be used instead of ethanolic potassium hydroxide, and the yields
are usually better since there is less tendency to form by-products, e.g.,

— CH 2 «CBr s — + 2 NaNH 2 > -CSC- + 2NaBr + 2NH,

3. By the electrolysis of a concentrated solution of sodium (or potassium)
salt of maleic or fumaric acid (q.v.) :

CO a Na-CH=CH-C0 2 Na + 2 H 2 — > CH^CH + 2 C0 2 + 2NaOH + H 2

4. By heating a trihalogen derivative of methane with silver powder;
e.g., iodoform gives acetylene when heated with silver powder :-

2CHI3 + 6Ag — > C 2 H 2 + 6AgI

UNSATURATED HYDROCARBONS gi

Acetylene may be synthesised from its elements by striking an electric
arc between carbon rods in an atmosphere of hydrogen. It is also formed
by the incomplete combustion of hydrocarbons, e.g., when a bunsen burner
" strikes back ".

Acetylene is a colourless gas, b.p. — 84 , and has an ethereal smell when
pure. It is sparingly soluble in water but readily soluble in acetone. When
compressed or liquefied acetylene is explosive, but its solution under pressure
(10 atm.) in acetone adsorbed on some suitable porous material can be
handled with safety. Acetylene burns with a luminous smoky flame (due
to the high carbon content), and hence is used for lighting purposes. It is
also used in the oxy-acetylene blow-pipe, a temperature above 3000° being
reached. Acetylene is used for the preparation of a large number of com-
pounds, e.g., acetaldehyde, ethanol, acetic acid, etc. (see text).

Owing to the presence of a triple bond, acetylene is more unsaturated
than ethylene, and forms addition products with two or four univalent
atoms or groups, never one or three (cf. ethylene). A triple bond consists
of one CT-bond and two «-bonds. When two univalent atoms add on to a
triple bond the digonal arrangement changes into the trigonal, and the
further addition of two univalent atoms changes the trigonal into the
tetrahedral arrangement. Under suitable conditions it is possible to isolate
the intermediate olefin.

1. Acetylene adds on hydrogen in the presence of a catalyst, the reaction
proceeding in two stages :

CzH2 "iir* CaH * nr* CaH «

The intermediate product, ethylene, can be obtained in very good yield
if the hydrogenation is carried out with a measured amount. of hydrogen
in the presence of Adams' platinum-platinum oxide catalyst (p. 65), or
palladium carried on a barium sulphate support.

The partial catalytic reduction of a triple bond in a wide variety of
acetylenic compounds has been carried out, but great difficulty has been
encountered in partially reducing a triple bond conjugated with a double
bond, i.e., the following reduction has proved difficult:

— C^C— C=C<— hr — CH=CH— C=C<

Lindlar (1952), however, has now developed a catalyst for the partial hydro-
genation of a triple bond in a wide variety of compounds; it consists of a
Pd — CaCO s catalyst partially inactivated by treatment with lead acetate, or
better, by the addition of quinoline.

Dialkylacetylenes may be catalytically reduced to a mixture of cis- and trans-
olefins, the former predominating. On the other hand, reduction with sodium
in liquid ammonia produces the trans-ole&n. Acetylenes are also reduced to
trans-olefms by lithium in aliphatic amines of low molecular weight (Benkeser
et al., 1955) cis — Reduction may be carried out with diborane (p. 76).

R

i

No/" R \c/ H

I R/ \H H/ \R

R cis trans

Lithium aluminium hydride may also be used to partially reduce a triple bond
provided the molecule contains the grouping — C=C — C(OH)C^ •

92 ORGANIC CHEMISTRY

2. Acetylene adds on gaseous chlorine or bromine in the dark to form

acetylene di- and tetrahalides; the addition is catalysed by light and

metallic halides (cf. olefins) :

ci ci
C 2 H 2 > C 2 H 2 C1 2 >- C 2 H 2 Cl4

Direct combination of acetylene with chlorine may be accompanied by
explosions, but this is prevented by the presence of a catalyst.

Acetylene reacts with dilute bromine water to produce acetylene di-
bromide:

CH=CH + Br 2 — '->- CHBr=CHBr

With liquid bromine and in the absence of a solvent, acetylene forms acetylene
tetrabromide:

C 2 H 2 + 2Br 2 — >■ C 2 H 2 Br 4

Acetylene adds on iodine with difficulty, but if the reaction is carried out in
ethanolic solution, acetylene di-iodide is obtained:

_ __ _ ethailOl _ «__ — ~-rT-r

ch=ch + 1 2 > CHI=CHI

Acetylene also undergoes substitution with halogen provided the right
conditions are used, e.g., dichloroacetylene is formed when acetylene is
passed into sodium hypochlorite solution at o° in the absence of air and
light:

C 2 H 2 + 2NaOCl — -> C a Cl 2 + 2NaOH

Similarly, if acetylene is passed into a solution of iodine in liquid ammonia,
di-iodoacetylene is formed (Vaughan and Nieuland, 1932) :

C 2 H 2 + 2l 2 + 2NH3 — > C 2 I 2 + 2 NH 4 I

These substitution reactions of acetylene are characteristic of hydrogen
only in the =CH group. Thus, for example, but-i-yne, but not but-2-yne,
can undergo these substitutions. Furthermore, it should be noted that the
halogen of the =CX group is very unreactive (cf. vinyl halides, p. 266).

3. Acetylene can add on the halogen acids, their order of reactivity
being HI>HBr>HCl>HF; HF adds on only under pressure (cf ethylene).
The addition of the halogen acids can take place in the dark, but is catalysed
by light or metallic halides. The addition is in accordance with Markowni-
koff's rule, e.g., acetylene combines with hydrogen bromide to form first
vinyl bromide, and then ethylidene bromide:

CH=CH + HBr — h CH 2 =CHBr— -> CH 3 -CHBr 2

Peroxides have the same effect on the addition of hydrogen bromide to
acetylene as they have on olefins (p. 68).

The mechanism of the addition of halogens and halogen acids is probably
the same as that for the olefins, e.g., the addition of hydrogen bromide may
be as follows :

CHHECH H— Br — > Br CH=CH a — > CHBr=CH,
Vinyl bromide may undergo the electromeric effect in two ways :

(i) CH 2 =CH— Br — > CH,— CH— Br
(ii) CH a =CH— Br — > CH 2 — CH— Br

UNSATURATED HYDROCARBONS 93

At the same time vinyl bromide is also capable of existing as a resonance hybrid
(cf., e.g., chlorobenzene, p. 546) :

CH,=CH— Br : <-> CH,— CH=Br :
6 • • • •

Considering the high electron-affinity of bromine, (ii) would seem to be more
likely than (i), and the bromine atom would therefore cause the electron drift to
take place towards itself. Considering the resonance effect, the tendency would
be to drive the electrons in the opposite direction to that of (ii), and since the
resonance effect is much stronger than the inductive effect (p. 519) direction (i)
will be the result; hence the addition of a molecule of hydrogen bromide to vinyl
bromide is :

+ vTV
- CH 3 -CHBr Br — ^ CH 3 -CHBr 2

4. When passed into dilute sulphuric acid at 60 ° in the presence of
mercuric sulphate as catalyst, acetylene adds on one molecule of water to
form acetaldehyde. The mechanism of this hydration probably takes place
via the formation of vinyl alcohol as an intermediate (cf. p. 70). Vinyl
alcohol has not yet been isolated; all attempts to prepare it result in the
formation of acetaldehyde (p. 265). Since we are suggesting it is an inter-
mediate product, but that it has never been isolated in this reaction, we
indicate this by enclosing vinyl alcohol in square brackets :*

CH=CH + H 2 ~> [CH 2 =CHOH] — > CH 3 -CHO

The conversion of acetylene into acetaldehyde is very important techni-
cally, since acetaldehyde can be used for the preparation of many important
compounds (see text).

The homologues of acetylene form ketones when hydrated, e.g., methyl-
acetylene gives acetone :

H SO

CH 3 -C=CH + H.O ' ■ *> [CH 3 -C(OH)=CH 2 ] — >- CH 3 -COCH 3

Hg'+

5. When acetylene is passed into dilute hydrochloric acid at 65 ° in the
presence of mercuric ions as catalyst, vinyl chloride is formed:

CH=CH + HC1— -> CH 2 =CHC1

Acetylene adds on hydrogen cyanide in the presence of barium cyanide as
catalyst to form vinyl cyanide:

CH=CH + HCN Ba(CN) ' > CH 2 =CHCN

Vinyl cyanide is used in the manufacture of Buna N synthetic rubber,
which is a copolymer of vinyl cyanide and butadiene.

When acetylene is passed into warm acetic acid in the presence of mercuric
ions as catalyst, vinyl acetate and ethylidene acetate are formed:

CH^CH + CH 3 -C0 2 H — -> CH 2 =CH-OOOCH 3

CH,=CIKKX>CH, + CH 3 -C0 2 H -^> CH s -CH(OOC-CH 3 ) 2

Vinyl acetate (liquid) is used in the plastic industry. Ethylidene acetate
(liquid), when heated rapidly to 300-400 , gives acetic anhydride and
acetaldehyde.

* In this book any compound that is suggested as an intermediate will be enclosed
in square brackets provided that it has not been isolated in the reaction shown.

94 ORGANIC CHEMISTRY

Acetylene reacts with nitric acid in the presence of mercuric ions to form
nitroform, CHfNO^. Acetylene combines with arsenic trichloride to
form Lewisite (p. 345). When acetylene is passed into methanol at 160-
200 in the presence of a small amount (1-2 per cent.) of potassium methoxide
and under pressure just high enough to prevent boiling, methyl vinyl ether
is formed:

CH=CH + CH 3 OH CH '° K > CH 2 =CH-OCH 3

This is used for making the polyvinyl ether plastics.

This process whereby acetylene adds on to compounds containing an active
hydrogen atom (p. 350) to form vinyl compounds is known as vinylation.

Acetylene and formaldehyde interact in the presence of copper acetylide
as catalyst to form butynediol, together with smaller amounts of propargyl
alcohol, CH=OCH 2 0H:

CH=CH + HCHO -^> CH=C-CH 2 OH

CH=CH + 2HCHO -^> CH 2 OH-C=C-CH 2 OH

Propargyl alcohol is used to prepare allyl alcohol, glycerol, etc. Butynediol
is used to prepare butadiene, etc.

This reaction in which acetylene (or any compound containing the =CH
group, i.e., a methyne hydrogen atom) adds on to certain unsaturated links
(such as in the carbonyl group), or eliminates a molecule of water by reaction
with certain hydroxy-compounds, is known as ethinylation. Thus the above
reactions with formaldehyde are examples of ethinylation ; another example
is the following:

R 2 N-CH 2 0H + CH=CH — > R 2 N-CH 2 -C=CH + H 2

Alkyl bromo- and chloro-methyl ethers (p. 354) add to acetylenes in the
presence of the corresponding aluminium halide to form olefins (Binda.cz
etal , i960); e.g.,

A1C1

Et0-CH 2 C1 + CH=CH — i-l> Et0-CH 2 -CH=CHC1

6. When acetylene is passed into hypochlorous acid solution, dichloro-
acetaldehyde is formed:

HOC1

CH=CH + H0C1 — > [CHC1=CH0H] >

[CHCVCH(OH) J — > CHC1 2 -CH0 + H 2

Dichloroacetic acid, CHC1 2 'C0 2 H, is also formed by the oxidation of di-
chloroacetaldehyde by the hypochlorous acid.

7. Acetylene and its homologues form ozonides with ozone, and these
compounds are decomposed by water to form diketones, which are then
oxidised to acids by the hydrogen peroxide formed in the reaction:

R-C=OR' + 3 — > R-C— C-R' — °-> R-C— OR' + H 2 2 — >

I I II II R-C0 2 H + R'-C0 2 H

0—0

Acetylene is exceptional in that it gives glyoxal as well as formic acid (Hurd
and Christ, 1936):

/°\ h 8 o
CH=CH + O, — > CH— CH — ^ CH-CH

II II II

o — o o o

UNSATURATED HYDROCARBONS 95

The triple bond in acetylenes is usually oxidised by potassium perman-
ganate to give acid fission products:

R-CEEC-R' -^?V R-COCOR' -^% R-C0 2 H + R'-C0 2 H

The intermediate a-diketone can be isolated if the oxidation is carried out in
the presence of magnesium sulphate.

8. When passed through a heated tube acetylene polymerises, to a small
extent, to benzene.

3 C 2 H 2 — > ||

\//

Homologues of acetylene behave in a similar manner, e.g., methylacetylene
polymerises to s-trimethylbenzene, and dimethylacetylene to hexamethyl-
benzene :

CH 3

3 CH 3 C=CH _► CH.1JJCH,

CH 3

3CH 3 C==CCH 3 >- prj Irxi

CH 3

Under suitable conditions acetylene polymerises to cyc/ooctatetraene
(q.v.):

/CH=CH X

4 C 2 H 2 — > || ||

HC X .CH

In addition to the above type of cyclic polymerisation, acetylene undergoes
linear polymerisation when passed into a solution of cuprous chloride in
ammonium chloride to give vinylacetylene and divinylacetylene :

CH^CH + CH=CH — > CH 2 =CH— C=CH — '->

CH 2 =CH— C=C— CH=CH 2

Compounds containing both a double and triple bond are named systematically
as alkenynes. The double bond is always expressed first in the name, and
numbers as low as possible are given to the double and triple bonds, even though
this may give " yne " the lower number. Thus vinylacetylene is but-i-en-3-yne,
and divinylacetylene is hexa-i : 5-dien-3-yne. The following compound,
CH S *CH=CH>C=CH, however, is pent-3-en-i-yne.

Vinylacetylene adds on one molecule of hydrogen chloride to the triple
bond to form chloroprene or 2-chlorobuta-i : 3-diene, the addition taking
place in accordance with Markownikoff 's rule:

CH 2 =CH— O^CH + HC1 — > CH 2 =CH— CC1=CH 8

Chloroprene readily polymerises to a rubber-like substance known as neoprene.
9. Acetylene forms metallic derivatives by replacement of one or both
hydrogen atoms, e.g., if acetylene is passed over heated sodium, both the
monosodium and disodium acetylides are formed:

Na Na

CHeeCH > CH=CNa > NaC=CNa

96 ORGANIC CHEMISTRY

By using a large excess of acetylene the main product is monosodium
acetylide, which is also obtained by passing acetylene into a solution of
sodium in liquid ammonia until the blue colour disappears. By treatment
of a fine dispersion of sodium in xylene at 100-105 ° with acetylene, sodium
acetylide can be obtained in 98 per cent, yield (Rutledge, 1957). The
monosodium derivative possesses the interesting property of being able to
absorb dry carbon dioxide to form the sodium salt of propiolic acid (q.v.) :

CH=ONa + C0 2 — > CH=OC0 2 Na

The alkali metal acetylides react with carbonyl compounds to form
acetylenic alcohols:

>CO + NaC=CR — > >C(OH)-C=CR

Acetylene also forms Grignard reagents by reaction with alkylmagnesium
halides (p. 351).

When acetylene is passed into an ammoniacal solution of cuprous chloride
or silver nitrate, cuprous acetylide Cu a C 2 (red) or silver acetylide Ag 2 C 2
(white) is precipitated. Both these compounds, when dry, explode when
struck or heated. When they are treated with potassium cyanide solution
pure acetylene is obtained, but on treatment with inorganic acid, the acetyl-
ene liberated is impure.

The acidic nature of hydrogen in acetylene is characteristic of hydrogen in
the group =CH, and it has been suggested that this is because the C — H bond
has considerable ionic character due to resonance.

H— C^C— H -<-> H— C=CH -<-> HC=C— H -<--> HC^CH

There is, however, evidence to show that the electronegativity of a carbon atom
depends on the number of bonds by which it is joined to its neighbouring carbon
atom (Walsh, 1947). Since re-electrons are more weakly bound than a-electrons,
the electron density round a carbon atom with re-bonds is less than that when
only o-bonds are present. Thus, a carbon atom having one re-bond has a slight
positive charge compared with a carbon atom which has only o-bonds. Thus,
the electronegativity of an sp 3 hybridised carbon atom is greater than that of
an sp 3 hybridised carbon atom. Similarly, a carbon atom which has two re-bonds
carries a small positive charge which is greater than that carried by a carbon
atom with only one re-bond. Thus the electronegativity of an sp hybridised
carbon atom is greater than that of an sp 2 hybridised carbon atom. It therefore
follows that the more s character a bond has, the more electronegative is that
carbon atom. Thus the attraction for electrons by hybridised carbon will be
sp>sp 3 >sp 3 . Therefore the ionic character of a C — H bond depends on the
state of hybridisation of the carbon atom and is greatest for sp hybridisation and
least for sp 3 hybridisation. Hence, in acetylene, the hydrogen atoms have a
large amount of ionic character (relative to ethylene and methane) and conse-
quently are more readily released as protons (than hydrogen in ethylene and
methane).

This interpretation of change in electronegativity with change in hybridisation
has a very important bearing on the problem of hyperconjugation (p. 271).

The structure of metallic carbides, i.e., compounds formed between carbon and
metals, is still a matter of dispute. It appears certain, so far, that the carbides
of the strongly electropositive metals: Na, K, Ca, Sr, Ba, are ionic — X-ray
crystal analysis has shown the lattice of these carbides to be ionic, containing the

ion C^C. These carbides react with water to produce acetylene. Copper and
silver carbides are not affected by water, and are explosive when dry. On
account of these differences it seems likely that the carbides of these two metals
are covalent, e.g., Cu — C^C — Cu. Thus these compounds may be regarded as
acetylides.

UNSATURATED HYDROCARBONS 97

In addition to the above carbides, there are a number of carbides which react
with water or dilute acids to produce methane, e.g., aluminium carbide; or a
mixture of hydrocarbons, e.g., uranium carbide which gives acetylene and other
unsaturated hydrocarbons; iron carbide which gives methane and hydrogen.
Many authors believe these carbides to be ionic, and include carbides of copper
and silver in this group.

There is also one other group of carbides which are highly refractory, and
which are extremely stable chemically; e.g., vanadium carbide is not attacked
by water or hydrochloric acid even at 6oo°. These carbides are believed to be
interstitial compounds, i.e., their lattice is not composed of ions, but resembles a
metallic or atomic lattice.

Structure of acetylene. By reasoning similar to that used for ethylene,
the structure of acetylene is shown to be H — CeeeeC — H, and may be repre-

H — C^-;C^H

(a) (b)

Fig. 4.2.

sented as Fig. 2(a). This representation, however, appears to be inadequate.
According to Coulson (1952), the two re-bonds form a charge cloud which has
cylindrical symmetry about the carbon-carbon axis (Fig. 2&).

HOMOLOGUES OF ACETYLENE

Homologues of acetylene may be prepared by any of the following methods :

1. By the action of ethanolic potassium hydroxide on vie- or gem-dihalogen
derivatives of the paraffins (cf. preparation of acetylene, method 2), e.g., methyl-
acetylene from propylene bromide :

etbanol

CH 3 -CHBr-CH 2 Br + 2KOH >■ CH 3 -CeeeecH + 2KBr + 2H a O

Since gem-dihalides are not usually readily accessible, and the wc-dihalides are,
the latter are used.

This method affords a simple means of introducing a triple bond into an
organic compound, e.g., w-butanol is catalytically dehydrated to but-i-ene, which
on treatment with, bromine gives 1 : 2-dibromobutane, and this, when heated
with ethanolic potassium hydroxide, yields but-i-yne:

CHa-CHj-CHj-CHjOH _^%. CH 3 -CH 2 -CH=CH 2 — V

35°° KOH

CH.-CH,-CHBr-CH 2 Br > CH 3 -CH 2 -Cee=CH

2. Monosodium acetylide (see reaction 9 of acetylene) is treated with an alkyl
halide, preferably a bromide, whereupon an acetylene homologue is produced :

CEEEECH + Na -i2^> CH=CNa -^> CHEEEEC-R + NaX

NH,

This reaction may be carried further as follows :

R-CEEEECH + Na - h ? uld > R-CEEEECNa -^> R-CeeeeC-R' + NaX

NH,

In practice this method is limited to the use of primary alkyl halides, since
higher sec. and tert. halides give mainly olefins when they react with the mono-
sodium derivatives of acetylene or its homologues.
E

9& ORGANIC CHEMISTRY

3. By the action of acetylene on a Grignard reagent (q.v.) and then treating the
resulting magnesium complex with an alkyl halide :

CH=CH + R— Mg— Br y RH + CH=C— Mg— Br -^-> CH=OR

The properties of the homologues of acetylene are very similar to those of
acetylene, particularly when they are of the type R-C=CH, i.e., contain the
=CH group. A very interesting reaction of the acetylene homologues is their
ability to isomerise when heated with ethanolic potassium hydroxide, the triple
bond moving towards the centre of the chain, e.g., but-i-yne isomerises to but-
2-yne:

-> [CH 3 -CH=C=CH 2 ] > CH 3 -CEEEOCR

There is a great deal of evidence to show that an allene is formed as an inter-
mediate.

On the other hand, when alkynes are heated with sodamide in an inert
solvent, e.g., paraffin, the triple bond moves towards the end of the chain; e.g.,
but-2-yne gives the sodium derivative of but-l-yne, which is converted into
but-i-yne by the action of water:

CH 3 -CEEOCH 3 + NaNH 2 -^?-> NH 3 -|- CH 3 -CH 2 -C=CNa-^->

CH 3 >CH 2 -C=CH

This reaction affords a means of stepping up the alkyne series by method 2 above.
A very useful acetylene derivative for synthetic work is ethoxyacetylene.
This may conveniently be prepared by the action of sodamide on chloroacet-
aldehyde diethyl acetal (Jones et al., 1954) :

NaNH

CH 2 Cl-CH(OC 2 H 5 ) 2 L> CH=OOC 2 H 6

Lithium derivatives of monosubstituted acetylenes can be prepared by inter-
action of the acetylene and lithium amide in dioxan (Schlubach et al., 1958).

QUESTIONS

1. Write out the structures of the isomeric pentenes and name them (use three
methods of nomenclature).

2. Give an account of the evidence for the structure of propylene.

3. Give as many methods as you can for separating a mixture of isobutane and
but-i-ene into its constituents.

4. Show how you would distinguish experimentally between the three isomeric
butylenes.

5. Name the compounds and indicate the conditions under which they are formed
when but-i-ene is treated with: — (a) bromine, (6) hydrogen bromide, (c) hydrogen
chloride, (d) hypochlorous acid, (e) ozone, (/) cone, sulphuric acid, (g) hydrogen,
(h) chlorine, (i) nitrosyl chloride, (j) peracetic acid, (k) heat.

6. Define and give examples of: — (a) unsaturation, (6) dehydration, (c) dehydro-
halogenation, (d) conjugation, (e) 1 : 4-addition, (/) polymerisation, (g) hydroxylation of
a double bond, (h) peroxide effect, (i) ozonolysis, (;') Prileschaiev's reaction, (A) poly-
condensation, (I) plastics, {m) thermoplastic, (n) thermo-setting plastic, (0) plasticiser,
(p) ethinylation, (q) vinylation, (r) Wittig reaction, (s) hydroboronation.

7. Write out the structures and names (by two methods) of the isomeric pentynes.

8. Give an analytical table to show how you would distinguish between ethane,
ethylene and acetylene.

9. Name the compounds and indicate the conditions under which they are formed
when acetylene reacts with the reagents named in question 5.

10. Prepare methylpropylacetylene using only acetylene and methyl iodide and any
inorganic compounds you wish.

11. Give an account of the evidence for the structure of acetylene.

12. Discuss the structures of butadiene and acetylene from the M.O. point of view.

UNSATURATED HYDROCARBONS 99

READING REFERENCES
Oilman, Advanced Organic Chemistry, Wiley (1942, 2nd ed.).

Vol I. (a) Ch. 1. The Reactions of the Aliphatic Hydrocarbons.

(b) Ch. 7. Unsaturation and Conjugation.

(c) Ch. 8. Synthetic Polymers.

Mayo and Walling, The Peroxide Effect, Chem. Reviews, 1940, 27, 357.

Riebsomer, Reactions of Nitrogen Tetroxide with Organic Compounds, ibid., 1945,

36, 197-211.
Levy, Scaife et al.. Addition of Dinitrogen Tetroxide to Olefins, J.C.S., 1946, 1093,

1096, 1100; 1948, 52.
Gorin, Kuhn, and Miles, Mechanisms of the Catalysed Alkylation of isoButane with

Olefins, Ind. Eng. Chem., 1946, 38, 745.
Groll and Hearne, Substitution in Straight-Chain Olefins, ibid., 1939, 31, 1530.
Bailey, The Reactions of Ozone with Organic Compounds, Chem. Reviews, 1958, 58, 925.
Organic Reactions, Wiley. Vol. II (1944), Cn - 8 - Periodic Acid Oxidation.
ibid., 1945, 42, 143-145. Oxidation with Lead Tetra-acetate.
De la Mare, Kinetics of Thermal Addition of Halogens to Olefinic Compounds, Quart.

Reviews (Chem. Soc), 1949, 3, 126.
Organic Reactions, Wiley. Vol. V (1949), Ch. I. The Synthesis of Acetylenes.
Raphael, Acetylenic Compounds in Organic Synthesis, Butterworth (1955).
Organic Reactions, Wiley. Vol. VII (1953), Ch. 7. Epoxidation and Hydroxylation

of Ethylenic Compounds with Organic Per-acids.
Lynch and Pausacker, The Oxidation of Olefins with Perbenzoic Acids. A Kinetic

Study, J.C.S., 1955, 1525.
Trapnell, Specificity in Catalysis by Metals, Quart. Reviews (Chem. Soc), 1954, 8, 404.
Bond, The Mechanism of Catalytic Hydrogenation and Related Reactions, ibid., 1954,

8, 279.
Cadogan and Hey, Free-Radical Addition Reactions to Olefinic Systems, ibid., 1954,

8, 308.
Bu'Lock, Acetylenic Compounds as Natural Products, ibid., 1956, 10, 371.
Kharasch, The Unique Properties of 2 : 4-Dinitrobenzenesulphenyl Chloride, /. Chem.

Educ, 1956, 33. 585-
Langford and Lawson, Characterisation of Organic Compounds with 2 : 4-Dinitro-
benzenesulphenyl Chloride, ibid., 1957, 34, 510.
Crawford, Infrared Spectra of Adsorbed Gases, Quart. Reviews (Chem. Soc), i960, 14,

378-

Advances in Organic Chemistry, Interscience, Vol. I (i960). Trjppett, The Wittig Re-
action, p. 83. Gunstone, Hydroxylation Methods, p. 103.

Bradley and Ledwith, The Reaction of Carbene with Alkyl Halides, J.C.S., 1961, 1495.

CHAPTER V

HALOGEN DERIVATIVES OF THE PARAFFINS

Halogen derivatives of the paraffins are divided into mono-, di-, tri-, etc.,
substitution products according to the number of halogen atoms in the
molecule.

Nomenclature. Monohalogen derivatives are usually named as the halide
of the corresponding alkyl group, e.g., C 2 H 6 C1 ethyl chloride; CH 3 "CHBr-CH 3
isopropyl bromide; (CH 3 ) 3 CC1 tert. -butyl chloride.

Dihalogen derivatives, (i) When both halogen atoms are attached to
the same carbon atom, they are said to be in the germinal (gem-) position.
Since the loss of two hydrogen atoms from the same carbon atom gives the
alkylidene radical, gm-dihalides are named as the alkylidene halides, e.g.,
CH 3 -CHBr 2 ethylidene bromide; CH 3 -CC1 2 - CH S isopropylidene chloride.

(ii) When the two halogen atoms are on adjacent carbon atoms they are said
to be in the vicinal (vie-) position, and these dihalides are named as the halide
of the olefin from which they may be prepared by the addition of halogen,
e.g., CH 2 ChCH 2 Cl ethylene chloride; (CH3) 2 CBr-CH 2 Br wobutylene bromide.

(iii) When there is a halogen atom on each of the terminal carbon atoms
of the chain, i.e., in the aoi-position, the compound is named as the poly-
methylene halide, e.g., CH 2 C1-CH 2 -CH 2 -CH 2 C1 tetramethylene chloride.

(iv) When the two halogen atoms occupy positions other than those
mentioned above, the compounds are named as dihalogen derivatives of
the parent hydrocarbon, the positions of the halogen atoms being indicated
by numbers (use principle of lowest numbers), e.g.,

CH 3 -CHC1-CH 2 -CHC1-CH 2 -CH 3 2 : 4-dichlorohexane

The alkyl halides, gem-, vie- and au-dihalides may also be named
by this (I.U.P.A.C.) system, e.g., CH s -CH 2 -CHBr-CH 3 2-bromobutane;
CH 2 C1*CH 2 'CH 2 C1 1 : 3-dichloropropane.

Polyhalogen derivatives are best named by the I.U.P.A.C. system (method
iv), and the names of the halogens (and any other substituents present) are
arranged alphabetically:

CH 2 Br-CHCl-CHCl-CH 3 i-bromo-2 : 3-dichlorobutane

CH 2 Cl-CHPCH(CH 3 )-CH 2 -CH 2 Br 5-bromo-i-chloro-2-iodo-3-methylpentane

ALKYL HALIDES

The alkyl halides have the general formula QjH^^X or RX, where X
denotes chlorine, bromine or iodine. Fluorine is not included, since fluorides
were, until fairly recently, chemical curiosities, and also do not behave like
the other halides. Fluorine compounds are discussed separately at the end
of this chapter.

General methods of preparation. 1. The method most widely used is to
replace the hydroxyl group of an alcohol by an X atom. This may be
done by means of various halogen reagents, and the accessibility of the
reagent is usually the factor deciding which one is used for the preparation
of a particular alkyl halide.

(i) Alkyl chlorides may be prepared by passing hydrogen chloride into
the alcohol in the presence of anhydrous zinc chloride (Groves' process),
e.g., ethyl chloride from ethanol:

C 2 H 6 OH + HC1— V C 2 H 5 C1 + H 2

HALOGEN DERIVATIVES OF THE PARAFFINS IOI

The yield of alkyl chloride depends on the nature of the alcohol. Primary
alcohols of the type R-CH 2 -CH 2 OH and R 2 CH-CH 2 OH usually give good
yields of alkyl chloride, but when the alcohol is of the type R S OCH 2 OH, the
main product is a tertiary alkyl chloride formed by rearrangement of the
molecule (see wsopentyl alcohol, below). Secondary alcohols, except for
t'sopropanol and sec.-butanol, react with hydrogen chloride to give mixtures
of chlorides, e.g., pentan-2-ol gives a mixture of 2- and 3-chlorides. Second-
ary alcohols containing highly branched radicals attached to the carbinol
group tend to give tertiary chlorides by molecular rearrangement, e.g.,
pinacolyl alcohol:

HC1

(CH 3 ) 3 OCHOH-CH 3 > (CH S ) 2 CC1-CH(CH 3 ) 2

Tertiary alcohols give very good yields of tertiary chloride with concen-
trated hydrochloric acid in the absence of zinc chloride.

The reaction between an alcohol and hydrogen chloride is reversible, and
it is believed that the zinc chloride functions as a dehydrating agent, thus
inhibiting the backward reaction.

The formation of the various halides by rearrangement is one example of
class of rearrangements often referred to as 1,2-sbifts. In this type of re-
arrangement a group migrates from a carbon atom (the migration origin)
to an adjacent atom (the migration terminus), which is usually carbon or
nitrogen. The rearrangement may be expressed in the following general
form:

z z z

i_ B ^ Y — > Y + 1-B+ -> A-B re a g ent > P™^

When Y ionises, B is left with only six valency electrons (open sextet), and
consequently carries a positive charge. Z migrates with its bonding pair of
electrons, and A now has an open sextet. Attack now takes place at A by Y -
or some other nucleophilic reagent to form the products. This mechanism
was first suggested by Whitmore (1932) and is often referred to as the
Whitmore mechanism.

The nature of Z is quite varied; the " key " atom (i.e., the atom joined
to the bonding pair which migrates) may be carbon, halogen, oxygen, etc.
If Z has a lone pair of electrons, e.g., halogen, then this pair may be used in
the migration to give a cyclic intermediate, i.e., a bridged ion.

1-

A

-B + — >■ A B — > products

If Z is an aryl group, then this group can supply an electron pair to form a
bridged iort (see, e.g., p. 172). If Z is an alkyl group, then there are no lone
pairs or multiple bonds, but again it is believed that a bridged ion is possible,
i.e.,

R ,- R -.

— c C + — >- >c=c<

I I

This type of bridged ion contains three partial bonds formed from one pair
of electrons. This idea of bridged carbonium ions was first suggested by
Nevell et al. (1939), but it must be admitted that the existence of such an
ion has not yet been definitely established when the migrating group is an

102 ORGANIC CHEMISTRY

alkyl group. The formation of these bridged ions in the 1,2-shifts is an
example of neighbouring group participation, and when the rate of the re-
arrangement is increased because of this effect, the rearrangement is said
to be anchimerically assisted (Winstein et al., 1953; see also p. 415).

A very important point about 1,2-shifts is that they are intramolecular,
i.e., Z is never actually free. If the mechanism were such that Z was free
during its migration, then the rearrangement would be called intermolecular.
The intramolecular nature of the 1,2-shifts has been established by using a
group Z which contains an asymmetric carbon atom attached to A, i.e.,
have type abcC — A. If Z actually separated during the migration, then it
means that the free migrating unit is a carbanion, i.e., dbcC'r. Examination
of such carbanions has shown that none ever retains its configuration; it
racemises, i.e., half the molecules have their configuration inverted, and so
the product is no longer optically active (the student should read pp. 413-
416 before proceeding further). Thus if Z moves with its bonding pair and
the configuration is retained in the product, then Z can never have been free,
i.e., the rearrangement is intramolecular.

We can now illustrate the foregoing discussion with weopentyl alcohol as
our example. Whitmore et al. (1932) showed that the reactions undergone
by Meopentyl alcohol and weopentyl halides are of two types: (i) replacement
reactions, which occur very slowly (if at all) and produce weopentyl com-
pounds; (ii) replacement (and elimination) reactions, which occur very fast
and produce tf-amyl compounds.

Hughes and Ingold (1946) have shown that the slow reactions take place
by the S N 2 mechanism (p. 106) and consequently without rearrangement,
e.g., Meopentyl bromide reacts with ethanolic sodium ethoxide to give ethyl
wtfopentyl ether:

Me 3 C Me 3 C „ Me.C

T +OEt-^ s - 7 ,- ^E^ I

BrCH 2 Br CH 2 OEt CH 2 OEt

On the other hand, when Meopentyl alcohol reacts with hydrobromic acid,
the reaction is fast, proceeds by the S^i mechanism (p. 106), and produces
i-amyl bromide by rearrangement (a 1,2-shift) :

Me 3 C-CH 2 OH^^Me s C-CH 2 — 0H 2 + ^
Me
H 2 0+ hi + +

Me 2 C— CH 2 — > Me 2 C— CH 2 Me > Me 2 CBr-CH 2 Me

There is no definite evidence that the intermediate is a bridged ion rather
than the classical carbonium ion.

A point of interest here is that the " driving force " of the rearrangement
is probably due to the stabilities of carbonium ions being tertiary >
secondary > primary due to the delocalisation of the charge by the inductive
effect of the alkyl groups increasing with the number of alkyl groups attached
to the positively charged carbon atom (in the classical carbonium ion; cf.

P- 33):

R H H

R-*-G--<-R> R-+-C-<-R > R>-G— H

+ + +

Alkyl bromides may be prepared:

(a) By refluxing the alcohol with excess constant-boiling hydrobromic
acid (48 per cent.) in the presence of a little sulphuric acid, which must
behave catalytically, since in its absence the reaction is slow.

HALOGEN DERIVATIVES OF THE PARAFFINS 103

The yield of alkyl bromide is usually excellent; when a secondary or ter-
tiary alcohol is used, it is better to omit the sulphuric acid, since this would
dehydrate these alcohols to olefins and thereby reduce the yield.

(V) By heating the alcohol with potassium bromide and concentrated
sulphuric acid in excess :

ROH + KBr + H 2 S0 4 — > RBr + KHS0 4 + H 2

The yield is very good for primary alcohols only; secondary and tertiary
are readily dehydrated to olefins under these conditions.

Alkyl iodides may be prepared in good yield by refluxing the alcohol with
excess of constant-boiling hydriodic acid (57 per cent.). Stone et al. (1950)
have shown that alkyl iodides may be prepared in good yields by heating
alcohols, ethers or olefins with sodium or potassium iodide in 95 per cent,
phosphoric acid.

(ii) Any alkyl halide may be prepared by the action of a phosphorus
halide on the alcohol. Phosphorus pentachloride gives variable yields
depending on the alcohol:

ROH + PC1 5 — > RC1 + HC1 + POCI3

Phosphorus trichloride gives poor yields of alkyl chloride except with alcohols
which tend to react by an S N I mechanism, e.g., tert. -alcohols behave in
accordance with the equation :

PCI3 + 3ROH — > 3RCI + P(OH) 3

The yields of alkyl halide with phosphorus tribromide or tri-iodide are
v.g.-ex. for primary alcohols; less for secondary alcohols, and still less for
tertiary. These phosphorus trihalides are usually prepared in situ ; bromine
or iodine is added to a mixture of red phosphorus and alcohol, and warmed,
(iii) Thionyl chloride (one molecule) refluxed with alcohols (one molecule)
forms alkyl chlorides in the presence of pyridine (one molecule) [Darzens
procedure] (see also p. 415) :

ROH + SOCl 2 pyri<Une > RC1 + S0 2 + HC1

In a number of cases only a small amount of pyridine need be used to
give the same yield of alkyl chloride as in the Darzens procedure.

2. By the addition of halogen acids to an olefin.

3. By direct halogenation, i.e., substitution reactions of the paraffins
with halogen, and this may be brought about by (a) light, (b) catalysts
or (e) heat.

(a) Photohalogenation is carried out by treating the paraffin with chlorine
or bromine at ordinary temperature in the presence of light. The reaction
is believed to take place by a free-radical chain mechanism, the initiation
being brought about by the formation of chlorine atoms by the U.V. part
of the spectrum of the light:

Cl 2 ^aCl-
Cl- +CH 4 — ^HC1 + CH 3 -
CH S - + Cl 2 — > CH 3 C1 + CI-

CI- + CH3CI — > HC1 + CH 2 C1-

CH 2 C1- + Cl 2 — > CH 2 C1 2 + CI- etc.

The termination of the chain reaction may take place by adsorption of
the chlorine atoms on the walls of the containing vessel, or by two chlorine
atoms combining with each other to form a chlorine molecule. There

104 ORGANIC CHEMISTRY

appears to be no evidence to show that CH 3 -, -CH 2 C1, etc., radicals combine
to terminate the chain reaction (c/. addition of chlorine to ethylene, p. 66).
Ritchie and Winning (1950) suggest that the chain-ending involves the forma-
tion of Clg molecules:

CI- + Cl 2 -^ Cl 3 -; 2 C1 3 - — > 3 C1 2

(b) Catalytic halogenation is carried out by treating the paraffin with
halogen in the presence of various metallic halides, e.g., cupric chloride
catalyses chlorination; ferric bromide, bromination.

(c) Thermal halogenation. Thermal chlorination has been studied in
great detail by Hass, McBee and their co-workers (1935 onwards), and as
a result of their work they suggested a number of rules for chlorination:

(i) If high temperature is avoided, no carbon skeleton rearrangements occur
in either thermal or photochemical chlorination.

(ii) Every possible monochloride is formed, and over-chlorination, i.e., chlorina-
tion beyond monosubstitution, may be suppressed by controlling the ratio of
chlorine to paraffin (this rule also holds good for photochemical chlorination).

(in) The order of ease of substitution is tertiary hydrogen> secondary>
primary. At 300°, with reaction in the vapour phase, the relative rates of
substitution of primary, secondary and tertiary hydrogen atoms are
1-00:3-25: 4-43.

(iv) As the temperature rises above 300 the relative rates of substitution tend
to become equal, i.e., 1:1:1. Increased pressure causes an increase in the
relative rate of primary substitution.

(v) In vapour-phase chlorination the presence of a chlorine atom on a carbon
atom tends to prevent further reaction on that carbon atom during the second
substitution. According to Tedder et al. (1960), this generalisation is a poor
approximation to the truth. These authors have shown the effect of the
halogen atom already present in the molecule is to retard substitution at a (3-
carbon atom, and also affects substitution at an a-carbon atom.

In the past it was believed that paraffins always tended to complete sub-
stitution. Rule (ii) shows this is not the case. Methane, however, has
been found to be an exception; chlorination of methane always results in
a mixture of all four substitution products, their relative amounts depending
on the ratio of chlorine to methane (see, e.g., methyl chloride).

Bromination takes place with greater difficulty than chlorination, and
there is less tendency for polysubstitution.

The mechanism of thermal halogenation is believed to take place by a free-
radical chain reaction (cf. photohalogenation) . The initiation of the chain
reaction is brought about by the thermal dissociation of chlorine molecules into
separate atoms:

heat

Cl a > CI- + CI-

CI- + CH 4 — > HC1 + CH 3 -, etc.

4. Direct chlorination of paraffins may be effected by means of sulphuryl
chloride. Sulphuryl chloride in the absence of light and catalysts does not
react with paraffins even at their boiling points, but in the presence of light
and a trace of an organic peroxide the reaction is fast :

RH + S0 2 C1 2 — y RC1 + S0 2 + HC1

The mechanism of this reaction is still obscure, but in view of the fact that
it is catalysed by organic peroxides (which are known to generate free
radicals), it is quite likely that chlorination with sulphuryl chloride proceeds
by a free-radical chain reaction.

HALOGEN DERIVATIVES OF THE PARAFFINS 105

In general, the products obtained by chlorination with sulphuryl chloride are
the same as those obtained by photochemical or thermal chlorination, and it has
been found that when sulphuryl chloride is used :

(i) The order of ease of replacement of a hydrogen atom is tertiary >
secondary> primary.

(ii) The second chlorine atom tends to substitute on that carbon atom
which is as far away as possible from the carbon atom already joined to the
chlorine atom.

(iii) It is difficult to get two, and impossible to get three, chlorine atoms
on the same carbon atom.

Alkyl chlorides and bromides are generally obtained fairly easily, iodides
not so easily. Fluorides have to be prepared by special means (see later in
this chapter). In many cases the iodide may be obtained from the corre-
sponding chloride or bromide by treating the latter in acetone or methanol
solution with sodium iodide (see p. 54) :

RC1 + Nal — - > RI + NaCl

Iodides may also be prepared from the corresponding bromide via the
Grignard reagent :

RBr + Mg -^> R— Mg— Br -^ RI

5. Hunsdiecker et al. (1935) found that various salts of the fatty acids
are decomposed by chlorine or bromine to form the alkyl halide, e.g.,

R-C0 2 Ag + Br 2 — > RBr + CO a + AgBr

The silver salt appears to give the best yield, but the yield also depends on
the solvent used; Stoll et al. (1951) found that trichloroethylene is a better
solvent than either carbon tetrachloride or carbon disulphide. The yield of
halide is primary>secondary>tertiary, and bromine is generally used,
chlorine giving a poorer yield of alkyl chloride, and iodine tending to form
esters.

aR-CO-jAg + I 2 — > R-C0 2 R + C0 2 + aAgl

Apart from preparing alkyl halides, this reaction also offers a means of
stepping down the fatty acids and alcohols.

The reaction considered above is often referred to as the Hunsdiecker
reaction or the Borodine— Hunsdiecker reaction.

6. Rydon et al. (1954) have shown that alkyl halides may be prepared, in
good yield, by the addition of halogen to a mixture of an alcohol and tri-
phenyl phosphite :

(C 6 H 5 0) 3 P + ROH + X 2 — > RX + XPO(OC 6 H 5 ) 2 + C 6 H 5 OH

General properties of the alkyl halides. The lower members methyl
chloride, methyl bromide and ethyl chloride are gases; methyl iodide and
the majority of the higher members are sweet-smelling liquids. The order
of the values of the boiling points (and densities) of the alkyl halides is
iodide >bromide>chloride>fluoride. In a group of isomeric alkyl halides,
the order of the boiling points is primary>secondary>tertiary. Many of
the alkyl halides burn with a green-edged flame. The chemical reactions
of the alkyl halides are similar, but they are not equally reactive, the order
of reactivity being iodide >bromide>chloride; the reactivity of alkyl
fluorides depends on the nature of the fluoride (p. 122). Alkyl iodides are
sufficiently reactive to be decomposed by light, the iodide darkening due
to the liberation of iodine :

2RI — > R— R + I 2

106 ORGANIC CHEMISTRY

Alkyl halides (and the polyhalides) are covalent compounds, insoluble in
water, with which they cannot form a hydrogen bond, but soluble in organic
solvents.

The alkyl halides are classified as primary, secondary and tertiary,
according as the halogen atom is present in the respective groups — CH 2 X,
=CHX and ^CX, i.e., according as the halogen atom is joined to a primary,
secondary or tertiary carbon atom.

It has already been pointed out that alkyl groups have a +1 effect.
Before considering the reactions of the alkyl halides, it is important to
explain why an alkyl group is electron-repelling. Several explanations
have been offered; the following is a highly favoured one. As we have
seen, methane and ethane are non-polar (p. 54). Thus the methyl group has
a zero inductive effect ; and this is true for all alkyl groups since all paraffins,
whether straight- or branched-chain, have a zero dipole moment. When,
however, one hydrogen atom in a paraffin is replaced by some polar atom
(or group), the alkyl group now exerts a polar effect which is produced by the
presence of the polar atom. Thus, in an alkyl halide, the alkyl group
possesses an inductive effect, but it is one which is produced mainly by the
mechanism of interaction polarisation {cf. p. 20). It therefore follows that
the alkyl group is more polarisable than a hydrogen atom, and since most
of the groups attached to the alkyl group are electron-attracting groups, the
alkyl group thus usually becomes an electron-repelling group, i.e., alkyl
groups normally have a -f-I effect.

General reactions of the alkyl halides. The alkyl halides are extremely
important reagents because they undergo a large variety of reactions that
make them valuable in organic syntheses.

1. Alkyl halides are hydrolysed to alcohols very slowly by water, but
rapidly by silver oxide suspended in boiling water, or by boiling aqueous
alkalis (see also 5) :

RX + KOH — > ROH + KX

This type of reaction is an example of nucleophilic substitution (S N ) since the
attacking reagent is a nucleophilic reagent (p. 33). It has been assumed
that this type of heterolytic reaction in solution can take place by two
different mechanisms, unimolecular or bimolecular.

Unimolecular mechanism. This is a two-stage process, the first stage con-
sisting of slow heterolysis of the compound to form a carbonium ion, followed
by rapid combination between the carbonium ion and the substituting
nucleophilic reagent. Since the rate-determining step is the first one, and
since in this step only one molecule is undergoing a covalency change, this
type of mechanism is called unimolecular, and is labelled S N i (Ingold et al.,
1928, 1933). Thus, if the hydrolysis of an alkyl halide is an S^i reaction,
it may be written :

R^-^X- + R+^>ROH

fast

Bimolecular mechanism. This is a one-stage process, two molecules simul-
taneously undergoing covalency change in the rate-determining step. This
type of mechanism is called bimolecular, and is labelled S N 2. Since the rate-
determining step in this reaction is the formation of the transition state, the
hydrolysis of an alkyl halide by an S M 2 reaction may be written:

Ho"^ R— X — > HO— R— X — > HO— R + X~

Any factor that affects the energy of activation of a given type of reaction
will affect the rate and/or the mechanism. Attempts have been made to

HALOGEN DERIVATIVES OF THE PARAFFINS IOJ

calculate E in terms of bond strengths, steric factor, heats of solution of
ions, etc., but apparently the results are conflicting. The following dis-
cussion is therefore largely qualitative, and because of this, one cannot be
sure which are the predominant factors in deciding the value of the energy
of activation.

The main difference between the two mechanisms is the kinetic order of
the reaction. S N 2 reactions would be expected to be second order, whereas
S N i reactions would be expected to be first order. These orders are only
true under certain circumstances, e.g., in a bimolecular reaction, the order is
second-order if both reactants are present in small and controllable con-
centrations. If, however, one of the reactants is the solvent, this will be in
constant excess, and so such a bimolecular mechanism will lead to first-
order kinetics.

Another important difference is that in the S N 2 mechanism, the molecule
is always inverted. In the S N i mechanism, the configuration of the resulting
molecule depends on various factors (see p. 415).

Let us first consider the nature of R. There are two effects to be con-
sidered, the polar factor and the steric factor.

Polar factor. Consider the series of ethyl, wopropyl, and *-butyl halides.
Since the methyl group has a +1 effect, the larger the number of methyl
groups on the carbon atom of the C— X, the greater will be the electron
density on this carbon. If we use one arrow-head to represent (qualitatively)
the electron-repelling effect of a methyl group, then we have the following
state of affairs:

Me X Me X

Me^CH 2 -^-X JCH-»-X Me->-C-»>-X

Me * ,, V

Me ^

This increasing electron density on the central carbon atom increasingly
opposes attack at this carbon atom by a negatively charged nucleophilic
reagent. Thus the formation of the transition state for the S N 2 mechanism
becomes increasingly difficult. It can therefore be anticipated that the
S N 2 mechanism is made more difficult in passing from EtX to *-BuX. On
the other hand, since the S N i mechanism involves ionisation as the first step,
then it can be expected that as the electron density increases on the central
carbon atom, the bonding pair in C — X becomes more and more displaced
towards the X atom, and consequently ionisation of X as a negative ion
will become easier. It therefore follows that the tendency for the S N i
mechanism should increase from EtX to *-BuX. These anticipated results
have been realised in practice. Hughes, Ingold et al. (1935-1940) examined
the hydrolysis of alkyl bromides in alkaline aqueous ethanol and showed
that MeBr and EtBr undergo hydrolysis by the Sw2 mechanism, ZsoPrBr by
both S N 2 and S N i mechanisms, and <-BuBr by the S N i mechanism only. It
has also been shown, however, that the actual position where the mechanism
changes over from S N 2 to S N i in a given graded polar series (such as the one
above) is not fixed but also depends on other factors such as the concentra-
tion of the nucleophilic reagent and the nature of the solvent (see below).

Steric effects. The steric effect was originally thought to be a spatial
effect brought into play by mechanical interference between groups, and was
described as steric hindrance (see p. 686). The term steric hindrance con-
sidered the geometry of the reactant molecule. When, however, a molecule
undergoes chemical reaction, it does so via a transition state. Consequently
the geometry of both the initial and transition state must be taken into
consideration. Thus steric factors may affect the speed and/or the

108 ORGANIC CHEMISTRY

mechanism of a reaction. It is, however, often very difficult to distinguish
between steric and polar factors. Nevertheless, the effects of the steric
factor can often be assessed in some sort of qualitative manner. When
steric effects slow down a reaction, that reaction is said to be subject to
steric hindrance {or retardation), and when they speed up a reaction, that
reaction is said to be subject to steric acceleration.

When there is repulsion between non-bonded atoms in a molecule due to
their close proximity, forces of steric repulsion are said to be acting. When
the stability of a molecule is decreased by internal forces produced by inter-
action between the constituent parts, that molecule is said to be under steric
strain. There are three sources of steric strain : (i) repulsion between non-
bonded atoms (steric repulsion) ; (ii) distortion of bond angles {angle strain) ;
(iii) dipole interactions.

In an S N 2 reaction there will be five groups " attached " to the carbon
atom at which reaction occurs (p. 414). Thus there will be " crowding " in
the transition state, and the bulkier the groups, the greater will be the
compression energy, and consequently the reaction will be hindered sterically.

Let us consider the following example. Hughes et al. (1946) examined
the rate of exchange of iodide ions in acetone solution with MeBr, EtBr,
isoPvBr, and <-BuBr under conditions where only the Su2 mechanism
operated. The relative reactivities were found to be 10,000 ; 65; 0*50; 0-039.

-/~V rv *- *-

I R— Br — > I R Br — >- I— R + Br"

Thus increasing the number of methyl groups on the central carbon atom

increases the steric retardation.

The problem is somewhat different for the S N i mechanism. Here the

transition state does not contain more than four groups attached to the

central carbon atom, and hence one would expect steric hindrance to be

less important in the S N i mechanism. If, however, the molecule contains

bulky groups, then by ionising, the molecule can relieve the steric strain,

since the carbonium ion produced is flat (trigonal hybridisation) and so

there is more room to accommodate the three alkyl groups. Thus, in such

S N i reactions, there will be steric acceleration. Brown et al. (1949) showed

that the solvolysis of tertiary halides is subject to steric acceleration {solvo-

lysis is a nucleophilic substitution reaction in which the solvent is the nucleo-

philic reagent).

/~* H,0
R 8 C X -^ R S C + + X-

(tetrahedral; (trigonal; planar;

large strain) small strain)

These authors showed that as R increases in size, the rate of solvolysis
increases. However, the larger the R groups are, the more slowly will the
carbonium ion be expected to react with solvent molecules, since steric
strain will be re-introduced into the molecule. In cases Eke this, the
carbonium ion tends to undergo an elimination reaction to form an olefin
(c/. ethylene), and Brown et al. (1950) have shown that this elimination
process increases as the alkyl groups become larger (see p. 112).

It should be noted that a fundamental part of the S N i mechanism is the
postulate of the transient existence of carbonium ions. Symons et al. (1959)
have shown that monoaryl-carbonium ions are stable in dilute solutions of
sulphuric acid. These authors have also concluded, from a spectroscopic
examination of *-butanol in sulphuric acid, that there is the trimethylcarbonium
ion, CMe 3 + , in solution and that this ion is probably planar (c/. p. 366). It is
interesting to note that of the ions Me + , MeCH 2 + , Me a CH + , Me 3 C*, the order of
stability would be expected to increase from left to right, since the charge is

HALOGEN DERIVATIVES OF THE PARAFFINS IO9

being increasingly neutralised by the +1 effect of the methyl groups. On the
other hand, the triphenylmethyl carbonium ion, (C,H 6 ) 3 C + , has been isolated as
its perchlorate and borofluoride (Dauben et al., i960).

Now let us consider the effect of the nature of the halogen atom. Experi-
mental work has shown that the nature of X has very little effect, if any, on
mechanism, but it does affect the rate of reaction for a given mechanism, e.g.,
it has been found that in S N i reactions, the rate follows the order
RI>RBr>RCl. This may be explained by consideration of the C — X bond
energy, the values of which are C — CI, 77 k.cal. ; C — Br, 65 k.cal. ; C — I,
57 k.cal. Thus the C — I bond is most easily broken and the C — CI bond
least easily. These energy differences also explain the same order of in-
creasing rate in S N 2 reactions. As we have seen, the stronger the C — X
bond, the greater will be the energy of activation.

The more pronounced the nucleophilic activity of the attacking reagent,
i.e., the greater the availability of its unshared pair, then the more the S N 2
mechanism will be favoured, since in the S N i mechanism the reagent does
not enter into the rate-determining step of ionisation. However, it can also
be expected that as the nucleophilic activity of the reagent decreases, a
point will be reached where the nucleophilic activity is so low that the
mechanism will change from S N 2 to S^i. Hughes, Ingold, et al. (1935)
examined the mechanism and rate of decomposition of various trimethyl-
sulphonium salts (Me 3 S + X~; see also p. 335) and showed that for X = OH,
the mechanism is Su2, and for X = I, CI, or Br, the mechanism is S H i and
the rate decreases in this order.

Mechanism and rate of reaction are very much affected by the nature of
the solvent. It has been found that the ionising power of a solvent depends
on its dielectric constant and its power of solvation, and it appears that the
latter is more important than the former. Solvation is due to the " attach-
ment " between solvent and solute molecules, and one important con-
tributing factor is attraction of a charge for a dipole. Since electrostatic
work is done in the process of solvation, energy is lost by the system, and
consequently the system is more stable.

Although the solute molecules have ionised, the oppositely charged pair
of ions may become enclosed in a " cage " of the surrounding solvent
molecules, and may therefore recombine before they can escape from the
cage. Such a complex is known as an ion-pair, and their recombination is
known as internal return. It has now been shown that many organic
reactions proceed via ion-pairs rather than dissociated ions (see p. 273)
Thus we have the possibility of the following steps:

RX ^ R+X- =^ R + + X-

ion-pair dissociated ions

Many attempts have been made to correlate reaction rates and nature of
the solvent. Hughes and Ingold (1935, 1948) proposed the following
qualitative theory of solvent effects:

(i) Ions and polar molecules, when dissolved in polar solvents, tend to
become solvated.

(ii) For a given solvent, solvation tends to increase with increasing
magnitude of charge on the solute molecules or ions.

(iii) For a given solute, solvation tends to increase with the increasing
dipole moment of the solvent.

(iv) For a given magnitude of charge, solvation decreases as the charge is
spread over a larger volume.

(v) The decrease of solvation due to the dispersal of charge will be less
than that due to its destruction.

110 ORGANIC CHEMISTRY

Since the rate-determining step in an S N i reaction is ionisation, any
factor that assists this ionisation will therefore facilitate S N i reactions.
Solvents with high dipole moments (i.e., high polarity) are usually good
ionising media and, in general, it has been found that the more polar the
solvent, the greater is the rate of the S N i reaction. Let us consider the
following S N i reaction:

R-X ^±R--X-^U R + + X- -^> R-OH

^ fast

Since the transition state has a larger charge than the reactant molecule,
the former will be more solvated than the latter (rule ii) . Thus the transition
state is more stabilised than the reactant molecule, i.e., the energy of
activation is lowered, and so the reaction proceeds faster than had there
been no (or less) solvation.
The rates of S N 2 reactions are also affected by the polarity of the solvent :

-/*V Hv s- 8-

HO R— X ^ HO R X — > HO— R + X"

A solvent with a high dipole moment will solvate both the reactant ion
(nucleophilic reagent) and the transition state, but more so the former than
the latter, since in the latter the charge, although unchanged in magnitude
(S - = 1/2), is more dispersed than in the former. Therefore solvation tends
to stabilise the reactants more than the transition state (rule iv). Thus the
activation energy is increased, and so the reaction is retarded.

The rate of the following S N 2 reaction is increased as the polarity of the
solvent increases.

R 3 N R— X ^ R 3 — N R X — ^R 4 N + X"

Here the charge on the transition state is greater than that on the reactants.
Hence the transition state is more solvated than the reactants and con-
sequently stabilised, and so the activation energy is lowered.

The polarity of the solvent may change the mechanism of a reaction, e.g.,
Olivier (1934) showed that the alkaline hydrolysis of benzyl chloride in
50 per cent, aqueous acetone proceeds by both S N 2 and S N i mechanisms.
When water was used as solvent, the mechanism was now mainly S N i. The
dipole moment of water is greater than that of aqueous acetone, and so
ionisation of the benzyl chloride is facilitated.

2. Alkyl halides are reduced by nascent hydrogen (Zn/Cu or Na and
ethanol, Sn and HC1, etc.) to form the corresponding paraffins:

RX + 2[H] — > RH + HX

Lithium aluminium hydride also effects this reduction.

Alkyl iodides may be reduced by heating them with concentrated hydriodic
acid and a small amount of red phosphorus at 150 .

RI + HI — > RH + I a

Since alcohols are readily converted into the corresponding halide, this
offers a method of converting alcohols into paraffins with the same number
of carbon atoms.

3. Alkyl halides undergo the Wurtz reaction to form paraffins (p. 51) :

2RX + 2Na — > R— R + 2NaX

On the other hand alkyl halides, when heated with certain metallic alloys,
form organo-metallic compounds, e.g., ethyl chloride heated with a sodium
lead alloy under pressure gives tetraethyl-lead :

4 C 2 H 5 C1 + 4 Na/Pb — -> (C 2 H 5 ) 4 Pb + 4 NaCl + 3 Pb

HALOGEN DERIVATIVES OF THE PARAFFINS III

4. When an alkyl halide is heated at about 300°, and at a lower tempera-
ture in the presence of aluminium chloride as catalyst, the alkyl halide under-
goes rearrangement, e.g., i-bromobutane rearranges to 2-bromobutane :

CH 3 -CH 2 -CH 2 -CH 2 Br — > CH 3 -CH 2 -CHBr-CH 3

If there is no hydrogen atom on the carbon adjacent to the C— X group, an
alkyl group migrates; e.g., weopentyl chloride rearranges to 2-chloro-2-
methylbutane (see also p. 102) :

(CH 3 ) 3 OCH 2 Cl — > (CH 3 ) 2 CC1-CH 2 -CH 3

5. When alkyl halides are boiled with ethanolic potassium hydroxide,
olefins are obtained, e.g., propyl bromide gives propylene:

CHs-CH^CHaBr + KOH ethanol > CH 3 -CH=CH 2 + KBr + H 2

According to Hughes and Ingold (1941), the elimination reaction between
an alkyl halide and aqueous potassium hydroxide may take place by either
a unimolecular (Ei) or a bimolecular (E2) mechanism.

' -— ' slow

(i) Ei: H— CH 2 — CH 2 — X — -> H— CH 2 — CH 2 + + X"

fast

-CH 2 — CH 2 + -~^-> H 2 + CH 2 =CH 2

(ii) E2: HO- H— CH 2 — CH 2 — X — > H 2 + CH 2 =CH 2 + X"

On the other hand, in ethanolic potassium hydroxide, which contains
potassium alkoxide RO _ K + , the reaction tends to take place by an E2 (iii)
or an S N 2 (iv) mechanism.

r\ ^ — v rv

(iii) RO- H— CH — CH 2 — X — > ROH + CH a =CH 2 + X~

r\ 3 r\ 3

(iv) RO- CH 2 — X — y RO— CH 2 + X"

The course of the reaction depends on the nature of the alkyl halide and
on the conditions used. Under the most favourable conditions for ethylene
formation, ethyl halides give only 1 per cent, ethylene, the main products
being replacement products. z'soPropyl halide gives up to 80 per cent,
propylene, and tertiary butyl halide 100 per cent, isobutene if the reaction
is carried out in ethanolic potassium hydroxide, which favours the bimolecular
mechanisms (iii) and (iv).

When two olefins may be formed by dehydrohalogenation of an alkyl
halide, the one that predominates is that which is the most substituted
olefin, i.e., the one carrying the largest number of alkyl substituents
(Saytzeff's rule, 1875), e.g.,

CH,-CHBr-CH,-CH,

p> CH 3 -CH=CH-CH 3 (I)
CH 2 =CH-CH 2 -CH 3 (II)

(I) predominates (this is a disubstituted ethylene, whereas (II) is a monosubsti-
tuted ethylene).

Another way of stating Saytzeff's rule is that hydrogen is eliminated from that
carbon atom joined to the least number of hydrogen atoms. Ingold et al. (1941)

112 ORGANIC CHEMISTRY

have offered the following explanation for this rule. If (I) and (II) are produced
by an E2 mechanism, their formation may be written as follows :

CH 3 CH 3

(v) HO- H— CH— CH— Br >• H a O + CH=CH + Br~

CH 3 CH 3 (I)

rv

HO" H— CH 2 CH,

^1 V* II

(vi) CH 2 — CH— Br — >■ H a O + CH a — CH + Br~

CH„ CHg (II) __

In the transition state of (v), the CH 3 group can enter into hyperconjugation
with the partly formed double bond (p. 269), thereby lowering the energy of
the transition state. In (vi), however, the CH 3 group cannot enter into
hyperconjugation with the partly formed double bond, and so the energy of this
transition state is higher than that of (v), and consequently the latter path is
favoured.

On the basis that the hyperconjugative effect determines the stability of the
olefin produced, it can be deduced that the ease of formation of olefin from an
alkyl halide by the E2 mechanism should be t-R>s-R>p-R, e.g., t-Bu (2 Me
groups) > isoPr (1 Me group) >Et (no Me groups) :

H H H

H— C-j- ••H"--OEt H— C H OEt H— C"j H OEt

Me— O BrS- Me— 6 Br^ H— 6 Br»-

a

Me H H "

Saytzeff's rule also applies when the mechanism is Ei. In this mechanism
the formation of the carbonium ion is the rate-determining step (as for the S N i
mechanism), and then the carbonium ion stabilises itself by elimination of a
proton to form an olefin, the stability of which will be largely determined by the
hyperconjugative effect.

Steric factors also appear to play a part in the direction of olefin formation.
Thus Brown et at. (1953) have shown that potassium 2.-butoxide gives 53-4
per cent, of (II), and potassium ethoxide 19 per cent, of (II). Brown et al. (1956)
have also shown that in the molecule R*CH 2 -CBr(CH 3 ) a , as R increases in
branching (methyl<ethyl<«opropyl<i.-butyl), a regular increase is obtained
in the ratio of i-/2-olefin in the product. These authors also showed that when
the potassium salts of ethanol, /.-butanol, i.-pentanol and triethylmethanol are
used with secondary and tertiary alkyl bromides, the yield of i-plefin is increased
from the left to the right alkoxide.

A point of interest here is that substitution and elimination often occur
together. Experimental results have shown that for a given compound, the
olefin yield is higher for the E2 mechanism than for the El. Since a high
concentration of the base with high nucleophilic power favours the E2
mechanism, olefin preparation is best carried out in concentrated solutions
of strong bases. Furthermore, since ionising solvents favour the Ei
mechanism, the preparation of olefins is best carried out in solvents with
low ionising power. Thus ethanolic sodium hydroxide solution (solvent of
low ionising power and a very strong base, OEt") gives better olefin yield
than does aqueous sodium hydroxide (solvent of high ionising power and a
base, OH", which is weaker than OEt"). Also, a base of weak nucleophilic
power will have little affinity for hydrogen, and consequently olefin forma-

HALOGEN DERIVATIVES OF THE PARAFFINS 113

tion will be decreased and substitution encouraged. The acetate ion is a
very weak base, and hence its use will give an ester as the main product
rather than olefin. Thus the conversion of an alkyl halide into the alcohol
is often best carried out via the ester, which is then hydrolysed.

6. When alkyl halides are heated with ethanolic ammonia under pressure,
a mixture of amines, i.e., substituted ammonias, is obtained, e.g.,

RX + NH S — > R-NH 2 + HX — > R-NH 2 -HX or [R-NH 3 ]+X-

Only primary alkyl halides give good yields of amines; secondary —
except j'sopropyl halide — and tertiary halides form mainly olefins.

7. When alkyl halides are heated with aqueous ethanolic potassium
cyanide, alkyl cyanides are obtained:

RX + KCN — > RCN + KX

Tertiary alkyl cyanides cannot usually be prepared this way, since tertiary
alkyl halides tend to eliminate hydrogen halide very readily when treated
with potassium cyanide.

Alkyl cyanides are very important compounds, since they may be used
to prepare many other compounds, e.g., acids, amines, etc.

When alkyl halides are heated with aqueous ethanolic silver cyanide,
cyanides and the isocyanides are formed:

RX + AgCN — > R-CN + AgX
RX + AgCN — > R-NC + AgX

8. When alkyl halides are heated with an ethanolic solution of a silver
salt, esters are obtained:

R-COjsAg + R'X — > R-C0 2 R' + AgX

The acid may be organic or inorganic. If the salt is silver nitrite two
isomeric compounds are obtained, the nitrite (ester) and the m>o-compound:

RX + Ag— O— N=0 — > R— O— N=0 + AgX
RX + Ag— O— N=0 — > R— N< + AgX

9. When alkyl halides are heated with sodium alkoxides, i.e., sodium
derivatives of the alcohols, ethers are obtained. This is the Williamson
synthesis (see p. 141) :

RONa -f- R'X — > R-O-R' + NaX

A modification of this reaction is to heat together an alkyl halide and an alcohol
in the presence of dry silver oxide :

2ROH + 2R'I + Ag a O — -> 2R-0-R' + 2AgI + H a O

This reaction is known as the Purdie method of alkylation, and is very important
in sugar chemistry.

10. Alkyl halides heated with aqueous ethanolic sodium hydrogen
sulphide form thioaicohols :

RX + NaSH — >■ RSH + NaX

11. Alkyl halides heated with an ethanolic solution of a mercaptide,
i.e., a metallic derivative of a thioalcohol, form thioethers:

RX + R'SNa — V R-S-R' + NaX

114 ORGANIC CHEMISTRY

12. When alkyl halides are heated with sodium, potassium or ammonium
sulphite, a sulphonate is formed. This is known as the Strecker reaction:

RX + Na 2 SO a — > R-S0 3 Na + NaX

13. Alkyl halides may be used in the Friedel-Crafts reaction, e.g., benzene
reacts with methyl iodide in the presence of anhydrous aluminium chloride
to form toluene:

Aici
C 6 H 6 + CH3I V C 6 H 6 -CH 3 + HI

14. Alkyl halides are used to prepare Grignard reagents:

RX + Mg -^> R— Mg— X

Methyl chloride, CH 3 C1, is prepared industrially :

(i) By the action of hydrogen chloride on methanol in the presence of
anhydrous zinc chloride :

CH3OH + HC1 -^> CH3CI + H 2

(ii) By heating trimethylamine hydrochloride with hydrochloric acid
under pressure :

(CH 3 ) 3 N-HC1 + 3HCI — > 3CH S C1 + NH 4 C1

(iii) By chlorinating methane with chlorine diluted with nitrogen, the
ratio by volume of CH 4 : Cl 2 : N 2 being 8 : 1 : 80. The reaction is carried
out in the presence of partly reduced cupric chloride as catalyst. All four
chloromethanes are obtained, the methyl chloride comprising 90 per cent, of
the chlorine used:

CH 4 + Cl 2 4- (N 2 ) -^ CH 3 C1 + HC1 + (N 2 ) 4- (CH 2 C1 2 4- CHC1 3 4- CC1 4 )

By adjusting the ratio of chlorine to methane, each chloromethane can
be obtained as the main product.

Methyl chloride is a colourless gas, b.p. —24°. It is fairly soluble in
water, and readily in ethanol. It is used in the manufacture of aniline
dyes, as a refrigerating agent, local anaesthetic and as a fire extinguisher.

Methyl iodide, CH 3 I, is prepared industrially:

(i) By warming a mixture of methanol and red phosphorus with iodine :

6CH 3 OH 4- 2P 4- 3l 2 — > 6CH 3 I 4- 2H 3 P0 3

(ii) By the action of methyl sulphate on potassium iodide solution in the
presence of calcium carbonate:

KI 4- (CH 3 ) 2 S0 4 -+ CH 3 I + K(CH 3 )S0 4 (90-94%)

Methyl iodide is a sweet-smelling liquid, b.p. 42'5°. Since it is a liquid,
it is easier to handle than methyl chloride, and so is used a great deal as a
methylating agent in laboratory organic syntheses, but the chloride is used
industrially, since it is cheaper.

Ethyl chloride, C 2 H B C1, is prepared industrially:

(i) By the action of hydrogen chloride on ethanol in the presence of
anhydrous zinc chloride :

C 2 H 6 OH + HC1 — V C 2 H B C1 4- H 2

HALOGEN DERIVATIVES OF THE PARAFFINS 1 15

(ii) By the addition of hydrogen chloride to ethylene (from cracked
petroleum) in the presence of aluminium chloride as catalyst :

C 2 H 4 +HCl-^Vc 2 H 6 Cl

Ethyl chloride is a gas, b.p. 12-5°. It is used in the preparation of tetra-
ethyl-lead, sulphonal, etc., and as a refrigerating agent.

DIHALOGEN DERIVATIVES
General methods of preparation. gem-Dihalides may be prepared :
(i) By the action of phosphorus pentahalides on aldehydes or ketones;

e.g., acetone gives wopropylidene chloride when treated with phosphorus

pentachloride :

CH 3 -COCH 3 + PC1 5 — ► CH 3 -CC1 2 -CH 3 + POCl 3

(ii) By the addition of halogen acids to the acetylenes, e.g., ethylidene
bromide from acetylene and hydrogen bromide :

HBr

CH^CH + HBr — > CH 2 =CHBr > CH 3 -CHBr 2

Tic-Dihalides may be prepared by the addition of halogen to olefins, e.g.,
propylene bromide from propylene and bromine:

CH 3 -CH=CH 2 + Br 2 — > CH 3 -CHBr-CH 2 Br

The method of preparing an a<»-dihalide is special to the particular halide,
e.g., trimethylene bromide may be prepared by the addition of hydrogen
bromide to allyl bromide at low temperatures :

CH 2 :CH-CH 2 Br + HBr— > CH 2 Br-CH 2 -CH 2 Br

Oldham (1950), however, has found that aco-dibromides may be prepared
by the action of bromine in carbon tetrachloride on the silver salt of a dibasic
acid (cf. p. 105) :

(CH 2 )„(C0 2 Ag) 2 + 2Br 2 — > Br-(CH 2 )„-Br + 2 AgBr + 2 C0 2 .

The yields are very good provided « is 5 or more.

General properties and reactions of the dihalides. The dihalides are sweet-
smelling, colourless liquids. gem-Dihalides are not as reactive as the
alkyl halides. It has been found that the polarity of the C — CI bond de-
creases progressively from methyl chloride, methylene chloride, chloroform
and carbon tetrachloride. Therefore the reactivity of chlorine decreases
progressively in these compounds in the same order (i.e., from methyl
chloride to carbon tetrachloride; cf. p. 105).

gww-Dihalides are hydrolysed by aqueous alkalis to the corresponding
carbonyl compound (aldehyde or ketone), e.g., isopropylidene chloride
gives acetone :

CH»-CC1 2 -CH 3 + 2KOH — > 2KC1 + [CH 3 -C(OH) 2 -CH 3 ] -^

CH 3 -COCH 3 + H 2

It has generally been found that a compound containing two (or more)
hydroxyl groups attached to the same carbon atom is unstable, and readily
eliminates a molecule of water (see p. 168).

gem-Dihalides give olefins when treated with zinc dust and methanol
(p. 64), and acetylenes when treated with ethanolic potassium hydroxide
(p. 90).

Il6 ORGANIC CHEMISTRY

vie- and ato-Dihalides are just as reactive as the alkyl halides. When
heated with zinc and methanol, w'c-dihalides give olefins (p. 64), but mu-
dihalides in which the two halogen atoms are in the 1 : 3 to the 1 : 6 positions
give cyclic compounds, e.g., trimethylene bromide gives cyc/opropane :

/CH 2 Br /CH 2

CH/ + Zn — > CH 2 ( I + ZnBr 2

\CH a Br \CH 2

wc-Di-iodides tend to eliminate iodine, particularly at raised temperatures,
to form olefins, e.g., propylene iodide gives propylene:

CH 3 -CHI-CH 2 I -^VcH 3 -CH==CH 2 + I 2

This property has been used to regenerate the double bond from w'c-dichlorides
or dibromides. These are heated with sodium iodide in ethanol, and the vic-di-
iodide formed decomposes into the olefin :

>CBr-CB< + 2NaI etha °°'> zNaBr + [>CI-CI<] — > >C=C< + I 2

Methylene chloride, CH 2 C1 2 (liquid, b.p. 40 °), was formerly prepared
industrially by partially reducing chloroform with zinc and hydrogen
chloride in ethanolic solution :

CHCL, + 2[H] — >■ CH 2 C1 2 + HC1

It is now also prepared industrially by the direct chlorination of methane
(see methyl chloride). It is used as an industrial solvent.

Methylene bromide, CH 2 Br 2 (b.p. 97 ), and methylene iodide, CH 2 I 2
(b.p. 181 ), are prepared by the partial reduction of bromoform and iodo-
form, respectively, with sodium arsenite in alkaline solution (the yield of
CH 2 Br 2 is 88-90 per cent. ; CH 2 I 2 , 90-97 per cent.), e.g.,

CHI3 + Na 3 As0 3 + NaOH — > CH 2 I 2 + Nal + Na 3 As0 4

All the methylene halides are used in organic syntheses.

Ethylene chloride, CH 2 C1-CH 2 C1 (b.p. 84 ), and ethylidene chloride,
CH 3 'CHC1 2 (b.p. 57°), are isomers; the former is prepared from ethylene
and chlorine, and the latter by the action of phosphorus pentachloride on
acetaldehyde (q.v.).

TRIHALOGEN DERIVATIVES

The most important trihalogen derivatives are those of methane, and
they are usually known by their trivial names: chloroform, CHC1 3 , bromo-
form, CHBr 3 and iodoform, CHI 3 .

Chloroform may be prepared in the laboratory or industrially by heating
ethanol or acetone with bleaching powder, or with chlorine and alkali (yield
about 40 per cent.). The reaction is extremely complicated, and the
mechanism is obscure (see chloral, p. 168).

The equations usually given for the action of bleaching powder on ethanol are :
(i) oxidation of ethanol to acetaldehyde; (ii) chlorination of acetaldehyde to
trichloroacetaldehyde; (iii) decomposition of trichloroacetaldehyde (chloral) by
free calcium hydroxide (present in the bleaching powder) into chloroform and
formic acid :

(i) CH s -CH 2 OH + Cl 2 > CH 3 -CHO + 2HCI

(ii) CH 8 -CHO + 3C1„ > CCl 3 -CHO + 3HCI

(iii) 2CCVCHO + Ca(OH) 8 — >-2CHCl 3 + (H-CO a ) 2 Ca

HALOGEN DERIVATIVES OF THE PARAFFINS 117

When acetone is used, the first product given is trichloroacetone, which is then
decomposed by the calcium hydroxide into chloroform and acetic acid :

(i) CH s -COCH 3 + 3C1 2 ^ CCl 3 'CO-CH 3 + 3HCI

(ii) 2CCVCOCH3 + Ca(OH) 2 > 2CHC1 3 + (CH 3 -C0 2 ) 2 Ca

Chloroform is also prepared industrially:

(i) By the chlorination of methane (see methyl chloride),
(ii) By the partial reduction of carbon tetrachloride with iron filings
and water :

CC1 4 +2[H] — ^CHC1 3 + HC1

When prepared this way chloroform is used for solvent purposes, and not
for anaesthesia (see later).

Pure chloroform may be prepared by distilling chloral hydrate with
aqueous sodium hydroxide:

CC1 3 -CH(0H) 2 + NaOH — > CHC1 3 + H-C0 2 Na + H 2

Chloroform is a sickly, sweet-smelling, colourless liquid, b.p. 6i°. It is
sparingly soluble in water but readily soluble in ethanol and ether. It
does not burn in air under usual conditions, but its vapour may be ignited,
when it burns with a green-edged flame. According to Hine (1950, 1954),
chloroform (and other haloforms) undergoes alkaline hydrolysis to produce
the formate ion and carbon monoxide by what Hine calls the alpha-elimina-
tion mechanism; this involves the removal of hydrogen and chloride ions
from the same carbon atom.

The mechanism proposed is as follows, involving the intermediate formation of
dichloromethylene :

chci 3 + oh- ^ :cci 3 - + H a O
:cci 3 - — > :cci a + cr

:cci, i^l> co + hco 2 -

H.O

Robinson (1961) has proposed, on kinetic evidence, that the intermediate di-
chloromethylene most likely decomposes to carbon monoxide as follows :
+ - -h+ - -a- -ci-

:cci 2 + h 2 o — >- h 2 o- CC1 2 — > ho— cci 2 — > ho— c— CI —^- CO

Formate is then formed as follows :

OH —

CO >■ H-CO'O-

slow

An interesting point about dichloromethylene is its reluctance to react with
hydroxide ion.

When chloroform is treated with zinc and hydrogen chloride in ethanolic
solution, methylene chloride (q.v.) is obtained; when treated with zinc and
water, methane is obtained:

CHCI3 + 6[H] -^V CH 4 + 3HCI
When chloroform is warmed with silver powder, acetylene is obtained:

2CHCI3 + 6Ag — > C 2 H 2 + 6AgCl

When treated with concentrated nitric acid, chloroform forms chloropicrin:

CHCI3 + HNO3 — > CC1 3 -N0 2 + H 2

Chloropicrin or nitrochloroform (liquid, b.p. 112 ) is used as an insecticide,
and has been used as a war-gas. Chloroform adds on to the carbonyl

Il8 ORGANIC CHEMISTRY

group of ketones in the presence of potassium hydroxide, e.g., with acetone
it forms chloretone (colourless needles, m.p. 97°), which is used as a drug:

KOH

(CH 3 ) 2 C=0 + CHCI3 y (CH 3 ) 2 C(0H)-CC1 3

Chloroform is employed in surgery as an anaesthetic, and for this purpose it
should be pure. In the presence of air and light, chloroform slowly forms
carbonyl chloride, which is extremely poisonous:

CHCI3 + |0 2 — > COCl 2 + HC1

Chlorine, water and carbon dioxide are also produced. Anaesthetic chloro-
form is therefore kept in well-stoppered dark-brown or blue bottles. Ethanol
is also added (1 per cent.), but its function is not quite clear. According to
some authors, it retards the decomposition of the chloroform. This is sup-
ported by the fact that infra-red measurements of such mixtures show the
absence of the carbonyl frequency.

A delicate test for chloroform is the " wocyanide test ". This is carried out
by heating chloroform with ethanolic potassium hydroxide and aniline,
Whereby phenyl wocyanide is formed, and is readily detected by its nauseat-
ing odour:

CHCI3 + 3KOH + C 6 H 5 -NH 2 — > C 6 H 5 -NC + 3KCI + 3 H 2

Chloroform is widely used in industry as a solvent for fats, waxes, resins,
rubber, etc.

Bromoform may be prepared by methods similar to those used for chloro-
form, but it is prepared industrially by the electrolysis of an aqueous solution
of acetone or ethanol containing sodium carbonate and potassium bromide
(acetone gives a better yield than ethanol). The solution is maintained
at about 20°, and hydrobromic acid is run in to neutralise the sodium
hydroxide produced during the electrolysis. Bromine is set free at the
anode, and probably reacts in the same way as does chlorine in the prepara-
tion of chloroform.

Bromoform is a liquid, b.p. 149-5°, and smells like chloroform, which it
closely resembles chemically.

Iodoform is prepared industrially by the electrolysis of an aqueous solu-
tion of ethanol or acetone containing sodium carbonate and potassium
iodide (ethanol gives a better yield than acetone). The solution is main-
tained at 60-70°, and a current of carbon dioxide is passed through the
solution to neutralise the sodium hydroxide formed.

Iodoform crystallises in yellow hexagonal plates, m.p. 119°. It is in-
soluble in water, but is readily soluble in ethanol and ether. It is used as
an antiseptic, but its antiseptic properties are due to the liberation of free
iodine, and not to iodoform itself. Iodoform chemically resembles chloro-
form and bromoform.

POLYHALOGEN DERIVATIVES

Carbon tetrachloride, CC1 4 , is prepared industrially in several ways :

(i) By the action of chlorine on carbon disulphide in the presence of
aluminium chloride as catalyst:

A1C1

CS 2 + 3C1 2 '-> CC1 4 + S 2 C1 2

The sulphur monochloride is removed by fractional distillation, and the
carbon tetrachloride is then shaken with sodium hydroxide, and finally
distilled.

(ii) By the chlorination of methane (see methyl chloride).

HALOGEN DERIVATIVES OF THE PARAFFINS II9

(iii) By chlorinolysis. This term was suggested by McBee, Hass and co-
workers (1941) to describe the process of chlorinating an organic compound
under conditions which rupture the carbon-carbon bond to yield chloro-
compounds with fewer carbon atoms than the original compound. Chlori-
nolysis may be effected with or without a catalyst, e.g., the hydrocarbon and
chlorine are heated at high temperature (300-400 °) and under high pressure
(about 1000 lb./sq. in.). The product is usually a mixture, e.g., propane
gives both carbon tetrachloride and hexachloroethane :

ci
CsH 8 — > CCI4 -f- C 2 C1 6

Carbon tetrachloride is a colourless liquid, b.p. 77°, which has a sickly
smell. It is insoluble in water but readily soluble in ethanol and ether.
Since its vapour is non-inflammable, carbon tetrachloride is widely used as
an industrial solvent (for fats, oils, resins, lacquers, etc.). It is also used
as a fire-extinguisher under the name of Pyrene.

Carbon tetrachloride is stable at red heat (about 500°), but when its vapour
comes into contact with water vapour at this temperature, some carbonyl
chloride is formed:

CC1 4 + H 2 > COCl 2 + 2HCI

Hence after using pyrene to extinguish a fire, the room should be well ventilated.

Carbon tetrachloride is reduced by moist iron filings to chloroform (q.v .).
The alkaline hydrolysis of carbon tetrachloride gives the same products
(formate and carbon monoxide) as chloroform, but the rate of reaction is
slower (Hine,i954).

Tetrachloroethane or acetylene tetrachloride, CHC1 2 -CHC1 2 , is prepared
by passing acetylene and chlorine into chambers filled with a mixture of
kieselguhr and iron filings. This method is used since the combination of
acetylene and chlorine is usually explosive unless a catalyst (and preferably
a diluent) is present.

Acetylene tetrachloride is a very toxic, colourless liquid, b.p. 146 . It
smells like chloroform; it is non-inflammable, and hence is widely used,
under the name of Westron, as a solvent for oils, fats, paints, varnishes,
rubber, etc.

When passed over heated barium chloride as catalyst, acetylene tetra-
chloride eliminates a molecule of hydrogen chloride to form trichloro-
ethylene :

CHC1 2 -CHC1 2 —V CHC1=CC1 2 + HC1

Trichloroethylene is a colourless liquid, b.p. 88-90 . It smells like chloro-
form, and is non-inflammable. It is more stable and less toxic than
acetylene tetrachloride, and hence is more widely used as an industrial
solvent, under the name of Westrosol, than Westron.
Hexachloroethane (perchloroethane) , C 2 C1 8 , may be prepared:

(i) By the chlorinolysis of propane (see carbon tetrachloride),
(ii) By passing ethylene mixed with 10 per cent, excess of chlorine through
a pyrex tube packed with activated charcoal at 300-350° :

C a H 4 + 5C1 2 — >- C 2 C1 6 + 4HCI

The excess chlorine prevents the formation of lower chlorinated products.

(iii) By passing a mixture of acetylene tetrachloride and chlorine over
aluminium chloride as catalyst :

A1CI

C 2 H 2 C1 4 + 2C1 2 '-> C 2 C1 6 + 2HCI

I JO

(iv) By treating trichloroethylene with chlorine and passing the penta-
caloroethane so produced over heated barium chloride, thus forming tetra-
Mloroethylene, which, in turn, gives hexachloroethane when treated with
cilorine:

ORGANIC CHEMISTRY

CHC1 2 -CC1 3 ^> CC1 2 =€CL

2 -%CCl 3 -CCl 3

CHC1=CC1 2 + Cl 2

Hexachloroethane is a solid, m.p. 187 . It smells like camphor and is
u|sed as a substitute for it.

s- Dichloroethylene, CHC1=CHC1, may be prepared by the action of finely
divided zinc on acetylene tetrachloride in the presence of water:

CHC1 2 -CHC1 2 + 2[H] Zn/Ha °> CHC1=CHC1 + 2HCI

II is a liquid and exists in two forms, the cis, b.p. 6o°, and trans, b.p. 48 (see
p. 427 for the meanings of cis and trans).

Dichloroethylene is used as a rubber solvent.

The halogen atom in the group =CHX is very unreactive (see p. 266). An
important property of the chlorinated unsaturated hydrocarbons is their ability
to add on chloroform or carbon tetrachloride in the presence of aluminium
chloride^ as catalyst, e.g., dichloroethylene forms 1:1:1:2:3: 3-hexachloro-
propane with carbon tetrachloride :

CHC1=CHC1 + CC1 4 ^±> CHC1 2 -CHC1-CC1 3

Paraffin wax has been chlorinated, and the products are used for dielectric
materials, protective coatings for fabrics, etc. Polychloro-derivatives of ethane,
propane and butadiene are used as dielectric materials, solvents (non-inflam-
mable), insecticides, plasticisers, etc.

FLUORINE DERIVATIVES OF THE PARAFFINS

Most organic compounds burn or explode when treated with fluorine
gas. Carbon heated in fluorine is attacked, sometimes explosively, with
the formation of mainly CF 4 , and small amounts of C 2 F 6 , C 2 F 4 , C 3 F 8 and
some other products.

Aliphatic fluorine compounds may be obtained in several ways :

(i) Direct fluorination of hydrocarbons may be carried out successfully
by diluting the fluorine with nitrogen, and carrying out the reaction in a
metal tube packed with copper gauze at a temperature of 150-350°. It is
very difficult to control the fluorination, and the product is usually a complex
mixture, e.g., methane gives CH 3 F, CH 2 F 2 , CHF 3 , CF 4 , C 2 F e and C 3 F 8 ;
ethane gives CF 4 , C 2 F 6 , CH 3 -CHF 2 , CH 2 F-CHF 2 ; no mono- or s-difluoro-
ethane is obtained.

When catalysts other than copper (actually CuF 2 ) are used, e.g., AgF,
CoF 2 , CeF 3 , MnF 2 , perfluoro-compounds are obtained, e.g., w-heptane gives
perfmoroheptane :

2AgF + F 2

It appears that when a catalyst is used, the perfluoro-compound obtained
usually has the same number of carbon atoms as the original compound;
if no catalyst is used, fluoro-compounds with fewer carbon atoms are usually
obtained.

The mechanism of direct fluorination is still obscure, but it appears that
the first step is the conversion of the catalyst into a higher fluoride, e.g.,
AgF into AgF 2 ; CoF 2 into CoF 3 , etc. Some of these higher fluorides' have
been isolated, e.g., AgF 2 .

HALOGEN DERIVATIVES OF THE PARAFFINS 121

(ii) Olefins and acetylenes add on hydrogen fluoride under pressure to
form fluoro-derivatives of the paraffins, e.g.,

C 2 H 4 + HF^C 2 H 5 F (f.g.-g.)
CH 3 -C^CH + 2HF — ► CH 3 -CF a -CH 3 (f.g.-g.)

If the unsaturated compound contains a halogen atom (other than fluorine),
this atom may be replaced by fluorine, e.g.,

R-CC1=CH 2 + HF — > R-CFC1-CH 3
R-CC1=CH 8 + 2HF — > R-CF 2 -CH 3 + HC1

Lead tetrafluoride (from lead dioxide and hydrogen fluoride) is particu-
larly useful for introducing two fluorine atoms into an olefin containing
chlorine, e.g.,

CC1 2 =CC1 2 + PbF 4 -^» CFC1 2 -CFC1 2 + PbF 2

It is also possible to add fluorine directly without a catalyst to highly
halogenated olefins, e.g.,

CFC1=CFC1 + F 2 — > CF 2 C1-CF 2 C1

(iii) By treating an alcohol with hydrogen fluoride, e.g.,

C 2 H 6 OH + HF — > C 2 H fi F + H 2

This reaction is very little used in practice. On the other hand, alkyl
fluorides may be prepared by heating alkyl toluene-/>-sulphonates with
potassium fluoride (Bergmann et al., 1958).

(iv) Fluorine compounds may be prepared indirectly by heating organic
halides with inorganic fluorides such as AsF 3 , SbF 3 , AgF, Hg 2 F 2 , etc., e.g.,

C a H 6 Cl + AgF — > C 2 H 6 F + AgCl (ex.)

This method was first used by Swarts (1898), and so is known as the Swarts
reaction.

When the organic halide contains two or three halogen atoms attached
to the same carbon atom, the best yield of fluoride is obtained when CoF 3
is used, but SbF 3 gives yields almost as good (and is more accessible), e.g.,

3CH 8 -CCl a -CH 3 + 2SbF 3 — > 3CH 3 -CF a -CH 3 + 2SbCl 3 (v.g.)

Alternatively, hydrogen fluoride may be used under pressure in the presence
of a catalyst, e.g.,

CC1 4 HF : 3 °°° > CF 2 Cl a + CFC1 3

4 (C + FeCl.) a z i <f

(v) A newer method of fluorination is the direct electrolytic method.
Nickel electrodes are used, and electrochemical fluorination takes place at the
anode, the reaction being carried out by the electrolysis of a solution of the
organic compound in anhydrous hydrogen fluoride, e.g.,

CH 3 -C0 2 H — ■> CF 3 -COF
C 2 H 6 -0-C 2 H 5 — -> C 2 H 6 -0-C 2 F 5

The particular merit of this method is that it usually leaves untouched
many types of functional groups.

Sulphur tetrafluoride is a very useful fluorinating agent, since it replaces
oxygen atoms by fluorine; thus:

R-CO a H-^VR-CF 3 ; R 2 CO — > R a CF 2 ; ROH— >RF

122 ORGANIC CHEMISTRY

The lower M-alkyl fluorides are gases. The first four members are stable,
and the higher members tend to decompose spontaneously into olefin and
hydrogen fluoride, e.g.,

CH 3 -CH 2 -CH 2 -CH 2 -CH 2 F — > CH 3 -CH 2 -CH 2 -CH=CH 2 + HF

Secondary and tertiary alkyl fluorides are so unstable that it is impossible
to prepare them free from olefin.

w'c-Difluorides are also usually unstable, e.g., ethylene fluoride decomposes
spontaneously at o° into hydrogen fluoride and butadiene :

2Ch 2 f-ch 2 f — > 4HF + ch 2 :ch-ch:ch 2

Alkyl fluorides are readily hydrolysed by strong acids to the corresponding
alcohols; alkalis have no effect. On the other hand, ethylene fluoride is
immediately hydrolysed by water to glycol:

CH 2 F-CH 2 F + 2H 2 — > CH 2 OH-CH 2 OH + 2HF

Fluorides with two or three fluorine atoms on the same carbon atom are
stable to water and strong acids, e.g., CHF 3 , CHF 2 *CHF 2 , etc.

Alkyl fluorides do not react with sodium, i.e., do not undergo the Wurtz
reaction, and do not form Grignard reagents. An interesting compound is
trifluoromethyl iodide, CF 3 I. It is converted into fluoroform, CHF 3 , by
potassium hydroxide, and it combines directly with many non-metals such
as P, As, Sb, S, Se, to give, e.g., with phosphorus, (CF 3 ) 3 P, (CF 3 ) 2 PI and
CF 3 -PI 2 .

Chlorofluoro-derivatives of methane and ethane are used as refrigerants
and for air-conditioning under the name of Freons, which are prepared by
the action of hydrogen fluoride on carbon tetrachloride, chloroform and
hexachloroethane.

Tetrafluoroethylene, C 2 F 4 (gas), is prepared by the action of antimony
trifluoride and hydrogen fluoride on chloroform, and then heating the
chlorodifluoromethane so produced at 8oo°:

CHC1 3 -^%- CHF a Cl -^> C 2 F 4 + 2HCI + other products

HF

When tetrafluoroethylene is polymerised, the plastic Teflon is produced.
Teflon is difficult to work, but is inert to chemical reagents, even to boiling
aqua regia.

Polychlorofluoroethylenes are valuable as oils and greases. Perfluoro-
heptane is useful in a process for the separation of uranium isotopes by
gaseous diffusion.

QUESTIONS

1. Write out the structures and names of all the dichloro-derivatives of butane and
tsobutane.

2. By means of equations show how you would convert ethanol into: (a) trichloro-
ethylene, (6) hexachloroethane, (c) s-dichloroethylene, (d) tetrachloroethylene, (e)
pentachloroethane.

3. Name the products and state the conditions under which thev are obtained when
ethyl iodide reacts with: (a) HI, (6) KCN, (c) KOH, (d) H 2 , (e) Mg, (/) Na, (g) NH S ,
(h) AgCN, (i) NaNO a , (j) AgN0 2 , (ft) NaHSO a , (I) C 6 H„.

4. Name the products and state the conditions under which they are obtained when
chloroform reacts with (a) nascent hydrogen, (6) KOH, (c) C 6 H 5 NH 2 , (d) O a , (e) Ag,
(/) HNO s , (g) CH 3 -COCH 3 , (A) CHC1=CHC1.

5. Define and give examples of: — (a) halogenation, (&) chain reaction, (c) Strecker
reaction, (d) Darzens procedure, (e) molecular rearrangement, (/) elimination reaction,
(g) alkylation, (h) Williamson synthesis, (i) Friedel-Crafts reaction, (j) Grignard reagent,
(k) hyperconjugation, (I) chlorinolysis, (m) Swarts reaction.

6. Discuss (i) the inductive effect of the alkyl group, (ii) S N and E mechanisms, (iii)
steric hindrance and steric acceleration, (iv) Saytzeff's rule.

HALOGEN DERIVATIVES OF THE PARAFFINS 123

READING REFERENCES
Hass, McBee et al., Chlorination of Paraffins, Ind. Eng. Chem., 1935, 27, 1190; 1936, 28,

333; 1941. 33. 137. 176, 181, 185; 1943, 35, 317.
Vaughan and Rust, Chlorination of Paraffins, /. Org. Chem., 1940, 5, 449.
Fredricks and Tedder, Free-radical Substitution in Aliphatic Compounds, J.C.S., 1960,

144.
Bethell and Gold, The Structure of Carbonium Ions, Quart. Reviews (Chem. Soc.), 1958,

12, 173.
Brown, Sulphuryl Chloride in Organic Chemistry, Ind. Eng. Chem., 1944, 36, 787.
Organic Reactions, Wiley, Vol. II (1944), Ch. 2. The Preparation of Aliphatic Fluorine

Compounds.
Bigelow, Action of Fluorine upon Organic Compounds, Chem. Reviews, 1947, 40, 51.
Ann. Reports (Chem. Soc), 1954, 51, 279. Perfluoroalkyl Compounds.
Cook (Ed.), Progress in Organic Chemistry, Butterworths. Vol. 2 (1953). Ch. 2.

Organic Fluorine Compounds.
Musgrave, The Reactions of Organic Fluorine Compounds, Quart. Reviews (Chem. Soc),

1954, 8. 33i-
Stacey et al. (Eds.), Advances in Fluorine Chemistry, Butterworths. Vol. 1 (i960).
Hughes, Reactions of Halides in Solution, Quart. Reviews (Chem. Soc), 1951, 5, 245.
Clark and Streight, Systematic Study of the Preparation of Alkyl Chlorides from the

Corresponding Alcohols, Trans. Roy. Soc, Can., 1929, [3], 23, 77.
Huntress, Organic Chlorine Compounds, Wiley (1949).
Johnson, The Degradation of Carboxylic Acids by means of Halogen : The Hunsdiecker

Reaction, Chem. Reviews, 1956, 56, 219.
Organic Reactions, Wiley. Vol. IX (1957), Ch. 5. The Reaction of Halogens with

Silver Salts of Carboxylic Acids.
Streitwieser, Solvolytic Displacement Reactions at Saturated Carbon Atoms, Chem.

Reviews, 1956, 56, 571.

CHAPTER VI

MONOHYDRIC ALCOHOLS

An alcohol is a compound that contains one or more hydroxyl groups, i.e.,
alcohols are hydroxy-derivatives of the paraffins. They are classified
according to the number of hydroxyl groups present. Monohydric alcohols
contain one hydroxyl group; dihydric, two; trihydric, three; etc. When
the alcohols contain four or more hydroxyl groups, they are usually called
polyhydric alcohols.

The monohydric alcohols form an homologous series with the general
formula CbH^+jjO, but, since their functional group is the hydroxyl group,
their general formula is more satisfactorily written as CJl^+^OR or ROH.

Nomenclature. The simpler alcohols are commonly known by their
trivial names, which are obtained by naming the alcohol as a derivative
of the alkyl radical attached to the hydroxyl group, e.g., CH 3 OH, methyl
alcohol; CH 3 -CH 2 -CH 2 OH. w-propyl alcohol; CH 3 -CH(OH)-CH 3 , iso-
propyl alcohol; (CH 3 ) 3 COH, te^.-butyl alcohol.

Another system of nomenclature considers the alcohols as derivatives
of methyl alcohol, which is named carbinol, e.g., CH 3 -CH 2 OH, methyl-
carbinol. The Chemical Society, however, now proposes to use methanol
instead of carbinol, e.g., CH 3 -CH 2 «CHOH-CH 3 , ethylmethylmethanol (both
methods have been used in this book).

In the I.U.P. A.C. system of nomenclature, the longest carbon chain contain-
ing the hydroxyl group is chosen as the parent hydrocarbon. The class
suffix is -ol, and the positions of side-chains and the hydroxyl group are indi-
cated by numbers, the lowest possible number being given to the hydroxyl
group (p. 63), e.g., CH 3 OH, methanol; C 2 H 5 OH, ethanol; CH 3 -CH 2 -CH 2 OH.
propan-i-ol; (CH 3 ) 2 CH-CHOH-CH 3 , 3-methylbutan-2-ol.

Monohydric alcohols are subdivided into primary, secondary and tertiary
alcohols according as the alkyl group attached to the hydroxyl group is a
primary, secondary or tertiary group, respectively. Primary alcohols
contain the primary alcoholic group — CHfiH, e.g., ethanol, CH 3 -CH 2 OH;
secondary alcohols, the secondary alcoholic group •CHftH)-, e.g., wopropanol,
(CH 3 ) 2 CHOH; and tertiary alcohols the tertiary alcoholic group —C(OH),
e.g., ter^.-butanol, (CH 3 ) 3 COH.

General methods of preparation. 1. By the hydrolysis of an alkyl halide
with aqueous alkali or silver oxide suspended in water:

RX + " AgOH '* — -> ROH + AgX

2. By the hydrolysis of esters with alkali:

R-C0 2 R' + KOH — > R-C0 2 K + R'OH

This method is important industrially for preparing certain alcohols that
occur naturally as esters.

As pointed out previously (p. 113), tertiary halides do not give a good
yield of alcohol on hydrolysis. A good yield of alcohol can, however, be
obtained by first heating the tertiary halide with silver acetate in ethanolic
solution, and then hydrolysing the ester so formed with alkali. Under these
conditions the tertiary alkyl radical shows little tendency to form olefin:

R 3 CX + CHa-COsAg — ► CH s -C0 2 CR 3 + AgX
CH S -C0 2 CR 3 + NaOH — > CH 3 -C0 2 Na + R 3 COH

124

MONOHYDRIC ALCOHOLS 125

3. By heating ethers with dilute sulphuric acid under pressure, e.g.,
diethyl ether forms ethanol:

(C a H s ) 2 + H 2 — V 2C 2 H 5 OH

This method is important industrially, since ethers are formed as by-
products in the preparation of certain alcohols (see ethanol and propanols).

4. By the reduction of aldehydes, ketones or esters by means of excess
sodium and ethanol as the reducing agent (Bouveault-Blanc reduction, 1903),
e.g.,

(i) Aldehydes: R-CHO + 2[H] — > R-CH 2 OH {g.-v.g.)
(ii) Esters: R-C0 2 R' + 4[H] — > R-CH 2 OH + R'OH {g.)
(iii) Ketones: R 2 CO + 2[H] — > R 2 CHOH {g.)

Hansley (1947) has improved the Bouveault-Blanc method by using
the theoretical quantity of sodium and ethanol (see below), and carrying
out the reaction in an inert solvent such as toluene or xylene. The yields
are usually 85-90 per cent.

The mechanism of the reaction is uncertain,
following :

Hansley has proposed the

O

R— C— OC 2 H 5 -

+
2Na

r ONa
R

.— C— OC 2 H 5

L Na

Na ester ketal

r OINa

CiH.OH

>C,H 5 ONa +

R-CHjjONa + C 2 H 5 ONa^

H,0

R-CH 2 OH + NaOH

C.H.OH

r ONa -\

•Na
L H

R 4-

2Na

R— C— lOQHsl

L H

Na derivative of
the hemiacetal

O 1

u

R— C + C 2 H 5 ONa
H

According to these equations, the theoretical amount of sodium is four atoms
per molecule of ester and two molecules of the reducing alcohol.

Darzens (1947) found that sodium hydride, NaH, gives a better yield of
alcohol than does sodium by the Bouveault-Blanc reduction of aldehydes,
ketones and esters. Nystrom and Brown (1947) have found that lithium
aluminium hydride, LiAlH 4 , also gives very good yields of alcohols from
aldehydes, ketones, esters, acids, acid chlorides and acid anhydrides, and one
advantage of this reagent is that it does not normally reduce the olefinic
bond, and hence an unsaturated aldehyde, ketone, etc., can be reduced to an
unsaturated alcohol. The most remarkable feature of this reagent is the
reduction of an acid to an alcohol, e.g., stearic acid is reduced to octadecan-
i-ol:

CHy(CHJ M -CO,H -^V CH 3 -(CH 2 ) 16 -CH 2 OH (91%)

Reductions with lithium aluminium hydride are usually carried out in
ethereal solutions, the compound in ether being added to the Uthium alu-
minium hydride solution. In certain cases the reverse addition is necessary,
i.e., the hydride solution is added to the solution of the compound to be
reduced.

126 ORGANIC CHEMISTRY

The reactions for the various compounds have been formulated as follows :

(i) 4RCOR + LiAlH 4 >■ (R 2 CHO) 4 LiAl -^L>

4 R 2 CHOH + LiOH + Al(OH) 3

(ii) 2R-C0 2 R' + LiAlH 4 >- (R-CH 2 0) 2 (0R') 2 L:A1 H '° >

2R-CH 2 OH + 2R'OH

(iii) 4 R-C0 2 H + 3LiAlH 4 > (R-CH 2 0) 4 LiAl + 2LiA10 2 + 4H 2

-^--> 4 R-CH 2 OH

(iv) 2R-C0C1 + LiAlH 4 > (R-CH 2 0) 2 LiAlCl 2 -^-> 2R-CH 2 OH

(v) (R-CO) 2 + LiAlH 4 — > (R-CH 2 0) 2 LiA10 -i^-> 2R-CH 2 OH

Aluminium trialkyls and dialuminium hydrides appear to be similar in
their reducing properties to lithium aluminium hydride (inter alia, Miller
et al., 1959).

Sodium borohydride, NaBH 4 , which is insoluble in ether but soluble in
water without decomposition, also reduces carbonyl compounds to alcohols,
but does not reduce" acids. On the other hand, lithium borohydride, which
is soluble in ether and is decomposed by water, behaves like lithium alu-
minium hydride but is not so vigorous and hence may be used to reduce a
more reactive group when the molecule contains two or more reducible
groups, e.g., when the compound contains both a carbonyl and a carboxyl
group, the former is reduced preferentially :

R-CO-CH 2 -CH 2 -C0 2 H -^V R-CHOH-CH 2 -CH 2 -C0 2 H

Aldehydes, ketones, esters, acid chlorides and acid anhydrides can be
reduced catalytically to alcohols in very good yields, e.g.,

acid chloride: R-COC1 + 2H 2 — — -> R-CH 2 OH + HC1

acid anhydride: (R-CO) 2 + 4H 2 cat ' > 2R«CH 2 OH + H 2

Catalytic reduction is particularly useful for preparing the higher alcohols
from esters. The ester is hydrogenated at 200° and at 150-200 atm. using
a copper catalyst, whereupon alcohols free from hydrocarbons are produced.
If a nickel catalyst is used and the temperature is above 250°, hydrocarbons
are the main product. Cadmium-nickel salts of acids may also be hydro-
genated under pressure in the presence of copper chromite to alcohols, the
lower members (C t to C 5 ) giving 70-95 per cent, yields (Adams et al., 1952).
On the other hand, carboxylic acids can be hydrogenated to primary
alcohols in the presence of a ruthenium or copper chromite catalyst (Guyer
et al., 1955). Aldehydes, ketones, carboxylic acids, acid chlorides, and esters
are readily reduced to alcohols by diborane (Brown et al., 1957).

5. Primary, secondary and tertiary alcohols may be prepared by means
of a Grignard reagent and the appropriate carbonyl compound (see p. 352).

6. A number of alcohols are obtained by fermentation processes (see
later).

7. In recent years synthetic methods have become very important for
preparing various alcohols :

(i) By the hydration of olefins, e.g., ethanol, wopropanol, etc.

(ii) By heating a mixture of carbon monoxide and hydrogen under pressure
in the presence of a catalyst, e.g., zinc chromite plus small amounts of alkali
metal or iron salts. A mixture of alcohols containing methyl, ethyl, n-

MONOHYDRIC ALCOHOLS I27

propyl, wobutyl and higher-branched alcohols is obtained, the individuals
being separated by fractional distillation (see below).

(hi) By the " oxo " process, which is the process whereby carbon monoxide
and hydrogen are added to olefins to yield aldehydes and alcohols (which
are separated by fractional distillation). Carbon monoxide, hydrogen and
the olefin are compressed to 200 atm. at 125-145°, and passed over a catalyst.
One catalyst consists of cobalt, thoria, magnesia and kieselguhr in the
proportions of 100 : 5 : 8 : 200 — this catalyst is also used in the Fischer-
Tropsch synthesis (p. 60); e.g., propylene, CO and H 2 give a mixture of
the two straight-chain butanols. The oxo-process is also known as the oxo-
synthesis and the carbonylation or hydroformylation reaction.

(iv) Methanol, ethanol, propanols and butanols are prepared industrially
by the oxidation of natural gas (p. 60).

Most of the methods given above can be used for the preparation of any
particular class of alcohol: it is only a question of starting with the appro-
priate compound. Primary alcohols may be prepared by the hydrolysis
of primary alkyl halides; by the reduction of aldehydes, esters, acids, acid
chlorides and acid anhydrides; and by means of a Grignard reagent and
formaldehyde:

^O /OMgX

H— C( + RMgX — ► H— C— R — '—> R-CH 2 OH
\H I

H

In addition to these, primary alcohols may be prepared, in variable yields,
by the action of nitrous acid on primary amines of the type R-CH 2 -NH 2 :

R-CH 2 -NH 2 + HN0 2 — > R-CH 2 OH + N 2 + H 2

Primary alcohols may also be prepared by the oxidation of primary
trialkylborons (p. 76).

Secondary alcohols may be prepared by the hydrolysis of secondary alkyl
halides, or better, via the ester (see above); by the reduction of ketones;
and by means of a Grignard reagent and any aldehyde other than form-
aldehyde:

S° /OMgX

R— C< + R'MgX — -> R— C( — --> R-CH(OH)-R'

H

Tertiary alcohols may be prepared by the indirect hydrolysis of tertiary
alkyl halides, viz., via the ester; and by means of a Grignard reagent and
a ketone (or an ester) :

/OMgX
R 2 C=0 + R'MgX — > R 2 C( — -* R 2 C(OH)-R'

General properties of the alcohols. The alcohols are neutral substances:
the lower members are liquids, and have a distinctive smell and a burning
taste; the higher members are solids and are almost odourless.

In a group of isomeric alcohols, the primary alcohol has the highest
boiling point and the tertiary the lowest, with the secondary having an
intermediate value. The lower members are far less volatile than is to be
expected from their molecular weight, and this is believed to be due to
association through hydrogen bonding extending over a chain of molecules,

128 ORGANIC CHEMISTRY

thus giving rise to a " large molecule " the volatility of which would be
expected to be low:

R R

i

'••ex '-ex

k k

The lower alcohols are very soluble in water, and the solubility diminishes
as the molecular weight increases. Their solubility in water is to be ex-
pected, since the oxygen atom of the hydroxyl group in alcohols can form
hydrogen bonds with the water molecules. In the lower alcohols the
hydroxyl group constitutes a large part of the molecule, whereas as the
molecular weight of the alcohol increases the hydrocarbon character of the
molecule increases, and hence the solubility in water decreases. This,
however, is not the complete story; the structure of the carbon chain also
plays a part, e.g., w-butanol is fairly soluble in water, but te^.-butanol is
miscible with water in all proportions.

General reactions of the alcohols, i. Alcohols react with organic and
inorganic acids to form esters :

R-C0 2 H + R'OH — >- R-C0 2 R' + H 2

Esters of the halogen acids are, as we have seen (p. ioo), the alkyl halides.

Alcohols, when heated with concentrated hydriodic acid and red phos-
phorus, are converted into paraffins.

The order of reactivity of an alcohol with a given organic acid is primary
alcohol>secondary>tertiary, but with a given halogen acid the order is
reversed. This implies that the mechanism of esterification (p. 187) of
organic acids is different from that of halogen acids. It has also been
observed that for a given alcohol the order of reactivity of the halogen
acids is HI>HBr>HCl. An explanation of all these observations is
offered later when the mechanism of esterification is discussed in detail.

2. Alcohols react with phosphorus halides to form alkyl halides (p. 103).
Treatment of alcohols with chlorine or bromine results in the formation of
halogen-substituted oxidised products (see also p. 168), e.g., chlorination
of ethanol gives trichloroacetaldehyde (cf. chloroform) :

CH 3 -CH 2 OH -^> CCl 3 -CHO

On the other hand, when chlorine is passed into an alcohol (primary or
secondary) in the presence of alkali, an organic hypochlorite is formed :

R 2 CHOH + Cl 2 Na ° H > R 2 CHOCl + HC1

These hypochlorites are unstable, and when heated or exposed to light, eliminate
hydrogen chloride:

R 2 CHOCl > R 2 CO + HC1

3. Alcohols combine with phenyl isocyanate to form phenyl-substituted
urethans :

C 6 H s -NCO + ROH — > C 6 H 5 -NH-C0 2 R

Urethans are well-defined crystalline solids, and so may be used to charac-
terise the alcohols.

MONOHYDRIC ALCOHOLS 129

Readily dehydrated alcohols, particularly tertiary alcohols, do not form
urethans but produce olefins and diphenylurea as follows :

R 2 C(OH)-CH 2 'CH 3 — >■ R 2 C=CH-CH 3 + H 2

C„H 5 -NCO + H a O > C 6 H 5 -NH 2 + C0 2

C 6 H B -NCO + C 6 H 5 -NH 2 > C,H 5 -NH-CO-NH-C 8 H 6

4. Alcohols are attacked by strongly electropositive metals, e.g., sodium
and potassium; hydrogen is liberated and the alkoxide is formed; e.g.,
ethanol reacts with sodium to form sodium ethoxide :

2C 2 H 5 OH + 2Na — > 2C 2 H 5 ONa + H 2

Sodium and potassium alkoxides are electrovalent compounds, i.e., their
formulae should be written, e.g., RO"Na + ; they are white deliquescent
solids, readily soluble in water with decomposition:

RONa + H 2 ^ ROH + NaOH
Alkoxides react with carbon disulphide to form xanthates:

RONa + C<

^S x SNa

The order of ease of formation of an alkoxide with sodium or potassium
is primary alcohol> secondary> tertiary. This may be explained as follows.
There are two possible ways in which the group COH may undergo fission:
C|— O — H and C — O — |H. Since oxygen has a higher electron-affinity than
either carbon or hydrogen, the shared electron pairs are displaced towards the

s+ s- s+
oxygen, i.e., we have C ->— O — <r- H. The greater the displacement of the
shared pair towards the oxygen atom in the C — O bond, the larger is the negative
charge on the oxygen atom, and consequently the weaker is the attraction of the
oxygen atom for the shared pair of the O — H bond. " Since alkyl groups are
electron-releasing, the larger the number of alkyl groups attached to the carbon
atom of the COH group, the greater will be the negative charge on the oxygen
atom. This may be represented as follows:

CH sX CH 3X

CH 3 ->-CH 2 ->-0— H; ^CH-»-0— H; CH 3 -M: > > > O— H

CH 3 X CH 3 X

Thus the displacement of the bonding pair of electrons in the C — O bond towards
the O atom is least in primary alcohols and greatest in tertiary, and consequently
the tendency of the COH group to break as C| — OH is greatest in tertiary
alcohols and least in primary; conversely, the tendency of the COH group to
break as C — O — |H is greatest in primary alcohols and least in tertiary. Hence
reactions involving the breaking of the C — O bond will take place most readily
with tertiary alcohols and least readily with primary; but those reactions
which involve the breaking of the O — H bond will take place most readily with
primary alcohols and least readily with tertiary. When sodium attacks an
alcohol, hydrogen is evolved, and since this involves the breaking of the O — H
bond, we can now understand why the order of ease of formation of alkoxides is
primary alcohol> secondary> tertiary.

A number of alkoxides are important as synthetic reagents; e.g., sodium
ethoxide, C 2 H 5 ONa; aluminium ethoxide, (C 2 H 5 0) 3 A1; aluminium tert.-
butoxide [(CH 3 ) 3 CO] 3 Al (see text for their uses). The aluminium alkoxides
may be conveniently prepared by the action of aluminium amalgam or aluminium
shavings on the alcohol.
F

130 ORGANIC CHEMISTRY

5. Primary and secondary alcohols may be acetylated with acetyl chloride,
e.g., ethanol gives ethyl acetate:

CH 3 -C0C1 + C 2 H 6 OH — ■>- CH 3 -C0 2 C 2 H 5 + HC1

With tertiary alcohols the reaction is often accompanied by dehydration
of the alcohol to olefin, or by the formation of a tertiary alkyl chloride;
e.g., fe^.-butanol gives a good yield of te^.-butyl chloride:

(CH 3 ) 3 COH + CH 3 -COCl — > (CH 3 ) 3 -CC1 + CH 3 -C0 2 H

6. Alcohols may be oxidised, and the products of oxidation depend on
the class of the alcohol (see below) .

7. Alcohols may be dehydrated to olefins by heat alone, but the tempera-
ture must be high (400-800 ). Dehydration, however, can be effected at
lower temperatures in the presence of catalysts, e.g., all three classes of
alcohols are dehydrated by passing over alumina at 350 ; primary alcohols
are dehydrated by concentrated sulphuric acid at about 170 , and secondary
and tertiary alcohols by boiling dilute sulphuric acid (this is used to avoid
polymerisation of the olefin). The mechanism when sulphuric acid is used
is described on p. 80. The mechanism with alumina is uncertain; a possi-
bility is :

R— OH A1 2 3 — > R— O— A1 2 3 — > R + + HO— A1 2 3

H

With secondary and tertiary alcohols, dehydration may occur in two ways,
e.g., _

_ H -^CHg'CHa'CH — CH 2
CH 3 -CH 2 -CH(OH)-CH 3

(1 : lH s SO,)

^CHo-CH=CH-CH,

Experiment shows that hydrogen attached to the adjacent carbon atom
joined to the least number of hydrogen atoms is eliminated most easily.
Thus, in the above reaction, the main product is but-2-ene (65-80 per cent.).
This elimination thus occurs in accordance with Saytzeff's rule for the
dehydrohalogenation of alkyl halides, and the reason is the same (see p. 112).
When alcohols containing no hydrogen atoms on the carbon atom
adjacent to the COH group are dehydrated, dehydration and molecular
rearrangement occur together, e.g., weopentyl alcohol gives 2-methylbut-2-
ene:

(CH 3 ) 3 OCH 2 OH ~"'°> (CH 3 ) 2 C=CH-CH 3

This is an example of the 1,2-shift, and the mechanism may be formulated
as follows (see also p. 101) :

CH 3 OH CH 3 >OH 2 CH 3

1 1 H+ 1^1 -H,0 I V+

CH 3 — C 6— H -> CH 3 — C C— H -> CH 3 — C C— H >-

J. I i I J- I

ch, h ch„ h ch, h

3H,

CH :
CH,

A:

I _h+ C H 3 \ yC.

C— H ■> )c=cC

K.\ CH 3 / \H

H a H

MONOHYDRIC ALCOHOLS 131

When a double bond is produced in the product and is accompanied by a
1,2-shift, the reaction is said to be a retropinacol rearrangement. This type
of rearrangement, when occurring in open-chain compounds, is also some-
times called the Wagner rearrangement.

Alcohols may also be converted into olefins via their methyl xanthates
(Tschugaev reaction, 1899):

cs, /^ ch.i
R 2 CH-CH 2 OH ?-> R 2 CH-CH 2 OCf >

NaOH \g Na

R 2 CH-CH 2 OC^ — ^> R 2 C=CH 2 + CH 3 SH + COS

\S-CH 3

A very important feature of this reaction is that no rearrangement occurs in the
formation of the olefin from alcohols which undergo rearrangement when
dehydrated by the usual dehydrating agents.

By means of dehydration, it is possible to convert a primary alcohol into
a secondary or tertiary, according to the structure of the primary alcohol,
e.g.,

(i) CH 3 -CH 2 -CH 2 OH -^5> CH 3 -CH=CH 2 -— >

35° (i actOH "

CH 3 -CHI-CH 3 — > CH 3 -CH(OH)-CH 3

(ii) (CH 3 ) 2 CH-CH 2 OH A ''°"> (CH 3 ) 2 C=CH 2 "" >

35° «• AffOH "

(CH 3 ) 2 CI-CH 3 '- > (CH 3 ) 2 C(OH)-CH 3

In the same way, a secondary alcohol of suitable structure can be converted
into a tertiary alcohol, e.g.,

(CH 3 ) 2 CH-CH(OH)-CH 3 -^> (CH 3 ) 2 C=CH-CH 3 — ^

350"

'AgOH

(CH 3 ) 2 CI-CH 2 -CH 3 "* > (CH 3 ) 2 C(OH)-CH 2 -CH 3

It is also possible to step down the alcohol series by means of dehydration,
e.g.,

R-CH„-CH,OH -^%- R-CH=CH„ °' -

350° 2

R'CH O CHo 7n/ rr ,-, [T

Zn/H^ R>CHQ _H 1 _ > R . CH QH

i — A

8. Alcohols also combine with acetylene in the presence of mercury com-
pounds as catalyst to form acetals:

2ROH + CH=CH -^» CH 3 -CH(OR) 2

If, however, the reaction is carried out in the presence of potassium alkox-
ides at high temperature and under pressure, vinyl ethers are obtained (p.

94)-

9. Alcohols can be made to undergo self-condensation in the presence of

sodium alkoxide at elevated temperatures:

2RCH 2 -CH 2 OH — > RCH 2 -CH 2 -CHR-CH 2 OH

This is known as the Guerbet reaction (1899), and can be used for primary
and secondary alcohols and also for mixed condensations.

!32 ORGANIC CHEMISTRY

Methods of Distinguishing between the Three Classes of Alcohols, i. By

means of oxidation. The nature of the oxidation products of an alcohol
depends on whether the alcohol is primary, secondary or tertiary.

(i) A primary alcohol on oxidation first gives an aldehyde, and this, on
further oxidation, gives an acid. Both the aldehyde and acid contain the
same number of carbon atoms as the original alcohol, e.g. :

CH 3 -CH 2 OH —-> CHyCHO J^> CH 3 -C0 2 H

(ii) A secondary alcohol, on oxidation, first gives a ketone with the same
number of carbon atoms as the original alcohol. Ketones are fairly difficult
to oxidise, but prolonged action of the oxidising agents produces a mixture
of acids, each containing fewer carbon atoms than the original alcohol, e.g.,
methyl-M-propylmethanol gives first pentan-2-one, and then a mixture of
acetic and propionic acids:

CH 3 -CHOH-CH 2 -CH 2 -CH 3 -™> CH 3 -CO-CH 2 -CH 2 -CH 3

-^> CH 3 -C0 2 H + CH 3 -CH 2 -C0 2 H

(hi) Tertiary alcohols are resistant to oxidation in neutral or alkaline
solution, but .are readily oxidised by acid oxidising agents to a mixture of
ketone and acid, each containing fewer carbon atoms than the original alcohol.

(CH 3 ) 2 C(OH)-CH 2 -CH 3 -^> (CH 3 ) 2 CO + CH 3 -CO a H

The oxidising agents usually used for oxidising alcohols are: acid di-
chromate, acid or alkaline potassium permanganate, and dilute nitric acid.

The mechanism of oxidation of alcohols is still uncertain. There is,
however, evidence to show, at least in the case of primary and secondary
alcohols, that the first step is dehydrogenation at the carbon of the C — OH
group. Westheimer et al. (1949) found that there is a kinetic isotope effect
(p. 36) when 2-deuterowopropanol, CH 3 'CDOH-CH 3 , is oxidised with
chromic acid; the rate of oxidation of this compound was found to be about
one-sixth the rate of oxidation of wopropanol. Thus the rate-determining
step is the fission of the C— H bond in the group H-C-OH. This is supported
by the fact that these authors found no kinetic isotope effect in the oxida-
tions of (CH 3 ) 2 CHOH and (CD 3 ) 2 CHOH. In tertiary alcohols, which are
very resistant to oxidation, there is no H-C-OH group. Consequently
it is logical to conclude that tertiary alcohols are oxidised by a different
mechanism. Since tertiary alcohols are very readily dehydrated, their
oxidation may therefore proceed via the oxidation of an intermediate
olefin.

2. The three classes of alcohols differ in their behaviour when the vapour
is passed over copper at 300 ° :

(i) A primary alcohol is dehydrogenated to an aldehyde, e.g.,

CH 3 -CH 2 OH — %► CH 3 -CHO + H 2

300

(ii) A secondary alcohol is dehydrogenated to a ketone, e.g.,

CH 3 -CH(OH)-CH 3 — ^ CH 3 -CO-CH 3 + H 2
(iii) A tertiary alcohol is dehydrated to an olefin, e.g.,
(CH 3 ) 2 C(OH)-CH 2 -CH 3 — ^> (CH 3 ) 2 C=CHCH 3 + H 2

MONOHYDRIC ALCOHOLS 133

3. The alcohol is converted by phosphorus tri-iodide into its corresponding
iodide, which is then heated with silver nitrite, and the resulting nitroparaffin is
treated with nitrous acid and alkali. Characteristic colours are obtained accord-
ing as the alkyl group is primary, secondary or tertiary (see p. 305).

Methyl alcohol, methanol (carbinol), CH 3 OH, is prepared industrially by
several methods. The earliest method was by the destructive distillation of
wood, whereby tar and an aqueous fraction known as pyroligneous acid are
obtained. Pyroligneous acid contains methanol, acetone and acetic acid,
and all three compounds may be obtained by suitable treatment (see acetic
acid, p. 181). It was this method which gave rise to the name " wood
spirit " for methanol. The modern methods are synthetic.

(i) Water gas mixed with half its volume of hydrogen — synthesis gas — is
passed at a pressure. of 200 atmospheres over a catalyst containing the oxides
of copper, zinc and chromium at 300° :

CO + 2H 2 — > CH3OH

If the proper precautions are taken, the yield of methanol is almost 100
per cent., and its purity is above 99 per cent. By changing the catalyst
and the ratio of carbon monoxide to hydrogen, methanol and a variety of
higher alcohols are produced (p. 126). Another commercial method uses
carbon dioxide instead of the monoxide; again a catalyst is required:

C0 2 + 3 H 2 — > CH3OH + H 2

(ii) By the catalytic oxidation of methane. A mixture of methane and
oxygen (ratio by volume of 9 : 1) at a pressure of 100 atmospheres is passed
through a copper tube at 200 °:

CH 4 + £O a ^CH 3 OH

Methanol is a colourless, inflammable liquid, b.p. 64 , and .is poisonous.
It is miscible with water in all proportions, and is also miscible with most
organic solvents. It burns with a faintly luminous flame, and its vapour
forms explosive mixtures with air or oxygen when ignited. It combines
with calcium chloride to form CaCl 2 *4CH 3 OH, and hence cannot be dried
this way (cf. ethanol).

Methanol is used as a solvent for paints, varnishes, shellac, celluloid
cements, etc.; in the manufacture of dyes, perfumes, formaldehyde, etc.
It is also used for making methylated spirit and automobile antifreeze
mixtures.

Structure of methanol. Analysis and molecular-weight determinations
show that the molecular formula of methanol is CH 4 0. Assuming that
carbon is quadrivalent, oxygen bivalent and hydrogen univalent, only one
structure is possible :

J

H— C— O— H or CH 3 OH

A

This is supported by all the chemical reactions of methanol, e.g., (i) only one
hydrogen atom in methanol is replaceable by sodium; this suggests that
one hydrogen atom is in a different state of combination from the other three,
(ii) Methanol is formed from methyl chloride by hydrolysis with sodium
hydroxide. Methyl chloride can have only the structure CH 8 C1. It is
reasonable to suppose that the methyl group in methyl chloride is unchanged

134 ORGANIC CHEMISTRY

by the action of dilute alkali, and that the reaction takes place by the
replacement of the chlorine atom by a hydroxyl group.

(iii) The presence of the hydroxyl group is confirmed, for example, by
the reaction between methanol and phosphorus pentachloride, when methyl
chloride, hydrogen chloride and phosphoryl chloride are formed. Thus one
oxygen atom (bivalent) and one hydrogen atom (univalent) have been
replaced by one chlorine atom (univalent). This implies that the oxygen
and hydrogen atoms exist as a univalent radical in methanol: the only
possibility is as a hydroxyl group, OH. It is the hydrogen of the hydroxyl
group which is displaced by sodium.

All these reactions indicate that the structure of methanol is CH 3 OH.

Ethyl alcohol, ethanol (methylcarbinol), C 2 H s OH, is prepared industrially
by several methods: (i) Ethylene (from cracked petroleum) is absorbed in
concentrated sulphuric acid (98 per cent.) at 75-80 , under pressure (250-
500 lb./per sq. in.). Ethyl hydrogen sulphate and ethyl sulphate are
formed :

C 2 H 4 + (HO) 2 S0 2 — > C 2 H 5 OS0 2 -OH

C 2 H 5 OS0 2 -OH + C 2 H 4 — >■ (C 2 H 5 0) 2 S0 2

The reaction mixture is then diluted with about an equal volume of water,
and warmed. Hydrolysis takes place and ethanol together with some diethyl
ether is formed:

C 2 H 6 OS0 2 -OH + H 2 — > C a H 5 OH + H 2 S0 4
(C 2 H 5 0) 2 S0 2 + 2H 2 — > 2C 2 H 5 OH + H 2 S0 4
(C 2 H 6 0) 2 S0 2 + C 2 H B OH — > (C 2 H 5 ) 2 + C 2 H 5 OS0 2 -OH

The ether is kept to a minimum by separating the ethyl sulphate from the
reaction products, and hydrolysing it separately.

The hydrolysed liquids are distilled, and the aqueous ethanol distillate
is concentrated by fractional distillation (see also below).

Ethanol is also manufactured by the direct hydration of ethylene with
steam under pressure in the presence of a suitable catalyst :

C 2 H 4 + H 2 —+ C 2 H 5 OH

(ii) Acetaldehyde (from acetylene) is catalytically reduced to ethanol
by passing its vapour, mixed with hydrogen, over finely divided nickel at
100-140 °:

CH 3 -CHO + H 2 — > CH s -CH 2 OH

(iii) The earliest method of preparing ethanol is by fermentation, and.
this is still used for the manufacture of beer, wine, brandy, etc., and also as
a source of ethanol. The starting material is starch, which is obtained from
sources depending on the particular country: common sources of starch
are wheat, barley, potato, etc. Recently, molasses (p. 460) has also been
used as the starting material for ethanol. The grain, e.g., wheat or barley,
is mashed with hot water, and then heated with malt (germinated barley)
at 50° for 1 hour. Malt contains the enzyme diastase which, by hydrolysis,
converts starch into the sugar, maltose (q.v.) :

(C 6 H 10 O 5 ) n + ^ H 2 ^V I C 12 H 32 O u

If molasses is used, then this step is unnecessary, since it contains carbo-
hydrates already present as sugars which can be fermented.

The liquid is cooled to 30 and fermented with yeast for 1-3 days. Yeast

MONOHYDKIC ALCOHOLS 135

contains various enzymes, among which are maltase, which converts the
maltose into glucose, and zymase, which converts the glucose into ethanol :

_ __ „ maltase _ ,, _

C 12 H 22 O u + H 2 ^2C 6 H 12 6

C 6 H 12 6 -=► 2C 2 H 6 OH + 2C0 2

The carbon dioxide is recovered and sold as a by-product. The fermented
liquor or " wort ", which contains 6-10 per cent, ethanol and some other
compounds, is fractionated into three fractions :

(i) First runnings, which consists mainly of acetaldehyde.
(ii) Rectified spirit, which is 93-95% w/w ethanol.
(iii) Final runnings or fusel oil, which contains »-propyl, »-butyl,
isobutyl, w-amyl, *soamyl and " active " amyl alcohol.

Industrial alcohol is ordinary rectified spirit. Methylated spirit is of two
kinds : (a) Mineralised methylated spirit is 90 per cent, rectified spirit, 9 per
cent, methanol and 1 per cent, petroleum oil, and a purple dye. (b) In-
dustrial methylated spirit is 95 per cent, rectified spirit and 5 per cent,
methanol, whose purpose is to " denature " the rectified spirit, i.e., make it
unfit for drinking purposes.

Absolute alcohol is 99-5 per cent, ethanol, and is obtained from rectified
spirit. When an aqueous solution of ethanol is fractionated, it forms a
constant-boiling mixture containing 96 per cent, ethanol from which 100
per cent, ethanol may be obtained by adding a small amount of benzene,
and then distilling. The first fraction is the ternary azeotrope, i.e., a
constant-boiling mixture containing three constituents, b.p. 64-8° (water,
7-4 per cent.; ethanol, 18-5 per cent.; benzene, 74-1 per cent.). After all
the water has been removed, the second fraction that distils over is the
binary azeotrope, b.p. 68-2° (ethanol, 32-4 per cent.; benzene, 67-6 per
cent.). After all the benzene has been removed, pure ethanol, b.p. 78-1°,
distils over.

Ethanol cannot be dried by means of calcium chloride, since a compound
(an alcoholate) is formed, e.g., CaCl 2 *3C 2 H 5 OH (c/. methanol). Distillation
of rectified spirit over calcium oxide, and then over calcium, gives absolute
alcohol. This method is often used in the laboratory, and was formerly
used industrially.

Ethanol is a colourless, inflammable liquid, b.p. 78-i°. It is miscible
with water in all proportions, and is also miscible with most organic solvents.
Ethanol and methanol resemble each other very closely, but they may be
distinguished (i) by the fact that ethanol gives the haloform reaction (q.v.),
whereas methanol does not; and (ii) ethanol gives acetic acid on oxidation;
methanol gives formic acid. These two acids are readily distinguished from
each other (p. 181).

Ethanol is used for the preparation of esters, ether, chloral, chloroform,
etc. It is also used as a solvent for gums, resins, paints, varnishes, etc.,
and as a fuel.

Structure of ethanol. Analysis and molecular-weight determinations
show that the molecular formula of ethanol is C 2 H 6 0. Assuming that
carbon is quadrivalent, oxygen bivalent, and hydrogen univalent, two
structures are possible:

CH 3 — CH 2 — OH CH— O— CH 3

(I) (H)

(i) Only one hydrogen atom in ethanol is replaceable by sodium or
potassium. This indicates that one hydrogen atom is in a different state of

136 ORGANIC CHEMISTRY

combination from the other five. In (I), one hydrogen atom differs from the
other five, but in (II) all the hydrogen atoms are equivalent.

(ii) When ethanol is treated with hydrochloric acid or phosphorus penta-
chloride, one oxygen atom (bivalent) and one hydrogen atom (univalent)
are replaced by one chlorine atom (univalent) to give ethyl chloride, C 2 H 6 C1.
This implies the presence of a hydroxyl group (cf. methanol).

(iii) When ethyl chloride is hydrolysed with dilute alkali, ethanol is ob-
tained. This reaction also indicates the presence of a hydroxyl group in
ethanol.

(iv) Ethanol may be prepared as follows :

C 2 H 6 — ?L> C 2 H s C1 -^> C 2 H 5 OH

The arrangement of the six hydrogen atoms in ethane is known, and it is
reasonable to suppose that five retain their original arrangement in ethyl
chloride and ethanol, since these five hydrogen atoms do not enter (pre-
sumably) into the above reactions. Thus there is an ethyl radical C 2 H 5 —
in ethanol. This is so in (I), but not in (II).

(v) Structure (II) is definitely eliminated, since it can be shown that it is
the structure of dimethyl ether (q.v.), a compound that has very little
resemblance, physically or chemically, to ethanol.

Thus (I) is accepted as the structure of ethanol, and it accounts for all the
known properties of ethanol.

It can be seen from the various examples given on structure determination,
e.g.. ethane, propane, ethylene, methanol and ethanol, that the method of
approach follows certain definite fines. First the molecular formula is
obtained. Then, if the compound is simple — in the sense that it contains
a small number of unlike atoms, and that the total number of atoms is also
small — the valencies of the atoms present are assumed, and various possible
structures are written down. If the compound is " simple " the number
of possible structures will not be large (four or five at the most). Then by
considering the chemical properties of the compound in question, the struc-
ture which best fits the observed facts is accepted as the correct one. If
the compound is not " simple ", the procedure is to detect the presence of
as many functional groups as possible; to degrade the compound into
simpler substances whose structures are already known or which may be
determined by further degradation; to build up structural formulae based
on the facts obtained; and then to choose that structure which best fits
the facts. Finally a synthesis is attempted, and if successful, will usually
give proof of the correctness of the structure suggested.

It can be seen from the above arguments that it is necessary to have
methods, preferably simple ones, for detecting the presence of functional
groups. The reader will become familiar with these methods as he reads
the text. At this stage we shall confine our attentions to detecting the
presence of a hydroxyl group. The usual tests are: (i) Treatment with
sodium; if hydrogen is evolved, a hydroxyl group is present, (ii) Treat-
ment with phosphorus pentachloride or acetyl chloride; the evolution of
hydrochloric acid fumes indicates the presence of a hydroxyl group.

Acetyl chloride is usually the most satisfactory (remember, however,
alcohols of the type R 3 COH, p. 130).

Propyl alcohols, C 3 H 7 OH. Two isomeric propyl alcohols are possible,
and both are known.

n-Propyl alcohol, propan-i-ol, n-propanol, CH 3 'CH 2 -CH 2 OH, was originally
obtained from fusel oil (see above), but it is now also produced by the

MONOHYDRIC ALCOHOLS 137

hydrogenation of carbon monoxide. A more recent method is by the cata-
lytic reduction of propargyl alcohol (from acetylene and formaldehyde) :

Ni

CH=OCH 2 OH + 2H 2 ► 2 CH 3 -CH 2 -CH 2 OH

w-Propanol is a colourless liquid, b.p. 97-4°, and is miscible with water,
ethanol and ether. It is used in the preparation of propionic acid, toilet
preparations such as lotions, etc.

isoPropyl alcohol, propan-2-ol, isopropanol, CH 3 , CH(OH)*CH 3 , is prepared
industrially:

(i) By the catalytic hydrogenation of acetone under pressure:

CH 3 -COCH 3 + H 2 -^CH 3 -CH(OH)-CH 3

(ii) By passing propylene (from cracked petroleum) into concentrated
sulphuric acid, then diluting with water, and distilling off the iso-
propanol. j'soPropyl ether is obtained as a by-product (cf. ethanol) :

CH 3 -CH=CH 2 + (HO) 2 S0 2 > (CH 3 ) 2 CHO-S0 2 -OH

(CH 3 ) 2 CHOS0 2 -OH + CH 3 -CH=CH 2 >- [(CH 3 ) 2 CHO] 2 S0 2

(CH 3 ) 2 CHOS0 2 -OH + H 2 > (CH 3 ) 2 CHOH + H 2 S0 4

[(CH 3 ) 2 CHO] 2 S0 2 + zH 2 > 2(CH 3 ) 2 CHOH + H 2 S0 4

[(CH 3 ) 2 CHO] 2 S0 2 + (CH s ) 2 CHOH > [(CH 3 ) 2 CH] 2 + (CH 3 )„CHOS0 2 -OH

tsoPropanol can also be prepared by direct hydration, and this can be
effected by passing a mixture of propylene and steam at 220-250 , and under
pressure (220 atm.), over a catalyst of tungsten oxide plus zinc oxide, on a
silica carrier.

woPropanol is a colourless liquid, b.p. 82-4°, and is soluble in water,
ethanol and ether. It is used for preparing esters, acetone, keten, as a
solvent, and for high-octane fuel.

Butyl alcohols, C 4 H 9 OH. Four isomers are possible, and all are known.
n-Butyl alcohol, butan-i-ol, n-butanol, CH 3 *CH 2 'CH a 'CH 2 OH, b.p. 117-4°,
is prepared industrially:

(i) By the Weizmann process (1911). Starch or molasses is fermented
with the micro-organism, Clostridium acetobutylicum, whereupon acetone
and n-butanol are obtained.

(ii) Synthetically from acetaldehyde :

2 CH 3 -CHO -^> CH 3 -CH(OH)-CH 2 -CHO -^>

aldol -H '°

CH 3 -CH=CH-CHO -^> CH 3 -CH 2 -CH 2 -CH 2 OH
crotonaldehyde

(iii) From propylene by the Oxo process (p. 127).

M-Butanol is widely used as a solvent.

iso Butyl alcohol, 2-methylpropan-z-ol, isobutanol, (CH 3 ) 2 CH'CH 2 OH, b.p.
108°, is obtained as a by-product in the preparation of methanol from
synthesis gas (p. 133). It behaves as a primary alcohol, but it readily
rearranges due to the presence of a branched chain near the COH group,
e.g., when it is treated with hydrochloric acid, wobutyl chloride and tert.-
butyl chloride are obtained (cf. p. 101) :

"HCI

(CH 3 ) 2 CH-CH 2 OH ^ (CH 3 ) 2 CH-CH 2 C1 + (CH 3 ) 3 CC1

sec-Butyl alcohol, butan-2-ol, sec.-butanol, CH 3 *CH 2 *CH(OH)*CH 3 , b.p.
100°, is prepared industrially by the hydration of 1- or 2-butene (from

138 ORGANIC CHEMISTRY

cracked petroleum) by means of concentrated sulphuric acid (cf. ethanol
and isopropanol).

sec.-Butanol is used for the preparation of butanone, esters, and as a
lacquer solvent.

tert.-Butyl alcohol, tert.-butanol {trimethylmethanol) , (CH 3 ) 3 COH, m.p.
25*5°, b.p. 83 , is prepared synthetically by the hydration of wobutene
(from cracked petroleum). It is mainly used as an alkylating agent.

Amyl alcohols, C 5 H u OH. Eight isomers are possible, and all are known:

1. CH 3 'CH 2 , CH 2 'CH 2 "CH 2 OH, w-amyl alcohol, pentan-i-ol, w-pentanol, b.p. 138°.

2. (CH 3 ) 2 CH>CH 2 'CH 2 OH, isoamyl alcohol, isopentanol, b.p. 130 .

3. CH 3 -CH 2 -CH(CH 3 )-CH 2 OH, " active " amyl alcohol, 2-methylbutan-i-ol,

b.p. 128 .

4. (CH 3 ) 3 OCH 2 OH, wopentyl alcohol, tert-butylcarbinol, b.p. 113°.

5. CH 3 'CH 2 , CH 2 *CH(OH)-CH 3 , pentan-2-ol, methyl-w-propylcarbinol, b.p. 120°.

6. CH 3 -CH 2 'CH(OH)-CH 2 -CH 3 , pentan-3-ol, diethylcarbinol, b.p. 117°.

7. (CH 3 ) 2 CH>CH(OH)*CH 3 , 3-methylbutan-2-ol, methylisopropylcarbinol, b.p.

114°.

8. (CH 3 ) 2 C(OH) , CH 2 l CH 3 , 2-methylbutan-2-ol, tert.-axs\y\ alcohol, tefl-pentanol,

ethyldimethylcarbinol, b.p. 102 .

Three amyl alcohols, viz., M-pentanol, wopentanol and " active " amyl
alcohol, have been isolated from fusel oil (q.v.). The last two are the chief
constituents of fusel oil, and all three are produced by the fermentation of
protein matter associated with the carbohydrates in starch. This mixture
of amyl alcohols (from fusel oil) is used for the preparation of esters (for
artificial essences), scents, and as a laboratory reducing agent with sodium
it is better than ethanol owing to its higher boiling point (this mixture of
amyl alcohols will be referred to in future as j'sopentanol).

A mixture of amyl alcohols, known as pentasol, is prepared industrially
by chlorinating at 200 °, and in the dark, a mixture of n- and isopentanes
(from petroleum) to the amyl chlorides which are hydrolysed with dilute
sodium hydroxide solution plus a little sodium oleate for emulsification,
to the amyl alcohols. Seven isomeric amyl chlorides are theoretically
possible by the chlorination of n- and wopentanes, but in practice six are
obtained — no 2-chloro-3-methylbutane is produced. Pentasol (the mixture
of six amyl alcohols) finds great use as a solvent in the lacquer industry.

Commercial " sec.-amyl alcohol " (80 per cent. 2- and 20 per cent. 3-
pentanol) is made by hydrating 1- and 2-pentenes (from cracked petroleum).

A number of higher alcohols occur as esters in waxes, e.g., cetyl (palmityl)
alcohol (hexadecan-i-ol), C 16 H 33 OH, m.p. 49 , occurs as the palmitate in
spermaceti (obtained from the oil of the sperm whale); carnaubyl alcohol
(tetracosan-i-ol), C 24 H 4e OH, m.p. 69 , as esters in wool-grease; ceryl alcohol
(hexacosan-l-ol), C 26 H 53 OH, m.p. 79 , as the cerotate in Chinese wax; myricyl
(melissyl) alcohol (triacontan-i-ol), C 30 H 61 OH, m.p. 88°, as esters in bees-wax.

A number of higher alcohols are now prepared industrially by the catalytic
reduction of the ethyl (or glyceryl) esters of the higher fatty acids, par-
ticularly the alcohols lauryl (dodecan-i-ol), C 12 H 25 OH, m.p. 24 ; myristyl
(tetradecan-i-ol), C U H 29 0H, m.p. 38°; palmityl (hexadecan-i-ol), C 16 H 33 OH,
m.p. 49°; and stearyl (octadecan-i-ol), C 18 H 37 OH, m.p. 59°. These alcohols
are used in the form of sodium alkyl sulphates, ROSCyONa, as detergents,
emulsifying agents, wetting agents, insecticides, fungicides, etc. These
sodium salts lather well, and are not affected by hard water, and hence can
be used as a soap substitute.

MONOHYDRIC ALCOHOLS 139

QUESTIONS

i. Write out the structures and names (three methods) of the isomeric amyl alcohols.
Indicate the class of each alcohol.

2. By means of equations show how you would convert: {a) w-butanol, (6) n-pentanol,

(c) hexan-2-ol, into butan-2-ol.

3. Give as many methods as you can for distinguishing between the three classes of
alcohols.

4. Define and give examples of: — (a) primary alcohol, (6) secondary alcohol, (c)
tertiary alcohol, (d) the Bouveault-Blanc reduction, (e) association, (/) dehydrogenation,
(g) fermentation, (h) the oxo process, (i) dehydration, (j) hydration, (k) Tschugaev
reaction, (/) Wagner rearrangement.

5. Name the products and indicate the conditions under which they are formed when
ethanol is treated with:— (a) AcOH, (b) HI, (c) PC1 3 , (d) PBr 6 , (e) PI 3 , (/) Na, (g) AcCl,
(h) H 2 S0 4 , (i) C 6 H 6 -NCO, (;) Br 2 .

6. Give an account of the evidence for the structure of (a) EtOH, (6) PrOH.

7. Starting from acetylene, how would you prepare: — (a) EtOH, (6) C a H„, (c) CH 4 ,

(d) EtNO a , (e) Et a O?

8. Starting with the following compounds, name the alcohol formed and indicate the
necessary conditions: — (a) acetaldehyde, (6) butan-2-one, (c) ethyl propionate, (d)
palmitic acid, (e) acetyl chloride, (/) propionic anhydride, {g) ^-propylamine, (h) EtMgl
and HCHO, (i) MeMgl and CH„-CHO, (;') MeMgl and Me 2 CO.

9. Write brief notes on the industrial preparation of: — (a) MeOH, (6) EtOH, (c)
PrOH, (d) the four butyl alcohols, (e) isoPrOH, (/) pentasol, (g) lauryl alcohol.

10. Suggest why the tertiary alcohols may be dehydrated or converted into alkyl
chlorides by AcCl.

READING REFERENCES

Killeffer, Butanol and Acetone from Corn, Ind. Eng. Chem., 1927, 19, 46.

Wynkoop, «-Butanol and Acetone, ibid., 1943, 35, 1240.

Lee, Fermentation, ibid., 1950, 42, 1672.

Gabriel, Butanol Fermentation Process, ibid., 1928, 20, 1063.

Backhaus, Ethyl Alcohol, ibid., 1930, 22, 1151.

Amyl Alcohols from Pentanes:

(i) Ayres, ibid., 1929, 21, 899.

(ii) Clark, ibid., 1930, 22, 439.
Brooks, The Manufacture of Alcohols and Esters, ibid., 1935, 27, 282.
Park and Donlan, Alcohols other than Butyl, ibid., 1943, 35, 1031.
Grave*, Higher Alcohols formed from Carbon Monoxide and Hydrogen, ibid., 1931,

23. T 38i.

Hansley, Sodium Reduction of Fatty Acid Esters, ibid., 1947, 39, 55.

Organic Reactions, Wiley. Vol. VI (1951), Ch. 10. Reductions by Lithium Alu-
minium Hydride.

Byrkit and Soule, Sodium Methylate and its Uses, Chem. Eng. News, 1944, 22, 1003.

Gaylord, Reduction with Complex Metal Hydrides, /. Chem. Educ, 1957, 34, 367.

Brown, New Selective Reducing Agents, /. Chem. Educ, 1961, 38, 173.

Becker, Base-catalysed Alkylations with Alcohols, /. Chem. Educ, 1959, 36, 119.

CHAPTER VII

ETHERS

The general formula of the ethers is CnH^+aO (which is the same as that
for the monohydric alcohols), and since their general structure is R — O — R',
they may be regarded as alkyl oxides or the anhydrides of the alcohols
(see below).

When the two alkyl groups in an ether are the same, the ether is said to
be symmetrical or simple, e.g., diethyl ether, C 2 H B — O— C 2 H 6 , is a simple
ether. When the two alkyl groups are different, the ether is said to be
unsymmetrical or mixed, e.g., ethyl methyl ether, CH 3 — O— C 2 H 5 , is a
mixed ether.

Nomenclature, i. In this system of nomenclature all the members are
known as ethers, and the individuals are named according to the alkyl
groups attached to the oxygen atom, e.g., CH 3 — O— CH 3 , dimethyl ether;
C 2 H 6 — — CH(CH 3 ) 2 , ethyl isopropyl ether.

2. According to the I.U.P.A.C. system of nomenclature, the ethers are
regarded as hydrocarbons in which a hydrogen atom is replaced by an alkoxyl
group, -OR, the larger radical being chosen as the alkane. For symmetrical
ethers, method i is to be used, e.g., C 2 H 5 -OC 2 H 6 , diethyl ether; CH 3 -0-C 2 H s ,
methoxyethane.

General methods of preparation, i. By heating excess of alcohol with
concentrated sulphuric acid or glacial phosphoric acid. One molecule of
water is removed from two molecules of alcohol — hence the reason for
regarding ethers as the anhydrides of alcohols:

2 ROH-^VR 2

-H,0 z

According to Van Alphen (1930), ether is formed from ethanol as follows
(c/. ethylene) :

C 2 H 8 OH + H 2 S0 4 — > [C a H 5 — OH^HSO^ — > C^ + H 2 + HSO~ 4

The ethyl carbonium ion may eliminate a proton to form ethylene (which
is obtained as a by-product), or react with a molecule of ethanol (which is
in excess) to form ether:

C 2 H-0 C 2 H 5 t-^C 2 H— O— C 2 H 6 + H +

H

On the other hand, Brooks (1935) believes that ethyl sulphate is formed
first, and this then reacts with the excess of alcohol to form ether (cf. ethanol) :

2C 2 H 6 OH + H 2 S0 4 — > (C 2 H 5 0) 2 S0 2 + 2 H 2
(C 2 H 6 0) 2 S0 2 + C 2 H 6 OH— >- (C 2 H B ) 2 + C 2 H 5 0-S0 2 -OH

The yields of ether are good from primary alcohols and fairly good from
secondary. Mixed ethers may also be prepared by this method, e.g.,
addition of tert-biityl alcohol to a boiling mixture of ethanol and aqueous
sulphuric acid gives tert.-bntyl ethyl ether in excellent yield.

140

ETHERS 141

2. By Williamson's synthesis, in which the sodium of potassium alkoxide
is heated with an alkyl hahde:

RONa + R'X — > R-O-R' + NaX (g.)

The mechanism is probably:

RO- R'— X — >■ RO— R'— X — > R-O-R' + X"

This method is particularly useful for preparing mixed ethers, and it is
best to use the alkoxide of the secondary or tertiary alcohol, and primary
alkyfhalide because secondary and tertiary alkyl halides readily decom-
pose into olefins.

Williamson's synthesis proves the structure of the ethers.

When R' is a methyl or ethyl radical, methyl or ethyl sulphate, respectively,
can be used instead of the corresponding alkyl halide, e.g.,

C 2 H 5 ONa + (CH 3 ) 2 S0 4 ^ — > CH 3 -0-C 2 H 5 + CH 3 NaS0 4

3. By heating alkyl halides with dry silver oxide:

2RX + Ag a O — > R 2 + 2AgX (v.p.)

4. By passing the alcohol vapour over a catalyst such as alumina, thoria,
etc., at 250 and under pressure:

2ROH — > R 2 + H a O

5. By passing an olefin into concentrated sulphuric acid, etc. (see ethanol
and isopropanol).

6. From halogeno-ethers and Grignard reagents (p. 354).

7. Ethers may also be obtained by refluxing alcohols with esters of toluene-
^>-sulphonic acid in the presence of sodium (Drahowzal et al., 1951).

ROH + CH 3 /~^SO s R' — ^> R-O-R' + CH 3 ^_^SO s H

General properties of the ethers. The lower members are gases or volatile
liquids, and their vapours are highly inflammable. Their boiling points
are much lower than those of the alcohols containing the same number of
carbon atoms, and this is probably due to the fact that ethers cannot
associate through hydrogen bonding. All the ethers are less dense than
water in which they are not very soluble, but their solubility is very much
increased in the presence of small amounts of alcohol.

The solubility of ethers is not as high as might have been expected in view of
the fact that the oxygen in ethers can form hydrogen bonds with water. This
low solubility may possibly be due to steric effects of the alkyl groups in the ether.

General reactions. 1. Ethers form stable salts (oxonium compounds)
with strong inorganic acids, e.g., [(C 2 H 5 ) 2 OH] + Cl,l(C 2 H 5 ) 2 OH]+HS0 4 -.
Because of this, ethers can be separated from paraffins and alkyl halides,
e.g., by shaking with concentrated sulphuric acid, ether is removed from a
mixture of ether and ethyl bromide.

2. When heated with dilute sulphuric acid under pressure, ethers form
the corresponding alcohols:

R 2 + H 2 -^% 2ROH

142 ORGANIC CHEMISTRY

3. When warmed with concentrated sulphuric acid, ethers form alkyl
hydrogen sulphates:

R 2 + H 2 S0 4 — > RHS0 4 + ROH
ROH + H 2 S0 4 — > RHS0 4 + H 2

4. Ethers are readily attacked by concentrated hydriodic or hydrobromic
acid, the final products depending on the temperature of the reaction.

(i) in the cold:

R 2 + HI — > RI + ROH

In the case of a mixed ether the iodine atom attaches itself to the
smaller alkyl radical :

CH 3 -0-C 2 H 5 + HI — > CH3I + C 2 H 5 OH

If both alkyl radicals contain the same number of carbon atoms, the
iodine atom attaches itself to the less complex group. Propyl wopropyl
ether, however, is an exception :

CH 3 -CH 2 -CH 2 -0-CH(CH 3 ) 2 + HI — > (CH 3 ) a CHI + CH ? -CH 2 -CH 2 OH

(ii) When heated:

R 2 + 2HI — > 2RI + H 2

Reactions (i) and (ii) are very useful for identifying the groups present
in an ether. Furthermore, since (i) occurs very easily, it is possible to
" protect " a hydroxyl group in a polyfunctional compound by converting
the hydroxyl group into an ether, which is later treated with concentrated
hydriodic acid to regenerate the hydroxyl group (see, e.g., p. 236).

Decomposition of ethers by concentrated hydriodic acid is the basis of
the Zeisel method for estimating methoxyl and ethoxyl groups (p. 314).

Although heating ethers with hydriodic acid results mainly in the formation
of alkyl iodides, some reduction products are also formed (p. 49). This reduction
may be avoided by heating the ether with potassium iodide and phosphoric acid
(Stone et al., 1950) :

R-O-R + 2KI + 2H 3 P0 4 — > 2RI + 2KH 2 P0 4 + H 2

Birch (1951) has shown that ethers are converted into paraffin and alcohol when
treated with sodium in liquid ammonia.

, > RH + R'OH

R-O-R' + 2[H] —

-> R'H + ROH

Which of these two directions occurs is decided by which transition state has the
lower energy.

5. When treated with chlorine or bromine, ethers undergo substitution,
the extent of which depends on the conditions. Hydrogen joined to the
carbon directly attached to the oxygen atom is most readily replaced, e.g.,
diethyl ether reacts with chlorine in the dark to form 1 : 1'-dichlorodiethyl
ether :

CH 3 -CH 2 -0-CH 2 -CH 3 -%■ CH 3 -CHCl-0-CH 2 -CH 3 — -V CH 3 -CHCl-0-CHCl-CH 3

In the presence of light, perchlorodiethyl ether is obtained:

(C 2 H 5 ) 2 O^V(C 2 Cl 5 ) 2
Halogeno-ethers may also be prepared in other ways (see, e.g., p. 354).

ETHERS 143

6. Acid chlorides react with ethers when heated in the presence of an-
hydrous zinc chloride, aluminium chloride, etc. ; e.g.,

R a O + CH 3 -COCl-^% RC1 + CH 3 -C0 2 R
Acid anhydrides also split ethers to form esters:

R 2 + (CH 3 -CO) 2 ZnC1 '> 2CH 3 -C0 2 R

7. Ethers react with carbon monoxide at 125-180 and at a pressure of
500 atmospheres, in the presence of boron trifmoride plus a little water :

R 2 + CO — >■ R-C0 2 R

Dimethyl ether (methyl ether) is prepared industrially by passing methanol
vapour at 350-400°, and at a pressure of 15 atmospheres, over aluminium
phosphate as catalyst (p. 141). It is a gas, b.p. — 23-6°, and is used as a
refrigerating agent.

Diethyl ether (ether, " sulphuric ether ") is prepared in the laboratory and
industrially by the " continuous etherification process", i.e., heating excess
ethanol with concentrated sulphuric acid. It is also obtained industrially:
(i) as a by-product in the preparation of ethanol from ethylene and sulphuric
acid (p. 134) ; (ii) by passing ethanol vapour, under pressure, over heated
alumina or aluminium phosphate (cf. dimethyl ether).

Diethyl ether is a colourless liquid, b.p. 34-5°. It is fairly soluble in
water, and is miscible with ethanol in all proportions. It is highly inflam-
mable and forms explosive mixtures with air; this is a great disadvantage
in its use as an industrial solvent for oils, fats, gums, resins, etc., and as an
extracting solvent. It is also used in surgery as an anaesthetic, and is the
usual solvent for carrying out Grignard reactions. In the presence of air and
light, ether forms ether peroxide (C 2 H 5 ) 2 O0 2 , whose structure, according to
Rieche and Meister (1936), is CH 3 -CH-OCH 2 *CH 3 . Ether peroxide is a

OOH
heavy, pungent, oily liquid, and is explosive. Since its boiling point is
higher than that of ether, it is left in the residue after ether distillations, and
may cause explosions. Addition of a small amount of a cuprous compound,
e.g., cuprous oxide, has been recommended for avoiding the formation of
ether peroxide. The chief impurity in ether is ethanol, and this has the
property of preventing the formation of ether peroxide.

An important derivative of ether is 2 : 2'-dichlorodiethyl ether, which is
prepared by heating ethylene chlorohydrin with concentrated sulphuric
acid at ioo°:

2C1CH 2 -CH 2 0H -^% ClCH 2 -CH 2 -OCH 2 -CH 2 Cl

2 i -H s O *

It may also be prepared by passing a mixture of ethylene and chlorine into
ethylene chlorohydrin at 8o° :

CH 2 0H-CH 2 C1 + C 2 H 4 + Cl 2 — > (CH 2 C1-CH 2 ) 2 + HC1

2 : 2'-Dichlorodiethyl ether is used as a solvent, as a soil fumigant and also
as the starting point of many chemicals.

2 : 2'-Dichlorodiethyl ether is also named as bis-(2-chloroethyl) ether.
The prefix bis (Latin, twice) indicates that there are two identical groups
attached to a given atom; it is generally used for complex groups.

Di-isopropyl'ether is obtained industrially as a by-product in the preparation of
isopropanol from propylene and sulphuric acid (p. 137). It is also prepared by

144 ORGANIC CHEMISTRY

passing propylene into 75 per cent, sulphuric acid at 75-125° under a pressure of
3-7 atmospheres; very little isopropanol is formed under these conditions.

Di-j'sopropyl ether is a colourless liquid, b.p. 69°. It is used as an industrial
solvent for extraction operations, and for decreasing the knocking properties of
petrol which, mixed with di-isopropyl ether, acquires a high octane number.

Di-isoamyl ether, [(CH 3 ) 2 CH'CH a 'CH a -] a O, is prepared by the action of con-
centrated sulphuric acid on isopentanol. It is a colourless liquid, b.p. 172°, and
has a pear-like odour. It is used as an industrial solvent, and as a solvent in
Grignard reactions in which higher temperatures are required than can be
obtained by using ether.

Mixed ethers of the following types have been prepared: primary-primary;
primary-secondary; primary-tertiary; secondary-secondary; secondary-
tertiary; tertiary-tertiary.

Isomerism within the ether series, and any other series in which the
isomers differ by the nature of the groups attached to a given atom, e.g.,
the amines (q.v.), is sometimes known as metamerism: thus there are three
metamers of formula C 4 H 10 O, viz., diethyl ether, C 2 H 5 'OC 2 H 5 ; methyl
propyl ether, CH 3 -0-CH 2 -CH 2 -CH 3 ; methyl t'sopropyl ether, CH 3 -OCH(CH 3 ) 2 .

The term metamerism was introduced by Berzelius, but he gave it a wider
meaning than that which is accepted today (see p. 399).

QUESTIONS

1. Write out the structures and names of the isomeric ethers having the molecular
formula 0511,20.

2. Name the compounds and state under what conditions they are formed when
ether is treated with:— (a) H a S0 4 , (b) HBr, (c) Br 2 , (d) PC1 6 , (e) Na, (/) AcCl, (g) CO,
(h) NaOH, (») 2 .

3. Define and give examples of: — (a) Williamson's synthesis, (6) the Zeisel deter-
mination, (c) metamerism.

4. A compound has the molecular formula C 4 H 10 O. How would you show whether
it was an alcohol or an ether? If it is an ether, how would you determine its structure?

READING REFERENCES
Van Alphen, The Formation of Ether from Alcohol, Rec. trav. chim., 1930, 49, 754.
Brooks, The Manufacture of Alcohols and Esters, Ind. Eng. Chem., 1935, 27, 284.
Norris and Rigby, Preparation and Properties of Mixed Aliphatic Ethers, /. A mer.

Chem. Soc, 1932, 54, 2088.
Evans and Edlund, Tertiary Alkyl Ethers, Ind. Eng. Chem., 1936, 28, 1186.

CHAPTER VIII

ALDEHYDES AND KETONES

Aldehydes and ketones both have the general formula CnH^O.
Aldehydes are the first oxidation products of primary alcohols, and their

/£>
functional group is the aldehyde group -CHO or -C , which can only occur

\H
at the end of a chain, since the carbon atom of the group has only one available
valency. Ketones are the first oxidation products of secondary alcohols,

C \
and their functional group is the ketonic group ^0=0, which cannot occur

C/

at the end of a chain, since in ketones the CO group has two available

valencies, each of which is joined to a carbon' atom (cf., however, ketens,

p. 287).

The C — O group is known as the carbonyl group, but when it occurs in
ketones, it is referred to as the ketonic group.

Nomenclature. Aldehydes. The lower members are commonly named
after the acids that they form on oxidation. The suffix of the names of
acids is -ic (the names of the trivial system are used, see p. 174) ; this suffix
is deleted and replaced by aldehyde, e.g.,

H-CHO — ^> H-C0 2 H
formaldehyde formic acid

(CH 3 ) 2 CH-CHO [Q] > (CH 3 ) 2 CH-C0 2 H

isohntyraldehyde jsobutyric acid

The positions of side-chains or substituents are indicated by Greek letters,
the a carbon atom being the one adjacent to the aldehyde group, e.g.,

CH 3 -CH(OH)-CH 3 -CHO {5-hydroxybutyraldehyde

According to the I.U.P.A.C. system of nomenclature, aldehydes are desig-
nated by the suffix -al, which is added to the name of the hydrocarbon from
which they are derived. The longest carbon chain containing the aldehyde
group is chosen as the parent hydrocarbon ; the positions of side-chains or
substituents are indicated by numbers, and the aldehyde group is given the
number 1, which may be omitted from the name if the aldehyde group is
the only functional group present in the compound (see p. 63), e.g.,

CH s -CHO ethanal

CH 3 -CH 2 -CH-CH(CH 3 ) -CH 2 -CH 3 2-ethyl-3-methylpentanal

CHO

Ketones. The lower members are commonly named according to the
alkyl groups attached to the ketonic group, e.g.,

CH 3 -CO-CH 3 dimethyl ketone

CH 3 'CH 2 'CO , CH(CH 3 ) 2 ethyl wopropyl ketone
145

I46 ORGANIC CHEMISTRY

The positions of side-chains or substituents are indicated by Greek letters,
the a carbon atom being the one adjacent to the ketonic group, e.g.,

CH 3 -CHC1-C0-CH 2 -CH 2 C1 a : p'-dichlorodiethyl ketone

If the two alkyl groups in a ketone are the same, the ketone is said to be
simple or symmetrical; if unlike, mixed or unsymmetrical (c/. ethers).

According to the I.U.P.A.C. system of nomenclature, ketones are desig-
nated by the suffix -one, which is added to the name of the hydrocarbons
from which they are derived. The longest carbon chain containing the
ketonic group is chosen as the parent hydrocarbon ; the positions of side-
chains or substituents are indicated by numbers, and the ketonic group is
given the lowest number possible, e.g.,

CH 3 -CO-CH 2 -CH 2 -CH 3 pentan-2-one

(CH 3 ) 2 CH-CO-CH(CH 3 )-CH 2 -CH 3 2 : 4-dimethylhexan-3-one

Since aldehydes and ketones both contain the carbonyl group, it might be
expected that they would resemble one another. It is therefore instructive
to compare their general methods of preparation and their general properties.

General methods of preparation of aldehydes and ketones

1. Aldehydes. By the oxidation of a. primary alcohol:

R-CH 2 OH — ^-> R-CHO

Oxidation may be effected by acid dichromate (yield: f.g.-g) ; or by passing
the alcohol vapour mixed with air over heated silver (250°) as catalyst
(yield: g.). Alternatively, the alcohol may be dehydrogenated by passing
the alcohol vapour over heated copper (300°) as catalyst :

R-CH 2 OH —^-> R-CHO + H 2 [ex.)
300°

fert.-Butyl chromate (prepared by adding chromium trioxide to tert. -hutanol)
oxidises primary alcohols to aldehydes almost quantitatively (Oppenauer
et al., 1949).

Ketones. By the oxidation of a secondary alcohol (using the same oxidising
agents as for aldehydes), or by dehydrogenation over heated copper:

R-CH(OH)-R'—^-> R-CO-R' (g.-ex.)
R-CH(OH)-R' — ^> R-CO-R' + H 2 {ex.)

There is, however, a specific reagent for oxidising secondary alcohols
to ketones, viz., aluminium ter t.-butoxide, [(CH 3 ) 3 C0] 3 A1 (the Oppenauer
oxidation. 1937). The secondary alcohol is refluxed with the reagent in
excess acetone solution ; it is the acetone which is reduced :

R-CH(OH)-R' + CH 3 -CO-CH 3 — >■ R-CO-R' + CH 3 -CH(OH)-CH 3 (g.-v.g.)

This reagent is particularly useful for oxidising unsaturated secondary
alcohols because it does not affect the double bond. On the other hand,
primary alcohols (particularly unsaturated alcohols) may also be oxidised
to aldehydes if acetone is replaced by ^-benzoquinone. In general, quinones
and aromatic ketones are better hydrogen acceptors than acetone.

ALDEHYDES AND KETONES 147

The mechanism of the Oppenauer oxidation is probably the reverse of that
of the Meerwein-Ponndorf-Verley reduction (p. 150).

AT-Bromosuccinimide (p. 375) oxidises primary and secondary alcohols to
aldehydes and ketones, respectively (Barakat et al., 1952), e.g.,

CH 2 -CO\ CH 2 -CO\

CH 3 -CH,OH + I >NBr — > CH 3 -CHO + I /NH + HBr

Primary and secondary alcohols, both saturated and unsaturated, may be
oxidised to the corresponding carbonyl compound by means of manganese
dioxide in acetone solution (inter alia, Bharucha, 1956). Ruthenium tetroxide
also oxidises saturated primary and secondary alcohols to the corresponding
carbonyl compound (Berkowitz et al., 1958).

2. Aldehydes. By heating a mixture of the calcium salts of formic acid
and any one of its homologues :

(R-C0 2 ) 2 Ca + (H-C0 2 ) 2 Ca — >- 2R-CH0 + 2CaC0 3

The yields are poor on account of many side reactions.

The mechanism of this decomposition has been studied by Bell et al. (1952),
who prepared acetaldehyde by heating a mixture of barium acetate containing
13 C and barium formate, and obtained acetaldehyde containing no 13 C.

x O-CO-| CH 3 H-CO| -(\

Ba \ * + ^Ba — > 2CH 3 -CHO + 2BaC0 3

\Q-CO- |CH 3 H-COJ -O/

Ketones. By heating the calcium salt of any fatty acid other than formic
acid (yield : variable due to many side reactions) :

(R-C0 2 ) 2 Ca — > R 2 CO + CaC0 3

If a mixture of calcium salts is used, mixed ketones are obtained:

(R-C0 2 ) 2 Ca + (R'-C0 2 ) 2 Ca — >■ 2R-COR' + 2CaC0 3

Apparently only methyl ketones have actually been prepared this way, i.e.,
calcium acetate is one of the salts used. This method of pyrolysis of calcium
salts as a general method seems to be of no value.

3. Aldehydes. By passing a mixture of the vapours of formic acid and
any one of its homologues over manganous oxide as catalyst at 300° :

R-C0 2 H + H-C0 2 H M °°> R-CHO + C0 2 + H 2 (f.g.-g.)

R 2 CO and R-CHO are obtained as by-products, and the reaction probably
proceeds via the manganous salt, the manganous carbonate which is formed
decomposing at 300 .

MnC0 3 — > MnO + C0 2

This mechanism postulating the decomposition of the intermediate manganous
salt is supported by experiments using isotopically enriched acid (Reed, 1955).

Ketones. By passing the vapour of any fatty acid other than formic acid
over manganous oxide at 300 :

2 R-C0 2 H ~^% R 2 CO + C0 2 + H 2 (g.)
A mixture of fatty acids gives mixed ketones:

R-C0 2 H + R'-C0 2 H — > R-COR' + C0 2 + H 2
R 2 CO and R' 2 CO are obtained as by-products.

I48 ORGANIC CHEMISTRY

4. Aldehydes. By the reduction of an acid chloride with hydrogen in
boiling xylene using a palladium catalyst suspended on barium sulphate
(Rosenmund's reduction, 1918) :

R-COC1 + H 2 — > R-CHO + HC1 {g.-v.g.)

Aldehydes are more readily reduced than are acid chlorides, and therefore
one would expect to obtain the alcohol as the final product. It is the barium
sulphate that prevents the aldehyde from being reduced, acting as a poison
(to the palladium catalyst) in this reaction. Generally, when the Rosen-
mund reduction is carried out, a small amount of quinoline and sulphur is
added; these are very effective in poisoning the catalyst in the aldehyde
reduction. Sakurai et al. (1944) have improved the Rosenmund reduction
by using BaS0 4 — Pd(OH) a as catalyst, and anhydrous acetone or ethyl
acetate as solvent (yield: 80-90 per cent.). On the other hand, Davies et al.
(1943) have shown that aliphatic acid chlorides may be reduced to aldehydes
in reasonable yield by means of sodium amalgam.

Brandt (1949) has shown that lithium hydride reduces acid chlorides,
anhydrides and thiolesters to the corresponding aldehyde, the yield being best
with thiolesters (70 per cent.).

When <-butanol is added to lithium aluminium hydride in ether, lithium tri-J-
butoxyaluminium hydride is precipitated. This, in diethylene glycol dimethyl
ether solution at — 78 , readily reduces acid chlorides to the corresponding
aldehydes (Brown et al., 1958). The yields are comparable with those obtained
by the Rosenmund reduction. Kuivila (i960) has also shown that tri-w-butyltin
reduces acid chlorides to aldehydes in good yield.

Ketones. There is no analogous method for the preparation of ketones.

5. Aldehydes. By the ozonolysis of olefins of the type R-CH!CH-R' :

_ R-CH— O— CH-R' „
R-CHICH-R'— ^-> I I f-> R-CHO + R'-CHO (g.-v.g.)

A A catalyst

Ketones. By the ozonolysis of olefins of the type R 2 CCR' 2 :

R 2 C:CR' 2 -^> 2 | I 2 — ^RsCO + R'sCO (g.-v.g.)

* * A A catalyst

6. Aldehydes. By the oxidation of 1 : 2-glycols of the type
R-CH(OH)-CH(OH)-R' with lead tetra-acetate or periodic acid (see p. 73):

R-CH(OH)-CH(OH)-R' — ^-> R-CHO + R'-CHO (g.-v.g.)

Ketones. By the oxidation of 1 : 2 glycols of the type R 2 C(OH)-C(OH)R' 2
with lead tetra-acetate or periodic acid:

R 2 C(OH)-C(OH)R' 2 — -^R 2 CO + R' 2 CO (g.-v.g.)

7. Aldehydes. Acetylene, when passed into hot dilute sulphuric acid in
the presence of mercuric sulphate as catalyst, is converted into acetaldehyde :

CH-CH + H 2 — > [CH 2 :CHOH] — > CH 3 -CHO (v.g.)

Ketones. All homologues of acetylene, treated in the same way as
acetylene, form ketones, e.g.,

RC-CH + H 2 — > [R-C(OH):CH 2 — ^ R-CO-CH 3 (g.)

8. Aldehydes. By Stephen's method (1925). An alkyl cyanide is dis-
solved in ether, or better, in ethyl formate or ethyl acetate (Stephen et al.,

ALDEHYDES AND KETONES I49

1956), and reduced with stannous chloride and hydrochloric acid, and then
steam distilled. Turner (1956), who has modified Stephen's method, also
showed that the yield of aldehyde from normal-cyanides increases with
increase in chain-length, and that branching in the a-position lowers the
yield.

According to Hantzsch (1931), Stephen's reaction proceeds via the imino-
chloride or aldimine hydrochloride (which is present as the stannichloride) :

r. C =N -^> [R-C=NH] + C1- -^% [R-CH=NH 2 ] 2 + SnCl 6 - " -^> R-CHO

HC1

Alkyl cyanides may also be reduced to aldehydes by means of the reverse
addition of the calculated amount of lithium aluminium hydride at low
temperature (see also p. 125).

4R-CN + LiAlH 4 — >■ (R-CH=N— ) 4 LiAl ^!% 4 R-CHO

Brown et al. (1959) have reduced amides and cyanides to aldehydes using ethoxy-
aluminium hydrides :

LiAlH.

2EtOH

>Li(EtO) 2 AlHi, + R-CONMe 2 -

3EtOH

- > Li(EtO) 3 AlH + RCN

-> R-CHO (70-90%)

Khuri (i960) has used lithium trimethylborohydride to reduce cyanides to
aldehydes.

Ketones. There is no analogous method for the preparation of ketones.

9. Aldehydes. By means of a Grignard reagent and formic ester (see

P- 354)-

Ketones. By means of a Grignard reagent and, for example, an alkyl
cyanide (see p. 355). Ketones are also formed by reaction between an alkyl
cyanide and Uthium-alkyls (p. 361).

10. Ketones. By the ketonic hydrolysis of the alkyl derivatives of
acetoacetic ester (see p. 228), and also from ethyl malonate derivatives

(P- 234).

Aldehydes. There is no analogous method for the preparation of alde-
hydes.

11. Aldehydes. Many aldehydes may be prepared by the oxo process
(p. 127), e.g., propionaldehyde (q.v.).

Ketones. There is no analogous method for the preparation of ketones.

12. Aldehydes. Aldehydes may be prepared (in yields up to 50 per cent.)
by means of the modified Sommelet reaction (Angyal et al., 1953 ; see benzal-
dehyde, p. 647); e.g., «-hexanal is produced by adding an aqueous solution
of w-hexylamine hydrochloride to a solution of hexamine in acetic acid
through which steam is passed.

CH S -(CH 2 ) 5 -NH 2 -HC1 + (CH 2 ) 6 N 4 — > CH^CH^-CHO (17%)

Ketones. There is no analogous method for the preparation of ketones.

Reactions common to aldehydes and ketones

It has been found that the reactivity of the carbonyl group depends on
the nature of the alkyl groups attached to it; the smaller the alkyl group,
the more reactive is the carbonyl group. Thus the order of reactivity is:

H\ CH a \ CH 3 \ C 9 H K \ C 2 H K

i\ CH 3 \ LH 3 \ L 2 H 6 \ C 2 ri 6 \

V^-r.^ V— ^ \^_^ )C=0> V

V C 2 H/

)c=o> ;c=o> ;c=o> /C=o> ;c=o

H/ H/ CH a / CH,/ C,h/

150 ORGANIC CHEMISTRY

If the alkyl groups are very large, e.g., tert. -butyl, the carbonyl group shows
very little reactivity. One contributing factor to this decreasing reactivity
with increasing size of the attached groups is steric hindrance (p. 107).

The carbonyl compounds undergo many nucleophilic addition reactions.
An important point in this connection is that the attack at a carbon atom is
always considered to be the factor that decides whether the attack is electro-
philic or nucleophilic; i.e., a negatively charged carbon atom is attacked by
an electrophilic reagent and a positively charged carbon atom by a nucleo-
philic reagent. It is not necessarily the order in which the fragments of the
addendum add on that decides electrophilic or nucleophilic addition; it is
whether the first carbon atom atom attacked is negative or positive, respec-
tively (see below) .

1 . Aldehydes and ketones are reduced catalytically or by nascent hydrogen,
aldehydes producing primary, and ketones secondary, alcohols, e.g.,

>C=0 + 2 [H]^^>CH0H (g.-v.g.)

Aldehydes and ketones are conveniently reduced by Raney nickel in aqueous
or ethanolic solution (yield: g.-v.g.). Reduction may also be effected by
means of the Meerwein-Ponndorf-Verley reduction (1925, 1926). The
carbonyl compound is heated with aluminium isopropoxide in wopropanol
solution ; the isopropoxide is oxidised to acetone, which is removed from the
equilibrium mixture by slow distillation:

R 2 CO +(CH 3 ) 2 CHOAl/ 3 ^ R 2 CHOAl/ 3 + CH 3 -CO-CH 3 ^g- R 2 CHOH (g.)

This reducing agent is specific for the carbonyl group, and so may be used
for reducing aldehydes and ketones containing some other functional group
that is reducible, e.g., a double bond or a nitro-group. The Meerwein-
Ponndorf-Verley reduction has been improved by Truett et al. (1951).
Aldehydes and ketones are also reduced to alcohols by lithium aluminium
hydride, lithium borohydride and sodium borohydride (p. 126).

Williams et al. (1953) used aluminium isopropoxide containing deuterium to
reduce cyc/ohexanone, and concluded that the mechanism involves the formation
of a cyclic complex:

\ A1 / \ A i/ OH O

C 3 H 7 o/ \oC 3 H 7 C 3 H 7 CX \OC 3 H,

When carbonyl compounds are heated with concentrated hydriodic acid
and red phosphorus, the carbonyl group is reduced to a methylene group
— CH 2 — . This may also be effected by means of the Clemmensen reduction
(1913) : the carbonyl compound is reduced with amalgamated zinc and con-
centrated hydrochloric acid:

R-CHO + 4 [H] -^?% R-CH 3 + H 2

HO

R-COR' + 4[H] -^JV R-CH 2 -R' + H 2

The Clemmensen reduction does not appear to work well for aldehydes,
but is reasonably good for many ketones.

ALDEHYDES AND KETONES 151

The mechanism of the Clemmensen reduction is uncertain. Poutsma et al.
(1959) have shown, by gas chromatography, that the reduction of acetophenone
proceeds as follows (Ph = C 6 H 6 ) :

PhCOMe — > PhCHOHMe — > PhCHClMe — > PhCH 2 Me

It was also shown that the alcohol, phenylmethylmethanol, was reduced to
ethylbenzene under the same conditions.

2. Aldehydes and ketones add on sodium hydrogen sulphite to form
bisulphite compounds :

/OH
>C=0 + NaHS0 3 ^ >C( or

\SO3Na

/OH
^SO^

Na +

These bisulphite compounds are hydroxysulphonic acid salts, since the sulphur
atom is directly attached to the carbon atom. This structure is supported
by work with isotope 34 S (Sheppard et al., 1954).

Most aldehydes form bisulphite compounds. Ketones of the type
CH 3 *CO - R, where R is a primary alkyl group, form bisulphite compounds
fairly readily; but where R is a secondary or tertiary alkyl group, the
bisulphite compound is formed very slowly. If the ketone is of the type
R'CO'R' where R and R' are ethyl or higher alkyl groups, the bisulphite
compound is not formed at all (cf. reactivity of carbonyl compounds,
above).

Bisulphite compounds are usually crystalline solids, insoluble in sodium
hydrogen sulphite solution. Since they regenerate the carbonyl compound
when heated with dilute acid or sodium carbonate solution, their formation
affords a convenient means of separating carbonyl compounds from non-
carbonyl compounds.

The mechanism of bisulphite formation is uncertain. Kinetic studies have
shown that it is probably sulphite ions which are the active species; thus a
possibility is :

NaHSO a ^=±: Na+ + H+ + SOa 2 "

>C=0 H + ^ >C— OH
O

o— s: >o— oh ^ >cc
o-

3. Aldehydes and ketones add on hydrogen cyanide to form cyanohydrins.
The carbonyl compound is treated with sodium cyanide and dilute sulphuric
acid:

/OH
>C=0 + HCN — >C( (g.-v.g.)

\CN

Cyanohydrins are important compounds in organic synthesis since they
are readily hydrolysed to a-hydroxy-acids:

R-CH(OH)-CN —^L^ R-CH(OH)-C0 2 H

All aldehydes form cyanohydrins; only the ketones acetone, butanone,
pentan-3-one, and pinacolone form cyanohydrins.

152 ORGANIC CHEMISTRY

Cyanohydrins are also prepared, if possible, indirectly from the bisulphite
compound, which is treated with sodium (or potassium) cyanide solution :

>C(OH)-S0 3 Na + NaCN — >■ >C(OH)-CN + Na 2 S0 2 (v.g.)

Lapworth (1903, 1904) showed that the addition of hydrogen cyanide to carbonyl
compounds is accelerated by bases and retarded by acids, and he concluded that
the addendum was the cyanide ion, and proposed the following mechanism :

PV slow ^ /°~ Ht-j fast „ / 0H

R 2 ok) + CN- ^=^ R 2 C( — '—>■ R 2 C(

\CN \CN

Svirbeley et al. (1955), however, have found that the addition of cyanide ion to,
e.g., propionaldehyde, is slightly subject to general acid catalysis (of. p. 168). In
this respect, the reaction is similar to semicarbazone formation (p. 153), and so
the following mechanism has been proposed :

fast

>C=0 +HA^±

„ slow . ,

>C=0— HA + CN- -> >C( + A-

\OH

4. Aldehydes and ketones form oximes when treated with hydroxylamine :
R 2 CO + NH,OH — >

/OH "I _ Ha o

> R,C=*

R 2 C( ^R 2 C=NOH (v.g.)

\nhohJ

The mechanism of this reaction is discussed in connection with semicarb-
azones (see below).

Oximes are usually well-defined crystalline solids, and may be used to
identify carbonyl compounds.

Aldoximes form cyanides when boiled with acetic anhydride, whereas ketoximes
form the acetyl derivative of the oxime :

(CH,CO) a O „

r-ch:n-oh — — % R-CN

-H,0

r„c:n-oh *E^% r 3 c:nococh 3

5. Aldehydes and ketones react with hydrazine to form hydrazines and
azines. The mechanism of their formation is probably similar to that of
oxime formation (see above) :

Hydrazone:

yc=o + h 2 n-nh 2 — ► [>c^ 0H 1 -=^ >c:n-nh 2

L x nh-nhJ

>c:n-nh 2 + o=c< — > h 2 o + >c:n— n:c<

By using suitable derivatives of hydrazine, more well-defined crystalline
products are obtained (and azine formation is avoided), e.g.,

(i) Phenylhydrazine forms phenylhydrazones : -

c 6 h 5 -nh-nh 2 + o=c< — > h 2 o + c 6 h 5 -nh-n:c<

ALDEHYDES AND KETONES 153

(ii) ^>-Nitrophenylhydrazine forms />-nitrophenylhydrazones:

no./ - \nh-nh 2 + o=c< — ^ no 2 / ^nh-n:c< + H 2

(iii) 2 : 4-Dinitrophenylhydrazine forms 2 : 4-dinitrophenylhydrazones :
_N0 2 _N0 2

NO,/ %NH'NH 2 + 0=C<-^N0 2 / ^NH-N:C< + H 2

) 2 ^ ^NH-NH 2 + 0=C<-

(iv) Semicarbazide forms semicarbazones:

NH 2 -CO-NH-NH 2 + 0=C< — > NH 2 -CO-NH-N:C< + H 2

The semicarbazone reaction has been studied in some detail by Conant et al.
(1932) and Westheimer (1934), an & has been shown to be a general acid-catalysed
reaction. The following mechanism has been proposed by Bartlett (1953) :

NH.-NH-CO-NH,

R a C=0 + HA^i R a C=0~-HA ~

•*TH a -NH-CONH a + A- * NNH-NH-CONH,,

(I)
H

/O-H— A +

X ^ R a C=NH-NH-CONH a + H a O + A- ^

NH-NH-CONH a

R»C=N-NH-CO-NH 2 + H a O + HA

a CX

Since the first step is formation of the acid complex, the rate of its formation
would be expected to increase with increasing pH. On the other hand, the
greater the pH, the greater will be the concentration of the semicarbazide salt :

NH a -NH-CONH 2 + H 3 + =^: NH 3 -NH-CO-NH a + H a O

This salt, however, no longer has a lone pair of electrons on the N atom, and so is
no longer a nucleophilic reagent. Thus the second step will be retarded or may
even be inhibited. It can therefore be expected that there will be an optimum
pH for the reaction. Both Conant and Westheimer found this to be the case.

This mechanism is also believed to operate in the formation of bximes and
phenylhydrazones. The addition compound (of type I) has been isolated in
certain cases, e.g., with chloral.

The carbonyl compound may be regenerated from the oxime, phenyl-
hydrazone or semicarbazone by boiling with dilute acid (hydrochloric or
sulphuric acid).

Wolff-Kishner reduction (1912). When hydrazones (or semicarbazones)
are heated with sodium ethoxide at 180 , nitrogen is eliminated, and a
hydrocarbon is obtained, i.e., by this means the carbonyl group is converted
into the methylene group:

>c:n-nh 2 -^>ch 2 + n 2 ( g .-v.g.)

An improved technique is to heat the carbonyl compound in ethylene or
trimethylene glycol with 85 per cent, hydrazine hydrate in the presence of three
equivalents of sodium, sodium hydroxide or potassium hydroxide, at 180-200 ;
the yield is usually above 65 per cent. (Huang-Minion, 1946). The yield of
hydrocarbon is better than that obtained by the Clemmensen reduction.
Reduction of sterically hindered carbonyl groups by either the Clemmensen
reduction or the Wolff-Kishner procedure is not easily effected. Barton et al.

154 ORGANIC CHEMISTRY

(1954, !955) have modified Huang-Minion's method (by excluding water), and
have thereby reduced sterically hindered carbonyl groups.

According to Szmant et al. (1952), the mechanism of the Wolff-Kishner re-
action is (B is the base) :

RjC^N-NHj + B ^i R 2 C=N'NH + BH+ — > R 2 CH + N a -^- R 2 CH 2

Weisberger et al. (1956) have suggested a free-radical mechanism.

Girard's reagents for carbonyl compounds (1934, 1936). Girard intro-
duced two reagents for carbonyl compounds: Girard's reagents " T " and
" P ", which are respectively trimethylaminoacetohydrazide chloride:
[(CHgJgN'CHyCO-NH'NHJ+Cl - , and pyridinium-acetohydrazide chloride,
[C 5 H 6 N-CH a -CO-NH«NH 2 ] + C1- (see p. 209) . These reagents react with carbonyl
compounds to form derivatives of the type [(CH 3 ) 3 N-CH 2 -CO-NH-N:Cc3 + Cl-.
These compounds are soluble in water, and have been found particularly
useful for isolating certain ketonic hormones.

6. Aldehydes and ketones condense with thio-alcohols (mercaptans) to
form mercaptals and mercaptols, respectively:

>C=0 + 2RSH — > >C(SR) a + H 2

When mercaptols (dithioketals) are heated with hydrazine hydrate and Raney
nickel at 90-135 , the compound undergoes desulphuration (Georgian et al.,
1959).

R 2 C(SR') 2 >- R 2 CH 2

7. Aldehydes and ketones add on a molecule of a Grignard reagent,
and the complex formed, when decomposed with water, gives a secondary
alcohol from an aldehyde (except formaldehyde, which gives a primary
alcohol), and a tertiary alcohol from a ketone (see p. 352) :

/OMgX HO /OH

>C=0 + RMgX — > >C( — =-► >C(

\R \R

8. Both aldehydes and ketones undergo the Wittig reaction to form olefins
(p. 64).

9. Phosphorus pentachloride reacts with simple carbonyl compounds to
form gm-dichlorides:

>C=0 + PC1 5 — > >CC1 2 + POCI3

The yield of gem-dichloride is usually good with aldehydes, but with ketones
it is poor. With more complicated carbonyl compounds chlorination also takes
place, substitution occurring on the a-carbon atom only; e.g., di-isopropyl ketone
gives a-chlorodiisopropyl ketone when treated with one molecule of phosphorus
pentachloride :

(CH 3 ) 2 CH-CO-CH(CH 3 ) 2 + PC1 5 ^->(CH 3 ) 2 CCl-CO-CH(CH 3 ) a + PC1 3 + HC1

Very little g^m-dibromides are obtained with phosphorus pentabromide ; the
main product from any carbonyl compound is the a-bromo-derivative. Phos-
phorus pentabromide dissociates more readily than phosphorus pentachloride,
and it is probable that the halogenation in the a-position is brought about by the
free halogen (see below) .

10. Chlorine or bromine replaces one or more a-hydrogen atoms in
aldehydes and ketones, e.g., acetone may be brominated in glacial acetic
acid to give monobromoacetone :

CH 3 -COCH 3 + Br 2 — > CH 3 -COCH 2 Br + HBr (43-44%)

The halogenation of carbonyl compounds is catalysed by bases and by
acids, but the details of the mechanisms are not certain. Let us consider

ALDEHYDES AND KETONES 155

the case of acetone. Now it is known that both bases and acids catalyse
enolisation of a ketone (see p. 221 for further details).

OH

base or acid |

CHo'CO'CH, s CH«*C == CH«

3 3S (stow) 3 2

keto enol

If the enol form is the one which is brominated, the bromination can be
considered to occur as follows :

OH O-rH O

I •— Sl r\ fast \* _H+ H

CH 3 -C=CH 2 Br— Br > CH 3 -C + -CH 2 Br + Br~ v ^ CH 3 -C-CH 2 Br

In both base and acid catalysed reactions it has been shown that the rate of
bromination (and of any other halogen) is independent of the concentration
of the bromine but is proportional to the concentration of ketone and base
or acid present. This is in agreement with at least a two-stage reaction,
the first and slow stage being enolisation, and the second and fast stage
being bromination.

A very interesting point about this mechanism is that if the change from
keto to enol could be speeded up to such an extent that it is now faster than
the halogenation, then the concentration of the halogen would enter into
the rate equation. This has actually been observed in, e.g., the iodination
of acetophenone in the presence of sulphuric acid (Hammett et al., 1939).
Another point that we shall discuss here is: How is it possible for a C — H
bond to be broken so readily? This must occur whatever the mechanism
of enolisation, and has been explained as follows. Owing to the inductive
effect of the carbonyl group (caused by the high electron-affinity of the
oxygen atom), the electrons of the C — H bonds on the a-carbon atom are
displaced towards the carbon atom, thus giving rise to a condition for an
incipient proton leaving C a .

Since the inductive effect falls off extremely rapidly from the
H O source, only a-hydrogen atoms will be affected by the induc-
•^ || tive effect of the carbonyl group. This offers an explanation
— C a ->C — for the unusual reactivity of hydrogen in a methyne group
-f (=CH — ) or in a methylene group ( — CH 2 — ) when adjacent

H to a carbonyl group or any other strongly electron-attracting

group. In such structures the incipient proton is readily re-
moved, and the compound is said to possess an " active " methyne or
methylene group.

Aldehydes and ketones with a-hydrogen atoms readily react with sulphuryl
chloride at room temperature in the absence of a catalyst, to replace a-hydrogen
atoms only, e.g.,

-i^V CH 3 -COCH 2 Cl

The halogen atom in the group CX-CO is very reactive, far more so than
the halogen atom in alkyl halides. It appears that the adjacent carbonyl
group plays a part, but its mode of operation is still not clear.

11. Aldehydes and ketones with a methyl or methylene group adjacent
to the carbonyl group are oxidised by selenium dioxide at room temperature
to dicarbonyl compounds; e.g., acetaldehyde forms glyoxal, and acetone
forms methylglyoxal:

CH 3 -CHO + Se0 2 — > CHO-CHO + Se + H 2
CH 3 -CO-CH 3 + Se0 2 — > CH 3 -CO-CHO + Se + H 2

156 ORGANIC CHEMISTRY

12. Aldehydes and ketones are converted into the formyl derivative
of the corresponding primary amine by excess ammonium formate or
formamide (Leuckart reaction. 1885) :

>C=0 + 2H-C0 2 NH 4 — > >CH-NH-CHO + 2H 2 + C0 2 + NH 3
>C=0 + 2H-CO-NH 2 — > >CH-NH-CHO + C0 2 + NH 3

13. Aldehydes and ketones undergo the Schmidt reaction (1924). This is
the reaction between a carbonyl compound and hydrazoic acid in the
presence of, e.g., concentrated sulphuric acid. Aldehydes give a mixture of
cyanide and formyl derivatives of primary amines, whereas ketones give
amides:

R-CHO + HN 8 ' *> R-CN + R-NH-CHO
R-COR + HN 3 — '—± R-CONH-R + N 2

The mechanism of the reaction is uncertain. It has been shown to be intra-
molecular, and Smith (1948) has proposed the following mechanism, which is an
example of the 1,2-shift {from carbon to nitrogen) ; for ketones :

OH
"""'-> R— C— R — ^>

\ c / h+ > \<y hn - . - x _ - h »°.

&>

OH HN— N=N:

R\ /R -N, + H.O + -H+

> O— R > H z O— CR >■ 0=CR

1,

N— NEEN RN RN NHR

(I)

For aldehydes the mechanism is the same, except that ion (I) is now (la), and
hence R or H can migrate :

H \ H,o

\C+ > H— C=0

NR NHR

/ R -H+

*C/ >- R— CN

HN

In ketones, if the two radicals are not identical, then two geometrical isomers of
(I) are possible. It is also reasonable to suppose that the anti group (to the
diazonium nitrogen) is the group that migrates (cf. the Beckmann rearrange-
ment, p. 667). In this way it is possible to explain how steric factors may
influence the isomer ratio of amides formed :

R \c/ R ' ■

'I +

N— N +

-> R— N=C— R' H '°> R-NH-COR'

R\ /R' + h.o
\c/ > R-£=NR' -+ R-CO-NHR'

+ II
+ N 2 — N

14. Aldehydes and ketones undergo condensation reactions, i.e., two or
more (identical or different) molecules unite with, or without, the elimination

ALDEHYDES AND KETONES 157

of water (or any other simple molecule). There is very little difference
between the terms condensation and condensation polymerisation or poly-
condensation (p. 79) ; generally, condensation is used for those reactions
in which the resulting compound is made up of a small number of the
reacting molecules.

Aldol Condensation. Acetaldehyde, in the presence of dilute sodium
hydroxide, potassium carbonate or hydrochloric acid, undergoes condensa-
tion to form a syrupy liquid known as aldol :

NaOH

2 CH 3 -CHO ■> CH 3 -CH(OH)-CH 2 -CHO (50%)

On heating, aldols eliminate water to form unsaturated compounds, e.g.,
aldol forms crotonaldehyde:

CH 3 -CH(OH)-CH 2 -CHO — > CH 3 -CH:CH-CHO + H 2

In many cases it is the unsaturated compound that is isolated, and not the
aldol, e.g., mesityl oxide and phorone (q.v.).

The aldol condensation can occur: (i) between two aldehydes (identical
or different); (ii) between two ketones (identical or different); and (iii)
between an aldehyde and a ketone. Whatever the nature of the carbonyl
compound, it is only the a-hydrogen atoms which are involved in the aldol
condensation.

(i) Generally, with two different aldehydes all four possible condensation
products are obtained; but by using different catalysts one product may
be made to predominate in the mixture, e.g.,

CH 3 -CHO + CH 3 -CH 2 -CHO ^^CH 3 -CH 2 -CH(OH)-CH 2 .CHO

CH.-CHO + CH 3 -CH 2 -CHO ~E=± CH 3 -CH(OH)-CH(CH 3 )-CHO

(ii) Acetone, in the presence of barium hydroxide, gives diacetone alcohol:

2CH 3 -CO-CH 3 , Ba(0H) i N (CH 3 ) 2 C(OH)-CH 2 -COCH 3

This equilibrium lies almost completely on the left, but the yield of diacetone
alcohol may be increased (71 per cent.) by boiling acetone in a Soxhlet with
solid barium hydroxide in the thimble. The acetone in the flask gets
richer and richer in diacetone alcohol, since the boiling point of the latter
is 164 , and that of the former is 56 .

When acetone is treated with hydrochloric acid, mesityl oxide and phorone
are formed:

2CH.-COCH, ^^ (CH 3 ) 2 C:CH-CO-CH 3 ^•• C °- CH ' ^

(ch 3 ) 2 c:ch-co-ch:c(ch 3 ) 2

(iii) When aldehydes condense with ketones, it is the a-hydrogen atom
of the ketone which is involved in the condensation, e.g.,

CH 3 -CHO + CH.-CO-CH, ^^ CH 3 -CH(OH)-CH 2 -CO-CH 3 (25%)

The yield of 4-hydroxypentan-2-one is low because aldol and diacetone
alcohol are also formed.

It is generally accepted that the base-catalysed aldol condensation of acet-
aldehyde takes place in two steps, the first being the formation of the carbanion
(I), and the second the combination of this anion with a second molecule of

158 ORGANIC CHEMISTRY

acetaldehyde to form the anion (II) of the aldol. The simplest mechanism that
embraces these facts is :

rv rv

HO- H— CH 2 -CHO ^e: H 2 + CH 2 -CHO

(I)

HyC

vT>

O- OH

H CH 2 -CHO ^i CH 3 -CH-CH 2 -CHO ^"' 0v - CH 3 'CH-CH 2 -CHO + OH~

(II)

Bell's work (1937) suggested that the reaction in dilute sodium hydroxide solution
was first-order with respect to the aldehyde, thereby indicating that the first step
is the rate-determining one. Bell also found that the reaction was not exactly
first-order in hydroxide ion. This is not in keeping with the mechanism given
above. More recent kinetic work led to a reaction order between first and
second, varying with the aldehyde concentration (Broche et al., 1955). This
suggests that the first and second steps of the reaction are of comparable rates,
and this conclusion was supported by the work of Bell et al. (1958). Bell et al.
(i960), however, have investigated the acidic properties of acetaldehyde in
alkaline solutions, where it is converted into the ion (III), CHyCHOH'O - .
This produces considerable reduction of the hydroxide ion concentration and
affects the apparent order of the reaction. Ion (III) is present in concentrations
comparable to the hydroxide ion concentration and may well contribute to the
formation of carbanion (I) in the first step given above. When the formation
of (III) was taken into account, Bell found that the corrected results showed the
aldol condensation is almost of the second-order in acetaldehyde. These results
have been interpreted by assuming the ionisation and condensation steps have
comparable rates (as already suggested above).

On the basis of analogy with acetaldehyde, the mechanism of the base-
catalysed condensation of acetone will be :

MeCOMe + OH" ^=i H 2 + CH 2 -COMe

<?

O O- OH

Me 2 C CH 2 -COMe =^= Me 2 OCH 2 -COMe ^ H, ° v Me 2 OCH 2 -COMe + OH"

There appear to be no detailed kinetic investigations on acid-catalysed aldol
condensations, but it is generally assumed that the condensation proceeds by
reaction between conjugate acid and the enol form of the carbonyl compound;
e.g., the formation of mesityl oxide:

Me 2 C=0 ^" + v Me 2 C— OH
Me
Me 2 C CH 2 =CMe— O— H ^ Me 2 OCH 2 — C=0 + H + ^

OH OH

Me Me

I + I -H + _,

Me 2 C-CH 2 — C=0 ^=i H 2 + Me 2 OCH 2 -C=0 "" -7" Me 2 C=CH-COMe

+ OH 2

Claisen condensation (1881). This is the condensation between an ester
and another molecule of an ester or ketone (see p. 224).

Claisen-Schmidt reaction (also known as the Claisen reaction) is the con-
densation between an aromatic aldehyde (or ketone) and an aldehyde or
ketone, in the -presence of dilute alkali to form an a (E-unsaturated compound.

ALDEHYDES AND KETONES 159

This reaction is similar to the aldol condensation, and is illustrated by the
formation of cinnamaldehyde from benzaldehyde and acetaldehyde:

C 6 H B -CHO + CH 3 -CHO Na ° H > C 6 H S -CH:CH-CH0 + H 2

The mechanism of the Claisen-Schmidt reaction is probably that of the base-
catalysed aldol condensation :

HO- H— CH.J-CHO ^±: H a O + CH a -CHO
>0 O- OH

C 6 H B -CH CH 2 -CHO ^=i C 6 H 5 CH-CH 2 CHO;p^C 6 H 6 -CH-CH 2 -CHO + OH"

The elimination of a molecule of water from the condensation product in alkaline
solution possibly proceeds as follows :

C 6 H 5 -CH— CH-CHO + OH- ^ H 2 + C 6 H 6 -CH— CH-CHO ^

£1 SI

)H H OH

OH- + C 6 H 5 -CH=CH-CHO

Other condensations involving aldehydes or ketones are the Knoevenagel
(p. 280) and Perkin (p. 651) reactions.

15. Darzens Glycidic Ester Condensation (1904). This reaction involves
the condensation of an aldehyde or ketone with an a-halogeno-ester to
produce an a : (3-epoxy-ester (glycidic ester) ; the condensing agent is usually
sodium ethoxide or sodamide:

„ C,H 5 ONa

R-CO-R' + R"-CHCl-CO,C s H 5 — ~ >

R"

OI

R >— I

•C0 2 C a H s + NaCl + C 2 H 6 OH

These glycidic esters, on hydrolysis, give the epoxy-acid, and this, on de-
carboxylation, gives a ketone, or an aldehyde which is formed by rearrange-
ment :

R" R"

R\ I (i)NaOH R\ I -CO,

)C C«WH, 7-^ pc— C-CO.H >

w/\ Q / (U)HC1 K" / \ Q /

R\ R\

)CH-CO-R"o>- R'-^C-CHO
R'/ R"/

Darzens (1936, 1937) nas a ^ so prepared glycidic esters by condensing a : a-di-
halogeno-esters with an aldehyde or ketone in the presence of dilute magnesium
amalgam :

Mg/Hg R\ H,0

H > x c _ c

R-CO-R' + CHC1 2 -C0 2 C 2 H 5 > )C— CHC1-C0 2 C 2 H

R'/|

OMgCl

R\ C 8 H 6 0Na R\

>C— CHC1-C0 2 C,H 5 > )C CH-CO a C 2 H 5

l6o ORGANIC CHEMISTRY

According to Ballester et al. (1955), the mechanism of the reaction is:

CH a Cl-C0 2 Et + B z^±z CHOC0 2 Et + BH+ RRC °>

RR'C— CH-C0 2 Et > RR'C CH-C0 2 Et + Cl~

This is an example of neighbouring group participation.

Reactions given by aldehydes only. 1. Aldehydes restore the magenta
colour to Schiff's reagent (rosaniline hydrochloride is dissolved in water,
and sulphur dioxide is passed in until the magenta colour is discharged).
The mechanism of this reaction is obscure.

2. Aldehydes are very easily oxidised, and hence are powerful reducing
agents. Aldehydes reduce Fehling's solution (an alkaline solution containing
a complex of copper tartrate) to red cuprous oxide; and Tollens' reagent
(ammoniacal silver nitrate solution) to metallic silver, which often appears
as a silver mirror on the walls of the containing vessel. In both cases the
aldehyde is oxidised to the corresponding acid :

R-CHO + [O] — > R-C0 2 H

It is worth noting that Fehling's solution and Tollens' reagent are both
weak oxidising agents, and can be used to oxidise readily oxidisable groups,
e.g. , they will oxidise unsaturated aldehydes to the corresponding unsaturated
acids; they are not sufficiently strong oxidising agents to attack an ethylenic
bond.

According to Veksler (1952), ketones with the structures Ar-CHj'COCH, and
Ar 2 CH*COCH 3 (Ar = aryl radical) reduce cold ammoniacal solutions of silver
oxide to a silver mirror; it appears that ammoniacal silver nitrate is not reduced.

According to Daniels et al. (i960), some aliphatic aldehydes, e.g., acetaldehyde,
crotonaldehyde, etc., do not give a positive test with Fehling's solution. These
authors suggest that Tollens' test is most satisfactory for distinguishing between
aldehydes and ketones.

3. All aldehydes except formaldehyde form resinous products when
warmed with concentrated sodium hydroxide solution. The resin is
probably formed via a series of condensations, e.g.,

2CH 3 -CHO — > CH 3 -CH(OH)-CH 2 -CHO - — --> CH 3 -CH:CH-CHO %

ch 3 -ch:ch-ch(oh)-ch 2 -cho ~ Ha °> ch 3 -ch:ch-ch:ch-cho, etc.

Aldehydes that have no a-hydrogen atoms undergo the Cannizzaro
reaction (1853), in which two molecules of the aldehyde are involved, one
molecule being converted into the corresponding alcohol, and the other into
the acid. The usual reagent for bringing about the Cannizzaro reaction is
50 per cent, aqueous or ethanolic alkali, e.g.,

2H-CH0 + NaOH — > H-C0 2 Na + CH 3 OH

The Cannizzaro reaction is mainly applicable to aromatic aldehydes (see
p. 649).

Although the Cannizzaro reaction is characteristic of aldehydes having no
a-hydrogen atoms, it is not confined to them, e.g., certain aliphatic a-mono-

ALDEHYDES AND KETONES l6l

alkylated aldehydes undergo quantitative disproportionation when heated with
aqueous sodium hydroxide at 170-200° (Hausermann, 1951):

200°

2(CH 3 ) a CH-CHO + NaOH -> (CH s ) a CH-CO a Na + (CH 3 ) 2 CH-CH a OH (100%)

The Cannizzaro reaction can take place between two different aldehydes,
and is then known as a " crossed " Cannizzaro reaction (see formaldehyde,
p. 165).

All aldehydes can be made to undergo the Cannizzaro reaction by treat-
ment with alurniniiim ethoxide. Under these conditions the acid and
alcohol are combined as the ester, and the reaction is then known as the
Tischenko reaction (1906); e.g., acetaldehyde gives ethyl acetate, and
propionaldehyde gives propyl propionate:

2 CH 3 -CHO * (0C|W V CH 3 -C0 2 C 2 H 5
2 CH a -CH i -CHO j -^^VCH 3 -CH 2 -C0 2 CH a -CH 2 -CH 3

Aldehydes give simple esters with aluminium ethoxide, the aldol condensation
with sodium ethoxide, and trimeric glycol esters with magnesium or calcium
ethoxides (Villain and Nord, 1947) :

3 R-CH a -CHO Mg(0 °' H,) '> R-CH a -CH(OH)-CHR-CH a O-CO-CH a -R

4. Aldehydes (except formaldehyde) react with ammonia in ethereal
solution to give a precipitate of aldehyde-ammonia, e.g.,

CH 3 -CHO + NH 3 — >■ CH 3 -CH(OH)-NH 2 (50%)

These aldehyde-ammonias are unstable and readily undergo cyclic polymeri-
sation, and it has been suggested, from X-ray crystal analysis, that acetalde-
hyde-ammonia is the trihydrate of trimethylhexahydrotriazine * (Moerman,
1938):

CH 3 -CHO + NH 3 — ► CH.-CHONH, P °' ymerises > (CH 3 -CHONH 3 ) 3 — >

/NH— CH-CH 3
CH 3 -CH( >NH-3H a O

\NH— CH-CH 3

5. Aldehydes combine with alcohols in the presence of dry hydrogen
chloride or calcium chloride to form first the hemi-acetal, and then the
acetal:

R-CHO + R'OH ~~^ R-CH(OH)-OR' ;==± R-CH(OR') 2 + H 2 (g.)

The hemi-acetal is rarely isolated since it very readily forms the acetal.
Acetals are diethers of the unstable 1 : i-dihydroxyalcohols, and may be
named as 1 : i-dialkoxyalkanes. Unlike the parent dihydroxyalcohols
these acetals are stable. They are also stable in the presence of alkali,
but are converted into the aldehyde by acid. Thus acetal formation may
be used to protect the aldehyde group against alkaline oxidising agents.
On the other hand, the aldehyde group can be protected in acid solution by
mecaptal formation (p. 334).

* The reaction between aldehydes and ammonia is usually quoted to be simple.
Although the first product is simple, the final one is complicated, and so this reaction
is not characteristic of aldehydes, since ketones also give complex products with
ammonia. In the latter case, however, it is doubtful whether a simple addition
product has ever been isolated.
G

l62 ORGANIC CHEMISTRY

A possible mechanism is :

OH OH

_tf

I EtOTT I' *t* H")"

R-CH + H+ :?=s R— C+ ^ R»C— OEt^T^T"

I I I ^

H H H

OH + OH 2

RC-OEt ^^ RC-OEt ^g± RC-OEt ^^
| , -TOT ,

H H H

HOEt

I — H+v

RCH-OEt ^ +H+ S " RCH(OEt) 3

6. Aldehydes react with aniline to form anils (azomethines) :

R-CHO + C 6 H 6 -NH 2 — > R-CH:N-C e H 6 + H a O
Aliphatic azomethines tend to polymerise; the aromatic analogues are stable.

7. The lower aldehydes polymerise with great ease (see formaldehyde
and acetaldehyde, below).

Reactions given by ketones only. 1. Ketones do not give Schiff's reaction.
Acetone, however, restores the magenta colour very slowly.

2. Ketones are not easily oxidised (c/. p. 132) ; they do not reduce Fehling's
solution or ammoniacal silver nitrate (see p. 160).

oe-Hydroxyketones, i.e., compounds containing the group •CH(OH) , CO,
readily reduce Fehling's solution and ammoniacal silver nitrate.

Strong oxidising agents, e.g., acid dichromate, nitric acid, etc., oxidise
ketones, only the carbon atoms adjacent to the carbonyl group being
attacked, and the carbon atom joined to the smaller number of hydrogen
atoms is preferentially oxidised :

CH 3 -CO-|CH 2 -CH 3 [ ° ] > 2 CH 3 -C0 2 H

If the adjacent carbon atoms have the same number of hydrogen atoms,
the carbonyl group remains chiefly with the smaller alkyl group; e.g., when
hexan-3-one is oxidised, the main product is propionic acid, and this is
accompanied by much smaller amounts of acetic and butyric acids :

CH 3 -CH 2 -qO-|CH 2 -CH 2 -CH s -^-> 2CH 3 -CH 2 -C0 2 H
CH 3 -CH 2 l-CO-CH 2 -CH 2 -CH 3 -^->CH 3 -C0 2 H + CH 3 -CH 2 -CH 2 -C0 2 H

All ketones containing the acetyl group, CH 3 *CO, undergo the haloform
reaction. This reaction is best carried out by dissolving the compound in
dioxan, adding dilute sodium hydroxide, then a slight excess of iodine in
potassium iodide solution, warming, and finally adding water. If the
compound contains the acetyl group, iodoform is precipitated (cf. iodoform,
p. 118).

The haloform reaction is very useful in organic degradations. A positive
iodoform test is given by all compounds containing the acetyl group attached
to either carbon or hydrogen, or by compounds which are oxidised under
the conditions of the test to derivatives containing the acetyl group, e.g.,
ethanol, isopropanol, etc.

ALDEHYDES AND KETONES 163

Booth and Saunders (1950) have shown that a number of other compounds
besides those mentioned above also give the haloform reaction, e.g., certain
quinones, quinols and m-dihydric phenols.

In practice, the base-catalysed iodination of aliphatic methyl ketones does not
occur quantitatively according to the equation :

CH 3 -COR + 3l 2 + 4NaOH — > CHI 3 + R-CO a Na + 3NaI + 3 H 2

For example, acetone and butanone consume more than the theoretical amount
of iodine, whereas, e.g., methyl isopropyl ketone consumes less than this
quantity (Cullis et al., 1956). Thus methods based on this reaction for the
quantitative estimation of methyl ketones are unsatisfactory.

3. Ketones react with ammonia to form complex condensation products,
e.g., if acetone is treated with ammonia and then followed by acidification of
the reaction products, diacetonamine (I), and triacetonamine (II), are formed.
If the reaction is carried out at room temperature (I) is the chief product ;
if heated, (II) is the chief product (Bradbury et al., 1947).

The mechanism of the reactions is uncertain, but it may be as follows.
Ammonia (base) causes acetone to undergo the aldol condensation to give
mesityl oxide and phorone, which then combine with ammonia to form
respectively (I) and (II) (see also p. 278) :

(CH 3 ) 2 C:CH-COCH 3 + NH 3 — > (CH 3 ) 2 C(NH 2 )-CH 2 -COCH 3

(I)

(Ch 3 ) 2 c:ch-co-ch:c(ch 3 ) 2 + 2NH3 — >

/C(ch 3 ) 2 -ch 2
^% nh( >co

\C(CH s ) 2 — CH 2

(II)

(CH 3 ) 2 C-CH 2 -CO-CH 2 -C(CH 3 ) 2 -
NH a NH 2

4. Ketones do not readily form ketals when treated with alcohols in the
presence of hydrogen chloride or calcium chloride (cf. acetals, above).
Ketals may, however, be prepared by treating the ketone with ethyl ortho-
formate (Helferich and Hauser, 1924) :

R 2 CO + H-C(OC 2 H 5 ) 3 — -> R 2 C(OC 2 H 5 ) 2 + H-C0 2 C 2 H 5 (g.)

5. When ketones are reduced catalytically or in acid solution, secondary
alcohols are obtained in good yields, but when they are reduced in neutral
or alkaline solution, pinacols are the main products; e.g., acetone reduced
with magnesium amalgam forms pinacol:

2CH 3 -COCH 3 + 2[H] -^(CH 3 ) 2 C(OH)-C(OH)(CH 3 ) 2

6. When ketones are treated with nitrous acid, the " half oxime " of
the a-dicarbonyl compound is formed, e.g., acetone gives oximinoacetone
(wonitrosoacetone; see p. 307):

CH 3 -COCH 3 + HN0 2 C ' H ''°' NO/HC '> CH 3 -COCH:N-OH + H 2

All compounds containing the •CH 2 -CO* group form the oximino-derivative
with nitrous acid, and this has been used to detect the presence of the
•CH 2 'CO group.

Benzaldehyde can also be used to detect the presence of the •CH 2 *CO*
group (p. 651) :

164 ORGANIC CHEMISTRY

7. Ketones condense with chloroform in the presence of potassium
hydroxide to form chlorohydroxy-compounds.

K0H /OH

CH 3 -CO-CH 3 + CHCI3 > (CH 3 ) 2 C(

X CC1 S

8. Ketones form sodio-derivatives when treated with sodium or sodamide
in ethereal solution (see p. 230), e.g.,

CH.-CO-CH, + NaNH„

lCH 3 '

o- 1

'C=ChJ

Na + NH 3

Formaldehyde [methanol), H-CHO, is prepared industrially:

(i) By passing methanol vapour over copper at 300° :

CH 3 OH — ^ H-CHO + H 2

(ii) By passing methanol vapour mixed with air over a copper or silver
catalyst at 250-360° :

CH 3 OH + §0 2 — > H-CHO + H 2

The amounts of methanol and air must be carefully controlled, otherwise
formic acid will result due to the further oxidation of. the formaldehyde:

H-CHO + J0 2 — > H-C0 2 H

The vapours are cooled, and the condensate obtained is a mixture of for-
maldehyde, methanol and water. It is freed from excess methanol by dis-
tillation, and the resulting mixture is known as formalin (40 per cent,
formaldehyde, 8 per cent, methanol, 52 per cent, water).

(iii) By passing a mixture of methane and oxygen over certain catalysts,
e.g., molybdenum oxides:

CH 4 + O a — > H-CHO + H a O

(iv) By the oxidation of natural gas.

Formaldehyde is a colourless, pungent-smelling gas, b.p. —21°, extremely
soluble in water. The low boiling points of aldehydes (and ketones) com-
pared with those of alcohols are probably due to the inability of carbonyl
compounds to associate through hydrogen bonding. On the other hand,
since the carbonyl oxygen atom is capable of forming hydrogen bonds with
water, the solubility of aldehydes and ketones in water is to be expected for
the lower members (cf. alcohols).

Formaldehyde is a powerful disinfectant and antiseptic, and so is used
for preserving anatomical specimens. Its main uses axe in the manufacture
of dyes, the hardening of casein and gelatin, and for making plastics.

Formaldehyde undergoes many of the general
reactions of aldehydes, but differs in certain
respects. When treated with ammonia, it does
not form an aldehyde-ammonia, but gives instead
hexamethylenetetr amine :

6H-CHO + 4NH 3 — > (CHa) 6 N 4 + 6H 2 (80%)

The structure of hexamethylenetetramine ap-
pears to be a complicated ring compound. It is a
crystalline solid, and has been used in medicine
under the name of urotropine or aminoform as
treatment for gout and rheumatism.

ALDEHYDES AND KETONES 165

Formaldehyde, since it has no a-hydrogen atoms, readily undergoes the
Cannizzaro reaction (see p. 160).

Formaldehyde is very useful for methylating primary and secondary
amines, e.g., it converts ethylamine into ethylmethylamine:

C 2 H 5 -NH 2 + 2H-CHO — > C 2 H 6 -NH-CH S + H-C0 2 H

Polymers of formaldehyde, (i) In dilute aqueous solution, formaldehyde
is almost ioo per cent, hydrated to form methylene glycol (Bieber et al.,
1947). This is believed to be the reason for the stability of dilute formalde-
hyde solutions (see also p. 168) :

CH 2 + H 2 =± CH 2 (OH) 2

(ii) When a formaldehyde solution is evaporated to dryness, a white
crystalline solid, m.p. 121-123 , is obtained. This is known as para-
formaldehyde, (CH 2 0)„-H 2 0, and it appears to be a mixture of polymers, n
having values between 6 and 50. Paraformaldehyde reforms formaldehyde
when heated.

Formaldehyde cannot be separated from methanol (in formalin) by fractiona-
tion; pure aqueous formaldehyde may be obtained by refluxing para-
formaldehyde with water until solution is complete.

(iii) When a formaldehyde solution is treated with concentrated sulphuric
acid, polyoxymethylenes, (CH 2 0) n 'H 2 — n is greater than 100 — are formed.
Polyoxymethylenes are white solids, insoluble in water, and reform form-
aldehyde when heated.

(iv) When allowed to stand at room temperature, formaldehyde gas slowly

polymerises to a white solid, trioxymethylene (metaform-

CH 2 aldehyde, trioxan), (CH 2 0) 3 , m. p. 61-62 . This trimer is

(y \q soluble in water, and does not show any reducing properties.

il Hence it is believed to have a cyclic structure (in which

CH there is no free aldehyde group).
\q/ 2 Trioxan is very useful for generating formaldehyde since:
(a) it is an anhydrous form of formaldehyde; (b) the rate
of depolymerisation can be controlled; and (c) it is soluble in organic
solvents.

(v) Formaldehyde polymerises in the presence of weak alkalis, e.g.,
calcium hydroxide, to a mixture of sugars of formula C s Hj 2 0,„ which is
known asformose or a.-acrose.

Condensation reactions of formaldehyde. Formaldehyde can participate
in the " crossed " Cannizzaro reaction, and the nature of the final product
depends on the structure of the other aldehyde. Aldehydes with no a-
hydrogen atoms readily undergo the crossed Cannizzaro reaction; e.g.,
benzaldehyde forms benzyl alcohol:

C 6 H 5 -CHO + H-CHO + NaOH — > C e H 5 -CH 2 OH + H-C0 2 Na

Aldehydes with one a-hydrogen atom react as follows :

NaOH /CH 2 OH

RR'CH-CHO + H-CHO > RR'CC

M3HO

This p-hydroxyaldehyde in the presence of excess formaldehyde forms a sub-
stituted trimethylene glycol:

/CH 2 OH /CH a OH

+ H-CHO + NaOH — > RR'CX"
M2HO \CHjOH

l66 ORGANIC CHEMISTRY

Thus the first step in the above reaction is the replacement of the oe-hydrogen
atom by a hydroxymethyl group, -CHaOH, and the second step is the crossed
Cannizzaro reaction.

In a similar manner, aldehydes with two oe-hydrogen atoms are converted first
into the hydroxymethyl, then into the bishydroxymethyl, and finally into the
trishydroxymethyl compound :

NaOH /CHO HCHO

R-CH a -CHO + H-CHO y R-CH( ->

\CH 2 OH

/CHO H . CH0 /CH 2 OH

R-C(-CH 2 OH — > R-C^-CH a OH + H-CO^Na

\CH 2 OH Na0H \CH 2 OH

A special case is acetaldehyde, which has three a-hydrogen atoms. This
reaction is best carried out by adding powdered calcium oxide to a suspension of
paraformaldehyde in water containing acetaldehyde. Tetrakishydroxy methyl-
methane (tetramethylolmethane) or pentaerythritol is formed :

Ca(OH) a

CH 3 -CHO + 4 H-CHO > C(CH 2 OH) 4 + H-C0 2 Ca/ 2 (55-57%)

Pentaerythritol is important industrially since its tetra-nitrate is a powerful
explosive.

Structure of formaldehyde, Analysis and molecular-weight determina-
tions show that the molecular formula of formaldehyde is CH 2 0. Assuming
the quadrivalency of carbon, the bivalency of oxygen and the univalency
of hydrogen, only one structure is possible :

H

I
H— C=0

This structure agrees with all the known properties of formaldehyde, and
has been proved by both infra-red and ulta-violet spectra measurements.
Acetaldehyde (ethanal), CH 3 *CHO, is prepared industrially:

(i) By passing ethanol vapour over copper at 300 :

CH 3 -CH 2 OH — > CH 3 -CHO + H 2

(ii) By passing ethanol vapour mixed with air over a silver catalyst
at 250 :

C 2 H s OH + |0 2 — > CH 3 -CHO + H 2

(iii) From acetylene:

H SO

C 2 H 2 + H„0 -^-V CH 3 -CHO

1 *■ ' 2 HgSO, •*

(iv) By the hydrolysis of methyl vinyl ether {q.v.) :

CH 3 -OCH:CH 2 + H 2 — > CH 3 -CHO + CH 3 OH
(v) By the oxidation of natural gas.

Acetaldehyde is a colourless, pungent-smelling liquid, b.p. 21 , miscible
with water, ethanol and ether in all proportions. It is used in the prepara-
tion of acetic acid, ethanol, paraldehyde, rubber accelerators, phenolic
resins, synthetic drugs, etc. Acetaldehyde is about 58 per cent, hydrated
in aqueous solution to form ethylidene glycol (Bell et al., 1952; see also p.
168):

CH 3 -CHO + H 2 ^ CH 3 -CH(OH) 2

ALDEHYDES AND KETONES 167

Polymers of acetaldehyde. (i) When acetaldehyde is treated with a few
drops of concentrated sulphuric acid, a vigorous reaction takes place and
the trimer paraldehyde, (CH 3 -CHO) 3 , is formed. This is pleasant-smelling
liquid, b.p. 128 , and is used in medicine as an hypnotic. When paralde-
hyde is distilled with dilute sulphuric acid, acetaldehyde is regenerated.
Paraldehyde has no reducing properties, and its structure is believed
to be (I):

1

CH

o/\o

CHo-CH-OCH-CHo

A A
1 1

CH 3 -CH-OCH-CH 3

(II)

1 1

CH.-HCv .CH-CH 3

(I)

(ii) When acetaldehyde is treated with a few drops of concentrated sul-
phuric acid at o°, the tetramer metaldehyde, (CH 3 -CHO) 4 , is formed. This
is a white solid, m.p. 246°, and regenerates acetaldehyde when distilled with
dilute sulphuric acid. Its structure may be (II).

Structure of acetaldehyde. Analysis and molecular- weight determinations
show that the molecular formula of acetaldehyde is C 2 H 4 0. Ethanol may
be oxidised to acetaldehyde, which, in turn, may be oxidised to acetic acid.
Both ethanol and acetic acid (g.v.) contain a methyl group, and it is therefore
reasonable to suppose that this methyl group remains intact during the
oxidation, and is therefore present in acetaldehyde:

CH 3 — CH 2 OH — !^L> CH 3 — CHO — ^-> CH 3 — C0 2 H

Phosphorus pentachloride reacts with acetaldehyde to form ethylidene
chloride, C 2 H 4 C1 2 , and no hydrogen chloride is evolved in the reaction.
This implies that there is no hydroxyl group present in acetaldehyde (cf.
p. 136), and since two univalent chlorine atoms have replaced one bivalent
oxygen atom, the inference is that there is a carbonyl group, ^>C=0,
present.

^°
Thus the structure of acetaldehyde is CH 3 — C'

\H

Study of the dipole moments of aldhydes and ketones, however, has presented
some difficulty in elucidating the structure of carbonyl compounds. It
has been suggested that the values of the dipole moments are larger than
can be accounted for by the inductive effect of the oxygen atom, and
it has therefore been proposed that aldehydes and ketones are resonance
hybrids :

>C=0 «-> >C— 6

Chloral (trichloroacetaldehyde), CCL/CHO, is prepared industrially by the
chlorination of ethanol. Chlorine is passed into cooled ethanol, and then
at 60 °, until no further absorption of chlorine takes place. The final product
is chloral alcoholate, CCl 3 'CH(OH)-(OC 2 H 5 ), which separates out as a crys-
talline solid which, on distillation with concentrated sulphuric acid, gives
chloral.

l68 ORGANIC CHEMISTRY

The mechanism of the reaction is obscure. The following set of equations have
been proposed (Fritsch, 1897):

CI /\Jli {• xj rtjj C w OH

CH 3 -CH 2 OH — ^-» CH 3 -CH\ * > [CH 3 -CH(OH)-(OC a H 6 )] -^ — =►

\C1

[CH 3 -CH(OC a H 5 ) a ] -%. CH 2 Cl-CH(OC 2 H 5 ) 2 -%. CHCl 2 -CH(OC a H„) 2

ci,

CH 2 Cl-CHCl-(OC 2 H 5 ) -%► CHCl 2 -CHCl-(OC 2 H 5 ) _^!%

CHCl a -CH(OH)-(OC 2 H 5 ) -^->CCl 3 -CH(OH)-(OC 2 H 6 ) H ' S ° 4 > CCl 3 -CHO

Chloral is a colourless, oily, pungent-smelling liquid, b.p. 98 , soluble in
water, ethanol and ether. When heated with concentrated potassium
hydroxide, it yields pure chloroform :

CClg-CHO + KOH — > CHCI3 + H-C0 2 K

Chloral is oxidised by concentrated nitric acid to trichloroacetic acid (I), and
is reduced by aluminium ethoxide to trichlorethanol (II).

CC1 3 C0 2 H <ff^i- CCI 3 -CHO A ' ( ° C ' H ° )a > CC1 3 -CH 2 0H
(I) (II)

Chloral undergoes the usual reactions of an aldehyde, but its behaviour
towards water and ethanol is unusual. When chloral is treated with water
or ethanol, combination takes place with the evolution of heat, and a crys-
talline solid is formed, chloral hydrate, m.p. 57 , or chloral alcoholate, m.p.
46 , respectively. These compounds are stable, and the water or ethanol
can only be removed by treatment with concentrated sulphuric acid. It
therefore seems likely that in chloral hydrate the water is present as water
of constitution, i.e., the structure of chloral hydrate is CCl s - CH(OH) a ;
similarly, the structure of chloral alcoholate is CCl 3 *CH(OH)-(OC 2 H 5 ). The
stability of these compounds is remarkable in view of the fact that the group
>C(OH) 2 in other compounds very readily eliminates water.

Bell et al. (1952) showed the forward and backward reactions in the hydration
of acetaldehyde are subject to general acid and general base catalysis {i.e., the
catalytic effect is due to all of the acids or bases present). Bell proposed the
following mechanism for general acid catalysis :

H H

fast v I „ . H,Q; slow \

CH 3 -C=0 + HA *- CH 3 -C=0-H— A

H

I f as J

CHj-COH + A-^=^CH 3 -CH(OH) 2 + HA
+ OH 2

Since the reaction is reversible, the conversion of the hydrate into the aldehyde
will involve the loss of a water molecule. If, however, the carbon atom of the

C — OH 2 group has a positive charge, then fission of this bond is strongly opposed,
and so the hydrate will be stabilised. In chloral, owing to the strong inductive

ALDEHYDES AND KETONES 169

effect of the chlorine atoms, the central carbon atom acquires a positive charge
(III). On the other hand, Davies (1940) has shown from infrared studies

•••' H \
C WH Cl\ /Q ~j

Cl-^C— OH Cl-^cf-H CH 3 ->-CH(

ClX \ H C / \> X ° H

''•w

(III) (IV) (V)

that chloral hydrate (and related compounds) contain hydrogen bonds, and he
suggests that this accounts for the stability of the hydrate (IV) . Acetaldehyde
hydrate (V) will be unstable because of the +1 effect of the methyl group, and
acetone hydrate still more unstable because of the +1 effect of two methyl
groups. In both of these hydrates the central carbon atom acquires a small

negative charge and so facilitates fission of the C — OH 2 bond. In formaldehyde
hydrate, there are no methyl groups, and so this compound would be expected
to be more stable than the corresponding acetaldehyde hydrate.

Propionaldehyde, CH 3 *CH 2 'CHO, b.p. 49°, and isobutyraldehyde,
(CH 3 ) 2 CH*CHO, b.p. 61 °, are prepared industrially by the isomerisation of
propylene oxide and isobutylene oxide, respectively, in the presence of steam and
an alumina-silica catalyst.

CH 3 -CH— fcHjs — > CH 3 -CH 2 -CHO

(CH a ) 2 (<— ^CHj — > (CH 3 ) 2 CH-CHO

Propionaldehyde is also prepared industrially by the oxo process by passing a
compressed mixture of ethylene, carbon monoxide and hydrogen over the catalyst
at 125-145 (p. 127).

Many higher aldehydes occur in nature, e.g., w-octaldehyde, C 8 H le O, and
w-nonaldehyde, C 9 H ls O, in rose-oil; w-decaldehyde, C 10 H 20 0, in rose-oil and
orange-peel oil.

Acetone {dimethyl ketone, profan-2-one), CH 3 # COCH 3 , is prepared in-
dustrially:

(i) By passing j'sopropanol vapour over copper at 300 :

CH 3 -CH(OH)-CH 3 — > CH 3 -COCH 3 + H 2

(ii) By passing acetic acid vapour over calcium oxide or manganous
oxide at 300-400° :

2CH 3 -C0 2 H — > CH 3 -COCH 3 + C0 2 + H 2

(iii) By passing ethanol vapour mixed with steam over zinc chromite
as catalyst, at 500°:

2C 2 H 6 OH + H 2 — > CH 3 -COCH 3 + C0 2 + 4H 2

A newer process is to pass a mixture of acetylene and steam over heated
pure zinc oxide as catalyst. Another new process is to heat a mixture of
propylene and steam under pressure in the presence of a suitable catalyst
(p. 137). Acetone is also manufactured by the oxidation of natural gas.

(iv) By fermentation (see w-butanol, p. 137), but this method is being
replaced by the above synthetic methods.

Acetone is a colourless, pleasant-smelling liquid, b.p. 56°, miscible with
water, ethanol and ether in all proportions. Pure acetone is best prepared

170 ORGANIC CHEMISTRY

by saturating acetone with sodium iodide at 25-30°. The solution is
decanted off from the excess solid, cooled to — io°, and the precipitate,
Nal'sCHg-COCHg, is warmed to 30°, pure dry acetone thereby being pro-
duced. Acetone is used as a solvent for acetylene, cellulose acetate and
nitrate, celluloid, lacquers, etc., and for the preparation of keten, sulphonal,
etc. Acetone is hydrated to only a very slight extent in water :

(CH 3 ) 2 CO + H 2 -< (CH 3 ) 2 C(OH) 2

The fact that acetone does form a hydrate is shown by the isolation of acetone
containing 18 when dissolved in water enriched with ls O (Urey et al., 1938) :

*

Me 2 CO + H 2 6 ;?=± Me 2 C( ^ Me 2 CO + H 2

\OH

Ketones do not polymerise, but readily undergo condensation reactions.
Acetone readily forms mesityl oxide, phorone and diacetone alcohol (p. 228).
but in addition to these condensations, acetone forms mesitylene when
distilled with concentrated sulphuric acid:

CH,

3CH,-CO-CH, — > CH || | CHs + 3H 2 (25%)

Structure of acetone. Analysis and molecular-weight determinations
show that the molecular formula of acetone is C 3 H 6 0. Acetone reacts with
phosphorus pentachloride to form isopropylidene chloride, C 3 H 8 C1 2 , and
no hydrogen chloride is evolved. This indicates that there is no hydroxyl
group present, and since a bivalent oxygen atom has been replaced by two
univalent chlorine atoms, this implies that a carbonyl group is present.
Assuming the quadri valency of carbon, the bivalency of oxygen and the
univalency of hydrogen, there are two structures possible for acetone :

(I) CH 3 -CH 2 -CHO CH 3 -CO-CH 3 (II)

(I) contains the aldehyde group, but since acetone does not behave like an
aldehyde, (II) must be its structure. This is confirmed by all the known
reactions of acetone (cf. acetaldehyde, for the structure of the carbonyl
group).

Ethyl methyl ketone, butan-2-one, CH 3 *COCH 2 *CH 3 , is prepared industrially :

(i) By passing sec.-butanol vapour over copper at 300° :

CH 3 -CH(OH)-CH 2 -CH 3 > CH 3 -COCH 2 -CH 3 + H 2

(ii) By the oxidation of sec.-butanol with air, using a silver catalyst
at 250° :

CH 3 -CH(OH)-CH 3 -CH 3 + £0 2 > CH 3 -COCH 2 -CH 3 + H 2

Butanone is a pleasant-smelling liquid, b.p. 8o°. It is widely used as a solvent
for vinyl resins, synthetic rubber, etc.

Methyl tsopropyl ketone, 3-methylbutan-2-one, CH 3 *COCH(CH 3 ) 2 , b.p. 94°,
may be prepared in several ways, but the best method is by treating tert. -pentanol
with bromine and hydrolysing the dibromo-derivative, whereupon a molecular
rearrangement takes place (see pinacone, below) :

(CH 3 ) 2 C(OH)-CH 2 -CH 3 Bf ' - > - (CH 3 ) 2 CBr-CHBr-CH 3 ■ H '° ^

[(CH 3 ) 2 C(OH)-CH(OH)-CH 3 ] ^-> (CH 3 ) 2 CH-COCH 3

ALDEHYDES AND KETONES 171

tert.-Butyl methyl ketone, pinacolone, pinacone, CH 3 *COC(CH 3 ) 3 , may be
prepared :

(i) By distilling pinacol or pinacol hydrate with sulphuric acid :

(CH 3 ) 2 C(OH)-C(OH) (CH 3 ) 2 -^% CH 3 -COC(CH 3 ) 3 (65-72%)

(ii) By passing pinacol dissolved in dioxan over a heated catalyst
of silica-gel impregnated with phosphoric acid, a 94 per cent, yield of
pinacolone being obtained (Emerson, 1947).

This conversion of pinacol into pinacone is an example of the 1,2-shift
(p. 101), and is known as the pinacoV-pinacone rearrangement.

Pinacolone is a colourless liquid, b.p. 119 , with a camphor-like odour.
It is oxidised by alkaline sodium hypobromite to trimethylacetic acid :

CH 3 -CO-C(CH 3 ) 3 -^^V (CH 3 ) 3 OC0 2 H (71-74%)

Many higher ketones occur in nature, e.g., heptan-2-one, CH 3 'CO"C 5 H u , in
clove-oil, and undecan-2-one, CH s *CO"C 8 H 19 , in oil of rue.

The pinacol-pinacone rearrangement is general for 1 : 2-glycols under acid
conditions ; the migrating group may be alkyl or aryl. The detailed mechanism
of this rearrangement is not certain, but it is generally accepted that the first
stage is the addition of a proton to one of the hydroxy-groups, followed by loss
of this as water. This leaves a carbonium ion, and it is the nature of this ion
and the steps leading to the product that are uncertain. According to Bunton
et al. (1959), the mechanism for the rearrangement of pinacol itself is:

Me 2 C— CMe 2 + H+ ^=± Me 2 C CMe 2 ^=i

II I vfl

OH OH OH + OH 2

(I)
Me

l>+ +
H a O + Me— C— CMe 2 > Me— C— CMe 3 > H+ + MeCOCMe 3

(II) OH H-A)

Kinetic measurements have shown that the conversion of (I) into (II) is the rate-
determining step.

Whitmore et al. (1939), using a related pinacolic deamination, showed that
when an optically active amine containing one asymmetric centre was used, this
asymmetric centre was predominantly inverted in the product (cf. Walden
inversion, p. 413; Ph = C 6 H 5 ) :

Ph Ph

4HN0„
CHMe > Ph— C CHMe -

II I cl

OH NH 2 OH v N 2

Ph Ph

Z— CHMe > H + + PhCOCI

N 2 + Ph— C— CHMe > H + + PhCOCHMe

OH

These results mean that the classical carbonium ion, which is flat, if formed at
all, must rearrange so fast that racemisation cannot occur (see p. 412). On the
other hand, the inversion may be explained by a synchronous migration of a
phenyl group and breaking of the C — N bond. In this case, the intermediate

172 ORGANIC CHEMISTRY

could be the phenonium ion, i.e., a bridged ion in which the configuration of the
asymmetric carbon atom is held in the inverted position :

/% /\ /%

V 8 .V

Ph— C CHMe > Ph— C CHMe >- Ph— C— CHMe

OH N 2 + OH OH

A pair of electrons of the benzene ring is used to form the bridge, and the rest of
the benzene molecule is represented as a pentadienyl cation (see p. 516). The
formation of this phenonium ion is supported by other experimental work, but
whether a bridged ion is formed when an alkyl group migrates is very doubtful.
When dealing with unsymmetrical pinacols, there are several other points to
be considered, e.g., which hydroxyl group is removed. Since the OH group is
removed as water, any polar effect in the molecule which weakens one C — O
bond more than the other will facilitate the release of this OH group. Consider
the two groups phenyl and methyl. The phenyl group has a powerful con-
jugative effect, whereas the methyl group has a weak inductive effect in com-
parison :

rv 1

<^>C C^-Me

W)H OH

Therefore the OH group lost should be the one on the phenyl side, and conse-
quently it will be the methyl group that migrates. This is borne out in practice,

eg-,

H,so 4
Ph 2 C(OH)'C(OH)Me 2 — > Ph 2 MeOCOMe

Another problem is that of " migration aptitude ", i.e., which of the two
groups migrates when the groups are different. A great deal of work has been
done with pinacols of the type (Ar = aryl) ArAr'C(OH)C(OH)ArAr', and it has
been shown that the aryl group with the greater electron-releasing property is
the one that migrates (Bachmann et al., 1932-1934). This is understandable
on the basis of the formation of an intermediate phenonium ion. Any polar
factor that helps to release a pair of electrons from a benzene ring will therefore
facilitate the formation of the bridged ion with this benzene ring. It has also
been found that migration aptitudes depend on steric factors.

QUESTIONS

1. Write out the structures and names of the isomeric aldehydes and ketones having
the molecular formula C B H 10 O.

2. What are all the possible oxidation products of: (a) »-pentanol, (6) hexan-2-ol,
(c) tert.-hvAa.uoV.

3. Give an analytical table to show how you could distinguish between the following
alcohols: methanol, ethanol, t'sopropanol and pentan-3-ol.

4. Starting with ethanol, how would you synthesise : (a) isobutyraldehyde, (6) crotyl
alcohol?

5. Show, by means of equations, how you would convert acetylene into isoprene.

6. Describe the industrial methods of preparation of: — (o) formaldehyde, (6) acet-
aldehyde, (c) chloral, (d) acetone, (e) butanone.

7. Give an account of the evidence for the structure of (a) propionaldehyde, (6)
butanone.

8. Name the compounds and state the conditions under which they are formed when
acetaldehyde or acetone, respectively, is treated with: — (a) "nascent" hydrogen, (6)
molecular hydrogen, (c) aluminium isopropoxide, (d) aluminium ethoxide, («) sodium
ethoxide, (/) NH 3 , (g) NaHSO s , (h) HCN, (i) N a H 4 , (;) RSH, (A) Br„ (I) PBr 6 , {m) SeO„
(n) NaOH, (0) HC1, (p) ammoniacal AgNO s , (?) CjHj-NH,, (»■) HN 8 , (s) ROH, (t) HN0 2 ,
(m) SO a Cl 2 , \v) Fehling's solution, (w) H-CHO.

9. Define and give examples of: — (a) the Oppenauer oxidation, (6) the Rosenmund
reduction, (c) ozonolysis, [d) Stephen's aldehyde synthesis, {e) the Meerwein-Ponndorf-

ALDEHYDES AND KETONES 173

Verley reduction, (/) the Wolfl-Kishner reduction, (g) the Clemmensen reduction, (h)
Girard's reagent, {«} the aldol condensation, (j) the Cannizzaro reaction, (k) the Claisen
condensation, (/) the Claisen reaction, (m) the Knoevenagel reaction, (n) the Perkin
reaction, (o) disproportionation, (p) the Tischenko reaction, (q) the haloform reaction,
(*•) the pinacol-pinacolone rearrangement, (s) the Schmidt reaction, (/) polymerisation,
(it) condensation, (») the crossed Cannizzaro reaction, (w) Darzens glycidic ester
condensation.

READING REFERENCES

Bevington, The Polymerisation of Aldehydes, Quart. Reviews (Chem. Soc), 1952, 6, 141.
Seelye and Turney, The Iodoform Reaction, /. Chem. Educ, 1959, 36, 572.

Organic Reactions (Wiley).

(i) Vol. I. (1942), Ch. 7. The Clemmensen Reduction,

(ii) Vol. II. (1944), Ch. 3. The Cannizzaro Reaction,

(iii) Vol. II. (1944), Ch. 5. Reduction with Aluminium Alkoxides.

(iv) Vol. IV. (1948), Ch. 7. The Rosenmund Reduction of Acid Chlorides to

Aldehydes,

(v) Vol. IV. (1948), Ch. 8. The Wolfi-Kishner Reduction,

(vi) Vol. V. (1949), Ch. 7. The Leuckart Reaction,

(vii) Vol. V. (1949), Ch. 8, Selenium Dioxide Oxidation,

(viii) Vol. V. (1949), Ch. 10. The Darzens Glycidic Ester Condensation,

(ix) Vol. VI. (1951), Ch. 5. The Oppenauer Oxidation,

(x) Vol. III. (1946), Ch. 8. The Schmidt Reaction.

(xi) Vol. VII. (1953), Ch. 6. The Nitrosation of Aliphatic Carbon Atoms,

(xii) Vol. VIII. (1954), Ch. 5. The Synthesis of Aldehydes from Carboxylic Acids.

Whitmore, The Common Basis of Intramolecular Rearrangements, /. Amer. Chem.

Soc, 1932, 54, 3274.
Davies, An Infra-Red Study of Chloral Hydrate and Related Compounds, Trans.

Faraday Soc, 1940, 63, 333, 11 14.
Walsh, Remarks on the Strengths of Bonds, ibid., 1947, 4^> 60, 158, 342.
Angyal et al.. The Preparation of Aliphatic Aldehydes, J.C.S., 1953. 1737.
Ballester, Mechanisms of the Darzens and Related Condensations, Chem. Reviews, 1955,

55. 283.
Bethell and Gold, The Structure of Carbonium Ions, Quart. Reviews (Chem. Soc), 1958,

12, 173-
Collins, The Pinacol Rearrangement, Quart. Reviews (Chem. Soc), i960, 14, 357.
Moulton et al., Mechanism of the Meerwein-Ponndorf- Verley Reduction, /. Org. Chem.,

1961, 26, 290.
Smith and Antoniades, The Interplay of Steric and Electronic Factors Affecting Geo-
metrical Isomerism of Diaryl Ketimine Derivatives, Tetrahedron, i960, 9, 210.

CHAPTER IX

FATTY ACIDS

The fatty acid series was so named because some of the higher members,
particularly palmitic and stearic acids, occur in natural fats. The general
formula of the fatty acids is C„H 2 „0 2 . As, however, their functional group
is the carboxyl group, — C0 2 H, they are more conveniently expressed as
CBH^-^COisH or R-C0 2 H, since these show the nature of the functional
group. Furthermore, since the fatty acids contain only one carboxyl
group, they are also known as the saturated monocarboxylic acids.

Only the hydrogen atom of the carboxyl group is replaceable by a metal;
therefore the fatty acids are monobasic. The structure of the carboxyl

//°
group is written as — C^ (or — COOH), but, as we shall see later, it does

\OH

not represent accurately the behaviour of the carboxyl group.

Nomenclature. The fatty acids are commonly known by the trivial
names, which have been derived from the source of the particular acid;
e.g., formic acid, H-C0 2 H, was so named because it was first obtained by
the distillation of ants; the Latin word for ant is formica. Acetic acid,
CH 3 -CO a H, is the chief constituent of vinegar, the Latin word for which
is acetum; etc. (see below).

Another system of nomenclature considers the fatty acids, except formic
acid, as derivatives of acetic acid, e.g.,

CH 3 -CH 2 -C0 2 H methylacetic acid

(CH 3 ) 3 OC0 2 H trimethylacetic acid

(CH 3 ) 2 CH-CH 2 -C0 2 H t'sopropylacetic acid

In the above two systems the positions of substituents in the chain are
indicated by Greek letters, the a-carbon atom being the one joined to the
carboxyl group, e.g.,

CH 3 -CH(OH)-CH a -CH 2 -CO a H y-hydroxyvaleric acid

CH 3 -CHC1-CH-C0 2 H

I p-chloro-a-methylbutyric acid

CH 3

According to the I.U.P.A.C. system of nomenclature, the suffix of the
mono-carboxylic acids is -oic, which is added to the name of the alk^ne corre-
sponding to the longest carbon chain containing the carboxyl group, e.g.,

H>C0 2 H methanoic acid

CHg-CHyCOaH propanoic acid

The positions of side-chains (or substituents) are indicated by numbers,
the carboxyl group always being given number i (see p. 63), e.g.,

CH 3 'CH-CH-CH 2 -C0 2 H 3 : 4-dimethylpentanoic acid

CH 3 CH 3

174

FATTY ACIDS 175

Alternatively, the caxboxyl group is regarded as a substituent, and is
denoted by the suffix carboxylic acid, e.g.,

4 3 2 1

CH 3 -CH 2 -CH*CH 2 *C0 2 H 2-methylbutane-i-carboxylic acid

CH S

General methods of preparation, i. Oxidation of alcohols, aldehydes
or ketones with acid dichromate yields acids:

R-CH 2 OH -%■ R-CHO -^> R-C0 2 H {g--v.g.)

R 2 CHOH -^> R 2 CO -^> R-C0 2 H + R'-C0 2 H (g.)

In certain cases the ester, and not the acid, is obtained by the oxidation
of alcohols (seep. 190).

2. A very good synthetic method is the hydrolysis of cyanides with acid
or alkali:

R. C =N^>

/OH"

R_C C

^R-Cf° J*V

^NH 2

R-C0 2 H + NH 3 (g.~v.g.)

The amide, R # CO'NH 2 , may be isolated if suitable precautions are taken
(p. 203).

3. By the reaction between a Grignard reagent and carbon dioxide:

RMgX + C0 2 — > R-C0 2 MgX -^> R«C0 2 H

4. Many fatty acids may be conveniently synthesised from alkyl halides
and ethyl malonate or acetoacetic ester. These syntheses will be discussed
in detail later (pp. 229, 233).

5. A number of higher fatty acids may be obtained by the hydrolysis of
natural fats, but it is usually difficult to obtain the acids pure from these
sources. On the other hand, higher fatty acids may be prepared by the
electrolysis of a methanolic solution of a mixture of a monocarboxylic acid
and a half-ester of a dicarboxylic acid (Linstead et al., 1950; cf. p. 52) :

R-CO a H + C0 2 H-(CH 2 )„-C0 2 CH 3 — > H 2 + 2 C0 2 + R-(CH 2 )„-CO a CH 3

6. The sodium salt of a fatty acid may be obtained by heating a sodium
alkoxide with carbon monoxide under pressure :

RONa + CO y R-CO a Na

It has also been found that the acid may be obtained by heating the alcohol
with carbon monoxide at 125-180°, under a pressure of 500 atmospheres, in the
presence of a catalyst of boron trifluoride and a little water:

ROH + CO — >■ R-CO a H

The sodium salts of acetic and propionic acids are formed by the interaction of
methylsodium or ethylsodium, respectively, and carbon dioxide, e.g.,

C a H 6 Na + CO a > C 2 H 6 -C0 2 Na

A recent method of preparing fatty acids is by the catalytic oxidation
of long-chain hydrocarbons. The hydrocarbons are obtained from the

I76 ORGANIC CHEMISTRY

wax fraction of the Fischer-Tropsch synthesis of hydrocarbons (p. 60),
which are oxidised by the passage of air at 120 in the presence of manganous
stearate as catalyst; or by oxidising with air in aluminium vessels, the
aluminium acting as catalyst. A mixture of fatty acids is obtained, and
this has been used for making fats. The actual composition of the mixture
of acids has not yet been ascertained.

Another recent method for manufacturing fatty acids is to heat an olefin
with carbon monoxide and steam under pressure at 300-400° in the presence
of a catalyst, e.g., phosphoric acid:

CH 2 =CH 2 + CO + H 2 — > CH 3 -CH 2 -C0 2 H
CH 3 -CH=CH 2 + CO + H 2 — >■ (CH 3 ) 2 CH-C0 2 H

General properties of the fatty acids. The first three fatty acids are
colourless, pungent-smelling liquids; the acids from butyric, C 4 H g 2 , to
nonoic, C 9 H 18 2 , are oils which smell like goats' butter; and those higher
than decoic acid, C 10 H 20 O 2 , are odourless solids.

The lower members are far less volatile than is to be expected from their

molecular weight. Nash et al. (1957) have reported that formic, acetic,

propionic, and w-butyric acid exist as dimers in aqueous solution. The

existence of these dimers can be explained by hydrogen bonding, and electron

diffraction studies (Pauling, 1934) have shown

r ^ \ r T? t ^ iat an eight-membered ring is present. On the

R \ n tt ryy other hand, crystalline formic acid (Holtzberg

O R— O ei a i ^ jgg^) an( j acetic acid (Jones et al., 1958)

exist as infinite chains in the extended form, whereas the long-chain fatty
acids (von Sydow, 1956) consist of cyclic dimers.

The melting points of the w-fatty acids show alternation or oscillation
from one member to the next, the melting point of an " even " acid being
higher than that of the " odd " acid immediately below and above it
in the series (see the physical constants of the individuals). A number
of homologous series follow this oscillation or " saw-tooth " rule; some
authors believe this to be connected with the zig-zag nature of the carbon
chain.

The first four members are very soluble in water, and the solubility
decreases as the molecular weight increases.

General reactions of the fatty acids. 1. The fatty acids are acted upon
by the strongly electropositive metals with the liberation of hydrogen and
formation of a salt :

R-C0 2 H + Na — > R-CO a Na + \H Z
Salts are also formed by the reaction between an acid and an alkali:
R-C0 2 H + NaOH — > R-C0 2 Na + H 2

2. Fatty acids react with alcohols to form esters:

R-C0 2 H + R'OH ^ R-C0 2 R' + H 2

3. Phosphorus trichloride, phosphorus pentachloride or thionyl chloride
act upon the fatty acids to form acid chlorides:

3 R-CO a H + PCI3 -^ 3 R-C0C1 + H3PO3
R-C0 2 H + PC1 5 — > R-COC1 + HC1 + P0C1 3
R-C0 2 H + SOCl 2 — > R-COC1 + HC1 + S0 2

FATTY ACIDS 177

The acid series may be " stepped up " via the acid chloride, but since
the process involves many steps, the yield of the higher acid homologue
is usually only f.-f.g.:

(a) R-C0 2 H -^V R-COC1 ■ "' > R-CHO "' > R-CH 2 OH -!%

x ' * catalyst catalyst

R-CH 2 Br -^V R-CH 2 -CN -^> R-CH 2 -C0 2 H

(b) R-C0 2 H — -> R-COC1 -^V R-CO-CN — ->-

_ Clemmensen „ ,-,,.» ,,.—. •*■*■

R-CO-C0 2 H > R-CH 2 -C0 2 H

reduction

Although method (b) involves fewer steps than (a), the final yield of acid
by route (a) is higher than that by (b), since the yields in each step of (b)
are generally low.

4. When the ammonium salts of the fatty acids are strongly heated, the
acid amide is formed by the elimination of water:

R-C0 2 NH 4 — > R-CO-NH 2 + H 2

5. When the anhydrous sodium salt of a fatty acid is heated with soda-
lime, a paraffin and other products are formed (p. 50).

R-C0 2 Na + NaOH(CaO) — > RH + Na 2 C0 3

The mechanism of this decarboxylation is uncertain, but there is much
evidence to show that salts decompose by an S B i reaction:

R-k< ->r: + <

r: + H — >- R— H
The decarboxylation oifree acids may be:

/O— H + *~\ V°- + -

R— C^ — >■ H + R-^C^ — > H + R: + CO s — > R~ H + CO,

The fatty acids may be " stepped down " by heating the calcium salt of
the acid with calcium acetate (see p. 147), and the methyl ketone produced
is oxidised with acid dichromate :

R-CH 2 -C0 2 Ca/ 2 + CH 3 -C0 2 -Ca/ 2 — >• R-CH 2 -CO-CH 3 -^>

R-C0 2 H + CH 3 — C0 2 H

It should be noted that the second step is based on the general rule that
when an unsymmetrical ketone is oxidised, the carbonyl group remains
chiefly with the smaller alkyl group.

Since very few fatty acids containing an odd number of carbon atoms
occur naturally, this method of descending the series affords a useful means
of obtaining " odd " acids from "even ".

When a mixture of the calcium salt (or any of the others mentioned above)
and calcium formate is heated, an aldehyde is obtained:

R-C0 2 Ca/ 2 + H-C0 2 Ca/ 2 — > R-CHO + CaC0 3

I78 ORGANIC CHEMISTRY

Metallic salts, particularly the silver salts, are converted by bromine into
the alkyl bromide. This reaction may also be used to " step down " the acid
series (p. 105).

6. When a concentrated aqueous solution of the sodium or potassium
salt of a fatty acid is electrolysed, a paraffin is obtained:

2R-C0 4 K + 2H 2 — > R—R + 2CO a + 2KOH + H 2

7. Fatty acids react slowly with chlorine or bromine in the cold,
but at higher temperatures and in the presence of a small amount of
phosphorus, reaction proceeds smoothly to give a-halogeno-acids (see also
p. 210); e.g.,

R-CH 2 -C0 2 H -^> R-CHBr-C0 2 H -^V R.CBr 2 -C0 2 H

8. All the fatty acids, except formic acid, are extremely resistant to'
oxidation, but prolonged heating with oxidising agents ultimately produces
carbon dioxide and water.

9. All the fatty acids are resistant to reduction, but prolonged heating
under pressure with concentrated hydriodic acid and a small amount of red
phosphorus produces a paraffin. Paraffins are also produced when a fatty
acid is heated with hydrogen at high temperature and under pressure, in the
presence of a nickel catalyst.

Ni

R-C0 2 H + 3H 2 > R-CH 3 + 2H 2

If the hydrogenation is carried out in the presence of a ruthenium or
copper-chromium oxide catalyst, carboxylic acids are converted into
primary alcohols (Guyer et al., 1955). Lithium aluminium hydride also
reduces acids to alcohols.

10. The fatty acids undergo the Schmidt reaction (p. 156) to form a primary
amine :

H SO

R-C0 2 H + HN 3 ' •> R-NH 2 + C0 2 + N 2

Schmidt's reaction with acids is a modification of the Curtius reaction (p. 209).
Since sulphuric acid accelerates the decomposition of the acid azide, this implies
that the conjugate acid of the azide decomposes more readily than the azide
itself. The following mechanism has been proposed [cf. p. r56) :

.CR

OH OH

H + -R—J-+ HN ». W-J—OTI -H.0

->

R— C " > R— C+ ""*> R— C— OH

I I I +

OH OH H— N— N=N

O R

R— C -n. 0=C— N— H H >

NH-NSN
0=C=NR H '°> RNH 2 + CO a

The rearrangement (r,2-shift) has been shown to be intramolecular, e.g., Kenyon
et al. (1939) showed retention of optical activity when a-phenylpropionic acid
underwent the Schmidt reaction :

C 6 H B -CH(CH 3 )-C0 2 H >■ C,H 6 -CH(CH 3 )-NH 2

FATTY ACIDS 179

This mechanism has been proposed for those acids which require warming to
complete the reaction. Newman et al. (1948) showed that mesitoic acid under-
went the Schmidt reaction at room temperature, whereas benzoic acid required
heating at 25°. Mesitoic acid is a sterically hindered acid, and for such acids the
reaction is believed to proceed through the acylium cation :

, Me
Me^CO^i^V
Me

_Me
> Me/ SnH.

NH— N=N me

* Formic acid (methanoic acid), H-C0 2 H, is prepared industrially by heating
sodium hydroxide with carbon monoxide at 210 , and at a pressure of 6-10
atmospheres :

NaOH + CO — > H-C0 2 Na

An aqueous solution of formic acid is obtained by distilling the sodium salt
with dilute sulphuric acid:

2H-C0 2 Na + H 2 S0 4 — > 2H-C0 2 H + Na 2 S0 4

Anhydrous formic acid is obtained by warming the sodium salt with con-
centrated sulphuric acid to which has been added some anhydrous formic
acid. Concentrated sulphuric acid dehydrates formic acid to carbon
monoxide :

H-C0 2 H ^> CO

* H,S0 4

On the other hand, concentrated sulphuric acid diluted with formic acid
shows very little tendency to dehydrate formic acid.

Formic acid may be prepared in the laboratory in several ways. One
method is to pass methanol vapour or formaldehyde mixed with air over a
platinum-black catalyst:

CH.OH + 2 — ^-> H-C0 2 H + H 2

I3WX1 -p \j 2 f "W|i

i0 2 -

H-CHO + i0 2 -^-> H-C0 2 H

Another method, which is mainly of academic interest, is the hydrolysis
of hydrogen cyanide :

HCN + 2H a -^> H-C0 2 H + NH 3

Carbon dioxide, when passed into an ethereal solution of lithium borohydride
at o", is reduced to formic acid (Burr et al., 1950) :

2C0 2 + 2LiBH 4 > 2H-C0 2 Li + B 2 H 6 (69-88%)

The most convenient laboratory preparation of formic acid is to heat
glycerol with oxalic acid at 100-110°. Glyceryl monoxalate is produced,
and decomposes into glyceryl monoformate (monoformin) and carbon

i8o
dioxide

ORGANIC CHEMISTRY

When the evolution of carbon dioxide ceases, more oxalic acid
is added, whereupon formic acid is produced:

CH 2 OH v

I COOH

CHOH + | — > C

COOH

ch 9 oh c:

CH 2 0-CO-COOH
HOH + H 2

H 2 OH

CH.O-CO-H

CO a + CHOH
CH,OH

(COOH),

CH 2 0-CO-COOH
HOH + H-C0 2 H

H„OH

The distillate contains formic acid and water. The aqueous formic acid
solution cannot be fractionated to give anhydrous formic acid because the
boiling point of the acid is 100-5°. The procedure adopted is to neutralise
the aqueous acid solution with lead carbonate, and concentrate the solution
until lead formate crystallises out. The precipitate is then recrystallised,
dried, and heated at ioo° in a current of hydrogen sulphide :

(H-CO^Pb + H 2 S — > 2H-C0 2 H + PbS

The anhydrous formic acid which distils over contains a small amount of
hydrogen sulphide, and may be freed from the latter by adding some dry
lead formate and redistilling.

The above procedure for obtaining the anhydrous acid from its aqueous
solution can only be used for volatile acids.

Formic acid is a pungent corrosive liquid, m.p. 8-4°, b.p. 100-5°, miscible
in all proportions with water, ethanol and ether. It forms salts which,
except for the lead and silver salts, are readily soluble in water (see p. 176).
Formic acid is a stronger acid than any of its homologues (see Table III).

TABLE III

Acid

Dissociation Constant,
A (at 25°)

Formic
Acetic
Propionic .
M-Butyric .
isoButyric
n- Valeric .
isovaleric .
Trimethylacetic
Caproic
w-Heptoic

2-4 X 10-*
1-845 X IO" 8

1-22 X IO -6
1-5 X IO -5

1-4 X IO" 6
1-56 X IO -6
1-68 X IO" 6
9-6 X io- 6
1-4 X io~ 6
1-3 X IO -6

Chloroacetic
Bromoacetic
Iodoacetic
Dichloroacetic .
Trichloroacetic .

1-55 X io-"
1-38 X IO -8
7-3 x 10-*
5-14 X IO" 8
1-2

Formic acid is dehydrated to carbon monoxide by concentrated sulphuric
acid (see above). When heated under pressure at 160°, formic acid is de-
composed into carbon dioxide and hydrogen:

H-COjjH -

i + H.

FATTY ACIDS l8l

The same decomposition takes place at room temperature in the presence
of a catalyst such as iridium, rhodium, etc.
When metallic formates are heated with an alkali, hydrogen is evolved.

H-COjjNa + NaOH — > H 2 + Na 2 C0 3

When calcium or zinc formate is strongly heated, formaldehyde is produced:

(H-C0 2 ) 2 Ca — > H-CHO + CaC0 3

When sodium or potassium formate is rapidly heated to 360 , hydrogen is
evolved and the oxalate is formed:

aH-COgNa — > (COONa) 2 + H 2

Formic acid forms esters, but since it is a relatively strong acid it is not
necessary to use a catalyst ; refluxing 90 per cent, formic acid with the alcohol
is usually sufficient.

Formic acid differs from the rest of the members of the fatty acid series
in being a powerful reducing agent; it reduces ammoniacal silver nitrate
and the salts of many of the heavy metals, e.g., it converts mercuric chloride
into mercurous chloride.

Structure of formic acid. Analysis and molecular-weight determinations
show that the molecular formula of formic acid is CH 2 2 . Assuming that
carbon is quadrivalent, oxygen bivalent and hydrogen univalent, two
structures are possible:

H-</° H V|

(I) (II)

Only one of the two hydrogen atoms in formic acid is replaceable by a metal.
This suggests that the two hydrogen atoms are not in the same state of
combination. Moreover, since hydrogen is evolved when formic acid is
treated with sodium, the implication is that a hydroxyl group is present.
Structure (I), but not (II), satisfies the above observations, and also indicates
the relationship of formic acid to aldehydes, thus accounting for its reducing
properties (see also acetic acid, below).

Acetic acid (ethanoic acid), CHyCOaH, is prepared industrially by the
oxidation of acetaldehyde with air in the presence of manganous acetate as
catalyst :

CH 3 -CHO + J0 2 — > CH 3 -C0 2 H

The function of the catalyst appears to be to prevent the formation of
peracetic acid (q.v.). Acetic acid is also manufactured by the oxidation of
natural gas.

One of the earliest methods for preparing acetic acid was by the destructive
distillation of wood to give pyroligneous acid. This contains about 10 per cent,
acetic acid, and was originally treated by neutralising with lime and then
distilling off the volatile compounds (these are mainly methanol and acetone).
On distillation with dilute sulphuric acid, the residue gives dilute acetic acid.
More recently, the acetic acid is extracted by means of solvents, e.g., isopropyl
ether.

Vinegar, which is a 6-10 per cent, aqueous solution of acetic acid, may
be made in several ways. Malt vinegar is prepared by the oxidation of
wort (p. 135) by means of the bacteria Mycoderma aceti:

CH 3 -CH 2 OH + O a — > CHj-COjjH + H 8

l82 ORGANIC CHEMISTRY

In the " quick vinegar process ", beech shavings, contained in barrels, are
moistened with strong vinegar containing the bacteria. A 10 per cent,
aqueous solution of ethanol containing phosphates and inorganic salts
(which are necessary for the fermentation) is then poured through the barrels,
and the ethanol is thereby oxidised to acetic acid. A plentiful supply of
air is necessary, otherwise the oxidation is incomplete and acetaldehyde is
produced.

It is only recently that vinegar has been used as a source of acetic acid,
and this is entirely due to the introduction of highly efficient methods of
fractionation.

Acetic acid is a pungent corrosive liquid, m.p. i6-6°, b.p. 118°, miscible
in all proportions with water, ethanol and ether (see also p. 176). It is stable
towards oxidising agents, and so is a useful solvent for chromium trioxide
oxidations. Acetic acid is commonly used as a solvent, and in the prepara-
tion of acetates, acetone, acetic anhydride, etc.

Most of the normal acetates are soluble in water, whereas most of the basic
acetates are insoluble. Calcium and manganese acetates are used in the prepara-
tion of acetone. Lead tetra-acetate is a very useful oxidising agent for 1 : 2
glycols. Aluminium acetate, which is known only in solution, is used for water-
proofing cloth, and as a mordant.

A solution of a neutral acetate gives a red coloration when treated with ferric
chloride. The same coloration is also produced by neutral formates, but acetates
and formates may be readily distinguished from one another by the fact that the
latter are powerful reducing agents.

Structure of acetic acid. Analysis and molecular-weight determinations
show that the molecular formula of acetic acid is C 2 H 4 2 . The presence
of a methyl group in acetic acid is indicated by the following considerations :

(i) Treatment of boiling acetic acid with chlorine gives halogen substituted
acids. The highest halogenated acid that can be produced is trichloroacetic
acid, C 2 H0 2 C1 3 . Of the four hydrogen atoms in the acetic acid molecule,
three have been replaced by chlorine and are, therefore, in a different state
of combination from the fourth. This suggests that acetic acid contains a
methyl group.

(ii) Methyl chloride may be converted into methyl cyanide, which, on
hydrolysis, gives acetic acid. Methyl chloride contains a methyl group,
and when it is converted into methyl cyanide the chlorine atom is replaced
by the cyano-group. Since there is no reason to suppose that the methyl
group in methyl chloride is affected by the substitution, it follows that methyl
cyanide contains a methyl group. Similarly, since the hydrolysis of a
cyanide affects the cyano-group only, when methyl cyanide is hydrolysed to
acetic acid, the methyl group remains intact. Thus the above reactions may
be formulated:

CH 3 — CI -^> CH 3 — CN -5^> CH 3 — C0 2 H

When acetic acid is treated with phosphorus pentachloride, acetyl chloride,
CH 3 — COC1, is formed and hydrogen chloride is evolved. This indicates
the presence of a hydroxyl group. Thus acetic acid contains a methyl
group and a hydroxyl group. Assuming the quadrivalency of carbon, the
bivalency of oxygen and univalency of hydrogen, only one structural formula

O

is possible for acetic acid, viz., CH 3 — C — OH. The presence of the carbonyl
group is supported by the fact that acetyl chloride may be catalytically
reduced to acetaldehyde, which has been shown to contain a carbonyl group
(p. 167).

FATTY ACIDS 183

Kolbe's synthesis of acetic acid (1845) is interesting in connection with the vital
force theory (p. 1). Carbon and sulphur were heated together, and the carbon
disulphide produced chlorinated :

C + 2S > CS 2

CS a _^>CCl 4

The carbon tetrachloride was then passed through a red-hot tube, whereupon
chlorine and tetrachloroethylene were formed and these, on cooling, combined to
form hexachloroethane; but in the presence of a little water and under the
influence of direct sunlight, some of the tetrachloroethylene was converted into
trichloroacetic acid:

CC1 2 =CC1 2 + Cl 2 + 2H a O ^ CC1 S -C0 2 H + 3HCI

Treatment of the trichloroacetic acid in aqueous solution with potassium
amalgam, or the electrolysis of the aqueous solution of the trichloroacetic acid
between electrodes of amalgamated zinc plates, produced acetic acid :

CC1 S -C0 2 H + 6[H] > CH 3 'C0 2 H + 3HCI

The structure of the carboxyl group is still not known with certainty. If

its structure were — C; , then the acid strength should be greater than

\OH
that of alcohols, due to the inductive effect of the carbonyl group which
tends to facilitate the release of a proton :

O

R^-C-^-O^-H

The acid strength of the fatty acids is greater than that of alcohols, but it
seems unlikely that the inductive effect alone would account for the large
difference. It has therefore been suggested that the carboxyl group is a
resonance hybrid:

/OH /OH //O /O

R-C' <->- R-Cv _ ?±H + + R-C< _ <-> R-CT

^O X) X) ^O

Owing to the positive charge on the oxygen atom of the hydroxyl group,
the electron pair of the — H bond is displaced towards the oxygen atom,
thereby facilitating the release of a proton. When the proton is released,
the two equivalent resonating structures contribute to the carboxylate ion
formed, and so the resonance energy of the ion will be greater than that of
the undissociated acid. Thus the driving force for the dissociation is this
tendency to achieve greater stabilisation. Since resonance is not possible
in alcohols, proton release is much more difficult than in carboxylic acids.

If acids are resonance hybrids, the C=0 bond will have some single-bond
character, and its length should therefore be longer than that of a " pure "
0=0 double bond; similarly, the C — OH bond will have some double-bond
character, and should be shorter than a " pure " C — OH single bond.
Electron diffraction experiments (Schomaker and O'Gorman, 1947) have
shown that the C=0 bond length in formic acid monomer is a trifle greater
than normal, and that the C — OH bond is considerably less than normal.
These results are in keeping with a resonating structure of the carboxyl
group.

The treatment of the carboxyl group and the carboxylate ion from the M.O.
point of view also leads to the same results as those obtained by the V.B. method.

I84 ORGANIC CHEMISTRY

the increased stability now being due to delocalisation (cf. p. 88). Let us first
consider the undissociated carboxylic acid (Fig. la). The carbon atom is linked
to the hydroxyl group by a a-bond, and to the oxygen of the carbonyl group by
one a- and one re-bond. This is shown in (b) . If the oxygen atom of the hydroxyl
group has sp z hybridisation, this would leave a p orbital perpendicular to the

plane of the carboxyl group, and it would contain a pair of electrons (c). This
orbital could now overlap with both the p orbital of the carbon atom and that
of the oxygen atom of the carbonyl group. The combination of these three
orbitals would give rise to M.O.s embracing three nuclei (Fig. d). Since four
electrons are involved, they must be accommodated, in the ground state, in the
lowest two M.O.s. The M.O. with the lowest energy level has one node (Fig. e),
and the next energy level has two nodes (Fig. /). Thus (g) represents the ground
state of the undissociated carboxyl group. Because of delocalisation, this
group has become more stable. Since a hydroxyl group is more electron-
attracting than an oxygen atom, the delocalisation of the pair of electrons in the
p orbital is not very large, i.e., although this lone pair embraces three nuclei,
this pair is far more likely to be found in the region of the donor atom than any-
where else. Since the oxygen atom of the hydroxyl group has lost " full-control "
of this lone pair, it will acquire a small positive charge, and since the oxygen atom
of the CO group has acquired a small share of this lone pair, this oxygen atom
acquires a small negative charge (actually this oxygen atom already has a small
negative charge due to the inductive effect). The greater electron-attracting
power of oxygen in a hydroxyl group over that of an oxygen atom alone is due to
the hydrogen atom sharing one pair of electrons in forming the bond, and thereby
decreasing, to some extent, the electron density on that oxygen atom.

Now let us consider the carboxylate ion (Fig. 2a). The oxygen atom of the
original hydroxyl group retains the a-electrons when the hydrogen atom is
removed as a proton. As in the case of the carboxyl group, one lone pair enters

R— C

./

V

(a)

into conjugation with the re-bond of the C=0 group (Fig. b). In the latter case,
however, delocalisation is complete, i.e., the lone pair is now likely to be found
equally well on either oxygen atom (each atom will have a charge of ■— £). Since
delocalisation is complete, the resonance (delocalisation) energy is much greater
than in the case of the undissociated carboxyl group.

In Figs, i and 2, the orbitals drawn in thicker type indicate that the atoms

FATTY ACIDS 185

concerned contain two electrons in these orbitals. This scheme has been used
throughout the book.

^°
Peracetic acid, CH 3 '(\ , may be prepared by treating acetic anhydride

\o— OH

with concentrated hydrogen peroxide, and then distilling under reduced pressure.

It may also be prepared by adding 90 per cent, hydrogen peroxide to cooled

glacial acetic acid in the presence of a small amount of sulphuric acid. Anhydrous

peracetic acid may be prepared by autoxidation of acetaldehyde at o° in the

presence of cobaltous ion (Phillips et al., 1957) :

Q y?0 HO\

2CH„«CHO '> CH 8 -Cf , ^CH-CH 8 21+ CH s -C00 2 H + CH 8 -CHO

Peracetic acid is an unpleasant-smelling liquid, f.p. +o-i°, soluble in water,
ethanol and ether. It explodes violently when heated above no°. It is a
powerful oxidising agent; it oxidises the olefinic bond to the oxide.

/°\
>C=C< + CHs-CO-0 2 H > >C C< + CH 3 -CO s H

It also oxidises primary aromatic amines to nitroso-compounds, e.g., aniline is
converted into nitrosobenzene:

C,H B -NH 2 + aCHs-COOjH — y C 6 H B -NO + 2CH 3 -CO a H + H a O

Infrared absorption spectra measurements of per-acids show that these acids
exist in solution very largely in the monomeric, intramolecularly hydrogen-
bonded form (inter alia, Minkoff, 1954).

CH <o-o

Propionic acid (propanoic acid), CHVCH^COaH, is prepared industrially
by the oxidation of w-propanol:

CH 3 -CH 2 -CH 2 OH -^i> CH 3 -CH 2 -CO a H

It is a colourless liquid with an acrid odour, m.p. —22°, b.p. 141 , miscible
with water, ethanol and ether in all proportions.

Butyric acids, C 4 H 8 2 . There are two isomers possible, and both are
known.

n-Butyric acid, CH 3 *CH 2 'CH 2 *C0 2 H, occurs as the glyceryl ester in butter,
and as the free acid in perspiration. It is prepared industrially by the
oxidation of »-butanol, and by the butyric fermentation of carbohydrates by
means of the micro-organism Bacillus butyricus.

M-Butyric acid is a viscous unpleasant-smelling liquid, m.p. — 4-7°, b.p.
162 , miscible with water, ethanol and ether. It is the liberation of free
M-butyric acid that gives stale butter its rancid odour.

tsoButyric acid, (CH 3 ) 2 CH>C0 2 H, occurs in the free state and as its esters
in many plants. It is prepared industrially by the oxidation of isobutanol:

(CH8) 2 CH'CH 2 OH -^-> (CH s ) 2 CH-C0 2 H
It may be prepared synthetically as follows :

(CH 3 ) 2 CHOH ^> (CH 3 ) 2 CHBr > (CH 3 ) 2 CH-CN — ^(CH 3 ) 2 CH-C0 2 H

acid

isoButyric acid is a liquid, m.p. — 47°, b.p. 154 . Its calcium salt is more
soluble in hot water than in cold, whereas calcium butyrate is more soluble
in cold water than in hot.

l86 ORGANIC CHEMISTRY

Valeric acids, C 8 H 10 O 2 . There are four isomers possible, and all are
known:

n-Valeric acid, CH 3 CH 2 -CH 2 -CH 2 -C0 2 H, m.p. —34-5°, b.p. 187 .

isovaleric acid, (CH s ) 2 CH>CH 2 -C0 2 H, m.p. — 51 , b.p. 175 .

Ethylmethylacetic acid or active valeric acid, CH 3 -CH 2 -CH(CH 3 )-C0 2 H,
b.p. 175 .

Trimethylacetic acid or pivalic acid, (CH 3 ) 3 OC0 2 H, m.p. 35-5°,
b.p. 164 .

The higher fatty acids which occur in nature usually have straight chains,
and usually contain an even number of carbon atoms. Caproic (hexoic),
C 6 H 12 2 (m.p. —9-5°, b.p. 205 ), caprylic (odoic), C 8 H 16 2 (m.p. 16 , b.p.
237°), and capric (decoic) acid, C 10 H 20 O 2 (m.p. 31-5°, b.p. 270 °), are present as
glyceryl esters in goats' butter. Laurie (dodecoic), C 12 H 24 2 (m.p. 44°),
and myristic (tetradecoic) acid, C 14 H 28 2 (m.p. 58°), occur as their glyceryl
esters in certain vegetable oils. The most important higher fatty acids are
palmitic (hexadecoic), C 16 H 32 2 (m.p. 64°), and stearic (octadecoic), C 18 H 36 2
(m.p. 72 ), which are very widely distributed as their glyceryl esters (together
with oleic acid) in most animal and vegetable oils and fats. The sodium
and potassium salts of palmitic and stearic acids are the constituents of
ordinary soaps.

Hansen et al. (1952, 1956) have shown the presence of branched-chain
acids in a number of animal fats, e.g., 13-methyltetradecanoic acid in butter
fat, and 14-methylpentadecanoic acid (wopalmitic acid) in hydrogenated
sheep fat.

Some still higher acids are found in waxes: arachidic (eicosoic), C 20 H 40 O 2 (m.p.
77 ), behenic (docosoic), C 22 H 44 2 (m.p. 82 ), lignoceric (tetracosoic), C 24 H 48 O a
(m.p. 83-5°), cerotic (hexacosoic) , C 26 H 62 2 (m.p. 87-7°), and melissic acid (tri-
acontoic), C 30 H 60 O 2 (m.p. 90 ).

The odd fatty acids may be obtained by degrading an even acid (see below).
Two odd acids which may be prepared readily are n-heptoic or cenanthylic acid,
C,H 14 O a (m.p. — io°, b.p. 223-5°), and nonoic or pelargonic acid, C 9 Hi 8 2 (m.p.
12°, b.p. 254°). w-Heptoic acid is prepared by the oxidation of w-heptaldehyde,
which is obtained by destructively distilling castor oil which contains ricinoleic
acid {q.v.) :

CH 3 -(CH 2 ) 5 -CHO + [O] —A- CH 3 -(CH 2 ) 6 -CO a H (68-70%)

rigSOj

Nonoic acid is obtained by the oxidation of oleic acid (q.v.).

Margaric acid (heptadecanoic acid), C 18 H 33 *C0 2 H, m.p. 6i°, is prepared by
heating a mixture of the calcium salts of stearic and acetic acids, and then
oxidising the heptadecyl methyl ketone so produced.

C 17 H M 'CO,ca + CH 3 -CO s c« >■ CaC0 3 + C 17 H 35 -COCH 3 -i^> C le H 33 -C0 2 H

Margaric acid has been used as a source of artificial fats for diabetics. Until
recently, margaric acid was considered to be a synthetic compound. In 1954,
however, Hansen et al. isolated w-heptadecanoic and w-pentadecanoic acids
from mutton fat, and w-pentadecanoic and M-tridecanoic acids from butter fat,
and in 1955 showed the presence of «-undecanoic acid in butter fat. Hansen
et al. (1954) have also shown that ox perenephric fat contains the consecutive
series of acids C 2 to C 10 .

ESTERS

Esters are compounds which are formed when the hydroxylic hydrogen
atom in oxygen acids is replaced by an alkyl group ; the acid may be organic
or inorganic. The most important esters are derived from the carboxylic
acids. The general formula of the carboxylic esters is CbHjrOj, which is

FATTY ACIDS 187

the same as that of the carboxylic acids. The structural formula of the
esters is RC— OR', and they are named as the alkyl salts of the acid, e.g.,

CH 3 -COOC 2 H 5 ethyl acetate

(CH 3 ) 2 CH-COOCH(CH 3 ) 2 wopropyl wobutyrate

Carboxylic esters are formed by the action of the acid on an alcohol:

acid + alcohol ^= ester -f- water

The reaction is reversible, the forward reaction being known as esterification,
and the backward reaction as hydrolysis.

General methods of preparation of the carboxylic esters. 1. The usual
method is esterification. The reaction is always slow, but is speeded up by
the presence of small amounts of inorganic acids as catalysts, e.g., when the
acid is remixed with the alcohol in the presence of 5-10 per cent, concen-
trated sulphuric acid:

R-COOH + R'OH ~^± R-COOR' + H 2 (v.g.)

Alternatively, hydrogen chloride is passed into the mixture of alcohol and
acid until there is a 3 per cent, increase in weight, and the mixture is refluxed
(the yields are very good). This is known as the Fischer-Speier method
(1895), and is more satisfactory for secondary and tertiary alcohols than the
sulphuric acid method, which tends to dehydrate the alcohol to olefin.
Klosa (1956) has shown that phosphoryl chloride is a good catalyst in the
esterification of acids with alcohols.

Esterification without the use of catalysts, and starting with one molecule
of acid and one molecule of alcohol, gives rise to about § molecule of ester.
The yield of ester may be increased by using excess acid or alcohol, the
cheaper usually being the one in excess. Increased yields may also be
effected by dehydrating agents, e.g., concentrated sulphuric acid behaves
both as a catalyst and a dehydrating agent. The same effect may be obtained
by removing the water or ester from the reaction mixture by distillation,
which is particularly useful for high-boiling acids and alcohols. On the
other hand, the water may be removed from the reaction mixture by the
addition of benzene or carbon tetrachloride, each of which forms a binary
mixture with water (and may form a ternary mixture with water and the
alcohol), the azeotropic mixtures boiling at a lower temperature than any
of the components.

The first point about esterification of a carboxylic acid is that it may be
formulated in two ways:

O O

R — C — OH H — OR' — > R — C + H a O acyl-oxygen heterolysis

A.

)R'
O O

t-C-O^H

R— C— O— H HO— R' — y R— C + H 2 alkyl-oxygen heterolysis

)R'

Jh

The second point is that esterification is usually carried out in the presence
of an acid catalyst (sulphuric or hydrochloric acid), and hence it is the

1 88 ORGANIC CHEMISTRY

conjugate acid of the carboxylic acid that is the substrate. The structure
of this conjugate acid has been formulated in two ways :

\h r— c— oh

Since carboxylic acids are resonance hybrids, it would appear more likely
that protonation occurs on the carbonyl oxygen:

S° /° H+ / 0H

\OH ^OH ^OH

This is supported by the work of Fraenkel (1961), who examined the nuclear
magnetic resonance spectrum of methyl formate in, e.g., 100 per cent, sul-
phuric acid, and concluded from his results that the ester is protonated
chiefly on the carbonyl oxygen.

Ingold (1941) has introduced the following notation : the letter A represents
the substrate in the conjugate acid of the carboxylic acid, and the subscripts
A0 and AL respectively denote acyl and alkyl bond heterolysis. The
numbers 1 and 2 represent the molecularity of the rate-determining step.

So far three different esterification mechanisms have been observed :

A ko t, A A0 2, and A Ah i.

The most common esterification mechanism is A A0 2, i.e., bimolecular acid
esterification with acyl-oxygen heterolysis. It has long been known that the
rates of hydrolysis of many esters are first-order both in ester and in hydrogen-
ion concentration, i.e.,

rate cc [ester] [H + ]

Since acid-catalysed hydrolysis of esters and esterification are the reverse of each
other (and reversible), then according to the principle of microscopic reversibility
(p. 32), the mechanism of acid-catalysed esterification will be that of acid-
catalysed hydrolysis, but in reverse.

Evidence for acyl-oxygen fission in acid-catalysed esterification has been
obtained in several ways, e.g., Roberts et al. (1938) esterified benzoic acid with
methanol enriched with w O and obtained water not enriched with ls O. There-
fore the oxygen in the water must have come from the benzoic acid; thus:

C 6 H s -CO— OH + CH s O— -H > C 8 H 6 -COOCH 3 + H 2

According to Ingold et al. (1939), the mechanism is:

H— OR'
R-C-OH + H + -^ R-C-OH a *'° H; slow ^ R-0-OH 2

H-hO— R'

-H.O; fast s xj 1 -H+; fast

R-^C -" T ' la5I > R-C-OR'

o o

FATTY ACIDS 189

However, if we assume that the carbonyl oxygen is protonated, then the
mechanism is :

H— OR'

R-C-OH ^ H+; fast N R-C-OH R ' 0H; *" ^ R-C-OH ^

a l l

OR' OR' OR'

R— C— OH 2 fast \ H 2 + R— C+ fast s r— c + H+

I * IV ll

OH O— H O

Since, in these mechanisms, the carbon atom which was originally joined to
three groups becomes joined to four (hybridisation changes from sp* to sp s ),
then one might anticipate steric retardation is possible in certain acids. The
following results of Newman et al. (1952), who examined the relative rates of
esterification of various acids with methanol, showed steric retardation was
operating: e.g.,

Acid: MeC0 2 H w-PrC0 2 H Me 3 C-C0 2 H Et s C-C0 2 H

Relative rates : 1 0-51 0-037 0-00016

Now let us consider unimolecular acid-catalysed esterification with alkyl-
oxygen heterolysis (A^i mechanism). Definite evidence for this mechanism
was obtained by Hughes, Ingold, et al. (1939), who showed that the esterification
of acetic acid with optically active octan-2-ol in the presence of sulphuric acid
gave a large amount of racemised ester. Thus a carbonium ion is produced
which then racemises. The mechanism may therefore be formulated:

R'OH jzr^ R'OH 2 + ^hrt H a O + R' +

+•

k /OR' /OR'

R-C^vO ^ R-< + H+

\o— H ^O

The A ^1 mechanism is common for ^-alcohols, and it is probably this type of
mechanism which operates for halogen acids :

ROH + HX^± ROH a + + X- ^t H a O + R+ + X~ ^ RX

This is in keeping with the fact that the rate of reaction is <-alcohol> s-> primary,
since this is the order in which the OH group is removed (see p. 129).

Unimolecular acid-catalysed esterification with acyl-oxygen fission (^4 AO i
mechanism) occurs with highly sterically hindered acids, e.g., mesitoic acid.
Hammett et al. (1937) showed that when mesitoic acid is dissolved in concentrated
sulphuric acid, the van't Hofi factor * is 4. When this solution is poured into
methanol, methyl mesitoate is formed (Newman, 194 1). A mechanism con-
sistent with these facts is that the reaction proceeds via an acylium ion :

O

II +/OH

R— C— OH + H 2 S0 4 ^ R— C< + HS0 4 ~

X)H

R— C=0 + MeOH > R— CX -f- H+

OMe

190 ORGANIC CHEMISTRY

Many primary alcohols, on oxidation with chromic acid, form esters in
addition to acids, e.g., w-butanol gives w-butyl rc-butyrate:

2 C 3 H,-CH 2 OH -^% C 3 H,-C0 2 C 4 H 9 (41-47%)

Several mechanisms have been proposed for this reaction, e.g., the direct
esterification of unchanged alcohol with acid formed on oxidation, or hemi-
acetal formation followed by oxidation to ester. Mosher et al. (1953) have
obtained evidence for the latter :

R-CH 2 OH — — -> R-CHO R-CH ' 0H > R-CH^ -^-> R-C0 2 CH 2 -R

H+ \OCH 2 -R

2. Acid chlorides or anhydrides react rapidly with alcohols to form
esters r

R-COC1 + R'OH — >- R-C0 2 R' + HC1 (v.g.-ex.)

(R-CO) 2 + R'OH — >- R-C0 8 R' + R-C0 2 H {v.g.-ex.)

The reaction with tertiary alcohols is very slow, and is often accompanied
by the dehydration of the alcohol to olefin. When the acid chloride is
used, there is also a tendency for a tertiary alcohol to form a tertiary alkyl
chloride (p. 130). Esters of tertiary alcohols may be conveniently prepared
by means of a Grignard reagent {see p. 356).

3. Esters may be prepared by refluxing the silver salt of an acid with an
alkyl halide in ethanolic solution :

R-C0 2 Ag + R'Br — > R-C0 2 R' + AgBr {v.g.-ex.)

This method is useful where direct esterification is difficult (c/. 2 above).

4. Esters are formed when a mixture of the vapours of an acid and an
alcohol is passed over a metallic oxide catalyst at 300°.

R-C0 2 H + R'OH ^% R-C0 2 R' + H 2

5. Methyl esters are very conveniently prepared by treating an acid with
an ethereal solution of diazomethane {q.v.) :

R-C0 2 H + CH 2 N 2 — > R-C0 2 CH 3 + N 2 {ex.)

6. Esters are readily obtained when an acid is treated with an olefin in the
presence of boron trifluoride as catalyst (Nieuwland et al., 1934) :

R-C0 2 H + C 2 H 4 —V R'C0 2 C 2 H 5

It has been found that the complex (CH a -C0 2 H) a "BF 3 is a very efficient catalyst
in the preparation of esters from acids and alcohols (Smith et al., 1940). Another
example of the use of boron trifluoride as a catalyst in organic chemistry is the
formation of esters by the interaction between an ether and carbon monoxide at
125-180 , under 500 atmospheres pressure, in the presence of boron trifluoride
plus a little water:

R 2 + CO >- R-C0 2 R

General properties of the esters. The carboxylic esters are pleasant-
smelling liquids or solids. The boiling points of the straight-chain isomers
are higher than those of the branched-chain isomers. The boiling points of
the methyl and ethyl esters are lower than those of the corresponding
acid, and this is probably due to the fact that the esters are not associated
since they cannot form intermolecular hydrogen bonds. The esters of low
molecular weight are fcflrly soluble in water — hydrogen bonding between
ester and water is possible — and the solubility decreases as the series is
ascended; all esters are soluble in most organic solvents.

FATTY ACIDS I9I

o

The structure of the esters is usually written as R — C — OR', but, as in
the case of the acids from which they are derived, there is a certain amount
of evidence to show that esters are resonance hybrids (or conjugated, p. 184) :

O O

11 I ♦

R— 0— OR' <--> R— C=OR

General reactions of the esters. 1. Esters are hydrolysed by acids or
alkalis :

R-0O 2 R' + H 2 add > R-C0 2 H + R'OH
R-C0 2 R' + NaOH >■ R-C0 2 Na + R'OH

When hydrolysis is carried out with alkali, the salt of the acid is obtained,
and since the alkali salts of the higher acids are soaps, alkaline hydrolysis is
known as saponification (derived from Latin word meaning soap) ; saponifica-
tion is far more rapid than acid hydrolysis.

Hydrolysis of secondary and particularly tertiary halides is accompanied
by the formation of olefin (p. 108). In many cases a better yield of alcohol
can be obtained by first converting the alkyl halide into the acetate ester
by heating with silver acetate in ethanolic solution, and then saponifying
the acetate ester:

CH 3 -C0 2 Ag + RX — > CH 3 -C0 2 R — H -> CH 3 -C0 2 Na -f ROH

Under the above conditions, secondary and tertiary halides show far less
tendency to form olefin than when hydrolysed directly by alkali.

In the case of hydrolysis of esters, not only can we have acid-catalysed
reactions (as for esterification), but hydrolysis may be effected in alkaline
solution (i.e., saponification). Hydrolysis may also be effected in neutral
solution, but since this is slow, it is not used. Using the symbols A, A0 >
al, 1, and 2 as for esterification (p. 188) for the protonated ester (in acid
solution), and using B for the unprotonated ester (in alkaline or neutral
solution), then there are eight possible mechanisms (jB ao i, -B A0 2, B^j,
B A7 jz; ^4 A0 i, -4ao2, A al x, A^i). All except two, the first and last, have
been observed.

The B A0 2 mechanism is a common one. Alkaline hydrolysis of esters has long
been known to be a second-order reaction, i.e., rate oc [ester][OH~]. Evidence
for acyl-oxygen fission has been obtained in several ways, e.g., Polanyi et al.
(1934) showed that the alkaline hydrolysis of w-pentyl acetate in water enriched
with ls O gave w-pentyl alcohol containing no ls O. Therefore acyl-oxygen fission
must have occurred :

CH 3 -CO— OC B H u + OH- > CH 3 -COOH + C 5 H u O- -^> C 5 H u OH

If the alcohol group R' is optically active, then if acyl-oxygen fission occurs, the
ion R'O - will be liberated and will retain its optical activity, since the bond
R' — O is never broken. Various examples of retention are known when the
reaction is bimolecular. A mechanism consistent with these facts is :

OXv O7 OO
HO" C— OR'^=^ HO— C- 1 - OR' ^=^ HO— C + R'0~ ^O— C + R'OH

R R R R

192 ORGANIC CHEMISTRY

The A AC 2 mechanism is also a common one for the hydrolysis of esters; this
is the reverse of the A^i mechanism for esterification (p. 188). The order has
been shown to be first in both ester and hydrogen-ion concentration, and evidence
for acyl-oxygen heterolysis has been demonstrated in several ways, e.g., Ingold
et al. (1939) showed that the acid-catalysed hydrolysis of methyl hydrogen
succinate in water enriched with ls O gave methanol with no extra ls O ; therefore
acyl-oxygen fission must have occurred :

H0 2 OCH 2 -CH 2 -CO— OMe + H a O ■
A mechanism consistent with the facts is :

R < oR , + H+ ^ KoH'^ *-^*

H— 6+~

O— H

-H
OH O— H O

R'C— OC ^ R'OH + R-C+ ^=±: R-C + H+

| ■ \h I I

OH OH OH

Ingold's mechanism is the reverse of the one given for esterification on p. 188.

The Afjj. mechanism for the acid-catalysed hydrolysis of esters is the reverse
of that for the Aj&i mechanism for acid-catalysed esterification (p. 189). Evi-
dence for alkyl-oxygen fission has been provided by, e.g., Bunton et al. (1951),
who hydrolysed /-butyl acetate with water enriched with 18 and obtained
i-butanol containing ls O.

MeCO'O— Bu 4 -(- H 2 > MeCO a H + Bu'OH

The A AO i mechanism for the acid-catalysed hydrolysis of esters is the reverse
of that for esterification (p. 189).

The B AL i mechanism for the base-catalysed hydrolysis of esters has been
shown to occur when the alkyl group of the alcohol is strongly electron-releasing,
the solvent has a high dipole moment, and the hydroxide-ion concentration is
very low. The mechanism is :

*~^
R-CO-O— R' =^ R-C0 2 - + R'+

R' + + H a O ^ R'OH 2 +

R-C0 2 - + R'OH 2 + >■ R'OH + R-CO a H

E.g., Kenyon et al. (1936) showed that the B Ah i mechanism was operating when
optically active i-phenylethyl hydrogen phthalate was hydrolysed in faintly
alkaline solution. The alcohol obtained, C 6 H 5 *CHOH-CH 3 , was racemised, and
this is in keeping with the intermediate formation of a carbonium ion.

The B AL 2 mechanism for the base-catalysed hydrolysis of esters is rare.
Bunnett et al. (1951) have shown that dimethyl ether is formed when methyl
benzoate is treated with methanolic sodium methoxide :

MeO- Me— 0-CO-C 6 H 6 > Me a O + C 6 H 5 -C0 2 -

2. Esters are converted into alcohols by the Bouveault-Blane reduction:
R-C0 2 R' + 4 [H] C,H ' 0H/Na > R-CH 2 OH + R'OH (g.)

The yield of alcohol from the acid portion of the ester increases with the
molecular weight of the alkyl radical R', e.g., the amyl ester gives a higher
yield of alcohol R-CH 2 OH than does the ethyl ester. Esters which are

:nh 3 + , .. - Q Q

FATTY ACIDS 193

difficult to reduce at the boiling point of ethanol are usually satisfactorily
reduced in w-butanol, which has a higher boiling point than ethanol, and
so permits the reduction to be carried out at a higher temperature. Lithium
aluminium hydride also reduces esters to alcohols.

Esters may also be reduced by molecular hydrogen at 100-300 atmospheres
in the presence of a copper chromite catalyst at 200-300 (the yields are
almost quantitative).

The formation of esters affords a convenient method of converting an
acid into its corresponding alcohol, and stepping up the acid series:

R . C 2 H -^V R-C0 2 R' -^-> R-CH 2 OH -^> R-CH 2 Br -^>

R-CH 2 -CN -^> R-CH 2 -C0 2 H

acid

3. Esters react with ammonia to form amides. This reaction is an
example of ammonolysis (which means, literally, splitting by ammonia) :

R-COOR' + NH 3 — > R-CO-NH 2 + R'OH (g.)

A possible mechanism for this reaction is :

-f C— OR' ^=i H.N— C-'-OR' =^

I I

R R + II 11

H 3 N-C + OR" — > H 2 N-C + R'OH

R R

With hydrazine, esters form acid hydrazides:

R-COOR' + H 2 N-NH 2 — > R-CO-NH-NH 2 + R'OH

4. By means of alcoholysis (splitting by alcohol), an alcohol residue in an
ester can be replaced by another alcohol residue. Alcoholysis is carried
out by refluxing the ester with a large excess of alcohol, preferably in
the presence of a small amount of acid or sodium alkoxide as catalyst.
Alcoholysis is usually effective in replacing a higher alcohol by a lower one,
e.g.,

CH 3 -C0 2 C 4 H 9 + C 2 H 5 OH ^=± CH 3 -C0 2 C 2 H B + C 4 H 9 OH

Alcoholysis of esters is known as transesterification.

In acidolysis, the acid residue is displaced from its ester by another acid
residue, e.g.,

CH 3 -C0 2 C 2 H 5 + C 5 H u -C0 2 H ^ C 5 H u -C0 2 C 2 H 5 + CH 3 -C0 2 H

Acidolysis is a useful reaction for converting the neutral ester of a dibasic
acid into its acid ester.

5. When an ester — preferably the methyl or ethyl ester — is treated with
sodium in an inert solvent, e.g., ether, benzene or toluene, and subsequently
with acid, an acyloin is formed (50-70 per cent, yield). It is important
that the reaction be carried out in the absence of any free alcohol. Acyloins
are <*$-keto-alcohols, and the mechanism of their formation is obscure.

H

194 ORGANIC CHEMISTRY

According to Kharasch and his co-workers (1939), the formation of acyloins
takes place via a free-radical mechanism, e.g., propionin from ethyl
propionate:

2C 2 H 5 -C0 2 C 2 H 5 + 2 Na — > 2C 2 H 5
/ONa

./ONa

-OC 2 H 5

C 2 H 5

0C s H *_ rxinxT , C2H ^ == °^ C 2 H 5 .C-ONa acid

or H =s=£b 2 C 2 H 5 ONa + I > >

UUa 5 C 2 H 6 -C=0 C 2 H 5 -C— ONa

C 2 H K -C— ONa

rc 2 H 5 -c— OHH

LC 2 H 5 -C— OHj

rearranges C 2 H 5 'CO

-'2 J - i 5" v ' ""j C 2 H 5 'CHOH

6. Carboxylic esters which contain a-hydrogen atoms react with sodamide
in liquid ammonia solution to form the acid amide and a condensation
product involving two molecules of the ester, e.g., ethyl acetate gives
acetamide and acetoacetic ester (see also p. 224) :

CH 3 -C0 2 C 2 H 5 + NaNH 2 — >■ CH 3 -CO-NH 2 + C 2 H 5 ONa

CH 3 -C0 2 C 2 H 5 + NaNH 2 — -> NH 3 + [CH 2 -C0 2 C 2 H 5 ]-Na + CH '" C0 ' C,H ' >

CH 3 -CO-CH 2 -C0 2 C 2 H 5 + C 2 H 5 OH

Of particular interest is the carbonation of esters, i.e., the introduction
of the carboxyl group. The ester is treated with sodamide in liquid ammonia
solution, the solvent is evaporated off, ether is added, and then solid carbon
dioxide :

R-CH 2 -C0 2 R' + NaNH 2 ' iquid NH '>

co R-CH-CO,R'
NH 3 + [R-CH-C0 2 R']~Na + ' ^ '

C0 5

Na

The yields of these malonic acid derivatives are 54-60 per cent, with acetates,
and become progressively lower as the molecular weight of the acid increases.

Esters are used as solvents for cellulose, oils, gums, resins, etc., and as
plasticisers. They are also used for making artificial flavours and essences,
e.g., isoamyl acetate — banana oil; amyl butyrate — apricot; isoamyl
tsovalerate — apple; methyl butyrate — pineapple; etc.

Ortho-esters are compounds of the type R-C(OR') 3 . They are derived
from the ortho-acids, R-C(OH) 3 , which have not yet been isolated, but
which are possibly present in aqueous solution (cf. ethylidene glycol, p. 168) :

R— Cf + H a O ^ R-C(OH) 3
x OH

The most important ortho-esters are the orthoformic esters, particularly
ethyl orthoformate. Ethyl orthoformate may be prepared by running
ethanol and chloroform into sodium covered with ether:

2CHCI3 + 6C 2 H 6 OH + 6Na — > 2 H-C(OC 2 H s ) 3 + 6NaCl + 3H 2 (70%)

FATTY ACIDS 195

Another method is to pass dry hydrogen chloride into a solution of hydrogen
cyanide and ethanol in ether:

^NH-HC1
H-C=N + C 2 H 5 OH + HC1 — > K-C(

X OC 2 H 5

The formidic ester hydrochloride is then allowed to stand in ethanol, where-
upon ethyl orthoformate is produced :

^NH-HC1
H-Gf + 2C 2 H 5 OH — > H-C(OC 2 H 5 ) 3 + NH 4 C1

\OC 2 H 5

Ethyl orthoformate may be used for preparing ketals (p. 163), and for
preparing aldehydes by means of a Grignard reagent (p. 355).

Esters of the Inorganic Acids

Alkyl halides, which can be prepared by the action of a halogen acid on
an alcohol, may be regarded as esters, but are exceptional in that they do
not contain oxygen.

Alkyl sulphates, (RO) 2 S0 2 . Only methanol and ethanol give a good yield
of the alkyl sulphate by reaction with concentrated sulphuric acid; the
higher alcohols give mainly olefins and ethers. On the other hand, all the
alcohols give a fair yield of alkyl hydrogen sulphate when a mixture of
alcohol and concentrated sulphuric acid is heated on a steam bath.

According to Barkenbus and Owen (1934), tne mos t useful method for prepar-
ing alkyl sulphates is by the action of an alkyl chlorosulphonate on an alkyl
sulphite :

(RO) 2 SO + C1S0 2 -0R MRO) 2 S0 2 + HC1 + SO s + R'-CH=CH 2

(RO) 2 SO + C1S0 2 OR >- (RO) 2 S0 2 + S0 2 + RC1

Methyl sulphate may be prepared :

(i) By heating methyl iodide with silver sulphate:

2CH 3 I + Ag 2 S0 4 -^ (CH 3 ) 2 S0 4 + 2AgI

(ii) By heating methanol with concentrated sulphuric acid, and then
distilling the methyl hydrogen sulphate under reduced pressure :

CH3OH + H 2 S0 4 — > CH 3 HS0 4 + H 2
2CH 3 HS0 4 — > (CH 3 ) 2 S0 4 + H 2 S0 4

(iii) By treating methanol with sulphur trioxide at low temperatures :

2SO3 + 2CH 3 0H — > (CH 3 ) 2 S0 4 + H 2 S0 4

(ii) and (iii) are industrial methods.

Ethyl sulphate may be prepared by the same methods as methyl sulphate,
but in addition, there is the industrial preparation by passing ethylene in
excess into cold concentrated sulphuric acid :

2C a H 4 + H 2 S0 4 — >- (C 2 H 5 ) 2 S0 4

Methyl sulphate, b.p. 188°, and ethyl sulphate, b.p. 208°, are heavy
poisonous liquids. They are largely used as alkylating agents since the
alkyl group will replace the hydrogen atom of the groups — OH, 'NH* or
— SH. The alkylation may be carried out by treating the compound with
the alkyl sulphate in sodium hydroxide solution. Usually only one of the

I9 6 ORGANIC CHEMISTRY

alkyl groups takes part in the reaction, e.g., methylation of a primary
amine :

R-NH 2 + (CH 3 ) 2 S0 4 + NaOH — > R-NH-CH 3 + CH 3 NaS0 4 + H 2

The methyl and ethyl esters of the carboxylic acids may be conveniently
prepared by treating the sodium salt of the acid with respectively methyl
or ethyl sulphate:

R-C0 2 Na + (C 2 H 5 ) 2 S0 4 — > R-C0 2 C 2 H 6 + C 2 H 5 NaS0 4

The sodium alkyl sulphates of the higher alcohols are used as detergents,
e.g., sodium lauryl sulphate:

Alkyl nitrates, R-ON0 2 . The only important alkyl nitrate is ethyl
nitrate, C 2 H 5 -ON0 2 . This may be prepared by heating ethyl iodide with
silver nitrate in ethanolic solution :

C 2 H 5 I + AgN0 3 — > C 2 H 5 -ON0 2 + Agl

When concentrated nitric acid is added to ethanol, the reaction is usually
violent ; part of the ethanol is oxidised, and some of the nitric acid is reduced
to nitrous acid. Apparently it is the presence of the nitrous acid which
produces the violent oxidation of the alcohol by the nitric acid. This
danger may be avoided by first boiling the nitric acid with urea, which
destroys any nitrous acid present, and then adding this mixture to cool
ethanol, any nitrous acid produced being immediately destroyed by the
urea:

CO(NH 2 ) 2 + 2 HN0 2 — > C0 2 + 2N 2 + 3 H 2

Ethyl nitrate is a pleasant-smelling liquid, b.p. 87-5°. When reduced
with tin and hydrochloric acid, it forms hydroxylamine and ethanol :

C 2 H 5 -0-N0 2 + 6[H] i^!> C 2 H 5 OH + NH 2 OH + H 2

Alkyl nitrites, R-ONO. Alkyl nitrites are isomeric with the nitro-
paraffins (p. 301). The only important alkyl nitrites are the ethyl and amyl
nitrites; the latter is actually mainly isoamyl nitrite, since the amyl alcohol
used is the wopentanol from fusel oil (p. 135). Ethyl and amyl nitrites are
prepared by adding, concentrated hydrochloric acid or sulphuric acid to
aqueous sodium nitrite and the alcohol, e.g.,

C 5 H u OH + HN0 2 — ► CjHu-ONO + H 2 (75-85%)

Ethyl nitrite, b.p. 17 , and amyl nitrite, b.p. 99°, are pleasantrsmelling
liquids, and are used as a means of preparing nitrous acid in anhydrous
media; e.g., an ethanolic solution of nitrous acid may be prepared by passing
dry hydrogen chloride into amyl nitrite dissolved in ethanol.

Trialkyl phosphates, (RO) 3 PO, may be prepared by refluxing an alcohol
with phosphoryl chloride in the presence of pyridine :

3ROH + POCl 3 + 3 C 5 H 5 N — > (RO) 3 PO + 3 C 6 H 5 N-HC1 (g.)

Triethyl phosphate, tributyl phosphate and tricresyl phosphate are widely
used as plasticisers.

Trialkyl borates, (RO) 3 B, may be prepared by fractionally distilling a
mixture of an alcohol and boric acid (Thomas, 1946) :

3ROH + H 3 B0 3 — > (RO) 3 B + 3 H 2 (ex.)

FATTY ACIDS 197

Trialkyl borates may be conveniently prepared in good yield by trans-
esteriiication (alcoholysis) of methyl borate (which is readily available) with
the appropriate alcohol (Brown et al., 1956).

4 (CH 3 0) 3 B + 3ROH — * (RO) 3 B + 3 [CH 3 OH,(CH 3 0) 3 B].

These authors have also shown that the following reaction readily occurs
with primary and secondary alcohols:

3ROH + NaBH 4 + CH 3 -C0 2 H — > (RO) 3 B + CH 3 -CO a Na + 4 H 2 .

With tertiary alcohols the reaction is:

2ROH + NaBH 4 + CH 3 -C0 2 H — > (RO) 2 BH + CH 3 -C0 2 Na + 3H 2

Tributyl borate is mainly used as a plasticiser.

ACID OR ACYL CHLORIDES

The general formula of the acyl radicals is R— C— . Acid chlorides,
which may be prepared by the replacement of the hydroxyl in the carboxyl
group by chlorine, are also known as acyl chlorides because they contain the

acyl radical. Their general formula may therefore be written R—C^ .

Nomenclature. According to the I.U.P.A.C. system the class suffix of the
acyl chlorides is -oyl, and the names may be illustrated by the following
examples :

CH 3 < CH 3 -CH 2 -C^

\ci X C1

ethanoyl chloride propanoyl chloride

This system is rarely used for naming the acyl chlorides. The more common
names are formed by changing the suffix -ic of the trivial name of the acid
into -yl, e.g.,

CH 3 -cf (CH 3 ) 2 CH-C0C1

acetyl chloride j'sobutyryl chloride

If the carboxyl group is considered as a substituent, then according to the
I.U.P.A.C. system, the nomenclature of all substances containing acyl
radicals is in all cases based on the name " carbonyl " for the CO group, e.g.,

CH 3 -CH 2 -CH 2 -COCl propane-i-carbonyl chloride

When naming amides (see below), the "yl" is elided before the suffix
amide, e.g.,

CH 3 'CH 2 'CH 2 -CONH 2 propane-i-carbonamide

General methods of preparation. 1. The acid is heated with phosphorus
trichloride or pentachloride, e.g.,

3R-COOH + PC1 3 — > 3R-COC1 + H 3 P0 3 (g.)

R-COOH + PC1 5 — > R-COC1 + HC1 + POCl 3 (v.g.)

I98 ORGANIC CHEMISTRY

The reaction with phosphorus trichloride is accompanied by the formation
of small amounts of volatile phosphorus compounds, e.g.,

R-C0 2 H + PC1 3 — > R-C0 2 PC1 2 + HC1

In some cases the acid anhydride is also formed due to the following reaction :

R-C0 2 H + R-COC1 ^ (R-CO) 2 + HC1

Thionyl chloride may be used instead of the phosphorus chlorides :

R-C0 2 H + SOCl 2 — > R-COC1 + S0 2 + HC1 (v.g.)

The inorganic chloride is chosen according to the boiling point of the acyl
chloride formed. Phosphorous acid decomposes at 200 ; the boiling point
of phosphoryl chloride is 107°, and that of thionyl chloride is 76 . Since
acetyl chloride boils at 52 , any of the three inorganic halides may be used,
but it is difficult to separate acetyl chloride from thionyl chloride (which
is generally used in excess) by fractionation. «-Butyryl chloride boils at
102 , and so phosphorus pentachloride cannot be used. Usually thionyl
chloride is the most convenient, but although it may be used with all the
monocarboxylic acids, it is not satisfactory for all dicarboxylic acids (p. 377).

2. By distilling the salt of the acid with either phosphorus trichloride,
phosphoryl chloride or sulphuryl chloride, e.g.,

3CH 3 -C0 2 Na + PC1 3 — > 3 CH 3 -C0C1 + Na 3 P0 3
2CH 3 -C0 2 Na + POCI3 — > 2 CH 3 -C0C1 + NaCl + NaP0 3
(CH 3 -C0 2 ) 2 Ca + S0 2 C1 2 — > 2CH 3 -C0C1 + CaS0 4

This method is used industrially since the salts are cheaper than the acid.

Acyl bromides may be prepared by the action of phosphorus tribromide or
pentabromide (red phosphorus and bromine) on the acid, or by the action of
excess hydrogen bromide on the acyl chloride :

R'COCl + HBr ^ R-COBr + HC1

Acyl iodides are prepared in the same way as the bromides. Acyl fluorides are
also known, and may be prepared by the action of hydrogen fluoride on acid
anhydrides :

(R-CO) 2 + HF >- R-COF + R-C0 2 H

General properties and reactions of the acyl chlorides. The lower acyl
chlorides are colourless liquids with irritating odours; the higher members
are colourless solids. The chlorine atom is very reactive and so the acid
chlorides are important reagents.

1. The acyl chlorides are readily hydrolysed by water, the lower members
reacting vigorously:

R-COC1 + H 2 — > R-C0 2 H + HC1

Acyl chlorides usually react rapidly with compounds containing " active "
hydrogen atoms, i.e., hydrogen attached to oxygen, nitrogen or sulphur;
e.g., esters are formed with alcohols:

R-COC1 + R'OH — > R-COOR' + HC1

Amides are formed with ammonia, and iV-substituted amides with primary
and secondary amines :

R-COC1 + 2NH 3 — > R-CONH 2 + NH 4 C1
R-COC1 + R'-NH 2 — ^ R-CONH-R' + HC1

FATTY ACIDS 199

Hydrazides are formed with hydrazine, and hydroxamic acids with hydroxyl-
amine :

R-COC1 + H 2 N-NH a — > R-CONH-NH 2 + HC1
R-C0C1 + NH 2 -OH — > R-CONH-OH + HC1

The mechanisms of all of these reactions are uncertain. Alcoholysis of acid
chlorides has been studied in some detail, and some mechanisms that have been
suggested are (i) and (ii), which differ in the nature of the intermediate (I) and
the transition state (II).

(i)

(ii) R— COC1 + R'OH

According to Hudson et al. (1955), the transition state is best represented as
intermediate between (I) and (II), and is stabilised by solvation. It is possible
that the other reactions— hydrolysis, ammonolysis, etc. — follow similar paths.

2. Acyl chlorides may be reduced catalytically to aldehydes or to alcohols:

r-COCI -^> R-CHO -^> R-CH 2 OH

Acid chlorides are reduced to esters of an enediol by sodium amalgam in an
inert solvent, e.g., ether (cf. esters, p. 194) :

R\ /R

4RCOCI + 4Na — > ) c==c \ + 4 NaC1

R-CO-O/ N>COR

Under suitable conditions, small yields of oc-diketones have been obtained.
Acid chlorides are reduced to alcohols by lithium aluminium hydride or sodium
borohydride, and to aldehydes by lithium hydride or lithium tri-«-butoxy-
aluminium hydride (p. 148).

3. Acyl chlorides react with the sodium salt of the fatty acids to form
acid anhydrides (q.v.) :

R-COC1 + R'-COONa — > R-COOCOR' + NaCl

4. Acyl chlorides may be used in the Friedel-Crafts reaction to produce
an aromatic ketone; e.g., acetyl chloride reacts with benzene in the presence
of anhydrous aluminium chloride to form acetophenone :

C 6 H 6 + CH 3 -COCl ^V C 6 H 5 -COCH 3 + HC1

5. Acyl chlorides react with Grignard reagents to produce ketones or
tertiary alcohols, according to the conditions (see p. 355); e.g., acetyl
chloride forms butanone with ethylmagnesium iodide:

CH s -COCl + C 2 H 5 -Mg-I — >. CH S .C0-C 2 H S + Mg-Cll

6. Acyl chlorides react with carboxylic acids as follows:

R-C0C1 + R'-CO a H ^ R-C0 2 H + R'-COCl

200 ORGANIC CHEMISTRY

If the acyl chloride, R'*C0C1, has the lowest boiling point, and the apparatus
is arranged so as to allow only this to distil, all of R-C0C1 will be converted
into R'-COCl. This reaction may therefore be used to prepare a volatile
acyl chloride.

A small amount of acid anhydride is formed as a by-product, due to the
reaction:

R-C0C1 + R'-C0 2 H ^ R-COOCOR' + HC1

There will also be present the anhydrides (R-CO) 2 and (R'-CO) 2 0.

7. Acyl chlorides are readily halogenated in the a-position (see p. 210).

8. Acyl chlorides form esters (and other products) when heated with an
ether in the presence of anhydrous zinc chloride as catalyst :

R-C0C1 + R' 2 -^% R-C0 2 R' + R'Cl

9. Acyl chlorides add on to the double bond of an olefin in the presence
of a catalyst, e.g., zinc chloride or aluminium chloride, to form a chloro-
ketone which, on heating, eliminates a molecule of hydrogen chloride to
form an unsaturated ketone:

CH 3 CH 3

C=CH 2 + CH a -COCl^>CH,— C- "" heat

CH 3 — C=CH 2 + CH 3 -C0C1 '-> CH 3 — C— CH 2 -CO-CH ;

CI

CH 3

CH 3 — C=CH-COCH 3 + HC1

Formyl chloride, H-C0C1, and formyl fluoride, H-COF, are said to exist
at low temperatures (—80°). There is no evidence of their existence at
ordinary temperature, but a mixture of carbon monoxide and hydrogen
chloride behaves as if it were formyl chloride in the Gattermann-Koch
aldehyde synthesis (p. 646).

Acetyl chloride, CH 3 *C0C1, is the most important acyl halide. It is a
colourless fuming liquid, b.p. 52°, and is soluble in ether and chloroform.
It is largely used as an acetylating agent, i.e., as a means of introducing the
acetyl group, into compounds containing " active " hydrogen atoms. It
is also used to detect the presence of hydroxyl groups in organic compounds,
and to estimate their number.

Acetyl chloride readily acetylates primary and secondary alcohols, but it
tends to replace the hydroxyl group of a tertiary alcohol by chlorine (p. 130) :

R 3 COH + CH 3 -C0C1 — > R3CCI + CH 3 -C0 2 H

The acetate of a tertiary alcohol may, however, be obtained by carrying out
the reaction in the presence of pyridine :

R 3 COH + CH 3 -C0C1 H^.% CH 3 -CO a CR 3 + HC1

The accepted abbreviation for the acetyl group is Ac; e.g., AcOH is
acetic acid; Ac 2 0, acetic anhydride; AcCl, acetyl chloride; C 6 H S 'NH-Ac,
acetanilide; etc.

FATTY ACIDS 201

ACID ANHYDRIDES

The acid anhydrides may be regarded theoretically as being derived
from an acid by the removal of one molecule of water from two molecules

of the acid:

R— C dH K U \

-h,o )o

R— C (OH

It is possible in practice to prepare many acid anhydrides by the direct
dehydration as indicated above, but this method is usually confined to the
anhydrides of the higher members of the acid series (see below).

Nomenclature. The acid anhydrides are named as the anhydride of the
acid radicals present, the trivial name of the acid being used, e.g.,

(CH 3 CO) 2 acetic anhydride

(CH 3 -CH 2 -CO) 2 propionic anhydride

The most important acid anhydride is acetic anhydride, but propionic
anhydride is increasing in importance due to its use in preparing cellulose
propionate. Formic anhydride is unknown, but mixed anhydrides contain-
ing the formic acid radical have been prepared, e.g., acetic formic anhydride,
H-COOCOCH 3 .

Acetic anhydride, (CH 3 -CO) 2 0, may be conveniently prepared by dis-
tilling a mixture of anhydrous sodium acetate and acetyl chloride:

CH 3 -C0 2 Na + CH 3 -C0C1 — > (CH 3 -CO) 2 + NaCl (80%)

Acetic anhydride is prepared industrially :

(i) By heating anhydrous sodium acetate with sufficient inorganic chloride
— phosphoryl chloride, thionyl chloride or sulphuryl chloride— to convert
half of the sodium salt into the acid chloride, which then reacts with un-
changed sodium acetate to form acetic anhydride.

(ii) By passing chlorine into a mixture of sodium acetate and sulphur
dichloride, and distilling:

8CH 3 -C0 2 Na + SC1 2 + 2C1 2 — > 4(CH 3 -CO) 2 + 6NaCl + Na 2 S0 4

(iii) By passing acetylene into glacial acetic acid in the presence of
mercuric ions as catalyst, and distilling the resulting ethylidene acetate:

heat

TTcr s +" _ _ ..—— . neat

C 2 H 2 + 2CH 3 -C0 2 H -^-> CH 3 .CH(OC0-CH 3 ) 2 >

(CH 3 -CO) 2 + CH 3 -CHO

(iv) By passing keten into glacial acetic acid:

CH 2 =C=0 + CH 3 -C0 2 H — >- (CH 3 -CO) 2

(v) By passing acetic acid vapour over a catalyst consisting of a mixture
of sodium ammonium hydrogen phosphate and boron phosphate, at 600-
620 :

2CH 3 -C0 2 H -^ (CH 3 -CO) 2

Anhydrides of the higher acids may be prepared by heating, and then
fractionating, a mixture of the acid and acetic anhydride:

2 R-C0 2 H + (CH 3 -CO) 2 ==i (R-CO) a O + 2CH 3 -C0 2 H

202 ORGANIC CHEMISTRY

This method is only satisfactory for anhydrides which have higher boil-
ing points than acetic acid. A much better method of preparation is to
treat the acid chloride with pyridine and benzene, add the acid, and then
heat:

R-C0C1 + R-C0 2 H + C 5 H 5 N — ■> (R-C0) 2 + C 5 H 5 N-HC1 (g.-ex.)

On the other hand, acid anhydrides may be prepared in very high yield by the
action of thionyl chloride on an ethereal solution of a mixture of acid and pyridine
(Gerrard et al., 1952) :

2R-C0 2 H + SOCl 2 + 2C 5 H 5 N > (R-CO) 2 + C 6 H 5 N-HC1 + C 5 H 5 NH-OSOC1

Properties of acetic anhydride. Acetic anhydride is a colourless liquid,
b.p. 139-5°, with an irritating smell. It is neutral when pure, and is slighty
soluble in water, but readily soluble in ether and benzene. It is hydrolysed
slowly in water, but rapidly by alkali:

(CH 3 -CO) 2 + H 2 — > 2CH 3 -C0 2 H

It undergoes reactions similar to those of acetyl chloride, but with less
vigour: only half of the acetic anhydride molecule is used in acetylation,
the other half being converted into acetic acid :

(CH 3 -CO) 2 + C 2 H 5 OH — > CH 3 -C0 2 C 2 H 6 + CH 3 -CO a H
(CH 3 -CO) 2 + NH 3 — > CH 3 -CONH 2 + CH 3 -C0 2 H

Acetylation with acetic anhydride is usually best carried out in the presence
of a small amount of sodium acetate or concentrated sulphuric acid as
catalyst. Acetic anhydride reacts with dry hydrogen chloride to form
acetyl chloride :

(CH 3 -CO) 2 + HC1 — > CH 3 -C0C1 + CH 3 -C0 2 H

It is readily halogenated, and may be used in the Friedel-Crafts reaction.
It reacts with aldehydes to form alkylidene acetates; e.g., with acetaldehyde
it forms ethylidene acetate :

(CH 3 -CO) 2 + CH 3 -CHO — > CH 3 -CH(OCOCH 3 ) 2

Acetic anhydride reacts with nitrogen pentoxide to form acetyl nitrate,
CH 3 'COON0 2 . This is a colourless fuming liquid, b.p. 22°jyo mm., which
explodes violently if heated suddenly. It is hydrolysed by water to acetic
and nitric acid :

CH 3 -COON0 2 + H 2 — > CH 3 -C0 2 H + HN0 3

Acetyl nitrate is very useful for preparing certain aromatic nitro-compounds,
but it is dangerous to handle (see p. 553).

Acetyl peroxide, CH 3 -COOOCOCH 3 , may be prepared by the action of barium
peroxide on acetic anhydride. It is colourless, pungent-smelling liquid, b.p.
63721 mm., and tends to explode on warming. It is a powerful oxidising agent.

ACID AMIDES

Acid amides are compounds in which the hydroxyl of the carboxyl group
has been replaced by the amino-group, -NHg, to form the amido-group :

<

\NH 2 .

FATTY ACIDS 203

There are three classes of amides: primary, R-CONH 2 ; secondary,
(R-C0) 2 NH; and tertiary, (R-C0) 3 N. Only the primary amides are im-
portant, and it can be seen from their formulae that all three classes may be
regarded as the acyl derivatives of ammonia.

Nomenclature. According to the I.U.P.A.C. system of nomenclature, the
suffix -ok of the parent acid is replaced by amide (see also acid chlorides,
above) :

H*CONH 2 methanamide

CHg-CO-NHa ethanamide or methylcarbonamide

The amides, however, are commonly named by replacing the suffix -ic of
the trivial name of the parent acid by amide, e.g.,

H-CONH 2 formamide

CH 3 -CONH 2 acetamide

General methods of preparation of the amides, i. By heating the am-
monium salt of the acid:

R-COONH 4 — > R-CONH 2 + H 2 (g.-v.g.)

Since the ammonium salts tend to dissociate on heating, the reaction is best
carried out in the presence of some free acid R'C0 2 H which represses the
hydrolysis and the dissociation of the ammonium salt.

Amides may also be prepared by heating an acid with urea (Cherbuhez
etal., 1946):

R-C0 2 H + CO(NH 2 ) 2 — > R-CONH 2 + C0 2 + NH 3 (g.-v.g.)

2. By ammonolysis, i.e., the action of concentrated ammonia solution
on acid chlorides, acid anhydrides or esters :

R-C0C1 + 2NH3 — > R-CONH 2 + NH 4 C1 (v.g.)

(R-CO) 2 + 2NH3 — > R-CONH 2 + R-C0 2 NH 4 (v.g.)
R-COOR' + NH 3 — > R-CONH 2 + R'OH (g.)

JV-Substituted amides may be prepared by using primary or secondary
amines instead of ammonia, e.g.,

R-C0C1 + R'-NH 2 — > R-CONH-R' + HC1

Philbrook (1954) has shown that amides may be readily prepared by adding a
benzene solution of an acid chloride dropwise to benzene through which is
passed a current of ammonia; for the lower aliphatic amides the yield is 65-
95 per cent.

Meyer (1906) found that ammonolysis of esters occurs most readily with esters
of strong acids and least readily with esters of weak acids or sterically hindered
acids. It has been found that ammonia and aliphatic amines react, in general,
more readily than aromatic amines. Stern (1956) has shown that acyl aryl-
amides may be readily prepared as follows (Ar = aryl radical) :

Na or

Ar-NH. >► Ar-NHNa

' NaNH,

R-C0 2 C 2 H 5 + Ar-NHNa — > R-CO-NH-Ar + C 2 H 5 ONa

3. By the graded hydrolysis of alkyl cyanides. The hydrolysis must be
carried out carefully, since if taken too far, the acid is produced. The
hydrolysis may be carried out satisfactorily by dissolving the alkyl cyanide

204 ORGANIC CHEMISTRY

in concentrated sulphuric acid and then pouring the solution into cold water,
or by shaking the alkyl cyanide with cold concentrated hydrochloric acid:

R-C=N + H 2 — > R-CONH 2 {g.)

A possible mechanism for this hydrolysis is :

R-C=N: + H — >- R-CEEN— H «->- R-C=N— H H '°>

6h 2 OH O

R-C=N— H ~ H+ > R-C=NH ^=s R-C— NH 2

Sometimes the conversion of an alkyl cyanide into the amide may be
effected by means of alkaline hydrogen peroxide :

R-CN + 2 H 2 O a -^> R-CONH 2 + 2 + H 2

General properties and reactions of the amides. Except for formamide,
which is a liquid, all the amides are colourless crystalline solids, and those
of low molecular weight are soluble in water (with which they can form
hydrogen bonds). The lower amides have much higher melting points
and boiling points than are to be expected from their molecular weights;
this indicates association (through hydrogen bonding) :

R R R

H 2 N— C=0 - - - H— NH— C=0 - - - H— NH— C=0 - - -

Infra-red spectroscopic studies indicate that in dilute solution (dioxan and
chloroform) amides are monomelic; in concentrated solution (chloroform)
and in the liquid state there is some association, and in the solid state
association is complete. Furthermore, the spectroscopic evidence appears to
indicate that amides are resonance hybrids (Richards and Thomson, 1947) :

R— C( . . ^-> R— CI +
X NH 2 ^nh 2

1. Amides are hydrolysed slowly by water, rapidly by acids, and far more
rapidly by alkalis :

R-CONH 2 + H 2 — > R-C0 2 H + NH 3

Possible mechanisms are:

Base hydrolysis

AO 6p O O

HO + C— NH 2 5pi HO— cQ*H 2 ^=± HO— C + NH, — >■ 6— C + NH 3

R R R R

Acid hydrolysis
O >o 6

H + L|l + H.O +Stj,Ow-

O— NH, + H ^=±: C— NH, ^=i H.O-h:— NH. ^

I II

R R R O O

H a O— C + NH 3 > HO— C + NH/

R R

FATTY ACIDS 205

2. Amides are very feebly basic and form unstable salts with strong
inorganic acids, e.g., R-C0-NH 2 -HC1. The structure of these salts may be
I or II:

L x nh 2 J L x nhJ

(i) (ii)

In the case of formamide (R=H), infra-red measurements have shown
that II is the actual state. Addition of a proton to the NH 2 group inhibits
resonance and consequently decreases stability, i.e., the amide is more stable
than the protonated molecule. Hence the weak basic properties of amides.

3. Amides are also feebly acidic; e.g., they dissolve mercuric oxide to
form covalent mercury compounds in which the mercury is probably linked
to the nitrogen:

2 R-CONH 2 + HgO — > (R-CO-NH) 2 Hg + H 2

"When amides are treated with sodium or sodamide in ethereal solution, the
sodium salt, [R-CO-NH]-Na+, is formed, and the structure of these salts may be
I or II :

[r-c'-nhJ Na+ U-c=nhJ

Na+

(I) (ID

When the dry sodium salt is heated with an alkyl iodide, the itf-alkyl derivative
is obtained; this corresponds to I. When, however, the silver salt is used, the
O-alkyl derivative is obtained; this corresponds to II. It is possible that I and
II are resonating structures.

4. Amides are reduced by sodium and ethanol, catalytically, or by lithium
aluminium hydride to a primary amine:

R-CO-NH 2 + 4 [H] Na/W ° H > R-CH 2 -NH 2 + H 2

5. When heated with phosphorus pentoxide, amides are dehydrated to
alkyl cyanides:

R-CO-NH 2 4% R-C-N

Alkyl cyanides are also formed when the amides of the higher fatty acids
are heated to a high temperature:

2 R-CO-NH 2 -^ R-CN + R-C0 2 H + NH 3

If this reaction is carried out in the presence of excess of ammonia, then all
the amide is converted into alkyl cyanide:

R-C0 2 H + NH 3 -^ R-C0 2 NH 4 — % R-CO-NH 2 ^> R-CN

Amides may also be converted into cyanides by phosphorus pentachloride, and
according to Wallach (1877), the reaction proceeds via the amido-chlonde I and
then the itnido-chloride (imino-chloride) II.

R. C ONH 2 ^V R-CC1 2 -NH 2 ^V R'CC1=NH ^V R-CN
(I) (II)

According to Kirsanov (1954). however, the reaction proceeds as follows, the
intermediate III being decomposed by heat or by hydrogen chloride.

R-CONH 2 + PC1 5 > 2HCI + R-CO-N=PCl 3 > R-CN + POCl 3

(III)

206

ORGANIC CHEMISTRY

Alkyl cyanides may be obtained from amides by treating the amide with the
boron trifluoride-amide complex in the presence of a small amount of a carboxylic
acid (Nieuwland, 1937), e.g.,

CT-T *CO H

CH 3 -CONH 2 -BF 3 + CH 3 -CONH 2 ' ' > CH 3 -CN

CH 3 -C0 2 H + NH 3 -BF 3

It is also interesting to note that the complex CH 3 *CONH 2 'BF 3 has been found
to be an effective acetylating agent for alcohols and phenols :

CH 3 -CONH 2 -BF 3 + ROH > CH 3 -C0 2 R + NH 3 -BF 3

6. When amides are treated with nitrous acid, nitrogen is evolved and
the acid is formed:

R-CONH 2 + HNO a — > R-C0 2 H + N 2 + H 2

7. Hofmann reaction or Hofmann degradation (1881). The Hofmann
reaction is the conversion of an amide into a primary amine with one carbon
atom less by means of bromine (or chlorine) and alkali. The overall equation
for the reaction may be written :

R-CONH 2 + Br 2 + 4KOH — > R-NH 2 + 2KBr + K 2 CO s + 2H 2

The mechanism of the Hofmann reaction has been the subject of a great deal
of work. A number of intermediates have been isolated: JV-bromamides,
R-CONHBr; salts of these bromamides, [R'CO'NBr] _ K + ; *'socyanates, R'NCO.
On this basis, a possible mechanism is the following, which is an example of the
1,2-shift:

/Br

R\

,NH,

Rs

->

NO

II
O

,-NC,

rR

^H

N— Br -1

L O

K+^>

O

(I)

-> C=NR _2i

H.O
■->■

R-NH,

co 2

O

o

(16)

The uncertainty of this mechanism lies mainly in the formation of (I). Hauser
et al. (1937) have shown that electron-releasing groups in the ^ara-position of a
migrating aryl group accelerate the reaction and electron-withdrawing groups
retard the reaction. This supports the formation of a
bridged ion (a phenonium ion), (la), certainly when the
migrating group is aromatic. Whether a bridged ion,
(I&), is produced when the migrating group is an alkyl
group, is uncertain (cf. p. 171). Furthermore, there is
much recent work to show that the rearrangement is
intramolecular, e.g., Wallis et al. (1931), using the
amide of optically active a-benzylpropionic acid, found
that there was retention of configuration in the product,
oc-benzylethylamine ; thus the migrating group is never free :

C,H 5 «CH 2 -CHMe-CONH 2 > C e H 6 -CH 2 -CHMe-NH 2

The Hofmann reaction can be carried out on all monocarboxylic acids,
and the yields are 70-90 per cent, for amides containing up to seven carbon
atoms; amides with more than seven carbon atoms in the chain give mainly
alkyl cyanide:

. R-CH 2 -CONH 2 Bra/K ° H > R-CN

The alkyl cyanide, however, may readily be reduced to the primary amine
R , CH 2 , NH 2 .

FATTY ACIDS 207

It is possible to obtain the amine in good yield from the long-chain amides
by modifying the procedure as follows. Bromine is added rapidly to a
methanolic solution of the amide containing sodium methoxide :

R-CO-NHjj + 2CH 3 0Na + Br 2 — >■ R-NH-C0 2 CH 3 + 2NaBr + CH 3 OH

The iV-alkyl-urethan, on hydrolysis with alkali, gives the amine:

R-NH-C0 2 CH 3 + 2NaOH — >- R-NH 2 + Na 2 C0 3 + CH 3 OH

The Hofmann reaction offers an excellent method of preparing primary
amines free from secondary and tertiary amines ; it also affords a means of
descending a series. A further interesting point about this reaction is that
if the amide contains an electronegative group, the product is a bromide,
e.g., heptafluorobutyramide gives bromoheptafluoropropane (Haszeldine
et al., 1956) :

C 3 F 7 -CO-NH alj ^C 3 F 7 Br (85%)

HYDROXAMIC ACIDS

The hydroxamic acids exhibit tautomerism; the keto form I is known as the
hydroxamic form, and the enol form II as the hydroximic form :

O OH

(I) II I (II)

R—C— NH-OH ^ R— C=N-OH

Hydroxamic acids may be prepared by the action of hydroxylamine on esters
or acid chlorides:

R— C< + NH 2 -OH — > R—C— NH-OH + C 2 H 5 OH

X OC 2 H 5

R-COC1 + NHj-OH >■ R—C— NH-OH + HC1

Hydroxamic acids give a red coloration with ferric chloride solution, a reaction
which is characteristic of enols. When treated with a strong inorganic acid,
hydroxamic acids undergo the Lossen rearrangement (1875), which results in the
formation of a primary amine. The mechanism of the Lossen rearrangement
is closely related to that of the Hofmann reaction, and so may be formulated :

H H

I I ^ + *^

R s ,N— OH R. y N— OH.+ „ „ R. ,N— H

c /t. w Hci x Xp/.: ~" a -H a o "x^/r: " -h+

-> Tr >

o o

&

R \^/ N: ho

\yL' ; > C=NR ' > RNH, + co 2

,|AU 11 — >

o o

This rearrangement (1,2-shift) has been shown to be intramolecular, and electron-
releasing groups in a migrating aryl group accelerate the reaction, thereby
supporting the formation of a phenonium ion.

The amides of the hydroxamic acids are known as amidoximes. These are
tautomeric substances :

/NH 2 #H

R-Qf ^ R-Cf

^N-OH \NH-OH

208 ORGANIC CHEMISTRY

ac

Aliphatic amidoximes are best obtained by the action of hydroxylamine on an
alkyl cyanide :

R-CN + NH 2 -OH y R-CT

^N-OH

IMIDIC ESTERS AND AMIDINES
Imidic esters, which are also known as imino-ethers, are best prepared by pass-
ing dry hydrogen chloride into a solution of an alkyl cyanide in anhydrous
alcohol; the imidic ester hydrochloride is slowly deposited (Pinner, 1877) :

x-NH-HCl

R-CjN + R'OH + HC1 y R-Cf ig.-v.g.)

\OR'

This reaction is believed to be analogous to the hydrolysis of cyanides to
amides:

R-CjN + H a O — y I R—cf 1 — y R-C<f *

M3HJ ^O

Tr-c^

There is, however, no evidence for the existence of the enol form in unsubstituted
amides.

Imidic esters form ortho-esters when allowed to stand in an alcohol:

^NH-HCl

R-Cf + 2 R'OH > R-C(OR') 3 + NH 4 C1

\OR'

Imidic esters are readily hydrolysed to esters :

^NH-HC1 /y O

R-Cf + H a O — > R— Cf + NH 4 C1

x OR' \OR'

When an ethereal solution of an imidic ester hydrochloride is treated with

potassium carbonate, the free ester R-C^q^, is obtained. When treated with an

ethanolic solution of ammonia, the imidic ester hydrochloride forms the amidine:

^NH-HC1 />NH

R-CT + 2 NH 3 — y R-Cf + NHX1

X)R' \NH 2

Formamidine acetate and acetamidine acetate may be prepared from ortho
esters (R = H or Me) as follows (Taylor et al., i960) :

R-C(OEt), + NH 3 + CH 3 -C0 2 H > H 2 N-CR=NH 2 +CH 3 -C0 2 -

These acetates are better precursors of their respective amidines than are the
corresponding hydrochlorides.

Amidines are strong monoacid bases, forming salts with strong acids. Their basic
strength may be explained by resonance. In the amidine, two different resonat-
ing structures contribute to the actual state of the molecule :

/•NH /NH

R-C( «^> R-< +

\NH 2 %H 2

In the amidine ion, the resonating structures are equivalent, and hence the
resonance energy is a maximum; the ion will therefore be stable :

^H s . /NH,

R— <X -<->R— Cx.

\NH 2 ^NH a

Amidines are readily hydrolysed to amides :

/NH , Q

R— C( + H 2 — > R—Cr + NH 3

H 2 \NH 2

FATTY ACIDS 200.

Amidines or their salts may be reduced to aldehydes by sodium and ethanol
in liquid ammonia, or by sodium and ethanol, but in the latter case the yields
are lower (Birch et al., 1954).

ACID HYDRAZIDES AND ACID AZIDES

Acid hydrazides may be prepared by the action of hydrazine on esters or acyl
chlorides :

R-COOC 2 H 5 + N a H 4 y R-CONH-NH 2 + C 2 H 6 OH

R-COC1 + N 2 H 4 y R-CONH-NH 2 + HC1

Girard's reagent " T " (p. 154) is trimethylaminoacetohydrazide, which may be
prepared by interaction between trimethylamine, ethyl chloroacetate and
hydrazine :

(CH 3 ) 3 N + C1CH 2 -C0 2 C 2 H 6 + N 2 H 4 >

[(CH 3 ) 3 N-CH 2 -CO-NH-NHJ+Cl- + C 2 H 5 OH

Girard's reagent " P " is formed in a similar manner, except that pyridine is used
instead of trimethylamine.

The acid hydrazides resemble the acid amides, but differ in the following ways :

(i) They are much more readily hydrolysed than the amides :

R-CONH-NH 2 + H a O > R-C0 2 H + N 2 H 4

(ii) They are reducing agents — hydrazine is a powerful reducing agent,
(hi) They form acid azides when treated with nitrous acid; no nitrogen is
evolved :

R-CONH-NH 2 + HNO a > R-CON 3 + 2 H a O

Acid azides may also be prepared by the reaction between an acyl chloride and
sodium azide:

R-COC1 + NaN 3 — > R-CON 3 + NaCl

The structure of the acid azides is best represented as a resonance hybrid :

O O

R— c— n=n=n: -<— >• r— c— i

-n— nsen:

When boiled with an alcohol, acid azides undergo rearrangement to form
AT-alkyl-substituted urethans :

R-CON 3 + R'OH y R-NH-CO a R' + N 2 (v.g.-ex.)

The mechanism of this rearrangement — the Curtius rearrangement (1894) — i s
similar to that of the Hofmann and Lossen rearrangements (see above), and so
may be formulated :

•-rv+
RXc/N-n^ _„.

o "l o

i^-g

R— N

/ x R'OH

C y R-NH-C0 2 R'

II
O

This mechanism is supported by the fact that when the terminal nitrogen
atom is 15 N in 3 : 5-dinitrobenzazide, all of this tracer is found in the nitrogen
eliminated in the reaction (Bothner — By et al., 1951).

The Curtius reaction may be used to step down a series (cf. the Hofmann
reaction) :

R.C0 2 H -™V R-CO a C 2 H 5 **> R-CO-NH-NH, ^2^

R.CON3 ^!^> R-NH-CO a CH 3 i^V R-NH,

The Curtius reactions offers a very good method for preparing primary amines
free from secondary and tertiary amines, and may also be used for preparing

210 ORGANIC CHEMISTRY

isocyanates and urethans; e.g., when heated in benzene or chloroform solution,
an acid azide rearranges to the alkyl wocyanate which can be isolated. If the
wocyanate is warmed with an alcohol, an iV-substituted urethan is formed :

R . C ON s benzen % R-NCO ^> R-NH-CO a R'

solution

HALOGEN DERIVATIVES OF THE FATTY ACIDS

The halogen derivatives of the fatty acids are compounds in which one
or more hydrogen atoms in the carbon chain have been replaced by halogen.

Nomenclature. The usual method of naming the halogenated fatty acids
is to use the trivial names of the acids, and to indicate the positions of the
halogens atoms by Greek letters, e.g.,

CH 2 C1-C0 2 H monochloroacetic acid

CH 3 -CHC1 # C0 2 H a-chloropropionic acid

(CH 3 ) 2 CCl-CHBr-C0 2 H a-bromo-p-chlorowovaleric acid

Preparation of the halogenated acids. There are no general methods of
preparing the various types of halogenated acids, i.e., the a-, (3-, y-, etc.,
halogeno-acids; the method depends on the position of the halogen atom.

a-Halogeno-acids. Although the fatty acids themselves are not readily
halogenated, their acid chlorides and anhydrides may be halogenated very
easily. Bromination takes place only at the a-position, but chlorination,
although it occurs mainly at the a-position, may also take place further in
the chain; e.g., chlorination of propionic acid results in the formation of
the a- and p-chloro-derivatives :

CH 3 -CH 2 -C0 2 H — ^- > CH 3 -CHC1-C0 2 H + CH 2 C1-CH 2 -C0 2 H

Fluorination of the fatty acids has not yet been studied in great detail,
but the work done so far seems to indicate that fluorine enters the chain
somewhat indiscriminately; e.g., fluorination of butyric acid in carbon
tetrachloride solution gives the (3- and y-fluoro-derivatives (Bockemuller,

1933)-

Iodo-derivatives are prepared from the chloro- or bromo-compound by
the action of potassium iodide in methanolic or acetone solution (c/. alkyl
iodides, p. 105).

The usual method for preparing a-chloro- or bromo-acids is by the Hell-
Volhard-Zelinsky reaction (H.V.Z. reaction), which is carried out by treating
the acid with chlorine or bromine in the presence of a small amount of red
phosphorus. The reaction possibly takes place as follows:

R-CH 2 -C0 2 H -^-V R-CH 2 -COBr -^-> R-CHBr-COBr
R*CHBr-COBr + R-CH 2 -C0 2 H ^=s R-CHBr-C0 2 H + R-CH 2 -COBr

R-CH a -COBr-^-> R-CHBr-COBr, etc.

The second a-hydrogen atom may be replaced by chlorine or bromine
by using excess of the halogen, but whereas bromination ceases when both
a-hydrogen atoms have been replaced, chlorination proceeds further in the
chain (see above). Since the H.V.Z. reaction with bromine is specific for
a-hydrogen atoms, it can be used to detect the presence of a-hydrogen in
an acid; e.g., trimethylacetic acid does not undergo the H.V.Z. reaction
with bromine.

FATTY ACIDS 211

The H.V.Z. reaction with bromine is applicable to dibasic and polybasic
acids, all the a-positions being substituted if sufficient bromine is used.

Sulphuryl chloride, in the presence of a small amount of iodine, chlorinates
aliphatic acids in the a-position; but in the presence of organic peroxides,
the a-, (3-, y-, etc., positions are substituted.

Another convenient method of preparing a-bromo-acids is by brominating
an alkyl-malonic acid and heating the bromo-acid, whereupon it is decar-
boxylated to the monobasic acid :

R-CH(C0 2 H) 2 ^_ > R-CBr(C0 2 H) 2 ^> R-CHBr-C0 2 H + C0 2

p-Halogeno-acids may be prepared by treating an a^-unsaturated acid
with halogen acid; e.g., acrylic acid forms (3-bromopropionic acid when
treated with hydrogen bromide in acetic acid solution:

CH 2 =CH-C0 2 H + HBr — > CH 2 Br-CH 2 -C0 2 H

This addition takes place contrary to Markownikoff's rule, and may be
explained by the inductive effect of the carboxyl group (see p. 282).

(3-Halogeno-acids may also be prepared by treating an ap-unsaturated
aldehyde with halogen acid, and oxidising the (3-chloroaldehyde produced;
e.g., acraldehyde gives |3-chloropropionic acid when treated with concentrated
hydrochloric acid, and the product then oxidised with concentrated nitric
acid:

CH 2 =CH-CHO + HC1 — y CH 2 C1-CH 2 -CH0 — -->

CH 2 C1-CH 2 -C0 2 H (60-65%)

The addition of all the halogen acids to a (3-unsaturated carbonyl compounds
takes place very readily, and in a direction contrary to Markownikoff's
rule.

When an olefin cyanohydrin is refluxed with 40 per cent, hydrobromic
acid, the p-bromoacid is obtained; e.g., ethylene cyanohydrin gives (3-
bromopropionic acid :

CH 2 (OH)-CH 2 -CN ■+ 2HBr + H 2 — > CH 2 Br-CH 2 -C0 2 H + NH 4 Br

(82-83%)

Alternatively, when an olefin halohydrin is oxidised with concentrated
nitric acid, the halogeno-acid is produced; e.g., trimethylene chlorohydrin
gives (3-chloropropionic acid:

CH 2 C1-CH 2 -CH 2 0H — ^ CH 2 C1-CH 2 -C0 2 H (78-79%)

y-Halogeno-acids may be prepared by the addition of halogen acid to a
Py-unsaturated acid; the addition occurs contrary to Markownikoff's rule
(cf. above) ; e.g., pent-3-enoic acid gives y-bromo-w-valeric acid when treated
with hydrobromic acid :

CH 3 -CH=CH-CH 2 -C0 2 H + HBr — > CH 3 -CHBr-CH 2 -CH 2 -C0 2 H

y-Halogeno-acids may also be prepared by heating a y-hydroxyacid with
concentrated halogen acid solution; e.g., y-hydroxybutyric acid gives
y-chlorobutyric acid when treated with concentrated hydrochloric acid:

CH 2 OH-CH 2 -CH 2 -C0 2 H + HC1 — > CH 2 C1-CH 2 -CH 2 -C0 2 H + H 2

S-Halogeno-acids, etc., are prepared by methods which are usually specific
for the particular acid.

212 ORGANIC CHEMISTRY

Properties of the halogeno-acids. The a-halogeno-acids undergo most of
the reactions of the alkyl halides, but the halogen atom in the acid is far
more reactive than that in the alkyl halide; the enhanced reactivity is due
to the adjacent carbonyl group. The reactions of the carboxyl group are
unchanged. (3-, y- and 8-Halogeno-acids undergo some of the reactions of
the alkyl halides, but tend to eliminate a molecule of hydrogen halide to form
an unsaturated acid or a lactone (p. 396). They do not form Grignard
reagents, but the halogeno-acid esters react with Grignard reagents to form
halogen-substituted tertiary alcohols. The halogeno-acids are reduced to
the corresponding fatty acids by sodium amalgam, but the reduction of
acids of the type RR'CChC0 2 H with lithium aluminium hydride gives
chlorohydrins (I), alcohols (II), aldehydes (III) and glycols (IV).

RR'CCl-CH 2 OH RR'CH-CH 2 OH RR'CH-CHO RR'C(OH)-CH 2 OH

(I) (II) (III) (IV)

The amount of each depends on the nature of R and R', e.g., very little glycol
is obtained unless at least one of these radicals is phenyl (Eliel et al., 1956).
The behaviour of an halogeno-acid with alkali depends on the position
of the halogen atom relative to the carboxyl group.

(a) a-Halogeno-acids are converted into the corresponding a-hydroxy-
acid:

R-CHX-C0 2 H + H 2 ^V R-CH(OH)-C0 2 H

On the other hand, if the a-halogeno-acid ester is heated with a tertiary
amine, the a^-unsaturated acid ester is formed by the elimination of a
molecule of hydrogen halide; e.g., ethyl a-bromobutyrate gives ethyl croto-
nate when heated with dimethylaniline, C 6 H 5 \N(CH 3 ) 2 :

CH 3 -CH 2 -CHBr-C0 2 C a H 5 — ^> CH s -CH==CH-C0 2 C 2 H 5

(b) p-Halogeno-acids are converted into the corresponding (3-hydroxyacid,
which, on continued reflux with alkali, eliminates a molecule of water to
form the a(3-unsaturated acid:

R-CHBr-CH 2 -C0 2 H -^> R-CH(OH)-CH 2 -C0 2 H ^V

' R-CH=CH-C0 2 H

(c) y- and 8-Halogeno-acids are converted into lactones; e.g., y-chloro-
butyric acid gives y-butyrolactone :

CH 2 C1-CH 2 -CH 2 -C0 2 H ^V CH 2 -CH 2 -CH 2 -CO

-HCl ( Q 1

(d) e-Halogeno-acids, etc., give the corresponding hydroxyacid; e.g.,
E-bromocaproic acid gives e-hydroxycaproic acid:

CH 2 Br-CH 2 -CH 2 -CH 2 -CH 2 -C0 2 H ^^> CH 2 OH-CH 2 -CH 2 -CH 2 -CH 2 -C0 2 H

The most characteristic reaction of the a-halogeno-acid esters is the
Reformatsky reaction (p. 363).

The halogeno-acids are all stronger acids than the parent acid (see Table
III, p. 180), and for a group of isomeric acids, the further the halogen
is removed from the carboxyl group the weaker is the acid. The increase
in acid strength may be explained by the high electron-affinity of the

FATTY ACIDS 213

halogen atom which exerts a strong inductive effect, thereby facilitating
the release of proton from the carboxyl group.

O O

o o

x x J! CH *\ II

X^ CHf

O

X \ II

X-H><-<-*-C << < 0-<-<-<-H

X^

The larger the number of halogen atoms on the a-carbon atom, the stronger
is the inductive effect, and consequently the stronger is the acid; also, the
further removed the halogen atom is from the carboxyl group, the weaker
is the inductive effect at the carboxyl group, and consequently the weaker is
the acid. In the same way, since an alkyl group is electron-repelling, in-
creasing their number on the a-carbon atom should decrease the strength of
the acid since release of the proton will be hindered (see Table III). This
explanation, based on the purely inductive effect through the chain of
atoms, appears to be too simple. Work by Grob et al. (1955) has led them
to conclude that the direct (or field) effect (p. 16) is the decisive factor in
determining the strengths of the carboxylic acids.

Chloroacetic acid, CH 2 C1-C0 2 H, may be prepared by the H.V.Z. reaction;
the reaction can be carried out in direct sunlight, and in the absence of
phosphorus. Chloroacetic acid is prepared industrially: (i) by agitating
trichloroethylene with 90 per cent, sulphuric acid:

rr en

CHC1=CC1 8 + 2H 2 -^> CH 2 C1-C0 2 H + 2HCI
(ii) By the oxidation of ethylene chlorohydrin with nitric acid :

CH 2 C1-CH 2 0H^^CH 2 C1-C0 2 H (84%)

Chloroacetic acid is a deliquescent solid, m.p. 6i°, soluble in water and
ethanol. It finds many uses in organic syntheses, and is used in the industrial
preparation of indigotin (q.v.).

Ethyl chloroacetate is converted into chloroacetamide when shaken with
aqueous ammonia:

CH 2 Cl-COOC 2 H B + NH 3 — > CH 2 C1-C0-NH 2 + C 2 H 5 OH (78-84%)

Dichloroacetic acid, CHC1 2 *C0 2 H, may be prepared in the laboratory
and industrially by adding calcium carbonate to a warm aqueous solution
of chloral hydrate, then adding an aqueous solution of sodium cyanide,
and finally heating the mixture:

NaCN

2CC1 8 -CH(0H) 2 + 2CaC0 3 >■ (CHCl 2 -C0 2 ) 2 Ca + 2 C0 2 + CaCl 2 + 2H 2

-^> CHCl 2 -CO a H (88-92%)

The action of the sodium cyanide is not understood.

214 ORGANIC CHEMISTRY

Dichloroacetic acid is a liquid, b.p. 194 . When carefully hydrolysed
with dilute alkali, it gives glyoxylic acid (q.v.) :

CHC1 2 -C0 2 H + H 2 Na ° H > CHO-C0 2 H + 2HCI

Vigorous hydrolysis with concentrated alkali gives oxalate and glycollate,
due to the glyoxylic acid undergoing the Cannizzaro reaction (p. 160) :

2CHOC0 2 H + 3NaOH — > (C0 2 Na) 2 + CH 2 (OH)-C0 2 Na + 2H a O

Trichloroacetic acid, CC1 3 -C0 2 H, is best prepared by oxidising chloral
hydrate with concentrated nitric acid :

CC1 3 -CH(0H) 2 + [O] -^> CC1 3 -C0 2 H + H 2

Trichloroacetic acid is a deliquescent solid, m.p. 58°, and is one of the
strongest organic acids.

The presence of three chlorine atoms on a carbon atom adjacent to a
carbonyl group causes the C — C bond to break very easily. Thus when
trichloroacetic acid is boiled with dilute sodium hydroxide, or even water,
chloroform is obtained :

CC1 3 -C0 2 H — y CHC1 3 + C0 2

The ready fission of the C~C1 bond may be attributed to the strong inductive
effect of the chlorine atoms, and a possible mechanism is :

CI

ivT> H +

CI -<- C — CO a - > CO a + CC1-T > CHCI3

I
CI

When boiled with concentrated alkali, formates are produced, due to the
hydrolysis of the chloroform which is formed first (see p. 117).

Trifluoroacetic acid, CF 3 *C0 2 H, is conveniently prepared by the oxidation
of ^-trifluoromethyltoluidine with chromic acid:

CF 3 ^ ^NH 2 -™-> CF 3 -C0 2 H

Trifluoroacetic acid is a liquid that fumes in the air, and is one of the
strongest organic acids known. It does not form fluoroform when heated
with alkali, and is reduced to trifluoroacetaldehyde by lithium aluminium
hydride. Peroxytrifluoroacetic acid, CF 3 -C0 3 H, is a useful oxidising agent
(Emmons, 1954; cf. peracetic acid).

Fluoroacetic acid may be prepared by heating a mixture of carbon mon-
oxide, hydrogen fluoride and formaldehyde under pressure.

CO + HF + H-CHO — >■ CH 2 F-C0 2 H

The simplest chloro-acid would be chloroformic acid, ChC0 2 H, but it is
unknown ; chlorination of formic acid results in the formation of hydrogen
chloride and carbon dioxide :

H-C0 2 H— ^L> 2HCI + C0 2

On the other hand, esters of chloroformic acid are known, and they may be
prepared by the action of carbonyl chloride on an alcohol in the cold, e.g.,
ethyl chloroformate from carbonyl chloride and ethanol :

COCl 2 + C 2 H s OH — y Cl-COOC 2 H 5 + HC1

FATTY ACIDS 215

QUESTIONS

1. Write out the structures and the names of the isomeric acids having the molecular
formula C 6 H 12 2 .

2. Name the compounds and state the conditions under which they are formed when
AcOH is treated with: (a) EtOH, (b) PC1 3 , (c) PC1 6 , (d) SOCl 2 , (e) Br 2 , (/) SO a Cl 2 , (g)
KMnO„ (h) Hi/red P, (i) LiAlH 4 , (j) H 2 O a , (k) CH 2 N 2 , (/) C 2 H 4 , (m) HN„, (n) AcCl.

3. How would you determine the structures of n- and isobutyric acids?

4. Suggest a synthesis of each of the valeric acids, starting with compounds con-
taining not more than three carbon atoms.

5. Name the compounds and state the conditions under which they are formed
when EtOAc, AcCl and Ac a O, respectively, are treated with: — (a) H a O, (b) NaOH,

(c) HC1, (d) nascent hydrogen, (e) molecular hydrogen, (/) NH 3 , (g) m-BuOH, (h) PC1 6 ,
(i) Br 2 , (;') NaNH 2 , (k) Na, (I) NH 2 -OH, (m) N 2 H 4 , (0) benzene, (p) Et a O, (q) isobutene.

6. Name the compounds and state the conditions under which they are formed when
CH 3 -CO-NH 2 is treated with: — (a) NaOH, (6) HC1, (c) Na, (d) nascent hydrogen, (e)
molecular hydrogen, (/) P a 5 , (g) PC1 6 , (A) HNO a , (j) BF 3 , (j) Br 2 /KOH.

7. Define and give examples of: — (a) acetylation, (b) estenfication, (c) saponification,

(d) ammonolysis, (e) alcoholysis, (/) acidolysis, (g) acyloin condensation, (h) alkylation,
(i) Hofmann degradation, {j) Curtius reaction, (A) Lossen rearrangement, (/) H.V.Z.
reaction, (m) Rosenmund's reduction.

8. Describe the industrial preparations of: — (a) HCO a H, (fc) AcOH, (c) EtOAc, (d)
AcCl, (e) Ac a O, (/) Me a S0 4 , (g) Et a S0 4 , (h) CH 2 Cl-CO a H, (i) CHCl 2 -CO a H, (j) CC1 3 -C0 2 H.

9. Show by means of equations how you would convert acetic acid into propionic
acid and vice-versa.

10. Show how you would distinguish between: — -(a) CH 2 C1*C0 2 H, CH 3 'COCl and
CHjCl'COCl; (6) paraffin, olefin, alkyl halide, alcohol, ether, carboxylic acid, carboxylic
ester, Ac a O, AcCl and AcNH 2 .

11. Discuss the mechanism of: — (a) esterification and hydrolysis, (6) the Hofmann,
Curtius and Lossen rearrangements, (c) ammonolysis of esters and acid chlorides, (d)
hydrolysis of cyanides and amides.

12. Discuss the methods of preparation and the properties of the halogeno-acids.

READING REFERENCES

Greenspan, Oxidation Reactions with Aliphatic Peracids, Ind. Eng. Chem., 1947,

39. 847-
Bell, The Use of the Terms " Acid " and " Base ", Quart. Reviews (Chem. Soc.), 1947,

1, 113-
Organic Reactions, Wiley. "Vol. IV (1948), Ch. 4. Acyloins.

Richards and Thomson, Spectroscopic Studies of the Amide Linkage, J.C.S., 1947, 1248.
Organic Reactions, Wiley. Vol. Ill (1946).

(i) Ch. 7. The Hofmann Reaction,
(ii) Ch. 8. The Schmidt Reaction,
(iii) Ch. 9. The Curtius Reaction.

Sidgwick, The Organic Chemistry of Nitrogen, Oxford Press. (Revised Edition by
Taylor and Baker, 1937.)

Ch. I. Esters of Nitrous and Nitric Acids.
Ch. V. Amides, etc.

Brown, The Mechanism of Thermal Decarboxylation, Quart. Reviews (Chem. Soc),

I95L 5. 131-

Sonntag, The Reactions of Aliphatic Acid Chlorides, Chem. Reviews, 1953, 52, 237.

Davies and Kenyon, Alkyl-Oxygen Heterolysis in Carboxylic Esters and Related
Compounds, Quart. Reviews (Chem. Soc), 1955, 9, 203.

Gunstone, Recent Developments in the Preparation of Natural and Synthetic Straight-
Chain Fatty Acids, ibid., 1953, 7, 175.

Gensler, Recent Developments in the Synthesis of Fatty Acids, Chem. Reviews, 1957, 57>
191.

Bender, Mechanisms of Catalysis of Nucleophilic Reactions of Carboxylic Acid Deriva-
tives, Chem. Reviews, i960, 60, 53.

CHAPTER X

TAUTOMERISM

Acetoacetic ester or ethyl acetoacetate (E.A.A.) is the ethyl ester of aceto-
acetic acid, CH 3 -CO-CH 2 -C0 2 H, a p-ketonic acid. Acetoacetic ester was
first discovered by Geuther (1863), who prepared it by the action of sodium
on ethyl acetate, and suggested the formula, CH 3 -C(OH):CH-CO a C„H 5
(p-hydroxycrotonic ester). In 1865, Frankland and Duppa, who, inde-
pendently of Geuther, also prepared acetoacetic ester by the action of sodium
on ethyl acetate, proposed the formula CH 3 -CO-CH 2 -C0 2 C 2 H 5 (p-ketobutyric
ester) . These two formulae immediately gave rise to two schools of thought,
one upholding the Geuther formula, and the other the Frankland-Duppa
formula, each school bringing forward evidence to prove its own claim, e.g.,

Evidence in favour of the Geuther formula {reactions of an unsaturated
alcohol), (i) When acetoacetic ester is treated with sodium, hydrogen is
evolved and the sodium derivative is formed. This indicates the presence
of a hydroxyl group.

(ii) When acetoacetic ester is treated with an ethanolic solution of bromine,
the colour of the latter is immediately discharged. This indicates the
presence of an olefinic double bond.

(iii) When acetoacetic ester is treated with ferric chloride, a reddish-
violet colour is produced. This is characteristic of compounds containing
the group — C(OH)IC< (cf. phenols). 6

Evidence in favour of the Frankland-Duppa formula (reactions of a ketone).
(1) Acetoacetic ester forms a bisulphite compound with sodium hydrogen
sulphite.

(ii) Acetoacetic ester forms a cyanohydrin with hydrogen cyanide.

(iii) Acetoacetic ester forms a phenylhydrazone with phenylhydrazine.

Thus the remarkable position arose where it was possible to show that
a given compound had two different formula, each of which was based on
a number of particular reactions. The controversy continued until about
1910, when chemists were coming to the conclusion that both formulae were
correct, and that the two compounds existed together in equilibrium in
solution (or in the liquid state) :

OH

CH 3 -CO-CH 2 -C0 2 C 2 H 5 ^ CH 3 -C=€H-C0 2 C 2 H 5

When a reagent which reacts with ketones is added to acetoacetic ester,
the ketone form is removed. This upsets the equilibrium, and in order
to restore the equilibrium mixture, the hydroxy-form of acetoacetic ester
changes into the ketone form. Thus, provided sufficient reagent is added,
acetoacetic ester reacts completely as the ketone form. Similarly, when a
reagent which reacts with olefins .or with hydroxy-compounds is added in
sufficient quantity, acetoacetic ester reacts completely as the hydroxy-form.
The problem was finally settled by Knorr (1911), who succeeded in
isolating both forms. He cooled a solution of acetoacetic ester in light
petrol to —78°, and obtained crystals which melted at —39°. This substance
gave no coloration with ferric chloride and did not combine with bromine,
and was therefore the pure ketone form corresponding to the Frankland-
Duppa formula. Knorr then suspended the sodium derivative of acetoacetic
ester in light petrol cooled to —78°, and treated this suspension with just
enough hydrogen chloride to decompose the sodium salt. He now obtained

216

TAUTOMEKISM 217

a product which did not crystallise, but set to a glassy solid when cooled.
This substance gave an intense coloration with ferric chloride, and was
therefore the pure hydroxy-form corresponding to the Geuther formula.

Thus acetoacetic ester is a substance that does the duty of two structural
isomers, each isomer being capable of changing rapidly into the other when
the equilibrium is disturbed, e.g., by the addition of certain reagents.
This is a case of dynamic isomerism, and the name tautomerism (Greek:
same parts) was given to this phenomenon by Laar (1885). The two forms
are known as tautomers or tautomerides, the ketone isomer being called the
keto form, and the hydroxy isomer the enol form. Hence this type of
tautomerism is known as keto-enol tautomerism. (In the I.U.P.A.C. system
of nomenclature the suffix -en indicates the presence of a double bond, and the
suffix -ol, a hydroxyl group. The word enol is a combination of these
suffixes and indicates the structure of this form.)

A simplified version for the occurrence of tautomerism is as follows. Every
structure has a certain amount of internal energy, and when dealing with two
structural isomers, a definite amount of energy must be supplied in order to
overcome the energy of transformation. If this amount of energy is small, then
each isomer will change spontaneously into the other, resulting in an equili-
brium mixture of the two forms. Calculations of the energy content of keto
and enol tautomers on the basis of bond energies, show that the keto form is
the more stable one. Thus, when the equilibrium mixture contains relatively
large amounts of the enol form, then there must be present some stabilising
factor for the enol form, e.g., hydrogen bonding and resonance (see p. 223).

When one tautomer is more stable than the other under ordinary con-
ditions, the former is known as the stable form, and the latter as the labile
form. In practice, it is generally difficult to say which is the labile form,
since very often a slight change in the conditions, e.g., temperature, solvent,
shifts the equilibrium from keto to enol or vice-versa (see below). Tauto-
merism in the solid state is rare, and hence, in the solid state, one or other
tautomer is normally stable, but in the liquid or gaseous state, or in solution,
the two forms usually exist as an equilibrium mixture.

It has been found that the enol form is more volatile than the keto, and
that the change from enol to keto is extremely sensitive to catalysts. Meyer
et al. (1920, 1921) found that traces of basic compounds were very effective
catalysts. Thus they found that soft glass vessels were unsuitable for the
separation of the keto and enol forms, since, when fractionated, the more
volatile enol form rapidly changed into the original keto-enol mixture under
the catalytic influence of the walls of the containing vessel. Meyer, however,
succeeded in separating the enol form from the keto by carrying out the
fractional distillation under reduced pressure in silica apparatus which had
been thoroughly cleaned (freed from dust, moisture, etc.) Distillation
under these conditions is known as aseptic distillation.

The greater volatility of the enol form is unexpected in view of the fact
that alcohols are less volatile than ketones containing ^

the same number of carbon atoms. This anomalous CH C'^H

behaviour of the enol form may be explained by assum- 3 11 :

ing that chelation takes place through hydrogen bonding. n—£ ,6
This is supported by various facts, e.g., (i) the enol \c"

form is more soluble in cyclohexane, and less soluble 1

in water, than the keto form. The presence of the OC.H

hydroxyl group should have made the enol form less 5

soluble in cyc/ohexane, and more soluble in water, than the keto (cf. alcohols),
(ii) The formation of the hydrogen bond intramolecularly prevents the
formation of the hydrogen bond intermolecularly , i.e., prevents association
which would have raised the boiling point of the enol form.

2l8 ORGANIC CHEMISTRY

The vibrational spectrum of liquid ethyl acetoacetate shows intermolecular
bonding between the enol and keto forms. In solvents such as carbon tetra-
chloride, and in dilute solution, intramolecular hydrogen bonding occurs in the
enolic form, and the tautomeric equilibrium is displaced in the direction of the
enol form (Shigorin, 1950).

In tautomerism in general there may be an equilibrium between two or
more forms. One may predominate, or all may be present to about the
same extent, the concentration of each form depending on the temperature
and the solvent (if in solution). Although it may not be possible to separate
tautomers owing to the ease and rapidity of their interconversion or, as in
many cases, due to one form being almost completely absent, the presence
of more than one compound may be shown by special properties of each
isomer. The refractive index of the mixture may be observed directly,
and the value obtained compared with those calculated for the various
tautomeric structures. Spectral analysis may show the presence of more
than one substance, since each structure will have its own characteristic
absorption binds. Furthermore, the intensities of the bands change as
the temperature changes, thus showing the displacement of the equilibrium.
The Raman effect may also be used to show the presence of tautomers;
e.g., the equilibrium mixture of acetoacetic esters shows Raman shifts due
to both the C=C and C=0 groups. Experiments using deuterium exchange
reactions have also shown the presence of keto-enol mixtures, e.g., Klar
(1934) showed that hydrogen is exchanged slowly by deuterium when
acetaldehyde is dissolved in D 2 0. Since the C — H bond in paraffins is
stable under these conditions, the inference is that a hydroxyl group is
present in acetaldehyde, i.e., some enol form is present. Acetone was found
to undergo this exchange more rapidly.

The methods used for the quantitative estimation of each form in any
tautomeric equilibrium mixture, of which the keto-enol system is only one
example, fall into two distinct groups, physical and chemical. Obviously,
whatever the method used, it should be one that does not disturb the
equilibrium of the mixture during the estimation.

Physical methods. Physical methods do not disturb the equilibrium,
for they do not depend on the removal of one form, and they should therefore
be used wherever possible.

(i) The refractive index of the equilibrium mixture is determined ex-
perimentally. The refractive indices of both the keto and enol forms are
calculated (from a table 01 atomic refractions), and from these figures it is
then possible to calculate the amount of each form present in the equilibrium
mixture. In some cases, the refractive index of each form may be obtained
directly by isolating it, e.g., acetoacetic ester.

(ii) If one form is an electrolyte, the electrical conductivity of the mixture
is determined experimentally, and the amount of this form present may be
calculated from the results, e.g., nitromethane (p. 304).

(iii) The composition of the equilibrium mixture may be determined by
means of optical rotation measurements (cf. mutarotation, p. 451).

(iv) Joshi and Tuli (1951) have introduced a new physical constant which they
have named the refrachor, and have used it to determine the percentage of
tautomers in an equilibrium mixture, e.g., they found that ethyl acetoacetate
contains 7-7% enol, and acetylacetone 72-4% enol.

(v) Jarrett et al. (1953) have determined keto-enol equilibria in two (3-diketones
by measurement of proton resonance.

Chemical methods. Since chemical methods cause the removal of one
form, it is necessary to use a reagent that reacts with this form faster than

TAUTOMEKISM 2IQ

the rate of interconversion of the tautomers. Meyer (1911, 1912) found
that in the case of keto-enol tautomerism, bromine reacts instantaneously
with the enol form, and so slowly with the keto form in comparison with the
enol, that the keto reaction may be ignored. Meyer introduced two pro-
cedures, the direct and the indirect method. In the direct method a weighed
sample of the keto-enol mixture dissolved in ethanol is rapidly titrated with
a dilute ethanolic solution of bromine at o° (to slow down the interconversion
of the tautomers). The first appearance of excess bromine indicates the
end point.(see also the bromination of acetone, p. 154).

The titration must be carried out rapidly; otherwise the keto form changes
into the enol during the time taken for the titration. In any case, it has
been found impossible to carry out the titration sufficiently rapidly to
avoid the conversion of some keto into enol, and so this method always
results in too high a value for the enol form.

More reliable results may be obtained by the indirect method. An excess
of dilute ethanolic solution of bromine is added rapidly to the weighed
sample dissolved in ethanol, and then an excess of 2-naphthoI dissolved in
ethanol is added immediately. By this means, the excess bromine is removed
almost instantaneously, and so the keto-enol equihbrium is not given time
to be disturbed. Potassium iodide solution and hydrochloric acid are now
added, and the liberated iodine- is titrated with standard thiosulphate :

KI + HC1
CH 3 -COCHBr-C0 2 C 2 H 5 + KI -
CH 3 -COCHI-C0 2 C 2 H 5 + HI -

-> KC1 + HI

-> CH 3 -COCHI-C0 2 C 2 H 5 + KBr

-> CH 3 -COCH 2 -C0 2 C 2 H 5 + I 2

Cooper and Barnes (1938) have suggested an improved indirect method using
methanol instead of ethanol, and di-isobutene instead of 2-naphthol (di-*sobutene
is a mixture of 2 : 4 : 4-trimethylpent-i-ene and 2:4: 4-trimethylpent-2-ene).

Schwarzenbach and Wittwer (1947) have introduced a new technique,
the flow-method, for the estimation of the enol content. The solution of
the keto-enol mixture and an acidified bromide-bromate solution are
simultaneously mixed and diluted in a mixing chamber, and the mixture
made to. flow past a platinum electrode. The relative amounts of the two
solutions are adjusted so that the potential measured at the platinum elec-
trode shows a sharp rise — this corresponds to the end-point in the titration
-of the enol form by the bromine. This method gives good results of enol
contents as low as io~ 5 per cent. ; e.g., the enol content of acetone in aqueous
solution was found to be 2*5 X io" 4 per cent.

TABLE IV

Compound

Per cent, enol
(in ethanol)

CH 3 -CO-CH,-CO a CH 8 .
CH 3 -CO-CH 2 -CO,C|H 6 .
CH 3 >CO-CH 2 -CO-CH s .
CH 3 -COCH(CH 3 )-COCH 3
C,H 6 -CO-CH 3 -CO-C,H 6 .
CH a (C0 3 CjH 6 ) a

Aldehydes of type R-CH 2 -CHO
Ketones of type R-CHj-COCHj-R

4-8

7-5
76

3i

96

trace

trace

trace

The enol content in keto-enol mixtures (in dipropyl ether) has been determined
by means of lithium aluminium hydride; the hydrogen liberated is estimated
(Honing et al., 1952). The values obtained are about 10 per cent, higher than
those obtained by the bromine titration or by physical methods.

220 ORGANIC CHEMISTRY

Enolisation. The phenomenon of enohsation is exhibited by compounds
containing either a methylene group, *CH 2 -, or a methyne group, ^>CH — ,
adjacent to a carbonyl group. The presence of one carbonyl group does
not always give rise to an appreciable amount of enol form, e.g., acetaldehyde,
acetone (see Table IV, above). If the compound contains a methylene or
methyne group attached to two carbonyl groups the percentage of enol
form is usually high, e.g., acetylacetone, CH 3 'COCH a *COCH 3 . When a
methyne group is attached to three carbonyl groups, the compound may
exist almost completely as the enol form, e.g., triacetylmethane,
(CH 3 -CO) 3 CH.

When the methylene or methyne group is attached to two or three carbonyl
groups, the hydrogen atom might migrate equally well to one or other
carbonyl group. This is not found to be so in practice for unsymmetrical
compounds, one enol form being present exclusively, or largely predominat-
ing; e.g., in acetoacetic ester the hydrogen atom migrates exclusively to
the acetyl carbonyl group (see also below). When two or more enol forms
are theoretically possible, ozonolysis may be used to ascertain the structure
of the form present; e.g., in hexane-2 : 4-dione, CHj'COCHg'COCHa'CHa,
the two possible enols are :

OH
CH 3 -C=CH-CO-CH 2 -CH 3

OH
CH 3 -CO-CH=C-CH 2 -CH 3

(I)

(")

Ozonolysis of I will give CH 3 -C0 2 H and CH 3 ;CH 2 -CO-CHO ; II will give
CH 3 -CO-CHO and CH 3 -CH 2 -C0 2 H. Identification of these compounds will
decide whether the enol is I or II, or both.

The type of tautomerism discussed above is known as the keto-enol triad
system. In this system a hydrogen atom migrates from atom 1 {oxygen)
to atom 3 [carbon) :

3 21 3 2 1

>CH— C=0 ^ >C=C— OH

Enols resemble phenols in a number of ways; e.g., both form soluble
sodium salts; both give characteristic colorations with ferric chloride;
and both couple with diazonium salts.

The keto-enol type of tautomerism is only one example of a triad system.
The triad system is the most important class of tautomeric systems, and
the following, which are exemplified in the text, are the commonest :

(i) Three-carbon systems :

CH— C=C ^ C=C— CH

(ii) Nitro-acinitro (pseudonitro) system:

CH— N0 2 ^ C=NO-OH

(iii) Nitroso-oximino system :

CH— N=0 ==i C=N-OH

(iv) Amidine system :

NH— C=N ^= N=C— NH

(v) Amido-imidol system :

NH— C=0 ^ N=C— OH

TAUTOMERISM 221

(vi) Azo-hydrazone system :

N=N— CH =^ HN— N=C

(vii) Diazo-amino (iriazen) system:

N =N— NH =^ NH— N=N

(viii) Diazo-nitrosamine system :

Ar— N=N-OH ^ Ar— NH— N=0

In addition to the "open" systems of tautomerism, there are also, e.g., ring-
chain tautomerism (see, e.g., aldol, p. 237; carbohydrates, p. 451, succinyl
chloride, p. 377) ; and valence tautomerism (see p. 485).

Modern theories of tautomerism. Ingold (1927) suggested the name
cationotropy for all those cases of tautomerism which involve the separation
of a cation; and the name anionotropy for those cases which involve the
separation of an anion. Lowry (1923) suggested the name prototropy for those
cases in which a proton separates, and called such systems prototropic systems.
Using Ingold's generalised classification of tautomeric systems, it can be
seen that prototropy is a special case of cationotropy. Braude and Jones
(1944) have proposed the term oxotropy for anionotropic rearrangements
involving only the migration of a hydroxyl group (see allylic rearrangement,

Laar (1885) attempted to account for keto-enol. tautomerism by postulating
that the hydrogen atom occupied a mean position with respect to the final
positions it would occupy in the keto and enol forms, and that by a lateral move-
ment in either direction the hydrogen atom formed the keto or enol form. Baly
^ and Desch (1904) extended this theory into the oscillation
■si-^ J? theory (isorropesis) , according to which the hydrogen atom
• ^-" U oscillated continuously between the two positions as shown.

CH C Tll j s t k eorv soon became untenable, since it did not agree

with many of the experimental observations (see below).

Jacobson (1887) objected to the use of the word tautomerism to describe the
above phenomenon because it involved the view that tautomers have no definite
structure, but are continually changing. J acobson thought that both forms were
present and that the change from one into the other was caused by the presence of
certain reagents. He therefore proposed the name desmotropy or desmotropism
(Greek: change of bonds). Hantzsch and Hermann (1887) suggested that both
the terms tautomerism and desmotropism should be used, the former to denote
that a compound exhibited a dual nature, and the latter to indicate those cases of
tautomerism where the two forms exist in different physical forms (i.e., those
cases of tautomerism in which both forms have been isolated).

Knorr (1896) introduced the term allelotropic mixture for equilibrium mixtures
whose composition varied with changes in temperature, i.e., an allelotropic
mixture is a tautomeric mixture in the liquid state or in solution; a mixture of
solid tautomers is not an allelotropic mixture. _ . _

The term pseudomerism has also been used for those cases of tautomerism in
which only one form, the keto or enol, may be shown to be present.

The actual steps involved in keto-enol tautomerism are still the subject
of much discussion. According to Hughes and Ingold, base-catalysed enolis-
ation of a ketone proceeds through an enolate anion I whose formation is
controlled largely by the inductive effects of the alkyl groups.

o-

B + R 2 CH-CO-R— > BH++ R 2 C=OR

(I)
Acid-catalysed enolisation involves the removal of a proton from the con-
jugate acid of the ketone, II, and this process is dependent mainly on the

CH 3 CH 3

222 ORGANIC CHEMISTRY

hyperconjugation (p. 270) by the alkyl groups in the transition state for the
formation of the carbon-carbon double bond.

H ^O H OH OH

H + + R 2 C— — C— R — y R 2 C— C— R — y H + + R a C=C— R

(II)

Evidence for these mechanisms is that alkyl groups depress base-catalysed
reaction rates.

The mechanisms described above are step-wise mechanisms, and they
appear to operate in certain cases. On the other hand, there is evidence
that in other cases, acid and base catalysis of enolisation take place by a
concerted or push-pull mechanism, i.e., the molecule undergoing change is
attacked simultaneously at two places. Thus the enolisation of, e.g.,
acetone, proceeds by the simultaneous removal of a proton from an a-carbon
and the addition of a proton to the oxygen of the carbonyl group. This may
be represented as follows (B is a general base, and HA is a general acid) :

B>H-r-CH 2 B--H--CH- CH,

^1 A jj ♦ JJ

6=0 < HA — > C— O - -HA — > BH + C— OH + A~

CH 3

transition state

In this type of mechanism, the solvent water molecules are believed to be
involved, one acting as a proton acceptor and the other as a proton donor.
Furthermore, the above mechanism is termolecular, and Swain (1950) has
presented evidence to show that both the acid- and base-catalysed enolisa-
tion of acetone are termolecular reactions. Emmons et al. (1956) have
obtained evidence supporting Swain's termolecular mechanism for ketone
j enolisation, but propose that the transition state for
, — < — , both acid- and base-catalysed enolisation may be re-
O "HA presented by III.
R*CH 2 ^ I j Under these conditions in either acid- or base-catalysed

/C~OR reactions, the transition state will be very close to the
R*CH/ j enol and hence will be stabilised by hyperconjugation.

B---H In the acid-catalysed reaction, bond b is relatively

' * ' tight and bond a is relatively loose. Hence in acid-

a (in) catalysed enolisation, steric hindrance around bond a
is not observed in most cases, and hyperconjugation is therefore a domin-
ating factor in the enolisation of a ketone. In the base-catalysed reaction,
bond a is relatively tight and bond b is relatively loose. Under these
conditions a steric factor becomes important and is probably the reason
why alkyl groups depress base-catalysed reaction rates.

The tremendous increase in speed by acid and base catalysts is well
illustrated by the fact that the pure keto and enol forms of ethyl acetoacetate
change very slowly into the equilibrium mixture (several weeks), whereas the
latter is obtained rapidly (several minutes) by the addition of acid or base
catalysts (Lowry et al., 1924).

It has been shown that in keto-enol equilibrium mixtures, the enol form
reacts extremely rapidly with halogens, is acidic and produces an anion
which is reactive both at the oxygen and a-carbon (to the hydroxyl group).
It has been pointed out above (see also Table IV, p. 219) that in keto-enol
tautomerism the enol content of the equilibrium mixture varies from com-
pound to compound; but the reason for this is still an open question. It

TAUTOMERISM 223

appears to be quite certain that the solvent plays a part, but the nature of
the other factors is uncertain. Some authors believe that resonance and
hydrogen bonding are involved; e.g., it has been suggested that the enol
form of acetylacetone is stabilised by chelation through hydrogen bonding:

O O O— H O

CH S -C— CH 2 — OCH 3 ^ CH 3 -C=CH— C-CH 3

Hydrogen bond formation is not possible in acetaldehyde, acetone, etc.,
and therefore the enol forms of these compounds cannot be stabilised this
way.

Now let us consider the case of ethyl acetoacetate:

O O O— H-—-O

CH 3 -C— CH 2 — C— OC 2 H 5 ^=i CH 3 -C=CH— C— OC 2 H 5

The enol form is stabilised by chelation; but ethyl acetoacetate contains
54 per cent, enol form in the vapour state, whereas acetylacetone contains
92 per cent. If chelation were the only factor involved, one might have
expected that the enol content of the two compounds would have been
about the same. A possible explanation for this difference may be due to the
resonating structure of the carbethoxy group:

o o-

I +
^^ — C=OC,H 5

-C— OC.H.

This effect will tend to stabilise the keto form, thereby reducing the tendency
to enolise. The large part played by the resonance effect of the carbethoxy
group in reducing the enol content may be inferred by considering ethyl
malonate. This compound contains only a trace of the enol form. Its
structure is IV, and its enol structure will be V.

Q O O— H----0

-C— OC 2 H 5 C 2 H 5 0— C=CH— C-

(IV) (V)

C 2 H fi O— C— CH 2 — C— OC 2 H 5 C 2 H 5 0— C=CH— C— OC 2 H 5

If both carbethoxy groups are affected by resonance, the keto form will be
stabilised, i.e., there will be very little tendency to enolise.

It was pointed out above that the nature of the solvent also affects the
position of equilibrium in keto-enol tautomerism. If we assume (as we
have above) that the enol form is, where, possible, stabilised by chelation,
then solvents that prevent chelation will reduce the enol content. Thus
hydroxylic solvents such as water, methanol, ethanol, acetic acid, etc., in
which the hydrogen atom of the hydroxyl group can form a hydrogen
bond with the oxygen atom of the carbonyl group of the keto form, will tend
to reduce the enol content; e.g., ethyl acetoacetate contains 54 per cent,
enol in the vapour phase ; in ethanol, the enol content is 7-5 per cent. On
the other hand, where hydrogen bond formation with the solvent is not
possible, e.g., in hexane, benzene, etc., the enol content will be large.

According to Arndt et al. (1946), the effect of the solvent on enol content
is to be explained as follows. Hydrophilic solvents favour the keto form,
whereas hydrophobic and lipophilic solvents favour the enol form. The

224 ORGANIC CHEMISTRY

above authors have shown that the more the hydrophobic groups pre-
dominate in a molecule of a tautomeric substance of a given structure, the
greater is the enol content in the equihbrium mixture. Thus they found
that the enol content of the estersof acetoacetic acid increased in the order
methyl, ethyl, propyl, butyl, 2-octyl and benzyl; this is the order in which
the whole molecule (ester) becomes more hydrophobic.

The preparation of ethyl acetoacetate. In the preparation of ethyl aceto-
acetate, condensation is effected between two molecules of the same ester.
This is one example of the Claisen condensation (1887), in which a keto-
ester is formed by the reaction between two molecules of an ester containing
a-hydrogen atoms. In certain cases the second ester does not contain an
a-hydrogen atom, e.g., ethyl formate, ethyl benzoate and ethyl oxalate (see
text for examples of their use). The Claisen condensation may also take
place between an ester and a ketone to form a 1 : 3-diketone. The con-
densation is brought about by sodium ethoxide, sodamide, triphenyl-
methylsodium, etc. Sodium ethoxide is the reagent usually employed in the
preparation of ethyl acetoacetate.

If sodium is added to ethyl acetate which has been very carefully freed
from ethanol, very little action takes place. When a small amount of
ethanol is added, a vigorous reaction sets in. This seems to indicate that
it is probably sodium ethoxide, and not sodium, which is the effective
reagent for the condensation (Snell and McElvain, 1931). The product
is the sodium enolate of ethyl acetoacetate and this, on treatment with
acid, liberates ethyl acetoacetate, which may be purified by distillation in
vacuo.

Many mechanisms have been proposed for the formation of ethyl aceto-
acetate. The one most widely accepted at the moment is (cf. aldol con-
densation).

rv rv

EtO- H— CH 2 -C0 2 Et ^ EtOH + CH 2 -C0 2 Et

Q

o O^

IV

vO

<A

OEt >OEt

>0 H ONa+

C_| I yf I I Me-CO,H

Me-C— CH'CO z Et + EtO~ ^ EtOH + Me-6=CH-C0 2 Et >

OH O

Me-C=CH-C0 2 Et ^=±: Me-C-CH 2 -C0 2 Et (28-29%)

Ethyl acetoacetate is now also being prepared industrially by polymerising
keten in acetone solution to diketen, which is then treated with ethanol.

2CH t =CF=0—+ ' I I i^- H ->CH 3 -CO-CH 2 -C0 8 C a H 5

CH 2 — C=0

In " mixed ethyl acetoacetate condensations " between two different esters, a
mixture of all four possible products is usually obtained; e.g., with ethyl acetate
and ethyl propionate :

CjHjONa

(i) CH 3 -CO |OC 2 H s + H| — CH 2 -CQ 2 C 2 H 5 ~>

CH 3 -CO-CH 2 -C0 2 C 2 H 5 + C 2 H 6 OH

TAUTOMERISM 225

CH 3

C,H,ONa

(ii) CH 3 CH 2 -C0 T0C 2 H 5 + H| — CH— C0 2 C a H 5 >

CH,

X

CH 3 -CH 2 -COCH-C0 2 C 2 H 6 + C 2 H B OH
CH,

C,H,ONa

(iii) CHs-CO JOCsH,, + H[ —CH-C0 2 C 2 H 5 -^ >

CH.

>i:

CH 3 -COCH-C0 2 C 2 H 6 + C 2 H 5 OH

C,H s ONa

(iv) CH,-CH,-CO lOC 2 H 5 + H j— CH 2 -CQ 2 C 2 H 5 ->

CH 3 -CH 2 -COCH 2 -C0 2 C 2 H 5 + C 2 H 5 OH

These reactions clearly show that only a-hydrogen atoms are involved in the
Claisen condensation.

Properties of ethyl acetoacetate. Ethyl acetoacetate is a colourless,
pleasant-smelling liquid, b.p. 181 (with slight decomposition), sparingly
soluble in water, but miscible with most organic solvents. It is readily
reduced by sodium amalgam, or catalytically to (3-hydroxybutyric ester:

CH 3 -CO-CH 2 -C0 2 C 2 H 5 + 2[H] — -> CH 3 -CH(OH)-CH 2 -C0 2 C 2 H 6

Lithium aluminium hydride reduces ethyl acetoacetate to butane-i : 3-
diol (Buchta et al., 195 1).

CH s -CO-CH 2 -C0 2 C 2 H 6 -i^V CH 3 -CHOH-CH 2 -CH 2 OH (30%)

On the other hand, ethyl acetoacetate may be reduced to ethyl butyrate by
treating it with methyl mercaptan in the presence of zinc chloride and sodium
sulphate, and then reducing the thioketal with Raney nickel in ethanolic
solution (Newman et al., 1950; cf. p. 334).

Ethyl acetoacetate is neutral to litmus, but is soluble in dilute sodium
hydroxide solution: it is the enol form which dissolves to form the sodium
salt. When ethyl acetoacetate is hydrolysed in the cold with dilute sodium
hydroxide, the solution then acidified, extracted with ether and the ether
removed in the cold under reduced pressure, free acetoacetic acid is obtained.
Krueger (1952) has prepared acetoacetic acid as a crystalline solid, m.p.
36-37 . As prepared in the usual way it is unstable, readily decomposing
into acetone and carbon dioxide:

CH 3 -COCH 2 -C0 2 H — > CH 3 -COCH 3 + C0 2

All p-ketonic acids readily decompose in this manner, but it is interesting
to note that trifluoroacetoacetic acid, CF 3 -COCH 2 -C0 2 H, is quite stable
and can be distilled without much decomposition. Ethyl acetoacetate
forms a-oximinoacetoacetate when treated with nitrous acid:

CH 3 -COCH 2 -C0 2 C 2 H 5 -HEL). CH S -COC(=NOH)-C0 2 C 2 H B

On the other hand, monosubstituted derivatives of ethyl acetoacetate are
split by nitrous acid; the nitroso-compound is formed first and then re-
arranges to the oximino form with fission of the molecule.

CH 3 -CO-CHR-C0 2 C 2 H 6 -^>CH 3 -CO-CR-C0 2 C 2 H 5 — > R-C-C0 2 C 2 H 6

NO NOH

226

ORGANIC CHEMISTRY

Ethyl acetoacetate forms a green copper compound which is soluble in
organic solvents. This indicates that the copper compound is not an ionic
but a chelate compound, and its structure is possibly:

CH 3 — C— O

// \ ,
CH Cu

/ \

O— C— CH,

%

CH

C 2 H 5 0— C=0

0=C— OC 2 H 6

Acetoacetic ester reacts with ammonia (and with primary amines), but
the structure of the compound produced is uncertain, there being two possi-
bilities, the imine form I, and the amine form II:

NH NH 2

(I) CH 3 -C-CH 2 -C0 2 C 2 H 6 CH 3 -C=CH-C0 2 C 2 H 5 (ii)

Most of the experimental work favours II. It is possible that the reaction
takes places through the enol form :

OH

CH 3 -C=CH-C0 2 C 2 H 5 + NH 3 — >

| OH H

CH 3 -C-

CH-C0 2 C 2 H 6

NH 2

-H a

NH 2
> CH 3 -C=CH-CO.C 2 H,

Monoalkyl derivatives of acetoacetic ester, CH 3 -CO-CHR-CO a C 2 H 5 , re-
act with ammonia in a similar manner but the dialkyl derivatives,
CH 3 -COCR 2 -C0 2 C 2 H 5 , form the amide CH 3 -COCR 2 -CONH 2 . This supports
the mechanism given above, since the monoalkyl derivatives can enolise,
whereas the dialkyl derivatives cannot. Ethyl acetoacetate also reacts
with hydroxylamine to form an /so-oxazolone.

CH 3 -COCH 2 -C0 2 C 2 H 5 + NH 2 OH — > CH 3 -C-CH 2 -C0 2 C 2 H

I

->

OH

CH 3 -C-

-CH 2
f + C 2 H 5 OH

to

^o^-

With phenylhydrazine, ethyl acetoacetate forms a pyrazolone (p. 755).

Acetoacetic ester undergoes the Knoevenagel reaction due to the presence
of an " active " methylene group (p. 155), and also couples with diazonium
salts. It reacts with Grignard reagents to form the hydrocarbon, which
indicates that it reacts in the enol form, e.g.,

OH
CH 3 -C=CH-C0 2 C 2 H 6 + CH 3 -Mg-I

OMgl
CH 3 -C=CH-C0 2 C 2 H 5 + CH 4

TAUTOMERISM 227

When acetoacetic ester is heated under reflux with a trace of sodium
hydrogen carbonate, ethanol is eliminated and dehydroacetic acid, m.p.
180 , is obtained:

/O H C 2 H 6 Q N /C\

CH -C/ \C=0 _ 2CaHsOH CH 3 -C/ \C=0

^c- 1oc 2 h 5 hK Y

O O (53%)

Dehydroacetic acid may also be prepared by the polymerisation of diketen:

2CH,=C O CH 3 — (XXX)

CH a — C=0 Hd /CH-CO-CH 3

O (60-80%)

THE USE OF ACETOACETIC ESTER IN THE SYNTHESIS
OF KETONES AND FATTY ACIDS

The synthetic use of acetoacetic ester depends on two chemical properties:
1. (a) When treated with sodium ethoxide, acetoacetic ester forms
sodioacetoacetic ester, i.e., the sodium derivative of the enolic form:

OH

I
CH 3 -C:CH-CO a C 2 H 5 + C a H 5 ONa — >

r "1 *

_CH 3 -C:CH-C0 2 C 2 H 5 J Na + C 2 H 6 OH {ex.)

(b) Sodioacetoacetic ester readily reacts with primary and secondary
alkyl halides (vinyl and aryl halides do not react) to produce alkyl derivatives
of acetoacetic ester in which the alkyl group is attached to carbon. This
fact has given rise to considerable speculation regarding the mechanism of
the alkylation process. The problem is still not settled, but a highly
favoured theory is that the negative ion is a resonance hybrid:

o- O

CH 3 -C:CH-C0 2 C 2 H 6 -<r+ CHj-C— CH-C0 2 C 2 H 6

(I) (II)

Thus in the actual state both the oxygen and carbon atoms have a negative
charge ; but since a negatively charged carbon atom is more reactive than
a negatively charged oxygen atom, it is the carbon atom that is the point of
attack. As the actual state of the molecule cannot be represented by a
single formula, the process of alkylation may be regarded as involving
structure II, i.e., a carbanion. There is also evidence to show that the
alkylation occurs by an S N 2 process. Thus :

CH 3 — €=0 CH 3 — C=0

Jr* <~* 1

HC R— X — > HC— R + X-

CO.C.H. CO.C.H.

228 ORGANIC CHEMISTRY

For the sake of simplicity the alkylation of sodioacetoacetic ester will, in
future, be represented as:

[CH 3 -COCH-C0 2 C 2 H S ]-Na + + RX — >

CH 3 -COCHR-C0 2 C 2 H B + NaX (v.g.)

After one alkyl group has been introduced, the whole process may then
be repeated to give the dialkyl derivative of acetoacetic ester:

CH 3 -COCHR-C0 2 C 2 H 6 C ' Hi ° Na > [CH 3 -CO-CR-C0 2 C 2 H 5 ]-Na + -^>

CH 3 -COCRR'-C0 2 C 2 H 5 (g.-v.g.)

Until recently, it was not considered possible to prepare disubstituted
derivatives of acetoacetic ester in one step. Sandberg (1957), however, has
now carried out' this one-step reaction in the preparation of ethyl p-aceto-
tricarballylate from acetoacetic ester (0*5 mole), ethyl bromoacetate (1-2
mole) and sodium hydride (1-2 mole) in benzene solution.

2NaH

CH 3 -COCH 2 -C0 2 C 2 H 5 + 2CH 2 Br-C0 2 C 2 H 5 >

/CH 2 - CO a C a H 6
CH 3 -CO-C^CO a C 2 H 5 (77%)

\CH 2 -C0 2 C 2 H 5

Potassium feri.-butoxide is usually best for preparing the metallo-acetoacetic
ester compounds, and generally alkyl iodides react faster than alkyl bromides
(Renfrew and Renfrew, 1946).

2. Acetoacetic ester and its alkyl derivatives can undergo two types of
hydrolysis with potassium hydroxide :

(a) Ketonic hydrolysis. Ketonic hydrolysis, so called because a ketone is
the chief product, is carried out by boiling with dilute aqueous or ethanolic
potassium hydroxide solution, e.g.,

CH 3 -COCH 2 -C0 2 C 2 H 8 + 2KOH — >

CH 3 -COCH 3 + K 2 C0 3 + C a H 5 OH (g.)

CH 3 -COCHR-C0 2 C 2 H 5 + 2KOH — >

CH 3 -COCH 2 R + K 2 C0 3 + C a H 8 OH (g.)

The ketone obtained is acetone or its derivatives, and the latter always
contain the group CH s 'CO — .

Dehn and Jackson (1933) found that 85 per cent, phosphoric acid was a
very good catalyst for the ketonic hydrolysis of acetoacetic ester and its
alkyl derivatives, the yield of ketone reaching 95 per cent.

The mechanism of ketonic hydrolysis is uncertain; a possibility is :

CH 3 — C — CH a v ** CH 3 *C'CH 3

OH O

It is the electron-attracting property of the carbonyl group that facilitates the
elimination of carbon dioxide.

(b) Acid hydrolysis. Acid hydrolysis, so called because an acid is the chief
product, is carried out by boding with concentrated ethanolic potassium
hydroxide solution, e.g.,

TAUTOMERISM 229

CH s -CO-CH 2 -CO !! C !! H 5 + 2KOH— >

CH S -C0 2 K + CH 3 -C0 2 K + C 2 H 8 OH (f.g.)

CH»-COCR 2 -C0 2 C 2 H 5 + 2KOH — »-

CH 3 -C0 2 K + R 2 CH-CO a K + C 2 H 6 OH (f.g.)

The acid obtained is acetic acid or its derivatives as the potassium salt.
From this the free acid is readily obtained by treatment with inorganic acid.

The mechanism of acid hydrolysis is possibly a reversal of the Claisen con-
densation. Only the ester of acetic acid undergoes condensation, and when the
condensation is reversed, the ester is converted into the potassium salt which
cannot recondense; hence the equilibrium is forced in the reverse direction to
that of condensation.

Examples of the synthesis of ketones. The formula of the ketone is
written down, and provided it contains the group CH 3 'CO — , the ketone can
be synthesised via acetoacetic ester as follows. The acetone nucleus is
picked out, and the alkyl groups attached to it are then introduced into
acetoacetic ester one at a time; this is followed by ketonic hydrolysis. It
is usually better to introduce the larger group before the smaller (steric effect) :

CH„-COCH 2 - CH.

(i) Butanone.

CH 3 -CO-CH 2 -C0 2 C 2 H 5 -^^> [CH 3 -CO-CH-C0 2 C 2 H 5 ]-Na + ^->

CH 3 -CO-CH(CH 3 )-C0 2 C a H 5 JgL* CH.-COCH.-CH.

CH,

(ii) j,-Methylpentan-2-one. |CH 3 'COCH- | CH 2 'CH

CH 3 -CO-CH 2 -C0 2 C 2 H 5 C,H '° Na > [CH 3 -COCH-C0 2 C 2 H 8 ]-Na + -^>
CH 3 -COCH(C 2 H 5 )-C0 2 C 2 H 5 ^^> [CH 3 -CO-C(C 2 H 5 )-C0 2 C 2 H 6 ]-Na +
-5L> CH 3 -CO-C(CH 3 )(C 2 H 5 )-C0 2 C 2 H 6 -££* CH.<XK»CH,-CH.

CH 3

Examples of the synthesis of fatty acids. The approach is similar to that for
ketones except that the acetic acid nucleus is picked out, and the acetoacetic
ester derivative is subjected to acid hydrolysis. The acetoacetic ester acid
synthesis is usually confined to the preparation of straight-chain acids or
branched-chain acids where the branching occurs on the a-carbon atom:

(i) n-Butyric acid. CH 3 *CH 2 '|CJHvO[}gH|

CH 3 -COCH 2 -C0 2 C 2 H 5 <W)W *> [CH 3 -COCH-C0 2 C 2 H 5 ]-Na + -^V

CH 3 -CO-CH(C 2 H 5 )-C0 2 C 2 H 6 "** , > CH 3 -CH 2 -CH 2 -C0 2 H

flydxOiysis

CH

3

(ii) a.-Methyl-n-valeric acid. CH 3 'CH 2 - CH 2 ' ICH'COgH

CH 3 -COCH 2 -C0 2 C 8 H 5 C,K '° m > [CH 3 -COCH-C0 2 C 2 H B ]-Na + -552*

CH a .CO-CH(C 3 H 7 )-C0 2 C 2 H 5 C,H '° Na > [CH 3 -CO-C(C 3 H 7 )-C0 2 C 2 H 5 ]-Na +

CH 3

I
-^>CH 3 -CO-C(CH 3 )(C 3 H 7 )-C0 2 C 2 H 5 ^^J CH 3 -CH 2 -CH 2 -CH-C0 2 H

230 ORGANIC CHEMISTRY

It has been found that ketonic hydrolysis and acid hydrolysis of aceto-
acetic ester always take place simultaneously, but one or the other can be
made to predominate by adjusting the concentration of the potassium
hydroxide. Thus, in the preparation of acids, there will always be some
ketone formed as by-product ; and vice-versa. For this reason it is better
to use ethyl malonate to synthesise acids since the yields are greater.

Ritter et al. (1962) have improved the preparation of acids via E.A.A. by
heating the mono-alkyl derivatives with catalytic amounts of alkoxide in
excess of absolute ethanol :

OEt"

MeCOCHR-C0 2 Et + EtOH y

MeC0 2 Et + R-CH 2 -C0 2 Et (70-90%)

In addition to the acetoacetic ester ketone synthesis, ketones may be prepared
in a somewhat analogous method as follows :

(i) From simpler ketones; e.g., acetone in ether solution is treated with
sodamide and the sodioacetone then treated with methyl iodide, whereupon ethyl
methyl ketone is produced :

CH 3 -COCH 3 NaNH > [CH 3 -CO-CHJ"Na+ — 'V CH.-COCEL-CH.

ether

Repetition of the process on ethyl methyl ketone gives a mixture of methyl
isopropyl ketone and diethyl ketone, which can be separated by fractional
distillation. By repeating the process hexamethylacetone (6i-tert.-bnty\ ketone),
(CH 3 ) 3 OCOC(CH 3 ) 3> can finally be obtained.

This synthesis has been modified to prepare &r/.-carboxylic acids, aceto-
phenone being the ketone used as starting material (Haller et al., 1914). The
yields are good for the simpler members of the series (Carter et al., 1946).

CH 3 .CO.C 6 H 6 ^!> R.CH a .CO.C 6 H 6 ^% R >H.COC 6 H 5
(11) RI (n) R I "R '/

(i) NaNH, R\ NaNH, R \ HNO, R \

,,, P „ T > R'^C-CO-C 6 H 5 V R'-^C-CONH 2 VR'^C-C0 2 H

<u) R I R „/ R „/ R „/

(ii) Ketones may also be prepared from certain esters. It has been found that
esters of the type R 2 CH>C0 2 C 2 H 5 do not undergo the normal Claisen condensation
with sodium ethoxide. In the presence of triphenylmethylsodium, (C„H 5 ) 3 C-Na + ,
however, they readily form the sodio-salt (Schlenk et al., 1931). These sodium
salts react with acid chlorides, and the resulting compounds, which are p-keto-
esters, yield ketones on ketonic hydrolysis :

(C,H.).CNa R'-COCl

R 2 CH-C0 2 C 2 H 6 -!-i-^ >■ [R 2 C-C0 2 C 2 H 5 ]-Na+ ->

„ _ ketonic

R'-COCR 2 -C0 2 C 2 H 5 > R'-COCHR 2

hydrolysis

(iii) p-Keto-acids, R'CO*CHR''C0 2 H, can be prepared by a-carboxylation of
ketones, R-COCH 2 R', with magnesium methyl carbonate. The intermediate
magnesium salt can be alkylated in situ and then decarboxylated to give ketones
R-COCHR'R" (Stiles et al., 1959).

Sodioacetoacetic ester reacts with many other halogen compounds besides
alkyl halides, and so may be used to synthesise a variety of compounds.

(i) 1 : z-Diketones. In the synthesis of 1 : 3-diketones the halogen
compound used is an acid chloride. Since acid chlorides react with ethanol,
the reaction cannot be carried out in this solvent in the usual way. The
reaction, however, may conveniently be carried out by treating acetoacetic
ester in benzene solution with magnesium and the acid chloride; e.g.,

TAUTOMERISM 231

pentane-2 : 4-dione may be obtained by the ketonic hydrolysis of the inter-
mediate product ethyl diacetylacetate:

CH 3 -COCH 2 -C0 2 C 2 H S + CH 3 -COCl — -> (CH 3 -CO) 2 CH-C0 2 C 2 H 5

ketonic ^ CH3 . C0 . CH2 . C0 . CH3

2

hydrolysis

If sodioacetoacetic ester or acetoacetic ester itself is treated with acetyl
chloride in pyridine as solvent, the O-acetyl derivative of acetoacetic ester,
acetoxycrotonic ester, is obtained, and not the carbon-linked compound
(as above) :

OCOCH3

[CH 3 -CO-CH-C0 2 C 2 H 5 ]-Na + + CH 3 -COCl Pyi ' dme > CH 3 -C:CH-C0 2 C 2 H B +NaCl

The reason for this unusual reaction is obscure.

(ii) Dicarboxylic acids. Dicarboxylic acids may be prepared by inter-
action of sodioacetoacetic ester and a halogen derivative of an ester, e.g.,
succinic acid from ethyl chloroacetate :

[CH 3 -COCH-C0 2 C 2 H 5 ]-Na + + C1CH 2 -C0 2 C 2 H 6 — >

CH 2 -C0 2 C 2 H 5 CH 2 -C0 2 H

CEL-COCH-COXJi, ^-> CH 2 -C0 2 H

3 lib hydrolysis * *

Ketonic hydrolysis of this acetoacetic ester derivative gives the y-keto-
acid ester (ethyl ester of lsevulic acid) :

CH 2 -C0 2 C 2 H 5

I Itctonic

CH 3 -COCH-C0 2 C 2 H 5 -> CH 3 -COCH 2 -CH 2 -C0 2 C 2 H S

hydrolysis

(iii) Long-chain fatty acids. An ingenious method of synthesising long-
chain fatty acids via acetoacetic ester involves a combination of methods
(i) and (ii) described above (Mrs. Robinson, 1930) :

(CH 2 )z - C0 2 C 2 H 5

[CH 3 -CO-CH-C0 2 C 2 H 5 ]-Na + Br -' CH ^' C0 ^ H L > CH 3 -COCH-CO a C 2 H 5

(CH 2 )« , CO a C 2 H B

(i)C ' H '° Na > CH 3 -CO-C-C0 2 C 2 H 6 -^->

(ii) CH,-(CH,)„-C0C1 J I "* 5 hydrolysis

CO(CH 2 )/CH 3
CH 8 -(CH 8 )/CO-CH 2 -(CH 2 ) a .-COgH

These keto-acids are readily reduced to the corresponding fatty acid by
means of the Clemmensen reduction (p. 150). Improved modifications of
this method have now been developed (see p. 363).

(iv) Ring compounds, (a) When sodioacetoacetic ester is treated with
certain dihalogen derivatives of the paraffins, carbocyclic compounds are
obtained (see p. 470).

(b) Acetoacetic ester may be used to prepare a number of heterocyclic
compounds (see Ch. XXX).

232 ORGANIC CHEMISTRY

MALONIC ESTER SYNTHESES

Malonic ester, CH 2 (C0 2 C a H 5 ) 2 , which is the diethyl ester of malonic acid
CH 2 (C0 2 H) 2 , is prepared by dissolving potassium cyanoacetate in ethanol,
adding concentrated hydrochloric acid, and warming the mixture on the
water-bath :

NOCH 2 — C0 2 K + 2C 2 H 5 OH + 2HCI — >

CH 2 (C0 2 C 2 H 5 ) 2 + KC1 + NH 4 C1 (v.g.)

Malonic ester may be prepared (yield 82-84 per cent.) by refluxing cyano-
acetic acid in ethanol in the presence of chlorosulphonic acid (Dvorruk et al.,
1953)-

It is a pleasant-smelling liquid, b.p. 199 °. Its use as a synthetic reagent
depends on two chemical properties.

1. With sodium ethoxide it forms a sodium derivative, sodiomalonic
ester, which reacts with compounds containing a reactive halogen atom,
e.g., alkyl halides, acid chlorides, halogen-substituted esters, etc. (cf. aceto-
acetic ester) . In all cases the yields are g.-v.g. The anion of the sodium salt
is probably a resonance hybrid, and when it reacts with a halogen reagent,
the entering radical becomes attached to the carbon atom :

OH O

C 2 H 5 0— C— CH 2 — C— OC 2 H s ^ C 2 H 5 0— C=CH— C— OC 2 H 5 C,H,0Na >

O O

C 2 H s 0-C=

=CH— C— OC„H s NaV-> C,H R 0— C— CH— C

O O

8 ^ ^ ^ --OC 2 H 5 .

O O

II II

RX

Na

> C 2 H 6 0— C— CHR— C— OC 2 H 5 + NaX

The process may then be repeated to produce the disubstituted derivative
of malonic ester:

R-CH(C0 2 C 2 H 5 ) 2 C ' H '° Na > [R-C(C0 2 C 2 H B ) 2 ]-Na + -^ RR'C(C0 2 C 2 H 6 ) 2

These disubstituted derivatives of malonic ester can readily be prepared
in one step by treating the ester with two equivalents of sodium ethoxide and
then with two equivalents of alkyl halide. This procedure is only used if it
is required to introduce two identical alkyl groups. The mechanism of this
reaction is obscure; it will be represented as follows:

CH 2 (C0 2 C 2 H 5 ) 2 2C,H '° Na > [C(C0 2 C 2 H s ) 2 ] 2 -2Na + ^>

R 2 C(C0 2 C 2 H 5 ) 2 + 2 NaX

2. Malonic acid and its derivatives eliminate a molecule of carbon dioxide
when heated just above the melting point of the acid (usually between
150° and 200°) to form acetic acid or its derivatives (the yields are v.g.-ex.) :

C0 2 H-CH 2 -C0 a H — y CH s -C0 a H + C0 2
C0 2 H-CHR-C0 2 H — > R-CH a -C0 2 H + C0 2

Decarboxylation may also be effected by refluxing malonic acid or its
derivatives in sulphuric acid solution.

In a number of cases ethyl cyanoacetate (b.p. 207 ) may be used instead
of malonic ester in many syntheses. The cyano-group is readily converted
into the carboxyl group on hydrolysis.

TAUTOMERISM 233

Synthesis of fatty acids. Malonic ester is preferable to acetoacetic ester
in synthesising acids, and should be used wherever possible. The structural
formula of the acid required is written down, the acetic acid nucleus picked
out, and the required alkyl groups introduced into sodiomalonic ester. The
substituted ester is then refluxed with potassium hydroxide solution, acidi-
fied with hydrochloric acid, and the precipitated acid dried and then heated
just above its melting point. Alternatively, the potassium salt may be
refluxed with sulphuric acid:

(i) n-Valeric acid. CH. 3 -CH 2 'CH 2 ' |CH 8 'CQ 2 H

CH 2 (C0 2 C 2 H 5 ) 2 ^^1> [CH(C0 2 C 2 H 5 ) 2 ]-Na + C,H ' Br >

KOH HC1

C 3 H 7 -CH(C0 2 C 2 H 5 ) 2 — ^> C 3 H 7 -CH(C0 2 K) 2 >-

C 3 H 7 -CH(C0 8 H) 2 I5< ~"° J > CH 3 -CH 2 -CH 2 -CH 2 -C0 2 H

CH 3

I

(ii) Dimethylacetic acid. CH 3 ' |CH>CQ 2 H

CH 2 (C0 2 C 2 H 5 ) 2 <WMfa > [CH(CO i C 1 H B ) J~Na + -^> CH 3 -CH(C0 2 C 2 H B ) 2
CWMu > [CH 3 -C(C0 2 C 8 H 5 ) 2 ]-Na + ^U (CH s ) 8 C(C0 2 C 2 H 5 ) a ^>

(CH J ),C(C0 1 K), -^> (CH 3 ) 2 C(C0 2 H) a I50 " 2 °°°> (CH 3 ) 2 CH-C0 2 H

Since two methyl groups are required for this synthesis, their introduction
can be carried out in one step (see above) :

CH 2 (C0 2 C 2 H 6 ) 2 2C,H '° Na > [C(C0 2 C 2 H 6 ) 2 ]*-2Na+ ^>

(CH 3 ) a C(CO a C 8 H 5 ) 2 K ° H,etc -> (CH 3 ) 8 CH-C0 2 H

Synthesis of dicarboxylic acids. Dicarboxylic acids of the type
RR'C(CO a H) 2 are readily prepared from malonic ester as shown above.
They are important only in so far as they are intermediates in the preparation
of fatty acids.

Dicarboxylic acids of the type C0 2 H'(CHa) B 'C0 2 H are very important,
and the malonic ester synthesis is particularly useful for their preparation.
The actual procedure depends mainly on the value of n, and the following
examples illustrate this point :

(i) Adipic acid. |C0 2 H-CH 2 -| CH g -CH 2 i-CH 2 -C0,HJ

The acetic acid is blocked off at one end, and the remaining fragment is
considered from the point of view of accessibility. In this example the
fragment required is y-bromobutyric ester. This is not readily accessible,
and so the procedure now is to block off the acetic acid nucleus at the other
end of the adipic acid molecule, and then to use too molecules of malonic
ester and the a.u>-dihalide of the polymethylene fragment that joins together the
two acetic acid nuclei. In this case it is ethylene bromide, and since this is
readily accessible, the synthesis of adipic acid may be carried out as follows:

[(C 2 H 5 8 C) 8 CH]-Na + + BrCH 2 -CH 2 Br + Na + [CH(CO a C 8 H 6 ) 2 ]- — >

(C 2 H B O a C) 2 CH-CH 2 -CH 8 -CH(C0 2 C 8 H s ) 2 -^^>

(H0 2 C) 2 CH-CH 8 -CH 2 -CH(C0 2 H) 2 ' 5 °" 20 ° > COsH-CHa-CHg-CHa-CHa'COgH

234 ORGANIC CHEMISTRY

Should the fragments required by either route be inaccessible, then a
method not involving the use of malonic ester may be more satisfactory
(see p. 379).

(ii) Succinic acid. |CQ 2 H-CH 2 - CH 2 -C0 2 H

If owe acetic acid nucleus is blocked off, the fragment required is ethyl
chloroacetate. This is readily accessible and hence succinic acid may be
synthesised from one molecule of malonic ester as follows :

[CH(C0 2 C 2 H 5 ) s ]-Na + + C1CH 2 -C0 2 C 2 H 5 — > (C 2 H 5 2 C) 2 CH-CH 2 -C0 2 C 2 H 5

W K ° H -> (H0 2 C) 2 CH-CH 2 -C0 2 H I5 °" 2 °°° > C0 2 H-CH 2 -CH 2 -C0 2 H

(ii) HC1

If the other acetic acid nucleus in succinic acid is blocked off, there is no
intervening fragment left. Hence if two moelcules of malonic ester could
be united directly, this would offer an alternative route for the synthesis
of succinic acid. Actually, the union of two malonic ester molecules may
be effected by means of iodine. Thus :

[(C 2 H 6 2 C) 2 CH]-Na + + I, + Na + [CH(C0 2 C 2 H 5 ) 2 ]^-».

(C a H 5 2 C) 2 CH-CH(C0 2 C 2 H 8 ) 2 ® K °" > (H0 2 C) 2 CH-CH(C0 2 H) 2

(l]J rlCI

I5 °~ 20 °° > COoH-CH,-CH,-C0 2 H

Synthesis of ketones. Bowman et al. (1952) have introduced the following
general synthesis of ketones and p-keto-esters (R' = tetrahydropyran-2-ol) :

R-CH(C0 2 R') 2 — " )Na > R"-COCR(C0 2 R') 2 CH '" C °' H >

v * '* (ii) R"-C0C1 ^ i ^ it reflux

R"-COCH 2 -R (50-92%)

Johnson et al. (1952) have used tert. -butyl esters of malonic acid (R' = tert.-
butyl):

(i)NaH t,,,™™,™™ CH »O S0 - H

R-CH(C0 9 R') 2 — > R"-CO-CR(C0 2 R') 2

V ' n ;(ii) R"-C0C1 k 2 n CH,-CO a H

R"-CO-CH 2 -R (56-85%)

Synthesis of higher ketonic acids. Sodiomalonic ester is treated with the
acid chloride-ester derivative of a dibasic acid, e.g., e-ketoheptoic acid:

[(C 2 H 5 2 C) 2 CH]-Na + + COCl-(CH 2 ) 4 -C0 2 C 2 H 5 — >
(C 2 H 5 2 C) 2 CH-CO(CH 2 ) 4 -C0 2 C 2 H 5 -^> (H0 2 C) 2 CH-CO(CH 2 ) 4 -C0 2 H

150-200

■> [C0 2 H-CH 2 -CO-(CH 2 ) 4 -C0 2 H] — > CH 3 -CO(CH 2 ) 4 -C0 2 H

The intermediate p-keto-dicarboxyhc acid is unstable and is readily de-
carboxylated. This might have been anticipated, since it may be regarded
as a derivative of acetoacetic acid, which is readily decarboxylated on
warming.

Synthesis of polybasic acids. It has been pointed out previously (p. 211)
that the monoalkyl derivatives of malonic acid are readily brominated, and
that these bromo-derivatives give oc-bromo-fatty acids on decarboxylation.

TAUTOMEEISM 235

Malonic ester is also readily brominated, and this monobromo-derivative
may be used to synthesise polybasic acids, e.g.,

CH 2 (C0 2 C 2 H 5 ) 2 -%- CHBr(C0 2 C 2 H 5 ) 2
CHBr(C0 2 C 2 H 5 ) 2 + [CH(C0 2 C 2 H 5 ) 2 ]-Na + — ->

Br,

NaBr + (C 2 H 5 2 C) 2 CH-CH(C0 2 C 2 H 5 ) 2 — =->
(C 2 H 5 2 C) 2 C— Br 2[CH {co 8 c ! H 5 ) a ]-Na+ (C 2 H 5 2 C) 2 C— CH(C0 2 C 2 H 5 ) 2
H K 2 C) 2 C— Br (C 2 H 5 2 C) 2 C— CH(C0 2 C 2 H 5 ) 2

(C,

2 ii 5 v -'2^2

Malonic ester may also be used for the preparation of unsaturated acids
(Knoevenagel reaction, p. 280), alicyclic compounds (p. 469), and hetero-
cyclic compounds (p. 386).

HYDROXYALDEHYDES AND HYDROXYKETONES

When naming a compound which contains more than one functional
group, it is necessary to choose one as the principal function. The compound
is then named by using the suffix of the principal function and the prefixes
of the other functions. The carboxylic and the sulphuric acid group are
always chosen as principal functions, and the usual order for choosing the
principal function is :

Carboxylic, sulphonic, acid halide, amide, imide, aldehyde, cyanide,
fsocyanide, ketone, alcohol, phenol, thioalcohol, amine, imine.

Examples. Acetoacetic acid contains a ketonic group and a carboxyl
group, and since the latter is the principal function, acetoacetic acid is
named as a ketone acid, viz., p-ketobutyric acid. Acetoacetic acid may
also be named propan-2-one-i-carboxylic acid — each functional group is
indicated by its appropriate suffix.

TABLE V

Function

Prefix

Suffix

Acid .......

carboxy

carboxylic or -oic

Alcohol ......

hydroxy

ol

Aldehyde ......

oxo, aldo (for
aldehydic O) or
formyl (for CHO)

al

Amine ......

amino

amine

Azo-derivative

azo

—

Azoxy-derivative .

azoxy

—

Carbonitrile (nitrile)

cyano

carbonitrile

Double bond

—

ene

Ether ....

alkoxy

—

Ethylene oxide, etc.

epoxy

—

Halogenide (halide)

halogeno (halo)

—

Hydrazine .

hydrazino

hydrazine

Ketone

oxo or keto

one

Mercaptan .

mercapto

thiol

Nitro-derivative .

nitro

—

Nitroso-derivative

nitroso

—

" Quinquevalent " nitrogen

—

onium, inium

Sulphide

alkylthio

—

Sulphinic derivative

sulphino

sulphinic

Sulphone

sulphonyl

—

Sulphonic derivative

sulpho

sulphonic

Sulphoxide

sulphinyl

—

Triple bond

—

yne

Urea .......

ureido

urea

236 ORGANIC CHEMISTRY

CH 2 OH>CH 2 *CHO. This is both an alcohol and an aldehyde, so that,
bearing in mind the order of preference, the name of the compound will be
[J-hydroxypropionaldehyde.

Table V indicates the prefixes and suffixes used for designating the
functions (in alphabetical order).

The simplest hydroxyaldehyde is glycolaldehyde (hydroxy ethanal),
CH 2 OH*CHO. It may be prepared by the oxidation of glycol with Fenton's
reagent (hydrogen peroxide and ferrous sulphate) :

CH 2 OH-CH 2 OH + H a 2 FeS °' -> CH 2 OH-CHO + 2H 2

It may also be prepared by ozonolysis of allyl alcohol:

CH 2 — O— CH-CH 2 OH H0
CH 2 :CH-CH 2 OH— ^->I I -^CHO-CH 2 OH

The most convenient means of preparing glycolaldehyde, however, is by
heating dihydroxymaleic acid, which is obtained from the oxidation of
tartaric acid (q.v.) :

HO a C— C— OH TH— C— OH"! CHO

J v

;— c— oh rH— c— OH-i

\\ __ >2 co 2 + n

:— c— OH Lh— c— OH J

hox— c— oh Lh— c— OH J CH 2 OH

A:

2

Glycolaldehyde exists in the solid form as the dimer, m.p. 96 ; but in
aqueous solution it exists as the monomer which forms the stable hydrate,
CH(OH) 2 -CH 2 OH (cf. chloral, p. 168). Careful oxidation of glycolaldehyde
with bromine water produces glycollic acid :

CH 2 OH-CHO + [O] — > CH 2 OH-C0 2 H

Glycolaldehyde is a powerful reducing agent, reducing ammoniacal silver

nitrate and Fehling's solution at room temperature. With phenylhydrazine

„ „ „ it forms the osazone. This osazone is identical with that

■ .JN ,JN "'W^s formed from glyoxal, and the mechanism of its formation is

w\r \rw r tt st ^ an °P en q 1168 ^ 011 ( see carbohydrates, p. 443).

n.JN-JNtfC 6 ti 5 Glycolaldehyde undergoes the aldol condensation in the
presence of alkali; with sodium hydroxide solution a tetrose sugar is formed,
and with sodium carbonate solution, a hexose sugar:

2 CH 2 OH-CHO Na ° H > CH 2 OH-CHOH-CHOH-CHO

3 CH 2 OH-CHO Na,C °' > CH 2 OH-CHOH-CHOH-CHOH-CHOH-CHO

Glycolaldehyde is formed in small amounts when formaldehyde is allowed
to stand in the presence of calcium carbonate :

2H-CHO CaC °'> CH a OH-CHO

Glycolaldehyde is a useful starting material in a number of organic syntheses,
but since it readily polymerises, the ethyl ether is used in synthetic work. The
ethyl ether may be prepared by catalytically dehydrogenating ethyl cellosolve :

Cu

C a H 5 OCH 2 -CH 2 OH- ->C 2 H s O-CH 2 -CHO + H a (43%)

250°

The ethyl group is easily eliminated when necessary by means of concentrated
hydriodic acid or hydrobromic acid (p. 142).

TAUTOMERISM 237

Aldol (acetaldol, §-hydroxybutyraldehyde, 3- hydroxybutanal),
CH 3 'CHOH-CH 2 -CHO, may be prepared by the aldol condensation of
acetaldehyde (p. 157) :

2CH S -CH0 Na ° H > CH 8 -CHOH-CH 2 -CHO

It is a colourless syrupy liquid, b.p. 83°/20 mm., miscible with water and
ethanol. When heated, it is dehydrated to crotonaldehyde:

CH 3 -CHOH-CH 2 -CHO — >- CH 3 -CH:CH-CHO + H 2

There is some doubt about the structure of aldol. Recent work suggests
it is an equilibrium mixture of p-hydroxybutyraldehyde and the cyclic
hemiacetal:

CH 3 -CHOH-CH 2 -CHO ^ CH 3 -CH-CH 2 -CHOH

This is an example of ring-chain tautomerism.

The simplest hydroxyketone is hydroxy acetone {acetol, pyruvic alcohol),
CH 3 -COCH 2 OH. The best method of preparation is by heating bromo-
acetone with potassium hydroxide in methanolic solution, and adding ethyl
formate:

H-C0 2 C 2 H B + KOH — >- H-C0 2 K + C 2 H 5 OH
CHj-CO-CHaBr + H-C0 2 K — > CH 3 -CO-CH 2 -OOC-H + KBr
CH 3 -COCH 2 -OOOH + CH 3 OH — >CH 3 -COCH 2 OH + H-C0 2 CH 3 (54-58%)

This is an example of alcoholysis, hydroxyacetone being replaced by
methanol. .

Hydroxyacetone is a colourless liquid, b.p. 145 , soluble in water, ethanol
and ether. It reduces ammoniacal silver nitrate, thereby being oxidised
to DL-lactic acid, and reduces Fehling's solution, thereby being oxidised
to a mixture of formic and acetic acids. Although ketones do not normally
reduce ammoniacal silver nitrate and Fehling's solution, a-hydroxyketones
are exceptions. Similarly, ketones form phenylhydrazones with phenyl-
hydrazine, but a-hydroxyketones f orm osazones ; thus hydroxyacetone gives
the same osazone as methylglyoxal (cf. glycolaldehyde) :

CH 3 CH 3

l c^-nh-nh^ ( i :N . NH . CgH5

CH 2 OH dH:N-NH-C 6 H 5

Hydroxyacetone appears to exist as an equilibrium mixture of hydroxy-
ketone and cyclic hemiacetal forms:

CH 3 -CO-CH 2 OH :?=* CH 3 -C—^CH 2

The structure of this cyclic hemiacetal is related to that of ethylene oxide
(p 250) which has the property of reducing ammoniacal silver nitrate. It
is therefore possible that the reducing properties of a-hydroxyketones are
due to the presence of the oxide-ring form. This is an example of ring-chain
tautomerism.

Diacetone alcohol (4-hydroxy-4-methytyentan-2-one),

238 ORGANIC CHEMISTRY

(CH 3 ) 2 C(OH) , CH 2 'CO , CH 3 , may be prepared by the condensation of acetone
in the presence of barium hydroxide :

2CH 3 -COCH 3 Ba( ° H) ' > (CH 3 ) 2 C(OH)-CH 2 -CO-CH 3

Diacetone alcohol is a colourless liquid, b.p. 164°. When heated with
a trace of acid or iodine, it eliminates a molecule of water to form mesityl
oxide:

(CH 3 ) 2 C(OH)-CH 2 -CO-CH 3 — -> (CH 3 ) 2 C=CH-CO-CH 3 + H 2

Diacetone alcohol is oxidised by sodium hypobromite to p-hydroxywovaleric
acid (cf. haloform reaction) :

(CH 3 ) 2 C(OH)-CH 2 -CO-CH 3 — W (CH 3 ) 2 C(OH)-CH 2 -C0 2 H
Diacetone alcohol is a very good solvent for cellulose esters.

DIALDEHYDES AND DIKETONES

Dialdehydes. The simplest dialdehyde is glyoxal (oxaldehyde, ethanedial) ,
CHO-CHO. It may be obtained by the oxidation of ethanol, acetaldehyde
or glycol with nitric acid, but the yields of glyoxal are poor. It is most
conveniently prepared, as the bisulphite compound, by refluxing a mixture
of paraldehyde, 50 per cent, aqueous acetic acid, dioxan, and selenious acid
(yield 72-74 per cent, on the selenious acid) :

(CH 3 -CHO) 3 + 3H 2 Se0 3 — > 3CHOCHO + 3Se + 6H 2 NaHS °* >

' CHO-CHO-2NaHS0 3 -H 2

Glyoxal is manufactured by the vapour-phase oxidation of glycol with air
at 250-300° in the presence of copper as catalyst.

Glyoxal exists as a colourless polymer giving, on distillation, the monomer.
This is a green vapour, which condenses to yellow crystals (m.p. 15°) which
polymerise on standing to a colourless solid of unknown molecular weight.
Glyoxal exists in aqueous solution as the dihydrate, CH(OH) 2 -CH(OH) 2 ,
which has been isolated (cf. chloral hydrate, p. 168).

Glyoxal undergoes many of the reactions of a dialdehyde; e.g., it reduces
ammoniacal silver nitrate, and forms addition compounds with two mole-
cules of hydrogen cyanide and sodium hydrogen sulphite. It does not
reduce Fehling's solution. Since it has no a-hydrogen atom, glyoxal under-
goes the Cannizzaro reaction in the presence of alkali to form glycollic acid,
one half of the molecule undergoing disproportionation at the expense of the
other half:

CHO-CHO + NaOH — > CH 2 OH-C0 2 Na

With phenylhydrazine, glyoxal forms the osazone (this is identical with
that from glycolaldehyde, see p. 236) :

cho ch:n-nh-c b h 5

i

+ 2C 6 H 5 -NH-NH 2 — > I + 2 H 2

HO ch:n-nh-c 6 h 5

It also combines with o-phenylenediamines to form quinoxaliries (hetero-
cyclic compounds); e.g., with o-phenylenediamine it forms quinoxaline
itself:

/W NH s Q^jj /V N VH4-2H0

■ _|_ j. 11 I V" + zn a u

I

^\nh. o^ 11 \/-W ch

*2

TAUTOMEEISM 239

It combines with ammonia to form the heterocyclic compound glyoxaline
(iminazole). The mechanism of this reaction is uncertain; one suggestion
is that one molecule of glyoxal breaks down into formic acid and formalde-
hyde, and the latter reacts as follows:

CHO NH 3 HC N

I + — > 5 \\ +3 H s°

CHO H-CHO HC V /CH

NH 3 NNH/

Glyoxal is one of the simplest coloured organic compounds, and when
reduced it forms the colourless compound glycol (see Ch. XXXI).

Methylglyoxal (pyruvaldehyde), CH 3 -COCHO, may be prepared by the
oxidation of hydroxyacetone or by the hydrolysis of oximinoacetone:

ch 3 -coch 2 oh -^-> ch 3 -cocho
ch 3 -coch:noh —> ch 3 -cocho

It is manufactured by the vapour-phase oxidation of propylene glycol with
air at 250-300 in the presence of copper as catalyst:

CH 3 -CHOH-CH 2 OH -^-> CH 3 -COCHO + 2H 2

It is a yellow oil with a pungent odour, and begins to boil at 72 to give a
light green vapour. The liquid form of methylglyoxal is the dimer, and
this slowly polymerises at room temperature to form a glassy mass of
unknown molecular weight (cf. glyoxal).

Succinaldehyde (succindialdehyde, butanedial), CHO'CH 2 'CH 2 -CHO, may
be prepared by the ozonolysis of hexa-i : 5-diene:

ch 2 :ch-ch 2 -ch 2 -ch:ch 2 ozonolysis > choch 2 -ch 2 -cho

Another method of preparation is to allow pyrrole and hydroxylamine to
interact, and to treat the succinaldoxime so formed with aqueous nitrous
acid:

ch=ch x NH . 0H ch 3 -ch:n-oh HNOj ch 2 -cho

I >NH -"'•"" >| ' HNO ' > |

£h==cw ch 2 -ch:n-oh t

H 2 -CHO

Succinaldehyde is a colourless oil, b.p. 170 , which readily polymerises.
It undergoes the usual reactions of a (di)aldehyde ; e.g., it reduces ammoniacal
silver nitrate and Fehling's solution, forms addition compounds with two
equivalents of hydrogen cyanide and sodium hydrogen sulphite ; etc. When
treated with phosphorus pentoxide, ammonia and phosphorus pentasulphide,
it gives respectively furan, pyrrole and thiophen:

CH=CH\
CH=CH X

CH a -CHO CH=CHOH

CH a -CHO "~~ CH=CHOH

P.O.

^>| H_C %H

p,s,

CH==QW
CH=CH\
CH=CH X

240 ORGANIC CHEMISTRY

Diketones. These are classified as a, p, y . . . diketones according as
the two carbonyl groups are in the 1:2, 1:3, 1:4,... positions
respectively.

Butane-2 : 3-dione, dimethylglyoxal (diacetyl), CH 3 *CO-COCH 3 , is the
simplest a-diketone. It may be prepared:

(i) By the hydrolysis of oximinobutan-2-one formed by the action
of nitrous acid on the ketone:

CH„-COCH 2 -CH 3 ^\ CH 3 -COC(:NOH)-CH 3 -^V CH 3 -COCO-CH 3
(ii) By the oxidation of butan-2-one with selenium dioxide :
CH 3 -CO-CH 2 -CH 8 + Se0 2 — >■ CH 3 -COCOCH 3 + Se + H 2
(iii) By the catalytic dehydrogenation of acetoin:

CH 3 -COCH(0H)-CH„ -^ CH 3 -COCOCH 3 + H 2

Acetoin may be readily oxidised by bismuth oxide in acetic acid; the yield of
dimethylglyoxal is almost quantitative (Rigby, 195 1). Air may be used as the
oxidising agent, with a small amount of bismuth oxide as catalyst (Rigby et al.,
i95i)-

Butane-2 : 3-dione is a yellow oil, b.p. 88°. It gives the usual reactions
of a diketone; e.g., it forms the monoxime and di-oxime with hydroxyl-
amine, the bisulphite compound with sodium hydrogen sulphite, etc. It
is oxidised by hydrogen peroxide to acetic acid:

CH 3 -COCO-CH 3 — V 2CH S -C0 2 H

This reaction is unexpected since hydrogen peroxide does not usually break
a carbon-carbon bond. Reduction with lithium aluminium hydride pro-
duces the corresponding diol.

Butane-2 : 3-dione forms a glyoxaline derivative with ammonia, and a
quinoxaline derivative with o-phenylenediamine (cf. glyoxal, p. 238).

The dioxime of butane-2 : 3-dione, i.e., dimethylglyoxime,
CH 3 'C(;NOH)'C(!NOH)'CH 3 , forms chelate compounds with many metals,
and is used to estimate nickel, whose salts produce a red precipitate when
treated with dimethylglyoxime. According to Brady et al. (1930), the
structure of bisdimethylglyoxime nickel is I. The planar configuration
round the nickel ion has been demonstrated (Sugden et al., 1935), and con-
firmed by X-ray analysis (Rundle et al., 1953).

xO-H-0\

~~=OCH„

(I)

CH 3 -C=N V / N=€-CH 3

ch s -c=n/ ^N=C-CH 3
x O-H~O x

Pentane-2 : 4-dione, acetylacetone, CH 3 'COCH 2 *COCH 3 , is the simplest
p-diketone. It may be prepared:

(i) By means of the Claisen condensation between ethyl acetate and
acetone (yield 38-45 per cent, on the acetone) :

CH 3 -C0 2 C 2 H 5 + CH 3 -COCH 3 <W>W * > CH 3 -CO-CH 2 'CO-CH 3

TAUTOMERISM 24I

(ii) By the ketonic hydrolysis of the acetyl derivative of acetoacetic
ester:

COCHg

CH 3 -CO-CH 2 -C0 2 C 2 H B -£-! > CH 3 -COCH-C0 2 C 2 H 5

ketonic > CH 3 -COCH 2 -CO-CH 3

hydrolysis

(iii) By the condensation of acetic anhydride with acetone in the
presence of boron trifluoride as catalyst (yield 80-85 per cent, on the
acetone) :

CHg-CO-CHs + (CH 3 -CO) 2 -^V CH 3 -COCH 2 -COCH 3 + CH 3 -C0 2 H
Hauser et al. (1956) have prepared (3-diketones as follows:

R'-COCl

R-COCH 3 + NaNH a > NH„ + R-COCHjNa >- R-COCH 2 -COR'

Pentane-2 : 4-dione is a colourless liquid, b.p. 139,7746 mm. It exhibits
tautomerism, e.g., gives a red coloration with ferric chloride (cf. acetoacetic
ester) :

OH

CH 3 -COCH 2 -COCH 3 ^ CH 3 -COCH=OCH 3

It is readily oxidised to acetic acid, and is converted into a mixture of
acetone and acetic acid when heated with potassium hydroxide solution
(cf. acid hydrolysis of acetoacetic ester, p. 228). It forms pyrazoles when
treated with hydrazine or its derivatives; e.g., with phenylhydrazine it
forms 3 : 5-dimethyl-i-phenylpyrazole:

C-CH, CH C-CH.

> CH 3 -C V /N +2H a O

\ N /

C«H S

Pentane-2 : 4-dione forms chelated compounds with various metals,
e.g., iron, aluminium, copper, etc.:

CH 3 — C=0 O— C— CH 3

HC Cu CH

\ - / \ /
CH a — C— O O— C— CH 8

Hexane-2 : 5-dione, acetonylacetone, CH 3 *COCH 2 *CH 2 *COCH 3 , is the
simplest y-diketone. It may be prepared by means of the acetoacetic
ester synthesis :

[CH 3 -CO'CH-C0 2 C 2 H 6 ]-Na + + BrCH 2 -COCH 3 — >
CH 2 -CO-CH 3

i-CH-i

CH 3 -CO-CH-C0 2 C 2 H 5 ket0Plc > CH 3 -CO-CH 2 -CH 2 .CO-CH 3

3 2 2 S hydrolysis 6 2 2 3

242 ORGANIC CHEMISTRY

An alternative and more convenient acetoacetic ester synthesis is as follows:

CH 3 -CO-CH-C0 2 C 2 H K

2[CH 3 -CO-CH-C0 2 C 2 H 5 ]-Na+

i.

CH.-CO

^3

k

-^"S

•C0 2 C 2 H 5

(i) KOH

CH 3 -CO-CH-C0 2 H heat CH 3 -CO-CH
(U)hci CH,-C0-CH-C0 2 H

x a '

'v-

> | +2C0 2

CH 3 -CO-CH 2

Hexane-2 : 5-dione is a colourless liquid, b.p. 192-194 . It exhibits
tautomerism:

OH OH

CH,-CO-CH„-CH a -CO-CH,

CH,-C=CH— CH=C-CH

:J>

It readily forms five-membered rings : with phosphorus pentoxide, 2 : 5-
dimethylfuran; with ammonia, 2 : 5-dimethylpyrrole; and with phosphorus
pentasulphide, 2 : 5-dimethylthiophen (cf. succinaldehyde, p. 239) :

CH

3

CH=

CH 3
,«CO

CH
.H 3 -CO

A

CH,-CO

H 3

CH 3

I
CH=C— OH

CH=C— OH

CH a

pA >r %

CH=C/
CH a

CH 9

CH=C

^>T" %H

CH=C^

CH,

CH 3

CH=

CH=c/^
CH 3

ALDEHYDIC AND KETONIC ACIDS

Gloxylic acid (glyoxalic acid, oxoethanoic acid), CHO*C0 2 H, is the simplest
aldehydic acid. It occurs in unripe fruits, e.g., gooseberries, and disappears
during ripening; it also occurs in animal tissues and fluids. It may be
prepared by the oxidation of ethanol, glycol or glycollic acid with nitric
acid (yields are poor). It may also be prepared by the reduction of oxalic
acid either electrolytically or by means of magnesium and sulphuric acid,
but it is most conveniently prepared by the hydrolysis of dichloroacetic
acid with water:

CHC1 2 -C0 2 H + H 2 — y CHO-C0 2 H + 2HCI

Glyoxylic acid crystallises from water with one molecule of water which
is combined as water of constitution, CH(OH) 2 -C0 2 H, i.e., dihydroxy-acetic
acid (cf. chloral hydrate). The anhydrous acid may be obtained as a thick
syrup by evaporating the aqueous solution over phosphorus pentoxide in
vacuo.

TAUTOMERISM 243

Glyoxylic acid gives all the reactions of an aldehyde and an acid; e.g.,
it reduces ammoniacal silver nitrate, forms a bisulphite compound, etc.
Since it has no a-hydrogen atoms, glyoxylic acid undergoes the Cannizzaro
reaction to form glycollic and oxalic acids :

2CHOC0 2 H Na ° H > CH 2 OH-C0 2 Na + (CO a Na) 2

When glyoxylic acid is reduced by means of metal and acid, tartaric acid
is obtained as well as glycollic acid:

CH(OH)-CO a H

[H]

> uijUn-LUsti -f- |

H(OH)-CO a H

CHOC0 2 H J^-> CH 2 OH-C0 2 H + I

In this respect glyoxylic acid resembles acetone which also gives a bimolecular
reduction product, pinacol (see p. 163).

Higher homologues of glyoxylic acid are known, but they are not important.

Pyruvic acid (acetylformic acid, pyroracemic acid, a.-ketopropionic acid,
2-oxopropanoic acid), CH 3 -COC0 2 H, is the simplest keto-acid. It may be
prepared :

(i) By heating tartaric acid alone, or better, with potassium hydrogen
sulphate at 210-220°. The reaction is believed to take place via the
formation of hydroxymaleic acid, I, which rearranges to oxalacetic
acid, II:

CH(OH)-CO a H _ HO CH-C0 2 H CH a -C0 2 H

1 — - — Hi — > I

CH(OH)-C0 2 H C(OH)-C0 2 H COC0 2 H

(I) (")

— ► CH.-CO-CO.H + C0 2 (50-55%)

This is the best method for preparing pyruvic acid, and it was this method
which gave rise to the name pyroracemic acid.

(ii) By the oxidation of lactic acid with silver oxide suspended in
water, or with Fenton's reagent :

CH 3 -CH(OH)-CO a H + [O] — > CH 3 -COC0 2 H + H 2

(iii) By the hydrolysis of a : a-dibromopropionic acid with water:

CH 3 -CBr 2 -CO a H + H 2 — > CH 3 -COC0 2 H + 2HBr

(iv) By the hydrolysis of acetyl cyanide formed by the action of
potassium cyanide on acetyl chloride :

TCCN TTC1

CH 3 -COCl > CH 3 -COCN > CH 3 -COC0 2 H

H s O

A general method for preparing a-keto-acids is to reflux a mixture of oxalic
ester, fatty acid ester and sodium ethoxide in ether for 16 hours, and then boil
the oxalo-ester with 10 per cent, sulphuric acid for 6 hours (Adickes and Andresen,
1943) :

R R

C.H.ONa^ ^ I ^ o tt H.SO,

COOC a H 6 + H 2 C-C0 2 C 2 H 5 > COCH-C0 2 C 2 H 6

COOCoH 5

R
CO-CH-COjH
C0 2 H

-CO s
>■ R-CH 2 -CO-CO a H (up to 94%)

244 ORGANIC CHEMISTRY

Shreiber (1954) has prepared a-keto-acids as follows:

R-CH 2 -CO a C 2 H 5 -™^> R-CHNa-C0 2 C 2 H B J£2^V

R-CH-C0 2 C a H 5 -|
_ |/OC 2 H 6

o— cc

Na +

s CO a C a H B _
— >■ C a H B ONa + CO a C 2 H 5 -CHRCO-CO a C a H B

hydrolysu^ ^j^qjj + R. C H a -CO-CO a H

Another general method for preparing a-keto-esters is by the action of acetic
anhydride on a glycidic ester (p. 159), followed by hydrolysis (Vogel et al., 1950) :

R \ /°\ (CH,-CO)-0 R \

)C CH-CO a C a H B > )C(0-COCH 3 )-CH(0-COCH 8 )-C0 2 C a H,

R'/ R'/

— CH *CO H \ H O \.

'—±-> >C=C(0-COCH 8 )-CO a C a H 5 — ?-> XH-CO-CO,C a H 5

R'/ R'/

Pjrruvic acid is a colourless liquid, b.p. 165° (with slight decomposition),
and smells like acetic acid. It is miscible in all proportions with water,
ethanol and ether. It behaves as a ketone and as an acid; it forms an
oxime, hydrazone, etc. It is reduced by sodium amalgam to lactic acid
and dimethyltartaric acid (cf. glyoxylic acid, above) :

CH 3 -C(OH)-C0 2 H
3CH S -COCO„H i^-> CH 3 -CH(OH)-C0 2 H +

CH3-C(OH)-C0 2 H

Pyruvic acid reduces ammoniacal silver nitrate, itself being oxidised to
acetic acid:

CH 3 -CO-C0 2 H + [O] — > CH 3 «C0 8 H + C0 2

It is oxidised by warm nitric acid to oxalic acid:

CH 3 -COC0 2 H + 4[0] — > (C0 2 H) 2 + C0 2 + H 2

Pyruvic acid is easily decarboxylated with warm dilute sulphuric acid to
give acetaldehyde:

CH 3 -COCO a H — > CH 3 -CHO + C0 2

On the other hand, when warmed with concentrated sulphuric acid, pyruvic
acid eliminates a molecule of carbon monoxide to form acetic acid:

CH 3 -CO-C0 2 H — > CH 3 -C0 2 H + CO

Both of these reactions with sulphuric acid are characteristic of a-ketoacids.
The methyl group is made reactive by the adjacent carbonyl group, and
so pyruvic acid undergoes many condensation reactions characteristic of
a compound containing an active methylene group; e.g., in the presence of
dry hydrogen chloride, pyruvic acid forms a-keto-y-valerolactone-y-
carboxylic acid:

C0 2 H

CH 3 -C— CH 2 — CO-C0 2 H

OH

C0 2 H

H '° > CH 3 -C— CH 2 — CQ-CO

C0 2 H

CH 3 -C=0 + CH 3 -CO-C0 2 H

-0-

TAUTOMERISM 245

Pyruvic acid is a very important substance biologically, since it is an
intermediate product in the metabolism of carbohydrates and proteins.

Laevulic acid ($-acetylpropionic acid, y-ketovaleric acid, butan-3-one-i-
carboxylic acid), CH 3 'CO*CH 2 *CH 2 'C0 2 H, is the simplest y-keto-acid. It
may be prepared via acetoacetic ester as follows:

[CH 3 -CO-CH-C0 2 C 2 H B ]-Na + + ClCH.j-CO.AH,; — >
CHg'COjCgHj

)-CH-i

NaCl + CH3-CO-CH-C0 2 C„H 5 -^^->CH3-CO-CH.-CH 2 -C0 2 H

hydrolysis

It may also be prepared by heating a hexose sugar, particularly laevulose,
with concentrated hydrochloric acid:

C 6 H 12 6 — > CH S -COCH 2 -CH 2 -C0 2 H + H-C0 2 H + H 2

In practice it is customary to use cane-sugar as the starting material:

C 12 H 22 O u — > 2CH 3 -COCH 2 -CH 2 -C0 2 H + 2H-C0 2 H + H 2 (21-22%)

By heating dilute solutions of sucrose under pressure with hydrochloric acid,
the yield is increased to 50 per cent.

Laevulic acid is a crystalline solid; m.p. 34 °, very soluble in water,
ethanol and ether. It behaves as a ketone (forms oxime, etc.) and as an
acid (forms esters, etc.). On the other hand, many of the reactions of
laevulic acid indicate that it exists in the lactol form, i.e., hydroxy-lactone
(this is an example of ring-chain tautomerism) :

CH.-CO-CH 3 CH 2 -C(OH)-CH 3

i — I >°

CH 2 -C0 2 H CH 2 -CO

When heated for some time, laevulic acid is converted into a- and 3-
angelica lactones:

CH=C— CH 3 CH— CH— CH 3

I >> II >>

CH.-CO CH— CO

It should be noted that a- and y-keto-acids differ from 0-keto-acids in
that their esters do not form sodio-derivatives, and the acids are not readily
decomposed by moderate heating (cf. acetoacetic acid).

A general method for preparing y-keto-acids is to treat a ketone with lithium
amide and lithium bromoacetate; the ketone undergoes a-carboxymethylation
(Puterbaugh et al., 1959) :

LiNH,

R-COCHjR' — > R-CO'CHR''CH a 'CO»H

Z CH.BrCO.Li a z

QUESTIONS

1. Show by means of equations how you would synthesise: — (a) hexane-2 : 5-dione,
(b) methyl »-amyl ketone, (c) methyl isopropyl ketone, (d) a-methylsuccinic acid,
(e) a : B-dimethylbutyric acid, (/) Et a CH-CO a H, (g) Me-CO-Pr,
(h) CH.-CH(CO a H)-CH(Me)-C0 8 H, (») CH s -CO-CH(Me)-CH,-CO a H,
(«) CH 3 -CH a -CH(Me)-CO-NH 2 , (ft) CH a -CO-CH,-CH 2 -CO a H,
(/) CH a OH-CH a -CH a -CO a H, (m) CH 3 -CH(CO a H)-CH(CO a H)-CH,-CO a H,
(») CH a 'CH 8 'CH(Me)-CH(Et)-CO a H, (0) Me a CH-CH s -CHj-CH(Me)-CH 8 -CO a H,
(p) Me a CH-CO-CHMe a , (?) Me 2 EtC-C0 8 H.

246 ORGANIC CHEMISTRY

2. Starting with compounds containing three or less carbon atoms, how would you
synthesise:— (a) CH 3 -COCH 2 -CHO, (6) CH 3 -COCH 2 -COC0 2 Et, (c) CHOCH 2 -CO a Et,
(d) CH 3 -CHOH-CH 2 -CH 2 OH, (e) CH 3 -CHBr-CH 2 -C0 2 H, (J) Me 2 CH-CH 2 -CH 2 -CH 3 ,
(g) CH 3 -CH 2 -CHC1-CH 2 -C0 2 H, (h) Me 2 CBr-CH 2 -C0 2 H?

3. Define and give examples of: — (a) tautomerism, (6) isomerism, (c) desmotropism,
(d) resonance, (e) enolisation, (/) cationotropy, (g) anionotropy, (h) prototropy, (i) the
Claisen condensation, (j) chelate compounds, (k) water of constitution.

4. Discuss the methods for determining the composition of keto-enol mixtures, and
discuss the relation between structure and enol content.

5. Name the compounds and state the conditions under which they are formed
when E.A.A. is treated with:— (a) NaOH, (b) H 3 P0 4 , (c) HN0 2 , (d) CuS0 4 , (e) NH 3 ,
(/) MeNH 2 , (g) MeMgl, (h) CH 3 -CHO, (i) AcCl, (j) ArN 2 Cl, (h) 3 , (I) heated under reflux.

6. Give one convenient method for preparing: — (a) glycolaldehyde, (b) aldol, (c)
acetol, (d) glyoxal, (e) methylglyoxal, (/) dimethylglyoxal, (g) succinaldehyde, (h) acetyl-
acetone, (i) acetonylacetone, (j) glyoxylic acid, (A) pyruvic acid, (/) lsvulic acid.

7. Which of the compounds in question 6 react with : — (a) NaHSO a , (6) C 6 H 6 -NH-NH 2 ,
(c) P 2 5 , (d) NH 3 , (e) P 2 S 5 ? Name the compounds and state under what conditions
they are formed.

READING REFERENCES

Organic Reactions, Wiley, Vol. I (1942), Ch. 9. The Acetoacetic Ester Condensation

and Certain Related Reactions.
Cooper and Barnes, An Improved Kurt Meyer Titration, Ind. Eng. Chem. (Anal. Ed.),

1938, 10, 379.
Schwarzenbach and Wittwer, The Bromometric Estimation of Enol Content by Means

of a Flow Apparatus, Helv. Chim. Acta, 1947, 60, 657, 653, 663, 669.
Barnes and co-workers, Direction of Enolisation, /. Amer. Chem. Soc, 1945, 37, 132,

134-

Zuffanti, Enolisation, /. Chem. Educ, 1945, 22, 230.

Baker, Tautomerism, Routledge & Sons (1934).

Braude, Anionotropy, Quart. Reviews (Chem. Soc), 1950, 4, 404.

Feigl and Suter, The Inner Complex Salts of Dimethylglyoxime, J.C.S., 1948, 378.

Joshi and Tuli, Refrachor: A New Physical Constant, ibid., 1951, 837.

Organic Reactions, Wiley, Vol. VIII (1954), Ch. 3. The Acylation of Ketones to form
jS-Diketones or j3-Ketoaldehydes. Vol. IX (1957), Ch. 1. The Cleavage of Non-
enolisable Ketones with Sodium Amide. Ch. 4. The Alkylation of Esters and
Nitriles.

Gunstone, Recent Developments in the Preparation of Natural and Synthetic Straight-
Chain Fatty Acids, Quart. Reviews (Chem. Soc), 1953, 7, 175.

Emmons and Hawthorne, Primary and Secondary Isotope Effects in the Enolisation
of Ketones, /. Amer. Chem. Soc, 1956, 78, 5593.

CHAPTER XI

POLYHYDRIC ALCOHOLS

DIHYDRIC ALCOHOLS OR GLYCOLS

Dihydric alcohols are compounds- containing two hydroxyl groups. They
are classified as a, p, y . . . glycols, according to the relative positions of the
two hydroxyl groups: a is the i : 2 glycol; p, 1 : 3; y, 1 : 4 . . . Although
it is unusual to find a compound with two hydroxyl groups attached to the .
same carbon atom, ether derivatives of these 1 : 1 glycols are stable, e.g.,
acetals (p. 161). The commonest glycols are the a-glycols.

Nomenclature. The common names of the a-glycols are derived from the
corresponding olefin from which they may be prepared by direct hydroxyl-
ation, e.g.,

CH 2 0H-CH 2 0H ethylene glycol

CH 2 -CH0H-CH 2 0H propylene glycol
(CH 3 ) 2 CH0H-CH 2 0H *'sobutylene glycol

p-, y- . . . Glycols are named as the corresponding polymethylene glycols,

CH 2 0H«CH 2 -CH 2 0H trimethylene glycol

CH 2 OH-CH 2 -CH 2 -CH 2 -CH 2 OH pentamethylene glycol

According to the I.U.P.A.C. system of nomenclature, the class suffix is -iiol,
and numbers are used to indicate the positions of side-chains and the two
hydroxyl groups, e.g.,

CH 3 -CH0H-CH 2 0H propane-i : 2-diol

CH S CH 3
CH 2 OH-CH 2 -CH-CH 2 'CH-CH 2 OH 2 : 4-dimethylhexane-i : 6-diol

Ethylene glycol, glycol, (ethane-i : 2-diol), CH 2 OH-CH 2 OH, is the simplest
glycol, and may be prepared as follows:

(i) By passing ethylene into cold dilute alkaline permanganate solution:

CH 2 :CH 2 + [O] + H 2 — >■ CH 2 OH-CH 2 OH

(ii) By passing ethylene into hypochlorous acid, and then hydrolysing
the ethylene chlorohydrin by boiling with aqueous sodium hydrogen car-
bonate :

CH 2 :CH 2 + HOC1 — >■ CH 2 C1-CH 2 0H

CH 2 C1-CH 2 0H + NaHC0 3 — > CH 2 0H-CH 2 0H + NaCl + C0 2

(iii) By boiling ethylene bromide with aqueous sodium carbonate:

CH 2 Br-CH 2 Br + Na 2 C0 3 + H 2 — >

CH 2 0H-CH 2 0H + 2NaBr -f C0 2 (50%)

The low yield in this reaction is due to the conversion of some ethylene
bromide into vinyl bromide:

CH 2 Br-CH 2 Br + Na 2 C0 3 — -> CH 2 :CHBr + NaBr + NaHC0 3

If aqueous sodium hydroxide is used instead of sodium carbonate, vinyl
bromide is again obtained as a by-product. The best yield of glycol from
ethylene bromide is obtained by heating ethylene bromide with potassium

247

248 ORGANIC CHEMISTRY

acetate in glacial acetic acid, and subsequently hydrolysing the glycol
diacetate with hydrogen chloride in methanolic solution:

CH 2 Br-CH 2 Br + 2CH 3 -C0 2 K — >

CH 2 -(OCOCH 3 )-CH 2 (OCOCH 3 ) + 2KBr -^> CH 2 OH-CH 2 OH
(90%) (83-84%)

(iv) By treating ethylene oxide with dilute hydrochloric acid:

I /O + H a O -^-> CH 2 OH-CH 2 OH
CH /

(v) By the reduction of any of the following compounds:

CHO CHO CH 2 OH C0 2 C 2 H 6

CHO CH 2 OH C0 2 CH 3 C0 2 C 2 H 5

glyoxal glycolaldehyde methyl glycollate ethyl oxalate

Glycol is prepared industrially by methods (ii) and (iv), and by the
catalytic reduction of methyl glycollate which is produced synthetically
(see p. 397).

Glycol is a colourless viscous liquid, b.p. 197°, and has a sweet taste (the
prefix glyc- indicates that the compound has a sweet taste: Greek glukus,
sweet). It is miscible in all proportions with water and ethanol, but is
insoluble in ether. It is widely used as a solvent and as an antifreeze agent.

The chemical reactions of glycol are those which might have been expected
of a monohydric primary alcohol. One hydroxyl group, however, is almost
always completely attacked before the other reacts. This is probably due
to the fact that the primary alcoholic group is more reactive in glycol itself
than in a compound of structure, e.g., R-O-CH^CHgOH. Thus one group
can be made to undergo one type of reaction and the other group another
type of reaction, to give complex products.

(i) When glycol is treated with sodium at 50°, only one alcoholic group
is attacked. To obtain the disodium derivative the temperature must be
raised to 160 :

CH g OH Na CH 2 ONa Na CH 2 ONa

CH.OH 5 °° CH.

OH l6o ° CH„ONa

(ii) Hydrogen chloride converts glycol into ethylene chlorohydrin at
160 . To obtain ethylene chloride it is necessary to raise the temperature
to 200°:

CH 2 OH HC1 CH 2 C1 HC1 CH 2 C1
CH 2 OH l6 °° CH 2 OH 2 °°° CH 2 C1

Ethylene chlorohydrin ($-chloroethyl alcohol, 2-chloroethanol) is a colourless
liquid, b.p. i28-8°. It is very useful in organic syntheses since it contains
two different reactive groups; e.g., by heating with aqueous sodium cyanide
it may be converted into ethylene cyanohydrin, which, on hydrolysis, gives
(3-hydroxypropionic acid (p. 407) :

CH 2 0H-CH 2 C1 + NaCN > CH 2 OH-CH 2 -CN + NaCl (79-80%)

HO

CH 2 OH-CH 2 -CN »- CH 2 OH-CH 2 -C0 2 H

POLYHYDRIC ALCOHOLS 249

(iii) When glycol is treated with phosphorus trichloride or phosphorus
tribromide, the corresponding ethylene halide is obtained:

CH.OHCH.Br

CH 2 OH CH 2 Br

Phosphorus tri-iodide, however, produces ethylene. Ethylene iodide is
formed as an intermediate, but readily eliminates iodine to form the cor-
responding olefin (see p. 116) :

CH 2 OH „ rCHJ-1 CH

LCH.lJ

CH 2 OH LCH 2 lJ CH a

(iv) When glycol is treated with organic acids or inorganic oxygen acids,
the mono- or di-esters are obtained, depending on the relative amounts of
glycol and acid; e.g., glycol diacetate is obtained by heating glycol with
acetic acid in the presence of a small amount of sulphuric acid as catalyst :

CH 2 OH HS0 CH g -0-CO-CH 3

I + 2CH 3 -C0 2 H H '^' > 1 +2H 2

CH 2 OH CH 2 -0-CO-CH 3

When glycol is heated with dibasic acids, condensation polymers are
obtained :

H0CH 9 -CH,O|H_+HOjOC-(CH 2 ) n -CO |OH + Hj OCH 2 -CH 2 0[H

+ HOjOC-(CH a ) re -CO |OH H — >

HOCH 2 -CH 2 -0-CO-(CH 2 ) B -CO-0-CH 2 -CH 2 -0-CO-(CH 2 )„-CO- - - -

(v) Glycol condenses with aldehydes or ketones in the presence of acid
to yield respectively cyclic acetals or cyclic ketals (1 : 3-dioxolanes) :

CH,OH /R CH 2 0\ /R

+ o=< — ► I )C( + H 2
H 2 OH X R' CH 2 Q/ \R'

i

Compounds of this type are useful in sugar chemistry (see p. 440).

(vi) When glycol is oxidised with nitric acid, glycollic and oxalic acids
may be readily isolated. All theoretically possible oxidation products have
beetf isolated, but since, except for glycollic and oxalic acids, they are more
readily oxidised than glycol itself, they have only been obtained in very poor
yields.

CH 2 OH CHO rol C0 2 H m CO.H ra C0 2 H

CH.

CO,H

'z l

Oxidation of glycol with acid permanganate, acid dichromate, lead tetra-
acetate or periodic acid results in fission of the carbon-carbon bond (see
p. 72).

250 ORGANIC CHEMISTRY

(vii) Glycol is converted in acetaldehyde when heated with dehydrating
agents, e.g.,

CH 2 OH-CH 2 OH Z " C1 ' > CH 3 -CHO + H a

On the other hand, when glycol is heated with a dehydrating agent such
as phosphoric acid, polyethylene glycols are obtained, e.g., diethylene glycol:

2CH 2 OH-CH a OH ~ H '°> CH a OH-CH 2 -0-CH 2 -CH 2 OH

By varying the amount of phosphoric acid and the temperature, the poly-
ethylene glycols U P to decaethylene glycol can be obtained. These con-
densation polymers contain both the alcohol and ether functional groups:
they are soluble in water (alcohol function) and are very good solvents
(ether function). They are widely used as solvents for gums, resins, cellu-
lose esters, etc.

Ethylene oxide, CH a — CH a . According to the I.U.P.A.C. system of nomen-
clature, an oxygen atom linked to two of the carbon atoms in a carbon
chain is denoted by the prefix epoxy in all cases other than those in which a
substance is named as a cyclic compound; thus ethylene oxide is epoxy ethane.
Epoxy-compounds contain the oxiran (oxirane) ring, and ethylene oxide is
also known as oxiran [oxirane). Oxiran compounds are also referred to as
cyclic ethers or alkylene oxides.

CH 2 0H-CH 2 C1 + KOH — > CH a — CH 2 + KC1 + H 2
The mechanism is possibly :

HO" H— O

I fast /°"V ("V slow /-°\

CH 2 — CH 2 C1 > H a O + CH 2 — CH 2 — CI y CH 2 — CH a + Cl~

This is an example of neighbouring group participation.

This method is used industrially, but the potassium hydroxide is replaced
by calcium hydroxide. A recent industrial preparation of ethylene oxide
is to pass ethylene and air under pressure over a silver catalyst at 200-
400 :

C 2 H 4 + J0 2 — > CH a — CH a

Ethylene oxide is a colourless gas, b.p. io^ , soluble in water, ethanol and
ether.

Ethylene oxide undergoes molecular rearrangement on heating to form
acetaldehyde :

CH a — CH a — >■ CH 3 -CH0

It reduces ammoniacal silver nitrate solution, and is reduced to ethanol by
sodium amalgam.

Epoxides are reduced by lithium aluminium hydride to alcohols, un-
symmetrical epoxides giving as main product the more highly substituted
alcohol; thus I and III are the main products:

R-CH-^CH 2 UAM ' > R-CHOH-CH 3 -f R-CH 2 -CH 2 0H

(I) (II)

R 2 C— CHR LlA ' H ' -> R 2 C(OH)-CH 2 -R + R 2 CH-CHOH-R

(III) (IV)

POLYHYDRIC ALCOHOLS 251

Eliel et al. (1956), however, have shown that the reverse takes place if
the reduction with lithium aluminium hydride is carried out in the pre-
sence of aluminium chloride or bromide, i.e., II and IV are now the main
products.

Ethylene oxide is used as an insecticide and in many laboratory syntheses.

Ethylene oxide is converted into ethylene glycol in dilute, acid solution
and into ethylene halogenohydrins with concentrated halogen acid solutions.
Ethylene oxide also forms mono-ethers with alcohols in the presence of a
small amount of acid as catalyst. The mechanisms of these reactions in
acid media probably follow the same pattern, e.g., with dilute acid (with
alcohols, replace H 2 by ROH) :

CH 2 \ CH 2 \ +

)0 + H 3 + ^ )OH + H 2

iH/ CH/

k

H 2 CH 2 x> H 2 0— CH 2 _ H+ CH 2 OH

T >OH — > I > I

CH/ CH 2 — OH CH 2 OH

Evidence from kinetic studies on solutions of epoxides under high pressure
strongly suggests that acid-catalysed hydrolysis is a bimolecular substitution
of the conjugate acid (Whalley et al., 1959).

With concentrated halogen acid :

rv

Cl- CH 2 -rv + CH 2 C1

I >H-^ I

ch/ ch 2 oh

Ethylene oxide also reacts with alcohols under the influence of basic catalysts.
A possible mechanism is:

EtO- CH.-cv x EtO— CH a CH 2 OEt

I >-> T r\ cv — >l +Eto-

CH/ CH 2 0- H— OEt CH 2 OEt

Of particular interest is the addition of hydrogen cyanide to form ethylene
cyanohydrin :

CH 2 — CH 2 + HCN — y CH 2 OH-CH 2 -CN

Sekino (1950) has obtained ethylene cyanohydrin (yield: 95-96 per cent.) by
reaction between liquefied ethylene oxide and liquefied hydrogen cyanide in the
presence of a catalyst such as an alkali-earth oxide.

The structure of ethylene oxide is uncertain; it may contain " bent "
bonds (see cyclopropane, p. 488).

When ethylene oxide is heated with methanol under pressure, the mono-
methyl ether of glycol is formed:

CH 2 — CH a + CH 3 OH — > CH 2 OH-CH 2 -0-CH 3

This is known as methyl cellosolve; the corresponding ethyl ether is known
as ethyl cellosolve. Cellosolves are very useful as solvents since they contain
both the alcohol and ether functional groups (cf. polyethylene glycols,
above).
Ethyl cellosolve may be used to prepare the ethyl ether of glycolaldehyde.

252 ORGANIC CHEMISTRY

The vapour of ethyl cellosolve is dehydrogenated by passing it over copper
heated at about 250 :

Cu

C a H 5 OCH 2 CH 2 OH >■ C 2 H B OCH 2 -CHO + H 2

This ethyl ether derivative does not polymerise so readily as glycolaldehyde,
and so may be used instead of the latter in certain syntheses. The ethyl
group is finally removed by means of concentrated hydrogen bromide.

The further action of ethylene oxide on cellosolves produces carbitols,
e.g.,

CH 2 OH-CH 2 -OCH 3 + CH 2 — CH a — ► CH a OH-CH 2 -OCH 2 -CH 2 -OCH 3
When glycol is treated with ethylene oxide, diethylene glycol is formed:

CH 2 OH-CH a OH + CH 8 — CH a — > CH a OH-CH a -0-CH a -CH 2 OH

Ethylene oxide reacts with ammonia to form a mixture of three amino-
alcohols which are usually referred to as the ethanolamines:

C a H 4 + NH 3 — -> CH a OH-CH 2 -NH 2 C,H '° >

(CH 2 OH-CH a ) 2 NH °' H, ° > (CH 2 OH-CH 2 ) 3 N
The ethanolamines are widely used as emulsifying agents.
The mechanism of these reactions is possibly:

• Ov +

H 3 N CH 2 x> H s N-<:H a H+ + CH 2 NH 2 CH a NH 2

CHj/ CUfi- CH 2 0~ CH 2 OH

In the presence of excess of ethylene oxide, the reaction proceeds with the
ethanolamine now being the amino reagent, etc.

Dioxan (1 : 4-dioxan, diethylene dioxide) may be prepared:

(i) By distilling glycol with a little sulphuric acid or concentrated phos-
phoric acid:

2 CH a 0H-CH a 0H — > 0( )0 + 2JLO

\CH,— CH,/

(ii) By heating 2 : 2'-dichlorodiethyl ether with aqueous potassium
hydroxide :

CH 2 Cl-CH a -0-CH a -CH 2 Cl + 2KOH — >

/CH 2 — CH 2 \
0( )0 + 2KCI + H a O

\CH a — CH/

(hi) Dioxan is prepared industrially by distilling glycol with phosphoric
acid (method i), and by dimerising ethylene oxide with 4 per cent, sulphuric
acid:

/CH 2 — CH a \

2 c 2 H 4 o — > o( ;o

\CH a — CH/

Dioxan is a colourless liquid, b.p. 101-5°, rniscible in all proportions with
water and most organic solvents. It is a useful solvent for cryoscopic and
ebullioscopic work.

POLYHYDRIC ALCOHOLS 253

General Methods of Preparation of 1 : 2 Glycols Other than Glycol itself

1. By the reduction of ketones with magnesium amalgam, e.g., pinacol may be
prepared by adding mercuric chloride in acetone to a mixture of magnesium and
benzene, refluxing the mixture, and then adding water to decompose the magnes-
ium " pinacolate " (the yield is 43-50 per cent, based on the magnesium) :

(CH 3 )X C(CH 3 ) 2 Hi0

2CH 3 -CO-CH 3 + Mg — > I I >

Mg
(CH 3 ) a C(OH)-C(OH)(CH 3 ) a + Mg(OH),

2. By the action of a Grignard reagent on oc-diketones, e.g.,

R R R R

CH 3 -COCOCH 3 J^*> CH 3 j-l>CH 3 -^ CH 3 -C-6CH 3
XMgO OMgX HO OH

3. By the action of a Grignard reagent on a-ketonic esters, e.g.,

R R R R

CH 3 -CO-CO a C a H 5 ^^> CH 3 -i— C-R -^% CH 3 -C-OR
XMgO OMgX HO OH

4. By the catalytic reduction of acy loins:

R-CH(OH)-CO-R — ^-> R-CH(OH)-CH(OH)-R

Ni

5. By hydroxylation of unsaturated compounds (see p. 72).

Pinacol {tetramethylethylene glycol, 2 : 3-dimethylbutane-2 : 3-diol),
(CH 3 ) 2 C(OH)-C(OH)(CH 3 ) 2 , is most conveniently prepared by reducing
acetone with magnesium amalgam (see above). It crystallises out of solu-
tion as the hexahydrate. The most important reaction of pinacol is the
rearrangement it undergoes when distilled with dilute sulphuric acid:

(CH 3 ) 2 C(OH)-C(OH) (CH S ) 2 H ' S °' > CH 3 -CO-C(CH 3 ) 3 +H 2

The pinacolr-pinacoUme rearrangement (p. 171) is general for pinacols (di-
tertiary alcohols).

When sulphuric acid is used as the catalyst, some diene is formed as
by-product:

CH 3 CH S CH 3 CH 3

CH 3 -C C-CH 3 ^% CH 2 =C C=CH 2 + 2 H a O

OH OH
If pinacol is heated with 48 per cent, hydrobromic acid, 22-25 per cent, of
pinacolone and 55-60 per cent, of diene are obtained. The yield of diene
is further increased (79-86 per cent.) by passing pinacol over alumina at
420-470°.

Polymethylene Glycols
A general method of preparing polymethylene glycols is to reduce «o> -di-
carboxylic esters with sodium and ethanol or lithium aluminium hydride, or

254 ORGANIC CHEMISTRY

catalytically, e.g., (i) ethyl sebacate refluxed with sodium in ethanol gives
decamethylene glycol :

C 2 H 5 2 C-(CH 2 ) 8 -C0 2 C 2 H 5 ^^^CH 2 OH-(CH 2 ) 8 -CH 2 OH (73-75%)

(ii) Ethyl adipate heated with hydrogen under pressure in the presence
of copper chromite as catalyst gives hexamethylene glycol :

C 2 H 5 2 C-(CH 2 ) 4 -C0 2 C 2 H 5 -^> CH 2 OH-(CH 2 ) 4 -CH a OH (85-90%)

Individual polymethylene glycols are usually prepared by special methods.

Trimethylene glycol (propane-i:3-diol), CH 2 OH-CH 2 'CH 2 OH, b.p. 214°
(with decomposition), may be prepared by the hydrolysis of trimethylene
bromide, or by fermentation of glycerol by Schizomycetes (together with
M-butanol).

Tetramethylene glycol (butane-i:4-diol), CH 2 OH'(CH 2 ) 2 -CH 2 OH, b.p.
230°, and hexamethylene glycol (hexane-i : 6-diol), CH 2 OH-(CH 2 ) 4 'CH 2 OH,
m.p. 42 , are most conveniently prepared by reducing the corresponding
aco-dicarboxylic esters (succinic and adipic, respectively). Tetramethylene
glycol is prepared industrially by hydrogenating butynediol. It is used for
preparing butadiene, y-butyrolactone and tetrahydrofuran.

Pentamethylene glycol (pentane-i: 5-diol), CH 2 OH-(CH 2 ) 3 -CH 2 OH, b.p.
239 , can be obtained from pentamethylene bromide, which is obtained
from piperidine by the von Braun reaction (p. 762).

The pentamethylene bromide is converted into the corresponding glycol
by heating with potassium acetate in glacial acetic acid, and subsequently
hydrolysing the diacetate with alkali (cf. glycol) :

CH 2 Br-(CH 2 ) 3 -CH 2 Br — '—1— > CH 3 -CO-0-CH 2 -(CH 2 )3-CH 2 -0-OC-CH 3

Na ° H > CH 2 OH-(CH 2 ) 3 -CH 2 OH (v.g.)

The four a : cu-diols C aa ,C 24 ,C 26 , and C 28 have been isolated from carnauba wax
(Murray et al., 1955). These authors have also isolated seven cu-hydroxyacids
from the same source: C 18 , C 20 , C 22 , C 24 , C 26 , C 28 , C 30 .

The polymethylene glycols are readily converted into the corresponding
mono- or di-halogen derivative by halogen acid, according to the amount
of halogen acid used. These polymethylene halides are useful reagents in
organic syntheses since they contain two reactive halogen atoms, one or
both of which may be made to undergo reaction. If the synthesis requires
reaction with one halogen atom only, the most satisfactory procedure is to
" protect " the other halogen atom by ether formation and subsequently
decompose the ether with concentrated hydrobromic acid, e.g.,

C 8 H 6 ONa + BrCH 2 -CH 2 -CH 2 Br — > NaBr + C 2 H 6 -OCH 2 -CH 2 -CH 2 Br

AgN0> ->. C 2 H 5 -OCH 2 -CH 2 -CH 2 -N0 2 - HBr > CH 2 Br-CH 2 -CH 2 -N0 2

Then, e.g.,

CH,Br-CH 2 -CH.-N0 2 ^> CN-CH 2 -CH 2 -CH 2 -N0 2

TRIHYDRIC ALCOHOLS

The only important trihydric alcohol is glycerol (propane-i: 2: 3-triol),
CH 2 OH*CHOH-CH 2 OH. It occurs in almost all animal and vegetable
oils and fats as the glyceryl esters of mainly palmitic, stearic and oleic acids
(see below) .

Glycerol is obtained in large quantities as a by-product in the manufacture

POLYHYDRIC ALCOHOLS 255

of soap, and this is still a commercial source of glycerol (see below) . Another
method of preparing glycerol is the fermentation of glucose to which sodium
sulphite has been added (the yield is 20-25 per cent.). Glycerol is now also
prepared synthetically as follows:

ch 3 -ch:ch, — - — >- ch„ci-ch:ch, aq,NaaC0 ' > ch,oh-ch:ch 2

480-500° ,. , ,, ., 150° . 12 atm. „ . . , ,

allyl chlonde allyl alcohol

HOC1

-> CH 2 OH-CHCl-CH 2 OH > CH 2 OH-CHOH-CH 2 OH

glycerol /3-monochlorohydrin

An alternative route that is used is:

Hon CH 2 C1-CH0H-CH 2 C1

ch 2 ci-ch:ch 2 > +

ch 2 c1-chc1-ch 2 0h

^^^ NaOH

CH 2 — CH-CH 2 C1 > CH 2 OH-CHOH-CH 2 OH

epichlorohydrin

A new process is to add osmium tetroxide and hydrogen peroxide to
acraldehyde; this produces glyceraldehyde which is then catalytically
hydrogenated to glycerol.

CH 2 :CH-CHO -^> CH 2 OH-CHOH-CHO -^->- CH 2 OH-CHOH-CH 2 OH

Glycerol is a colourless, syrupy liquid, b.p. 290 (with some decomposition).
It is miscible with water and ethanol in all proportions, but is almost in-
soluble in ether. It is used as an antifreeze, for making explosives, and,
because of its hygroscopic properties, as a moistening agent for tobacco,
shaving soaps, etc.

Glycerol contains one secondary and two primary alcoholic groups, and
it undergoes many of the reactions to be expected of these types of alcohols.

a ff a' or y

The carbon atoms in glycerol are indicated as shown : CH 2 OH>CHOH'CH 2 OH.
(i) When glycerol is treated with sodium, one a-hydroxyl group is readily
attacked, and the other a-group less readily; the p-hydroxyl group is not
attacked at all:

Na w a

CH 2 OH-CHOH-CH 2 OH ^ CH 2 ONa-CHOH-CH 2 OH y

CH 2 ONa-CHOH-CH 2 ONa

(ii) On passing hydrogen chloride into glycerol at no until there is the
theoretical increase in weight corresponding to the esterification of one
hydroxyl group, both a- and ^-glycerol monochlorohydrin are formed, the
former predominating (66 per cent.). Continued action of hydrogen
chloride at 110°, using 25 per cent, of acid in excess required by theory for
the esterification of two hydroxyl groups, produces glycerol a : a'-dichloro-
hydrin (a-dichlorohydrin) and glycerol a : (3-dichlorohydrin (p-dichloro-
hydrin), the former predominating (55-57 per cent.) ; some other products
are also formed:

CH 2 OH-CHOH-CH 2 OH — -> CH 2 Cl-CHOH-CH 2 OH + CH 2 OH-CHCl-CH 2 OH

HC1

> CH 2 Cl-CHOH-CH 2 Cl + CH 2 C1-CHC1-CH 2 0H

When either of these dichlorohydrins or glycerol itself is treated with
phosphorus pentachloride, glycerol trichlorohydrin (1:2: 3-trichloro-

256 ORGANIC CHEMISTRY

propane) is obtained. This is a liquid, b.p. 156-158 , which smells like
chloroform.

When glycerol a : a'-dichlorohydrin is oxidised with sodium dichromate-
sulphuric acid mixture, s-dichloroacetone (1 : 3-dichloropropan-2-one) is
obtained:

CH 2 C1-CH0H-CH 2 C1 -H> CH a Cl-COCH 2 Cl (68-75 %)

It is a solid, m.p. 45 °, and is a useful starting material in many syntheses.

When glycerol a : a'-dichlorohydrin is treated with powdered sodium
hydroxide in ether solution, epichlorohydrin (3-chloro-i : 2-epoxypropane) is
obtained (see p. 250 for the mechanism) :

CH 2 C1-CH0H-CH 2 C1 + NaOH — >

CH 2 C1-CH— CH 2 + NaCl + H 2 (76-81%)

Epichlorohydrin is also obtained by distilling an alkaline solution of the
a-dichlorohydrin under reduced pressure (yield: 90 per cent.). It is a
liquid, b.p. 117°, and smells like chloroform.

Hydrogen bromide and the phosphorus bromides react with glycerol
in the same way as the corresponding chlorine compounds, but the analogous
iodine compounds behave differently, the products depending on the amount
of reagent used. When glycerol is heated with a small amount of hydrogen
iodide or phosphorus tri-iodide, allyl iodide is the main product :

CH 2 OH-CHOH-CH 2 OH -^> [CH 2 I-CHI-CH 2 I] — > l % + CH 2 :CH-CHJ

When a large amount of phosphorus tri-iodide is used, the main product
is isopropyl iodide, which is formed by the following sequence of reactions
from the allyl iodide first formed:

TTT T

CH 8 :CH-CH 2 I > [CH 3 -CHI-CH 2 I] — ^>

HI

CH 3 -CH:CH 2 ^CH 3 -CHI-CH 3 (80%)

(iii) When glycerol is treated with monocarboxylic acids, esters are
obtained which may be mono-, di- or tri-esters, according to the amount of
acid used; high temperature and an excess of acid favour the formation
of the tri-ester (see the glycerides, below).

Nitroglycerine is manufactured by adding glycerol in a thin stream to a
cold mixture of concentrated nitric and sulphuric acids:

CH 2 OH CH 2 -ON0 2

CHOH + 3HN0 3 — > CH-0-N0 2 + 3H 2

CH 2 OH CH 2 -ON0 2

Nitroglycerine is an ester, not a nitro-compound; it is glyceryl trinitrate.
The incorrect name appears to have been introduced due to the use of
" mixed acid " which is normally used for nitration (see p. 553).

Nitroglycerine is a poisonous, colourless, oily liquid, insoluble in water.
It usually burns quietly when ignited, but when heated, rapidly struck or
detonated, it explodes violently. Nobel (1867) found that nitroglycerine
could be stabilised by absorbing it in kieselguhr. This was dynamite,
which is now, however, usually manufactured by using sawdust as the
absorbent, and adding solid ammonium nitrate. Blasting gelatin or
gelignite is made by mixing nitroglycerine with gun-cotton (cellulose nitrate).

POLYHYDRIC ALCOHOLS

257

The smokeless powder, cordite, is a mixture of nitroglycerine, gun-cotton and
vaseline.

When heated with formic acid or oxalic acid at 260 , glycerol is converted
into allyl alcohol (p. 267). With dibasic acids, glycerol forms condensation
polymers known as alkyd resins, the commonest of which is glyptal, formed
by heating glycerol with phthalic anhydride.

When boric acid is added to an aqueous solution of glycerol, a complex
is produced which has a higher electrical conductivity than boric acid itself,
i.e., is a stronger acid. This complex is believed to be a borospiramc acid
in which the two rings attached to the boron atom are perpendicular to
each other (hence name spiran) :

>C— Ox /O— Cc

1 >< 1

H +

Glycol does not increase the conductivity of boric acid. It is therefore
believed that the two hydroxyl groups in glycol are in the <ra»s-position,
whereas the hydroxyl groups in glycerol are in the cis-position.

(iv) Glycerol can theoretically give rise to a large variety of oxidation
products. The actual product obtained in practice depends on the nature
of the oxidising agent used:

CH,OH

c

HOH

Ah

,OH

CH 2 OH
-> CHOH -

CH 2 OH

->C1

.HOH

C0 2 H
■CHOH

CHO C0 2 H C0 2 H

glyceraldehyde glyceric acid tartronic acid

CH 2 OH C0 2 H

— ^CO — > CO *

CH 2 OH

dihydroxyacetone

C0 2 H

mesoxalic acid

Dilute nitric acid converts glycerol into glyceric and tartronic acids; con-
centrated nitric acid oxidises it to mainly glyceric acid (80 per cent.);
bismuth nitrate produces mainly mesoxalic acid. Bromine water, sodium
hypobromite and Fenton's reagent (hydrogen peroxide and ferrous sulphate)
oxidise glycerol to a mixture of glyceraldehyde (predominantly) and di-
hydroxyacetone ; this mixture is known as glycerose. These two compounds
are interconvertible in the presence of anhydrous pyridine, this being known
as the Lobry de Bruyn-van Ekenstein rearrangement (see also p. 450) :

CHO

A

_;hoh ^

CH S

„OH

"CHOH "

C'OH

I
CH,OH

CH 2 OH

CO

CH„OH

(v) When heated with potassium hydrogen sulphate, glycerol is dehy-
drated to acraldehyde :

CH 2 OH-CHOH-CH 2 OH KHS °' > CH 2 :CH-CHO + 2 H 2

(vi) The mixed ethers of glycerol may be conveniently prepared by the
action of sodium alkoxide on a glycerol chlorohydrin, e.g., triethylin, the

K

258 ORGANIC CHEMISTRY

triethyl ether of glycerol, is prepared by heating 1:2: 3-trichloropropane
with sodium ethoxide:

CH 2 C1 CH 2 -OC 2 H s

i

HC1 + 3C 2 H B ONa — > CH-OC 2 H 5 + 3 NaCl

I I

CH 2 C1 CH 2 -OC 2 H 5

Helferich and co-workers (1923) found that triphenylmethyl chloride (trityl
chloride), (C 6 H 5 ) 3 CC1, usually formed ethers only with primary alcoholic
groups. Thus the a-mono- and the a : a'-di-triphenylmethyl (trityl) ethers
of glycerol can be prepared, the latter offering a means of preparing (3-esters
of glycerol, e.g.,

CH 2 OH CH 2 -OC(C 6 H 5 ) 3 CH 2 -OC(C 6 H 5 ) 3 CH 2 OH

choh-^^->choh RC0C1 > CH-O-COR -^(WoR

CH„OH CH.-OCfC.H,), CH.-OaCJT), CH,OH

X 2 V

A better means of obtaining the p-ester is to protect the aa'-hydroxyl groups
by cyclic ether formation with benzaldehyde (Bergmann and Carter 1930).
This acetal-like compound is formed when glycerol and benzaldehyde are
heated together, or the cool mixture treated with hydrogen chloride :

CH 9 OH CH 2 0\

HOH + C 6 H 5 -CHO — -> CHOH ;CH-C 6 H 5 + H 2

i

CH 2 OH CHjO

(vii) Glycerol can be fermented to produce a variety of compounds,
e.g., propionic acid, succinic acid, acetic acid, M-butanol, trimethylene
glycol, lactic acid, w-butyric acid, etc. Fermentation by means of a par-
ticular micro-organism usually produces more than one compound, e.g.,
propionic acid bacteria produce propionic acid, succinic acid and acetic
acid.

Structure of glycerol. All the chemical reactions of glycerol indicate
that its structure is CH 2 OH-CHOH>CH 2 OH, and this is supported by many
syntheses, e.g.,

(i) From propylene (see above).

(ii) The following absolute synthesis:

CaC 2 — -> C 2 H 2 J^-> CH 3 -CHO -^-> CH 3 -C0 2 H ^11% CH 3 -COCH 3

2 2 2 HgSO, 3 3 3 salt 3 3

-E-U* CH 3 -CHOH-CH 3 ~ > CH 3 -CH:CH, Br ' /CC '' ->■ CH 3 -CHBr-CH 2 Br

CH 2 -OCOCH 3 CH 2 OH

Br,/Fe „ TT _ TT „ „ TT T-, CH 8 -CO,K J, _, „ TT NaOH 1

> CH 2 Br-CHBr-CH 2 Br — -^—> CH-OCOCH 3 > CHOH

CH 2 -OCOCH 3 CH 2 OH

Polyhydric Alcohols

Tetrahydric alcohols. Erythritol, CH 2 OH'CHOH-CHOH>CH 2 OH, exists in
three forms: dextrorotatory erythritol, m.p. 88°; lsevorotatory erythritol, m.p.

POLYHYDRIC ALCOHOLS 259

88°; weso-erythritol, m.p. i2i-5°. The meso-iorm occurs in certain lichens and
seaweeds. All three forms may be oxidised to tartaric acid,

C0 2 H-CHOH-CHOH'C0 2 H

Pentaerythritol, C(CH 2 OH) 4 , is prepared by the condensation of formaldehyde
with acetaldehyde (see p. 166).

Pentahydric alcohols (pentitols). There are four pentitols with the structure
CH 2 OH>(CHOH) s 'CH 2 OH, which may be prepared, by reducing the correspond-
ing aldopentoses (see p. 437). Two are optically active, forming a pair of
enantiomorphs — d- and L-arabitol — and the other two are optically inactive,
existing as meso forms — adonitol and xylitol.

Aldopentose Pentitol

D- and L-ribose adonitol (ribitol)

D- and L-xylose xylitol

S^T } B-arabitoKD-lyxitol)

L-f^ose 086 } L-arabitol (L-lyxitol)

Adonitol and D-arabitol occur naturally.

D-Rhamnitol, CH^CHOH^-CHjOH, may be prepared by reducing the
corresponding deoxyhexose, D-rhamnose.

Hexahydric Alcohols (hexitols). There are ten hexitols with the structure
CH a OH*(CHOH)4'CH 2 OH, which may be prepared by reducing the correspond-
ing aldohexoses (see p. 438). Eight exist as four pairs of enantiomorphs, and the
remaining two as meso forms.

Aldohexose Hexitol

D-glucose and L-gulose D-sorbitol (D-glucitol)

L-glucose and D-gulose L-sorbitol (L-glucitol)

D-mannose D-mannitol

L-mannose L-mannitol

D-idose D-iditol

L-idose L-iditol

D-talose and D-altrose D-talitol

L-talose and L-altrose L-talitol

d- and L-galactose dulcitol

D- and T.-allose allodulcitol (allitol)

v-Sorbitol, T>-mannitol and dulcitol occur naturally.

Rhamnohexitol, CH 3 '(CHOH) 6 -CH 2 OH, is prepared by reducing the corre--
sponding aldose, rhamnohexose.

The polyhydric alcohols chemically resemble glycerol in many ways. When
heated with concentrated hydriodic acid or with a mixture of red phosphorus,
iodine and water, they are converted into the corresponding 2-iodoparamn and
paraffin; e.g., any hexitol gives a mixture of 2-iodohexane and w-hexane when
heated with hydriodic acid (see also the sugars, p. 438) :

HI

CH s OH'(CHOH) 1 -CH 2 OH >- CH 3 -CH 2 «CH 2 -CH 2 -CHT-CH 3

OILS, FATS AND WAXES

Oils and fats are compounds of glycerol and various organic acids, i.e.,
they are glyceryl esters or glycerides. Oils, which are liquids at ordinary
temperatures, contain a larger proportion of unsaturated acids than do
the fats, which are solids at ordinary temperatures. The acids present in
the glycerides are almost exclusively straight-chain acids, and almost always
contain an even number of carbon atoms. The chief saturated acids are
lauric, fnyristic, palmitic and stearic (see p. 186). The chief unsaturated
acids are oleic, linoleic and linolenic (see p. 285). Palmitic acid is the
most abundant of the saturated acids, and acids containing less than
eighteen carbon atoms are usually present only as minor constituents, but

260 ORGANIC CHEMISTRY

sometimes they are present in appreciable amounts in insect waxes and
marine fats.

Glycerides are named according to the nature of the acids present, the
suffix -ic of the common name of the acid being changed into -in. Glycerides
are said to be " simple " when all the acids are the same, and " mixed "
when the acids are different:

CH 2 -OCOC l7 H 35 CH 2 -OCOC 1B H 31

CH-OCOC 17 H 35 CH-OCOC 17 H 33

CH a -OCOC 17 H 35 CH 2 -OCO-C 17 H 33

tristearin a-palmito-a' : p-diolein

(simple glyceride) (mixed glyceride)

It is still not certain whether simple glycerides occur naturally; if they
do, they are definitely not as common as the mixed glycerides.

According to Desnuelle et al. (1959), the structure of natural glycerides is
not random ; the position of the acid residue depends on the chain-length of
the unsaturated acids and on the degree of their unsaturation. In vegetable
oils the saturated acids occur mainly at the 1- and 3-positions and un-
saturated acids at the 2-position. In animal fats the distribution is not so
rigid.

Synthesis of glycerides. Simple triglycerides are readily prepared from
glycerol and an excess of acyl chloride, or from 1:2: 3-tribromopropane
and the silver or potassium salts of the fatty acids.

Mixed triglycerides are far more difficult to prepare. A number of methods
have been developed, e.g.,

(i) The sodium salt of an acid is heated with glycerol monochloro-
hydrin, and the monoester so formed is acylated :

ch 2 c1 ch 2 -0-co-r ch 2 -ocor

Jhoh^^hgh J^^h.o.co.r'
ch 2 oh ch 2 oh ch 2 -ocor'

The preparation of i-monoglycerides has been improved by Hartman
(i960). The 2- and 3-positions in glycerol are protected by the formation of
the wopropylidene derivative (cf. (ii) below; see also p. 440), and the pro-
tecting group is removed by boric acid in hot 2-methoxyethanol.

(ii) 1 : 3-Benzylidene-glycerol is treated with an acid chloride, the
benzaldehyde residue is removed by hydrolysis, and the monoester so
formed is then further acylated :

CH,0\ CH„Gv

R*COCl I ^~"""~"»«. TTC1

CHOH ^CH-C 6 H 5 -> CH-OCOR ^CH-C e H K — ->

CH.,0/

CH 2 -OCOR'

R'-COCl 1

> CH-OCO-R

CH 2 -OCOR'

POLYHYDRIC ALCOHOLS 26l

In both methods (i) and (ii) there is, however, still difficulty in knowing
with certainty the position of the acyl group introduced first, since the
acyl group tends to migrate in the monoester, producing the following
equilibrium :

CH.-0-CO-R CH 2 OH

CHOH ^ CH-OCOR

CH 2 OH CH a OH

This isomerisation has been shown to be intramolecular. When 2-
palmitin was isomerised to the i-isomer in alcohol containing glycerol
labelled with 14 C, no 14 C was found in the ester (Doerschuk, 1952). Martin
(io'tt) has shown that perchloric acid rapidly catalyses the change of 1- or
2-monoglycerides to the equilibrium mixture containing 90-92 per cent, ot
the i-isomer. Isomerisation of aliphatic 2-monoglycendes to the corre-
sponding i-isomer may even occur slowly in the solid state (Brokaw et al.,

I9 An interesting difference between the 1- and 2-isomers is that the former
form urea inclusion compounds (p. 387) whereas the latter do not (Aylward
et al., 1956).

Analysis of oils and fats. Oils and fats are characterised by means of physical
as well as chemical tests. The usual physical constants that are determined are
melting point, solidifying point, specific gravity and refractive index.

The chemical tests give an indication of the type of fatty acids present in the

01 The acid value, which is the number of milligrams of potassium hydroxide
required to neutralise 1 gram of the oil or fat, indicates the amount of free acid

^The^'aponification value is the number of milligrams of potassium hydroxide
required to neutralise the fatty acids resulting from the complete hydrolysis of
1 srram of the oil or fat. , . .. ,, , , . .,,

The iodine value, which is the number of grams of iodine that combine with
100 grams of oil or fat, gives the degree of unsaturate of the acids in
the substance. Several methods are used for determining the iodine value. In
HubVs method, a carbon tetrachloride solution of the substance is treated with a
solution of iodine and mercuric chloride in ethanol; in Wys method, iodine
monochloride in glacial acetic acid is used. Another method is to use a solution
of glacial acetic acid containing pyridine, bromine and concentrated sulphuric

acid (Dam's solution). „*„„.,:„„,

The Reichert-Meissl value, which is the number of ml. of o-iN-potassium

hydroxide required to neutralise the distillate of 5 grams of hydrolysed fat
• indicates the amount of steam-volatile fatty acids (i.e., acids up to launc) present

in the substance. , , .• A

The acetyl value, which is the number of milligrams of potassium hydroxide

required to neutralise the acetic acid obtained when 1 gram of an acetylated

oU or fat is hydrolysed, indicates the number of free hydroxyl groups present

in the substance.

Preparation of glycerol from oils and fats. When an oil or fat is hydrolysed
by superheated steam, glycerol and the free fatty acids are obtained:

CH 2 -OCOR CH 2 OH

CH-OCOR + 3H 2 — > CHOH + 3 R'CQ 2 H
CHs-OCOR CH 2 OH

262 ORGANIC CHEMISTRY

Alternatively, the hydrolysis may be carried out by means of very dilute

sulphuric acid in the presence of a catalyst, e.g., 2-(/3-sulphonaphthyl)-

CH *(CH ) "CH-CO H stea " c ac id. This method is rapidly gaining in-

3 ^ 2 ' 15 i a dustrial importance. The free fatty acids are used

s.\ yL in the manufacture of candles.

/ |f \S0 3 H If an oil or fat is saponified, i.e., hydrolysed with

L )L J alkali, soaps are obtained. Any metallic salt of a

^/ ^' fatty acid is a soap, but the term soap is usually

applied to the water-soluble salts since only these have detergent properties.

The saturated fats give hard soaps whereas the unsaturated fats, i.e., the oils,

give soft soaps. Ordinary soap is a mixture of the sodium salts of the even

fatty acids from octoic to stearic. The sodium salts of a given oil or fat are

harder and less soluble than the corresponding potassium salts. Thus soft

soaps are usually the potassium salts, particularly when they are derived

from oils.

The preparation of soaps and glycerol is carried out by saponifying the
oil or fat with sodium hydroxide solution, and then adding sodium chloride
to " salt out " the soap, i.e., help the soap to separate out by causing it to
rise to the top of the liquid. The lower aqueous layer is run off and glycerol
is obtained from it. The soap is again heated with sodium hydroxide
solution to ensure complete saponification, and allowed to separate out to
the top. The lower layer is run off, and the soap is then boiled with water
and allowed to set.

Glycerol is obtained from the aqueous layer by neutralising the excess of
sodium hydroxide with sulphuric acid, and evaporating in vacuo until the
liquid contains about 50 per cent, glycerol. Sodium chloride (which was
added for salting out the soap) is precipitated. The liquid is filtered and
the sodium chloride is used again. The filtrate is concentrated further by
evaporation in vacuo until the glycerol content is about 85 per cent. Glycerol
is then obtained in over 99 per cent, purity by steam distillation in vacuo.
A more recent method of purification of crude glycerol is by means of ion-
exchange resins .

Hardening of oils. Glycerides of the unsaturated acids are liquid at room
temperature and so are unsuitable for edible fats. By converting the un-
saturated acids into saturated acids, oils are changed into fats. This introduction
of hydrogen is known as the hydrogenation or hardening of oils, and is carried
out by the Sabatier-Senderens reduction (p. 38). The oil is heated to 150— 200
and hydrogen is passed in, under pressure, in the presence of a finely divided
nickel catalyst. The nickel is recovered by filtration. In the hydrogenation
process only a proportion of the unsaturated acids are converted into the
saturated acids, otherwise a fat as hard as tallow would be obtained; the
hydrogenation is carried out until a fat of the desired consistency is obtained.

Synthetic fats. Truly synthetic fats would be those prepared from synthetic
glycerol and synthetic fatty acids. The cost of synthetic glycerol is so high that
synthetic fats are not, so far, prepared industrially since they cannot compete,
in price, with natural fats. During the war (1939-1945), however, synthetic
fats were prepared in Germany by three methods :

(i) Glycerol and fatty acids, both of which are obtained as by-products
from natural fats, were recombined.

(ii) Fatty acids were esterified with alcohols other than glycerol, e.g.,
ethanol. These ethyl esters hydrolyse fairly easily, giving rise to an un-
pleasant taste.

(iii) Acids, obtained by the oxidation of hydrocarbons (p. 175), were
esterified with glycerol in the presence of tin or zinc, and under reduced
pressure; under these conditions the excess glycerol and the water formed
during the reaction were removed.

POLYHYDRIC ALCOHOLS 263

Phosphatides (phospholipids). These occur in all animal and vegetable cells,
and are glycerides in which one organic acid residue, the a- or |3-, is replaced by
a group containing phosphoric acid and a base. When the base is cholamine,
CH 2 -OH*CH 2 *NH 2 , the phosphatide is known as a kephalin; when the base is

choline, CH 2 OH-CH 2 'N(CH 3 ) 3 }OH, the phosphatide is known as a lecithin, e.g.,

CH.-O-CO-R

I CH 2 .OCOR O

CH-OCOR O I || +

I || CH-O P-O— CH 2 -CH 2 -N(CH 3 ) 3 }OH

CH 2 -0 P-O— CH 2 -CH 2 -NH 2 | |

I CH 2 -0-CO-R OH
OH
a-kephalin j8-lecithin

It is quite likely that both kephalins and lecithins have a betaine structure
(see p. 325).

Only a few monocarboxylic acids have been isolated from phosphatides:
stearic, oleic, linoleic and arachidonic from kephalins, and palmitic, stearic,
oleic, linoleic, linolenic and arachidonic from lecithins. Enzymic studies of
phospholipids have shown that saturated acids predominate at the 1 -position and
unsaturated acids at the 2-position (Hanahan et al., i960).

Drying oils. These are oils which, on exposure to air, change into hard
solids, e.g., linseed oil. All drying oils contain a large proportion of the un-
saturated acids linoleic and linolenic, and it is this " drying " property which
makes these oils valuable in the paint industry. The mechanism of drying is
not known. It appears to be a complicated process involving oxidation,
polymerisation and colloidal gel formation, and it has been found to be catalysed
by various metallic oxides, particularly lead monoxide.

Waxes. These are esters of the higher homologues of both the fatty acids and
monohydric alcohols, e.g.,

beeswax myricyl palmitate

spermaceti cetyl palmitate

carnauba wax myricyl cerotate C 25 H 51 'CO 2 C 30 H 61

Some waxes are also esters of cholesterol, e.g., cholesteryl esters occur in wool-
wax.

QUESTIONS

1. Name the compounds and state the conditions under which they are formed
when glycol is treated with:— (a) Na, (6) HBr, (c) PC1 6 , (d) PI 3 , (e) AcCl, (/) CH 3 -CHO,
(g) CH 3 -CO-CH 3 , (h) H 3 P0 4 , (») HNO3, (j) KMn0 4 , (k) ZnCl 2 , (I) C 6 H 5 -NCO.

2. Name the compounds and state the conditions under which they are formed when
glycerol is treated with:— (a) Na, (6) HC1, (c) PCl,s, (d) PI 3 , (e) Ac a O, (/) HNO„ (g)
H-CO a H, {h) (C0 2 H) 2 , (i) H 3 B0 3> (j) Br 2 , (ft) KHS0 4 , (/) C„H 6 -CHO.

3. How are glycol and glycerol prepared (i) in the laboratory; (ii) industrially?

4. By means of equations show how you would convert glycerol into: — (a) M-PrOH,
(6) dihydroxyacetone, (c) diallyl ether, (d) epichlorohydrin, (e) /S-ketoglutaric acid.

5. Show by means of equations how you would prepare: — (0) CH 2 OH'CH 2 -OMe,
(6) CH a -(OEt)'CH a -OEt, (c) CH a -(OAc)-CH 2 -0-CO-C 2 H 5 , {d) C a H 4 0,

/CH a — CH 2 \
(e) MeO-CH a -CH 2 -OCH a -CH 2 OH, (/) OC )o, (g) Et 2 C(OH)-C(OH)Et 2 ,

\CH a — CH/
(h) CH 2 OH-(CH a )„-CH a OH, where n = 1, 2, 3 and 4, respectively,
(t) CH a OH-(CH 2 ) 6 >CH 2 -CO a H, (_;) CH 2 OMe-CHOMe-CH 2 OMe,
(ft) CH 2 OH-CHOAc-CH 2 OH.

6. Define and give examples of: — (a) oils, fats and waxes, (6) acid value, (c) saponi-
fication value, (d) iodine value, (e) Reichert-Meissl value, (/) acetyl value, {g) soap,
(A) hardening of oils, (i) kephalins, (j) lecithins, (k) drying oils.

7. Show how you would distinguish between :

(a) a mineral oil and a vegetable oil;
\b) triolein and tristearin.

264 ORGANIC CHEMISTRY

READING REFERENCES

Davidson, Glycol Ethers, Ind. Eng. Chem., 1926, 18, 669.

Snell, Synthetic Detergents and Surface Activity, /. Chem. Educ, 1947, 24, 505.

The Structure of Cyclopropane and Ethylene Oxide.

Walsh, Nature, 1947, 159, 165, 712.

Robinson, ibid., 1947, *59> 4°°; J 947> !6o, 162.

Linnett, ibid., 1947, 160, 162.
Levey, Synthetic Glycerol, Ind. Eng. Chem. (News Ed.), 1938, 16, 326.
Eoff, Lander and Beyer, Glycerol from Sugar Fermentation, Ind. Eng. Chem., 1919, 11,

842.
Snell, Soap and Glycerol, /. Chem. Educ., 1942, 19, 172.
Wurster, Hydrogenation of Fats, Ind. Eng. Chem., 1940, 32, 1193.
Drying Oils.

Bradley, Ind. Eng. Chem., 1937, 3 9. 44° : I 94 2 , 34. 237.

Frilette, ibid., 1946, 38, 493.
Hilditch, Chemical Constitution of Natural Fats, Wiley (1947, 2n< i e< i)-
Elderfield (Editor), Heterocyclic Compounds, Wiley, Vol. I. (1949). Ch. 1. Ethylene

and Trimethylene Oxides.
Gilman, Advanced Organic Chemistry, Wiley (1953). Vol. Ill, Ch. 3. Lipids.
Hartman, Advances in the Synthesis of Glycerides of Fatty Acids, Chem. Reviews, 1958,

58, 845.
Parker and Isaacs, Mechanisms of Epoxide Reactions, Chem. Reviews, 1959, 59, 737.

CHAPTER XII

UNSATURATED ALCOHOLS, ETHERS, CARBONYL
COMPOUNDS AND ACIDS

UNSATURATED ALCOHOLS

The simplest unsaturated alcohol is vinyl alcohol, CH 2 ICHOH. This is,
however, unknown; all attempts to prepare it result in the formation of
acetaldehyde (together with a small amount of ethylene oxide), e.g., vinyl
bromide on treatment with silver oxide in boiling water gives acetaldehyde :

CH 2 :CHBr + " AgOH " — y AgBr + [CH 2 :CHOH] — > CH 3 -CHO

It thus appears that the group — CHXHOH is unstable. It should, however,
be noted that this group is unstable only when the CHOH group is at the
end of a chain; when it occurs in the chain, i.e., as — CHIC(OH) — , it is stable
(cf. keto-enol tautomerism, p. 220).

Although vinyl alcohol itself is unknown, many of its derivatives have
been prepared, and are quite stable. These derivatives may be prepared
by the interaction between acetylene and the other reactant in the presence
of a suitable catalyst (see also p. 93) ; e.g., vinyl chloride:

C 2 H 2 + HC1 ^> CH 2 :CHC1

Vinyl chloride may also be prepared directly by the action of concentrated
hydrochloric acid on calcium carbide in the presence of mercuric ions as
catalyst:

CaC 2 y CH,:CHC1

8 Hg'+ 2

Vinyl chloride is manufactured by the thermal decomposition of ethylene
chloride at 600-650°:

CH 2 C1-CH 2 C1 — y CH 2 =CHC1 + HC1

Vinyl chloride and bromide are conveniently prepared in the laboratory by
heating ethylene chloride and bromide, respectively, with ethanolic potass-
ium hydroxide :

CH 2 Br-CH 2 Br + KOH J!^ ^ CH 2 :CHBr + KBr + H 2

Many of the vinyl compounds are used to make plastics. Vinyon, which
is a thermoplastic, is a copolymer of vinyl chloride and vinyl acetate. Its
main use at the moment is the manufacture of industrial filter cloth which
is very resistant to the action of acids and alkalis, and which has great
strength both wet and dry.

Vinyl cyanide (acrylonitrile) is very useful for introducing the cyanoethyl
group, •CH 2 -CH 2 *CN, by reaction with compounds containing an active
methylene group. The best catalyst for cyanoethylation is benzyltrimethyl-

+
ammonium hydroxide, C 6 H 5 -CH 2 'N(CH 3 ) 3 }OH-, but in many cases an
aqueous or ethanolic solution of alkalis is effective (Bruson, 1942, 1943),
e.g.,

CH,-CH,-CN

J,

CH 3 -CO-CH 2 -CH 3 + 2CH 2 :CH-CN — > CH 3 -CO-C-CH 3 (90%)

CH 2 -CH 2 -CN
265

266

ORGANIC CHEMISTRY

This type of reaction offers a means of preparing a variety of compounds.
Vinyl cyanide normally causes di- or tri-cyanoethylation of compounds
which have an active methylene or methyl group respectively, but Campbell
et. al. (1956) have described conditions for monocyanoethylation. In addi-
tion to active methylene and methyl groups, compounds such as primary and
secondary amines, alcohols, phenols, etc., can also undergo cyanoethylation.
The halogen atom in vinyl halides is not reactive; vinyl halides do not
undergo the usual double decomposition reactions of the alkyl halides.
Vinyl chlorides and bromides readily form Grignard reagents in tetrahydro-
furan as solvent (Normant, 1954, 1957), and they react with lithium to form
lithium compounds which undergo the usual reactions (Braude, 1950-1952 ;
see p. 362), e.g.,

R 2 C=CHBr ^ R 2 C=CHLi

R'-CHO

> R 2 C=CH-CHOH-R'

The reason for the unreactivity of the halogen atom is not clear. Some
believe it to be due to resonance through which the halogen atom acquires
some double-bond character, and is thereby more strongly bound to the
carbon atom due to the shortening of the C — CI bond (p. 19) :

ch 2 =ch— ci : <-> ch 2 — ch=ci :

At the same time, resonance also stabilises the compound, and so the chlorine
atom will not be so reactive.

An alternative explanation is that due to Walsh (1947). As we have seen
(p. 96), an sp z hydridised carbon atom attracts electrons more than does an
sp 3 hydridised carbon atom. Thus the CI atom in the C — CI group is more
tightly bound in C=C — CI than in C — C — CI.

The non-reactivity of the chlorine atom in vinyl chloride may be explained from
the M.O. point of view as follows. If the chlorine atom has sp 3 hybridisation,
the C — CI bond will be a a-bond and the two lone pairs of electrons would occupy
the other two sp 2 orbitals. This would leave a p orbital containing a lone pair,
and this orbital could now conjugate with the 7T-bond of the ethylenic link (Fig. 1) .
Thus two M.O.s will be required to
accommodate these four ^-electrons
(cf. the carboxyl group, p. 184).
Furthermore, since chlorine is far
more electron-attracting than car-
bon, the electrons will tend to be
found in the vicinity of the chlorine Fig. 12. i.

atom. Nevertheless, the chlorine

atom has now lost " full control " of the lone pair, and so acquires a small positive
charge (or alternatively, it is less negative than it would have been had there been
no conjugation). Since two carbon atoms have acquired a share in the lone pair,
each carbon atom acquires a small negative charge. Hence, owing to de-
localisation of bonds (through conjugation), the vinyl chloride molecule has
an increased stability. Before the chlorine atom can be displaced by some
other group, the lone pair must be localised again on the chlorine atom. This
requires energy, and so the chlorine is more " firmly bound " than had no con-
jugation occurred.

Allyl alcohol (prop-2-en-z-ol) , CH 2 !CH - CH 2 OH, may be prepared as
follows :

(i) By boiling allyl chloride with dilute sodium hydroxide solution under
pressure (this is a commercial method).

CH 2 =CH-CH 2 C1 + NaOH — i-> CH a =CH-CH 2 OH + NaCl

UNSATURATED ALCOHOLS, ETHERS, CARBONYL COMPOUNDS AND ACIDS 267

Another commercial method is the isomerisation of propylene oxide in the
presence of lithium phosphate as catalyst (cf. p. 169).

(ii) By the controlled catalytic reduction of propargyl alcohol:

CH:OCH 2 OH -^->- CH 2 :CH-CH 2 OH

(iii) By heating glycerol with formic acid or oxalic acid at 260° :

CHgOH CH 2 -OCOCOOH CH 2 -OOOH

I COOH I _ COj I

CHOH+ I — ^CHOH >CHOH

I COOH I I

CH 2 OH CH 2 OH CH 2 OH

glyceryl glyceryl

monoxalate monoformate

- H,0 - COj-H,0

T

CH,-0-CO CH,

1 I -,co, II
CH-O-CO ^->CH

I I

CH 2 OH CH 2 OH

dioxalin

In practice it is better to use formic acid since this gives a higher yield
(45-47 per cent.).

Allyl alcohol is a colourless, pungent-smelling liquid, b.p. 97 , miscible
with water in all proportions. The presence of the allyl group produces a
pungent smell in compounds containing it, e.g., mustard oils contain allyl
wothiocyanate; onions and garlic, allyl sulphide.

Allyl alcohol has the properties of an unsaturated compound and a
primary alcohol. It is oxidised to glycerol by dilute alkaline permanganate:

CH 2 :CH-CH 2 OH + H 2 + [O] — > CH 2 OH-CHOH-CH 2 OH

Allyl alcohol (and other a, {3-unsaturated primary and secondary alcohols)
may be conveniently oxidised to the corresponding oxo compound with
manganese dioxide at room temperature (Morton et al., 1948):

CH 2 =CH-CH 2 OH Mn ° 2 > CH 2 =CH-CHO

In general, oxidising agents attack the double bond in allyl alcohol as well as
the alcoholic group. Oxidation, however, may be carried out by " protecting "
the double bond by bromination and then oxidising and debrominating with
zinc dust in methanolic solution, e.g.,

CH 2 =CH'CH 2 OH ^i> CH 2 Br-CHBr-CH 2 OH HN ° a >

Zn

CH,Br-CHBr-CO,H > CH 2 =CH-C0 2 H

2 a MeOH *

Allyl alcohol adds on chlorine or bromine to form the corresponding 2 : 3-
dihalogeno-propan-i-ol. Allyl alcohol also adds on halogen acids and
hypohalous acids, but the addition takes place contrary to Markownikoff's
rule, e.g., glycerol (i-monochlorohydrin is formed with hypochlorous acid:

CH,:CH-CH 2 OH + HOC1 — > CH 2 OH-CHCl-CH 2 OH

268 ORGANIC CHEMISTRY

This may be due to the presence of the oxygen atom which exerts a strong
inductive effect causing the electromeric effect to take place towards the
oxygen atom, i.e.,

CH 2 =CH-^-CH 2 -^OH

The allyl halides form an interesting group of compounds because of the
high reactivity of the halogen atom.

Allyl chloride (3-chloropropene) , b.p. 45 , is prepared industrially by the
chlorination of propylene at high temperature. It may be conveniently
prepared in the laboratory by warming allyl alcohol with hydrochloric acid:

CH 2 :CH-CH 2 OH + HC1 — > CH 2 :CH-CH 2 C1 + H 2

It is now important as a chemical raw material for the production of drugs,
plastics, etc.

Allyl bromide (3-bromopropene) , b.p. 70 , is best prepared in the laboratory
by distilling aqueous allyl alcohol with 48 per cent, hydrobromic acid in the
presence of sulphuric acid:

CH 2 :CH-CH 2 OH + HBr — -> CH 2 XH-CH 2 Br + H 2 (92-96%)

Allyl iodide (3-iodopropene) , b.p. 103-1°, may be prepared by heating
glycerol with a small amount of hydriodic acid (p. 256).

The halogen atom in the allyl halides is very reactive, and it has been
found experimentally that the position of the double bond with respect to
the halogen atom determines the reactivity of the halogen atom. In
compounds of the type CIC — X, the halogen atom X is unreactive (see
vinyl halides, above). In COC — X, X is more reactive than in the alkyl
halides. When the halogen atom in an unsaturated compound is further
removed from the double bond than in the allyl position, it behaves
similarly to the halogen atom in alkyl halides; e.g., 4-chlorobut-i-ene,
CH 2 ICH-CH 2 -CH 2 C1, undergoes the usual reactions of the alkyl halides with
reagents that do not affect the double bond.

The high reactivity of the allyl halides may be explained by the fact that
when the halogen atom ionises, the allyl carbonium ion produced is a reson-
ance hybrid, and consequently it is stabilised (see also p. 271) :

CH 2 =CH— CH 2 ■<-> CH 2 — CH=CH,

2

The more stable the ion produced, the more easily it is formed. Alkyl
carbonium ions cannot be stabilised by resonance, and hence the alkyl
halides are not so reactive as the allyl halides.

The allyl halides add on the halogen acids to form a mixture of the 1 : 2-
and 1 : 3-dihalides. The addition of hydrogen bromide to allyl bromide
has been studied in great detail (Kharasch, 1933) :

CH 2 :CH-CH 2 Br + HBr — > CH 2 Br-CH 2 -CH 2 Br + CH 3 -CHBr-CH 2 Br

1 : 3-dibromide 1 : 2-dibromide

In the presence of peroxides, the 1 : 3-dibromide is obtained in 90 per cent,
yield. In the absence of peroxides, the 1 : 2-dibromide is obtained in
90 per cent, yield. When allyl bromide, which has not been carefully
purified, is treated with hydrogen bromide, without special precautions
being taken to exclude all traces of air, a mixture of the 1 : 2- and 1 : 3-
dibromides is obtained.

The explanation for these results is as follows. The 1 : 3-dibromide is
formed by a free-radical mechanism, the peroxides producing atomic

UNSATURATED ALCOHOLS, ETHERS, CARBONYL COMPOUNDS AND ACIDS 269

bromine from the hydrogen bromide, thereby setting up a free-radical chain
reaction (cf. p. 68) :

(R-CO a ) 2 — > 2R-COO ^ 2R- + 2CO a

R- + HBr > RH + Br-

Br- + CH 2 =CH-CH 2 Br >■ CH 2 Br-CH-CH a Br i^>

CH 2 Br-CH 2 -CH 2 Br + Br-; etc.

Mayo (19.40) has suggested that the bromine atom adds to the terminal
carbon to give a secondary free radical which is more stable than a primary
free radical. Evidence for this free-radical mechanism is that it takes place
in the presence of peroxides; the formation of the i : 3-dibromo compound
is very largely suppressed in the presence of antioxidants (which destroy
peroxides) ; and its formation is catalysed by light (which is known to pro-
duce free radicals).

The mechanism for the formation of the 1 : 2-dibromide is polar (which
gives "normal" addition; see p. 68). At first sight, however, the polar
mechanism would be expected to produce the 1 : 3-dibromide because of the
inductive effect of the bromine atom, i.e.

rv +

CH 2 =CH— CH 2 ->~ Br + HBr — -> CH 2 -CH 2 -CH 2 Br + Br- — >■

CH 2 Br-CH 2 -CH 2 Br

Baker and Nathan (1953), however, have explained the 1 : 2-addition by
what is often referred to as the Baker-Nathan effect. As we have seen, the
general inductive effect of alkyl groups is Me 3 C > Me 2 CH > MeCH 2 > Me.
This inductive order has been used satisfactorily to explain various physical
data, etc. In some reactions, however, the inductive order is reversed, e.g.,
Baker and Nathan (1935) examined kinetically the reaction between p-
substituted benzyl bromides and pyridine. The reaction was carried out
in acetone, and was shown to be entirely S N 2. Thus the reaction may be
formulated:

^Br~

The greater the +1 effect of the R group, the faster should be the reaction,
but it was found that the rate order for the R groups was

Me > Et > isoVv ~ *-Bu

Thus the order of electron release is almost exactly the reverse of that given
above. Therefore alkyl groups must possess some mechanism of electron-
release in which the order is

Me > Et > isoYv > i-Bu

Further work showed that in all cases where the general inductive order of
alkyl groups was reversed, the alkyl group was attached to an aromatic
nucleus. This led Baker and Nathan (1935, 1939) to suggest that when the
H — C bond is attached to an unsaturated carbon atom, the o-electrons of
the H — C bond become less localised by entering into partial conjugation
with the attached unsaturated system, i.e.,

r\

H— C— C=C

270 ORGANIC CHEMISTRY

Thus there is conjugation between electrons of single and those of multiple
bonds. This type of conjugation is known as hyperconjugation, and is a
permanent effect (this name was given by Mulliken, 1941).

There are various ways of looking at the hyperconjugative effect. A
widely used one is to regard the H — C bond as possessing partial ionic
character due to resonance, e.g., propylene may be written as a resonance
hybrid as follows:

H H + H +

H— C— CH=CH,^-> H— C— CH=CH,^->H— C=CH— CH,«->

I I J

H H H

From this point of view, hyperconjugation may be regarded as " no bond
resonance ". The hydrogen atoms are not free; the effect is to increase the
ionic character of the C — H bond, the electrons of which become partially
delocalised through conjugation.

Various physical data have been explained by hyperconjugation, e.g.,
bond lengths :

CH3 CH 3 CH 3 — C^CH

I-54A I-46A

Because of hyperconjugation, the CH 3 — C bond has some partial double
bond character and consequently is shortened.

Also, since hyperconjugation stabilises a molecule (through resonance),
propylene, for example, should be more stable than " expected ". This
has been found to be so; the observed heat of hydrogenation of propylene
is less than the calculated value.

Using this scheme of hyperconjugation, we are now in a position to explain
the formation of the 1 : 2-dibromide from allyl bromide and hydrogen bromide
by the polar mechanism.

H H+

,=CH— CI

CH 2 =CH— CHBr *e^ CH 2 =CH— CHBr<->

H+ HBr

CH 2 — CH=CHBr y CH 3 -CHBr-CH 2 Br

The inductive effect of the bromine atom would tend to give the 1 : 3-di-
bromide (see above). If, however, the hyperconjugative effect is operating,
and is stronger than the —I effect of the bromine atom, then the 1 : 2-dibromide
will be formed.

An interesting point here is that with allyl alcohol, the —I effect of the
hydroxyl group is stronger than hyperconjugative effect (p. 519).

<& ZH)

c

<3>

(a)

From the M.O. point of view, hyperconjugation maybe explained as follows:
7t-Orbitals can overlap with other 7t-orbitals to produce conjugation (p. 88).
7t-Orbitals, however, can also overlap to a certain extent with adjacent o-orbitals
to form extended orbitals. When this occurs we have hyperconjugation, and this
phenomenon is exhibited mainly by the hydrogen bonds in a methyl group when

UNSATURATED ALCOHOLS, ETHERS, CARBONYL COMPOUNDS AND ACIDS 27I

this group is attached to an unsaturated carbon atom. The question is: how
does this overlapping occur? Coulson (1942) treats the methyl group (Fig. ■za)
as a " compound " atom to form a " group " orbital (Fig. b). An alternative
" group " orbital may also be formed as in (c). In this arrangement, the methyl
group behaves as a group with a re-orbital, and this can conjugate with an
adjacent re-bond in, e.g., propylene (Fig. d). Hence, owing to partial delocalisa-
tion of bonds (a and re) in this way, the propylene molecule is more stable than
" expected ".

The unusual reactivity of the halogen atom in allyl halides may be explained
as follows. If the chlorine were to ionise, the carbon atom to which it was
attached now has a positive charge and has only six electrons. Hence the re-bond
covering the other two carbon atoms can extend to embrace this third carbon
atom. The net result is an M.O. covering three carbon atoms, and so the de-
localisation energy will be increased, i.e., the new arrangement is stabilised,
and behaves as if the chlorine atom is ionic (Fig. 3).

CH-CH^r-CI -»-CH 2 — CH-CH 2 | + CI"^-CH 2 -CH-CH 2 r CI"

Fig. 12.3.

There is now, however, a certain amount of doubt that dipole moment
data are evidence for hyperconjugation. Since the electronegativity of a
carbon atom depends on its state of hybridisation, there will be a bond
moment when two differently hybridised carbon atoms are linked together.
Petro (1958) has calculated bond moments for several types of carbon-
carbon single bonds :

Bond Bond moment

C(spy—C(spz)- o-68D

C(sp*)+— C(sp)- i-i5D

C(sp*) + — C(sp)~ i- 4 8D

Ferreira (i960) has calculated the dipole moments of various molecules,
e.g., toluene, propylene, etc., on these principles, and has shown that the
calculated values agree well with the experimental values. Thus it appears
that dipole moment evidence for hyperconjugation is not valid.

It has also been found experimentally that all C — C single bonds are not the
same length, e.g.,

sp 3 — sp 3 i"54A

sp 3 — sp 2 i - 5oA

sp 3 — sp I-46A

This shortening has been attributed to hyperconjugation, but it can also be
explained in terms of different electronegativities of the two carbon atoms.

It is quite possible, however, since bond moment (and bond lengths) is a
ground-state property, that hyperconjugation exists in excited states and in
transition states of molecules, and therefore is important for the interpreta-
tion of reactivity of molecules.

Mulliken (1959) has concluded that the explanation offered above is partly
right, but does not believe the re-delocalisation shortenings are negligible.
He has made some calculations and has attributed 40 per cent, of observed
shortenings to it-electron resonance and 60 per cent, to hybridisation
differences.

Crotyl alcohol (crotonyl alcohol, but-2-en-i-ol) , CHg-CHtCH-CHgOH, may
be prepared by the reduction of crotonaldehyde with aluminium isopro-
poxide (p. 150):

ch 3 -ch:ch-cho " CH3) * CH03 ' A1 > ch 3 -ch:ch-ch 2 oh (85-90%)

272 ORGANIC CHEMISTRY

Crotyl alcohol is a colourless liquid, b.p. 118 , fairly soluble in water.
When treated with hydrogen bromide, crotyl alcohol produces a mixture
of crotyl and methylvinylcarbinyl bromide :

CH 3 -CH:CH-CH 2 OH -^> CH s -CH:CH-CH a Br + CH 3 -CHBr-CH:CH 2

The formation of the rearranged product is an example of anionotropy,
and the most widely studied example is the three-carbon system, i.e., the
allylic system. Nucleophilic substitution reactions of allylic halides may
occur by the S N 2 mechanism, and when this is operating, substitution
proceeds normally (see below). When, however, nucleophilic substitution
reactions are carried out under conditions which favour the unimolecular
mechanism, then the product may be a mixture of two isomers, the re-
arranged product being an example of the allylic rearrangement. The
mechanism by which this rearrangement occurs is known as S N i', and the
steps involved are believed to be as follows, via a resonance hybrid cation :

MeCH=CH-CH 2

Br-

V

; MeCH=CH-CH a 01V
<--> MeCH-CH=CH 2

I Br-

-H„ Q y
^ + H 2 C~

MeCH=CH-CH 2 Br

(SnI)

MeCHBr-CH=

(S N i')

=CH,

There is much evidence to support this S N i' mechanism, e.g., Hughes et al.
(1948) examined the conversion of a-methylallyl and crotyl chloride into
a-methylallyl and crotyl ethyl ether by reaction with ethanolic sodium
ethoxide. The concentration of the sodium ethoxide was made so small as
to give first-order kinetics, or so large as to give second-order kinetics. The
results were :

MeCHCl-CH=CH 2

MeCH=CH-CH 2 Cl

>

2nd
order

^

2nd
order

tleCHOEt-CH=CH 2

' MeCHOEfCH=CH 2 1

MeCH=CH>CH 2 OEt

(10

0%)

(13%)

+

MeCH=CH-CH 2 OEt

(87%)

(100

%)

With second-order substitution, each chloride gives its own unrearranged
ether (thus the S N 2 mechanism). With first-order substitution, however,
each chloride gives the same mixture of isomeric ethers. This latter result
may be explained by assuming that two mechanisms are operating S N i
(without rearrangement) and S N i' (with rearrangement), both proceeding
through a common intermediate cation.

An interesting point about this reaction is the predominance of the crotyl ether
whichever chloride is used as starting material. A possible answer is hyper-
conjugation, which can stabilise the crotyl cation but not the a-methylallyl
cation :

Me— CH— CH=CH,

Me— CH=CH— CH 2

In addition to the S^r' mechanism, there is also the S N 2', i.e., a bimolecular
nucleophilic substitution with allylic rearrangement. This has been difficult

6

UNSATURATED ALCOHOLS, ETHERS, CARBONYL COMPOUNDS AND ACIDS 273

to demonstrate, but a good example of it has been given by de la Mare et al.
(1953). These authors have suppressed the S N 2 reaction by steric factors
and reduced the S^i reaction by means of a reagent of high nucleophilic
activity and a solvent of low ionising power. They found that the reaction
between a : a-dimethylallyl chloride and sodium thiophenoxide in ethanol
gives 62 per cent, of rearranged product by a second-order reaction (i.e.,
S N 2') :

tCI

vQ
=CH 2 SPh — > Me 2 C=CH-CH 2 -SPh + NaCl

One other problem that will be discussed here is the case of ion pairs in the
allylic rearrangement (see also p. 109). Young et al. (1951) have shown that
a : a-dimethylallyl chloride in acetic acid undergoes acetolysis accompanied
by rearrangement to y : y-dimethylallyl chloride:

Me 2 CCl-CH=CH 2 Me<:o ' H > ffg + Me 2 C=CH-CH 2 Cl

The rate of the isomerisation was shown to be proportional to the concen-
tration of the starting chloride only. Thus a possible mechanism is the
S N i'. If this were so, then the rate of isomerisation would be affected by
addition of chloride ions (since the first step is ionisation). Experiment
showed that the rate was unaffected by added chloride ions, and it was also
shown that when an excess of radioactive chloride ion was added, the rate of
isomerisation was far greater than the incorporation of CI*-. It thus
appears that the chloride ion is set free during the isomerisation and is
mainly the " first ion " to recombine, even though the solution contains a
large excess of CI*- ions. Young, et al. explained this result by assuming
that the starting chloride undergoes ionisation to form an ion-pair, and this is
followed by internal return:

Me 2 CCl-CH=CH 2 =?=±: [Me 2 C=CH=-CH 2 ]+Cl- =^ Me 2 C=CH-CH 2 Cl

ion pair

Nevertheless, some ion-pairs escape from the solvent cage, and these " free"
carbonium ions are capable of undergoing acetolysis and combination with
CI*".

A number of unsaturated alcohols occur in essential oils. The essential oils are
volatile oils with pleasant odours, and are obtained from various plants to which
they impart their characteristic odours. They are widely used in the perfume
industry. Citronellol occurs in citronella oil, rose oil, etc. It appears to be a
mixture of the two structural isomers I and II :

(CH 3 ) 2 C:CH-CH 2 -CH 2 -CH(CH 3 )-CH 2 -CH 2 OH (I)

(II) CH 2 :C(CH 3 )-CH 2 -CH 2 -CH 2 -CH(CH 3 )-CH 2 -GH 2 OH

Geraniol and nerol are cis-trans isomers of the following structure (and possibly of
the structure corresponding to II above) :

(CH 3 ) 2 C:CH-CH 2 -CH 2 -C(CH 3 ):CH-CH 2 OH

Geraniol is a constituent of geranium oil, rose oil, etc. Nerol occurs in neroli oil,
etc. Linalool, which is an isomer of geraniol and nerol, occurs in linaloe oil,
lavender oil, coriander oil, etc. :

(ch 3 ) 2 c:ch-ch 2 «ch 2 -c(oh)-(ch 3 )-ch:ch 2

These alcohols are the oxygen derivatives of the group of compounds known as
the open-chain terpenes. It appears that all of these alcohols are mixtures of

274 ORGANIC CHEMISTRY

structural isomers (see citronellol, above). It is possible, however, that the two
structures form a three-carbon tautomeric system :

CH 3 CH 3

CH 3 — C=CH— - ;^=s CH,=C— CH,-~-

Farnesol, a sesquiterpene alcohol, occurs in rose oil, the oil from ambrette seeds,
etc., and, like the terpene alcohols, is a mixture of two forms, one of which is:

(CH 3 ) 2 C:CH-CH 2 -CH 2 -C(CH 3 ):CH-CH 2 -CH 2 -C(CH 3 ):CH-CH 2 OH

Phytol is an unsaturated alcohol which occurs in chlorophyll (the green colouring
matter of leaves and other parts of a plant). Its structure is:

(CH 3 ) 2 CH- (CH 2 ) 3 -CH- (CH 2 ) 3 -CH- (CH 2 ) 3 -C:CH-CH 2 OH
CHg CH 3 ^H 3

The simplest acetylenic alcohol is propargyl alcohol (prop-2-yn-i-ol),
CH:OCH 2 OH. This may be prepared from 1:2: 3-tribromopropane by
the following series of reactions, which clearly show the difference in re-
activity of the bromine atoms, and at the same time illustrate the use of
aqueous and ethanolic solutions of potassium hydroxide :

CH 2 Br-CHBr-CH 2 Br ^^-V CH a :CBr-CH 2 Br -^=>

2 * KOH i i KOH

CH 2 :CBr-CH 2 OH -^^-> CH:C-CH 2 OH

2 ■* KOH *

Propargyl alcohol (together with butyne-diol) is prepared by the inter-
action of acetylene and formaldehyde in the presence of silver or cuprous
acetylide as catalyst :

C 2 H 3 + H-CHO ^> CH:OCH 2 OH

C 2 H 2 + 2H-CHO ^%- CH 2 OH-C:C-CH 2 OH

Propargyl alcohol is a colourless liquid, b.p. 114°. It behaves in many
ways like acetylene, e.g., it adds on two or four bromine atoms, and forms
the silver or cuprous compounds when treated with an ammoniacal solution
of silver nitrate or cuprous chloride. It may be catalytically reduced to
allyl and w-propyl alcohols; it is now being used to prepare these two com-
pounds commercially.

UNSATURATED ETHERS

The simplest unsaturated ether is methyl vinyl ether, CH 3 'OCH:CH 2 ,
which is prepared industrially by passing acetylene into methanol at 160-
200° in the presence of 1-2 per cent, potassium methoxide, and under pressure
sufficient to prevent boiling:

c 2 h 2 + ch s oh — > ch 3 -och:ch 2

The acetylene is diluted with nitrogen to prevent explosions.

Methyl vinyl ether is a very reactive gas, b.p. 5-6 . It is hydrolysed
rapidly by dilute acid at room temperature to give methanol and acetalde-
hyde (cf. the ethers).

ch 3 -och:ch 2 + h 2 o — ^ ch 3 oh + ch 3 -cho

This is a potential industrial method for preparing acetaldehyde. Methyl
vinyl ether is stable in alkaline solution. It undergoes many addition
reactions at the double bond, e.g.,

UNSATURATED ALCOHOLS, ETHERS, CARBONYL COMPOUNDS AND ACIDS 275

CH 2 :CH-OCH 3 + HC1 > CH 3 -CHCl-OCH 3

CH „:CH-0-CH 3 + CH.OH — — > CH 3 -CH(OCH s ),

2 * 3 trace of HCl v '

It readily polymerises, and is used for making polyvinyl ether plastics.

Divinyl ether (vinyl ether), CH 2 .*CH>OCHrCH 2 , may be prepared by heat-
ing ethylene chlorohydrin with sulphuric acid, and then passing the product,
2 : 2'-dichlorodiethyl ether, over potassium hydroxide at 200-240°:

2Cich„-ch 8 oh ' • > cich 2 -ch 2 -o-ch 2 -ch 2 ci >■ ch 2 :ch-och:ch 2

(75%) ( - 2HC " (35%)

Divinyl ether is a colourless liquid, b.p. 39°. It is used as an anaesthetic,
being better for this purpose than diethyl ether.

UNSATURATED ALDEHYDES

The simplest unsaturated aldehyde is acraldehyde (acrolein, prop-2-en-
x-al), CH 2 ICH'CHO. This is conveniently prepared by heating glycerol
with potassium hydrogen sulphate, which is the most satisfactory dehyrat-
ing agent for this reaction :

CH 2 OH-CHOH-CH 2 OH — — -> CH 2 :CH-CHO (33-48%)

Acraldehyde may also be readily prepared by the oxidation of allyl alcohol
with manganese dioxide (p. 267).

Acraldehyde is prepared industrially in many ways, e.g.,

(i) by passing a mixture of acetaldehyde and formaldehyde vapour over
sodium silicate as catalyst :

CH 3 -CHO + CH 2 — > CH 2 =CH-CHO + H 2

(ii) By the direct oxidation of propylene over copper oxide :

CH 2 =CH-CH 3 + 2 — > CH 2 =CH-CHO + H 2 0.

(iii) By the pyrolysis of diallyl ether :

(CH 2 =CH-CH 2 — ) 2 — > CH 2 =CH-CHO + CH 3 -CH=CH 2 -f H 2

Acraldehyde is a colourless liquid, b.p. 52°, with a pungent irritating
odour. It is unstable, readily polymerising to a white solid. It is an
ap-unsaturated aldehyde, and undergoes many of the usual reactions of
an olefin and an aldehyde, e.g., it adds on two halogen atoms or a molecule
of halogen acid (contrary to Markownikoff's rule; cf. allyl alcohol) :

CH 2 :CH-CHO + Br 2 — > CH 2 Br-CHBr-CHO

CH 2 :CH-CHO + HCl — > CH 2 C1-CH 2 -CH0

The addition of hydrogen bromide is unaffected by the presence of peroxides.
Acraldehyde reduces ammoniacal silver nitrate, forms a cyanohydrin with
hydrogen cyanide, and a phenylhydrazone with phenylhydrazine :

ch 2 :ch-cho + [O] AgN0 ' > ch 2 :ch-co 2 h

CH 2 :CH-CHO + HCN ^ CH 2 :CH-CH(OH)-CN

ch 2 :ch-cho + c 6 h 5 -nh-nh 2 >■ ch 2 :ch-ch:n-nh-c s h 5 + H 2

Acraldehyde undergoes the Tischenko reaction (p. 161) to form allyl
acrylate :

2 ch 2 :ch-cho (C ' H °° )aA1 > ch,:ch-co 2 ch 2 -ch:ch„

276 ORGANIC CHEMISTRY

It does not, however, undergo the normal aldol condensation; instead, in
the presence of alkali, cleavage of the molecule takes place :

CH 2 :CH-CHO + H 2 ^===^ H-CHO + CH 3 -CHO

This reaction is actually the reversal of the aldol condensation, and is
characteristic of ap-unsaturated aldehydes.

When acraldehyde is reduced with metal and acid, there are obtained an
unsaturated alcohol, a saturated aldehyde, a saturated alcohol, and a com-
pound formed by bimolecular reduction :

ch 2 :ch-cho-^> ch 2 :ch-ch 2 oh + ch„-ch 2 -cho +

ch 3 -ch 2 -ch 2 oh + ch 2 :ch-choh-choh-ch:ch 2

A good yield of the bimolecular product is obtained by reducing acraldehyde
with magnesium amalgam (cf. pinacol, p. 253). Acraldehyde is reduced
by sodium amalgam to w-propanol and by aluminium isopropoxide (Meer-
wein-Ponndorf-Verley reduction, p. 150) to allyl alcohol. It should be
noted that metal and acid do not usually reduce a double bond; it is only
when the double bond is in the ap-position with respect to a carbonyl group
that it is reduced in this way (see also below). ap-Unsaturated aldehydes
maybe catalytically reduced to unsaturated alcohols by means of a platinum
catalyst in the presence of traces of ferrous sulphate and zinc acetate; the
former promotes reduction of the aldehyde group and the latter inhibits
reduction of the double bond (Adams et al.. 1927). When Raney nickel is
used as catalyst at fairly low temperatures and pressures, the product is
a saturated aldehyde, but at higher temperatures and pressures the product
is a saturated alcohol. Broadbent et al. (1959) have shown that both rhenium
heptaselenide and heptasulphide are good catalysts for hydrogenation.
Both are resistant to poisoning, but the former saturates the carbon-oxygen
double bond more easily than the carbon-carbon double bond.

ap-Unsaturated aldehydes (and ketones) may be reduced to the un-
saturated alcohols by lithium aluminium hydride, e.g.,

CH 3 -CH=CH-CHO UMH ' > CH 3 -CH=CH-CH 2 OH

If, however, a phenyl group is attached to the p-carbon atom, the double
bond is also reduced (see cinnamic acid, p. 690).

Acraldehyde adds on to butadiene to form a cyclic compound (see the
Diels-Alder reaction, p. 472) :

Crotonaldehyde (but-2-en-i-al) , CH 3 -CHXH*CHO, may be prepared by
heating aldol alone, or better, with a dehydrating agent, e.g., zinc chloride.
The best yield is obtained by distilling aldol with acetic acid as catalyst :

CH 3 -CHOH-CH 2 -CHO -^^- CH 3 -CH:CH-CHO

Crotonaldehyde is a colourless liquid, b.p. 104°. It closely resembles
acraldehyde in its chemical properties. It exists, however, in two geo-
metrical isomeric forms, cis and trans :

H— C— CH 3

CH 3 — C— H

11
H— C— CHO

II
H— C— CHO

cis

trans

Catalytic reduction (Raney nickel) of crotonaldehyde in the presence of
chloroform gives butyraldehyde; in the absence of chloroform, butanol is
formed (Cornubert et al., 1950).

UNSATURATED ALCOHOLS, ETHERS, CARBONYL COMPOUNDS AND ACIDS 277

A number of unsaturated aldehydes occur naturally in the essential oils, many
accompanying the corresponding unsaturated alcohol, e.g., citral or geranial,
neral, citronellal.

The simplest acetylenic aldehyde is propargylaldehyde (propiolaldehyde,
propynal), CH-OCHO. This may be prepared from acraldehyde, the aldehyde
group of which must be " protected " by acetal formation:

HCl Br,

CH 2 :CH-CHO + 2C s H 6 OH ^ H 2 + CH 2 :CH-CH(OC 2 H 5 ) 2 '-+

ethanolic dilute

CH a Br-CHBr-CH(OC s H 5 ) 2 > CH-C-CHtOCjHj)., > CH-C-CHO

KOH riCI

It may also be obtained by the controlled oxidation of propargyl alcohol.
Propargylaldehyde is a liquid, b.p. 60°, which undergoes cleavage when treated
with sodium hydroxide (c/. acraldehyde, above) :

CH-C-CHO + NaOH > CH;CH + H-CO a Na

UNSATURATED KETONES

The simplest unsaturated ketone is methyl vinyl ketone {but-^-en-2-one),
CH 3 -COCHXH 2 . This may be prepared by means of an aldol condensation
between formaldehyde and acetone, the product being dehydrated by heat.

H-CHO + CH 3 -C0-CH3-^VCH 2 0H-CH 2 -C0-CH 3 ^^>CH 2 :CH-C0-CH S

It may also be prepared by the action of acetic anhydride on ethylene in
the presence of zinc chloride as catalyst :

ch 2 :ch 2 + (ch 3 -co) 2 o -^V ch 2 :ch-coch 3 + CH 3 -C0 2 H

Methyl vinyl ketone may also be prepared by the action of acetic anhydride
on vinylmagnesium bromide in tetrahydrofuran at —60° to — 70 ° (Martin,
I 957) '• tms is a general method for preparing vinyl ketones :

2 CH 2 =CH-MgBr + 2 (CH 3 -CO) 2 — >

2CH 2 =CH-COCH 3 + MgBr 2 + (CH 3 -C0 2 ) 2 Mg (6o~8o%)

It is manufactured by hydrating vinylacetylene in the presence of dilute
sulphuric acid and mercuric sulphate (c/. acetaldehyde, p. 166) :

CH 2 :CH-C:CH + H 2 — > CH 2 :CH-COCH 3

Methyl vinyl ketone is a liquid, b.p. 79°, which polymerises on standing.
It is used commercially as the starting material for plastics.

Reduction of vinyl ketones with zinc or magnesium produces 1 : 6-diketones
(Kossanyi, i960) :

2R-CO-CR=CH 2 >- R-CO(CH 2 ) 4 -COR

Mesityl oxide (4-methylpent-3-en-2-one), (CH 3 ) 2 C:CH-CO-CH 3 , may be
prepared by distilling diacetone alcohol (q.v.) with a trace of iodine:

(CH 3 ) 2 C(OH)-CH 2 -CO-CH 3 -=^>(CH 3 ) 2 C:CH-CO-CH 3 (90%)

Stross et al. (1947) have shown that mesityl oxide prepared this way is a
mixture of mesityl oxide and isomesityl oxide, CH 2 :C(CH 3 )-CH 2 -CO'CH 3 .
Mesityl oxide may also be prepared as follows:

(CH 3 ) 2 C:CH 2 + CH 3 -COCl ZnC ' a > (CH 3 ) 2 CC1-CH 2 -C0-CH 3

" HC1 > (ch 3 ) 2 c:ch-co-ch 3

278 ORGANIC CHEMISTRY

Mesityl oxide is a liquid, b.p. 130°, with a peppermint smell. It is used
as a solvent for oils, gums, etc.

Phorone (2 : 6-dimethylhepta-2 : 5-dien-4-one), (CH 3 ) 2 C:CH-C<>CH:C(CH 3 ) 2 ,
may be prepared by the action of hydrochloric acid on acetone :

3CH 3 -COCH 3 -^> (ch 3 ) 2 c:ch-co-ch:c(ch 3 ) 2 + 2H 2

Phorone is a yellow, crystalline solid, m.p. 28 , b.p. 198 .

a(3-Unsaturated ketones show very marked additive properties at the
olefinic double bond. They may be reduced catalytically (platinum or
nickel catalyst) to saturated ketones, since the olefinic linkage is more rapidly
reduced than the keto group. If the hydrogenation is carried further,
saturated alcohols are produced:

r-ch:ch-cor' -5l> R.CH 2 -CH 2 -COR' -^-> R-CH,-CH 2 -CHOH-R'

Pt Pt a a

Cornubert et al. (1954) have shown that the double bond may be selectively
reduced by catalytic hydrogenation (Raney nickel) if the reaction is carried
out in ethylene chloride solution.

Reduction with metal and acid, or better, with magnesium amalgam, gives
bimolecular products (cf. acraldehyde, above) :

Mg/Hg (CH 3 )X-CH 2 -CO-CH 3
2(CH 3 ) 2 C:CH-COCH 3 — ^->-

(CH 3 ) 2 6CH 2 -COCH 3

Reduction of a : ^-unsaturated ketones with a mixture of lithium aluminium
hydride and aluminium chloride reduces the carbonyl group to a methylene
group.

The addition of halogen to a : ^-unsaturated ketones occurs by the normal
electrophilic mechanism across the 3 : 4-positions. The addition of halogen
acids, however, occurs by nucleophilic attack by the 1 : 4-mechanism (see
below).

The <x(3-unsaturated ketones add on ammonia, primary amines and
secondary amines to form (3-amino compounds; with ammonia, mesityl
oxide forms diacetonamine, and phorone forms triacetonamine :

NH 2

I
(CH 3 ) 2 GCH-COCH 3 + NH 3 — > (CH 3 ) 2 OCHyCOCH 3

(ch 3 ) 2 c:ch-co-ch:c(ch 3 ) 2 + 2NH3 — >

r(CH 3 ) 2 C-CH 2 -CO-CH 2 -C(CH 3 ) 2
L NH„ NH,

nh /CH 2 C(CH 3 ) 2

— -> co; ^NH

\CH 2 — C(CH 3 ) 2

All the evidence is in favour of a 1 : 4-addition mechanism when the addendum
is of the type H — Z, the product being an enol which then tautomerises to the
stable keto form, the overall result being 3 : 4-addition, e.g.,

4 S^7\ 2 ("V :NH 3

R— CH=CH— C£=0 > R— CH— CH=C— O >

I + lva I

R' H— -N— H R'

A

R— CH— CH=C— OH — >- R-CH-CH,-C=0

II II

NH, R' NH 2 R'

(enol form) (keto form)

UNSATURATED ALCOHOLS, ETHERS, CARBONYL COMPOUNDS AND ACIDS 279

With halogen acid, the mechanism will be:

0< OH O— H

•—■al IK hx s + I x- VT\ K
R'CH=CH— CR' === R-CH— CH=CR' > R-CHX-CH=CR' >

R-CHX-CH 2 -COR'

Similarly, <x(3-unsaturated ketones add on hydrogen cyanide to form
p-cyano-compounds, R*CH(CN) "CH^COR' ; and sodium hydrogen sulphite to
form p-sulphonic acids, R , CH(S0 3 Na)'CH 2 , CO , R'. The reaction with hydr-
oxylamine usually results in a mixture of the oxime, R , CH".CH > C(!N-OH) , R',
and the p-hydroxylaminoketone, R'CH(NH'OH)'CH 2 'CO"R'. In some
cases there is also formed the (3-hydroxylamino-oxime,

r-ch(nh-oh)-ch 2 -c(:n-oh)-r'.

One of the most important reactions that a(3-unsaturated ketones undergo
is the Michael condensation (1887). . This is the addition reaction between
an a|3-unsaturated keto-compound and a compound with an active methylene
group, e.g., malonic ester, acetoacetic ester, etc. The condensation is
carried out in the presence of a base, e.g., sodium ethoxide, or secondary
amines.

(i) (CH 3 ) 2 C:CH-COCH 3 + CH 2 (C0 2 C 2 H 6 ) 2 ^^-> (CH 3 ) 2 C-CH 2 -CO-CH 3

;h(co 2 c 2 h 5 ) 2

(ii) (CH 3 ) 2 C:CH-COCH 3 + CH 3 -CO-CH 2 -C0 2 C 2 H 5 CiH " NH ->

(CH 3 ) 2 OCH 2 -CO-CH 3

CH 3 -CO-CH-C0 2 C 2 H 5

By this means it is possible to synthesise a variety of compounds. The
description of -the Michael condensation given above is a limited one. The
term now includes a large variety of addenda and acceptors; the a : (3-
unsaturated compounds may be ketones, esters, cyanides, etc., and the
addenda may be malonic ester, E.A.A., cyanoacetic ester, nitroalkanes, etc.

The mechanism of the reaction is believed to be as follows (cf. the aldol and
Claisen condensations) ; e.g.,

CH 2 (C0 2 Et) 2 + EtCT ^ CH(CO a Et) 2 + EtOH

H5

Me 2 C=CH— -CMe + CH(CO a Et) 2 ^
O- O— H

I EtOH N Vf^ |<

Me,C— CH=CMe ;== Me„C— CH==CMe

I 1

CH(CO a Et) 2 CH(C0 2 Et) 2

Me 2 C— CH 2 -COMe

-> I

CH(CO a Et) 2

UNSATURATED MONOCARBOXYLIC ACIDS

Nomenclature. The longest carbon chain with the carboxyl group and
the double bond is chosen, and the position of the double bond with respect
to the carboxyl group is indicated by number, e.g.,

CH 3 'CH;CH'CH 2 'C0 2 H pent-3-enoic acid

280 ORGANIC CHEMISTRY

Alternatively, the acid is named as a substitution product of the olefin,
e.g.,

CH 3 -CHICH-C0 2 H prop-i-ene-i-carboxylic acid

In practice the unsaturated acids are usually known by their common
names (see text below).

Many methods are available for preparing unsaturated monocarboxylic
acids, but each one depends on the position of the double bond in the acid.
A general method for preparing afi-unsaturated acids is by the Knoevenagel
reaction (1898). This is the reaction between aldehydes and compounds
with active methylene groups in the presence of an organic base. The
reaction may take place in one of two ways :

(i) R-CHO + CH 2 (C0 2 C 2 H 5 ) 2 -^> R-CH:C(C0 2 C 2 H 6 ) 2 + H 2

(ii) R-CHO + 2CH 2 (CO.C 2 H 5 ) 2 — -> R-CH( + H 2

\CH(C0 2 C 2 H 6 ) 2

Reaction (i) is favoured by using equivalent amounts of aldehyde and ethyl
malonate in the presence of pyridine (Doebner, 1900). Reaction (ii) is
favoured by using excess ethyl malonate in the presence of piperidine, and
when the aldehyde is aliphatic. Furthermore, it appears that the term
Knoevenagel reaction is taken to mean the condensation when unsaturated
compounds are produced. Obviously then, to prepare a(3-unsaturated acids,
(i) must be used, followed by hydrolysis and heating.

R-CHO + CH 2 (C0 2 C 2 H 5 ) 2 JII^Hl^ R-CH:C(C0 2 C 2 H 6 ) 2 WK ° H >

(ii) HC1

r-ch:c(co 2 h) 2 - a °" 200 ° > r-ch:ch-co 2 h + co 2

In practice, it is usually sufficient to treat the aldehyde with malonic acid
in the presence of pyridine; e.g., acetaldehyde gives crotonic acid:

CH 3 -CHO + CH 2 (C0 2 H) 2 Pyndme > CHg-CHICH-COaH + C0 2 + H 2 (60%)

The mechanism of this reaction has been the subject of much discussion.
When a tertiary base, e.g., pyridine, is used as catalyst, then the mechanism
is believed to be similar to that of the aldol condensation (p. 157) :

CH 2 (C0 2 C 2 H 5 ) 2 + B ^ :CH(C0 2 C 2 H 5 ) 2 + BH +

-XO 0~

C ll - I

R— C + :CH(C0 2 C 2 H 5 ) 2 ^ R— C— CH(C0 2 C 2 H 5 ) 2 ^=^

H OH H

B + R— C— CH(C0 2 C 2 H 8 ) 2 -^^->R-CH:C(C0 2 C 2 H 5 ) 2
H
According to Zabicky (1961), the last stage proceeds as follows:

R-CHOH-CH(C0 2 C 2 H 5 ) 2 =^ R-CHOH-C(C0 2 C 2 H 5 ) 2 + H + — >

R-CH=C(C0 2 C 2 H 5 ) 2 + OH"

UNSATURATED ALCOHOLS, ETHERS, CARBONYL COMPOUNDS AND ACIDS 28l

When a primary or secondary base is used as catalyst, the mechanism may still
possibly be as given above, but a complication arises from the fact that carbonyl
compounds can form addition products with such bases, and these addition
products may therefore be intermediates in the reaction.

Ethyl malonate condenses with aldehydes only. On the other hand,
ethyl cyanoacetate condenses with ketones in the presence of acetamide
in glacial acetic acid solution, provided the water formed is removed con-
tinually by distillation; e.g., acetone forms j'sopropylidenecyanoacetic ester:

-ho / CN

(CH 3 ) 8 CO + CH 2 (CN)-C0 2 C 2 H 6 — > (CH 3 ) 2 C:c'

\C0 2 C 2 H 5

Since cyano-compounds are readily hydrolysed to the corresponding acid,
the above condensation may be used to prepare ocp-unsaturated acids of the
type R 2 C:CH-C0 2 H.

It can be seen from the foregoing that although the methylene group in
ethyl malonate will react with aldehydes, it is not sufficiently " active "
to react with ketones. Replacement of one carbethoxy group by a cyano-
group makes the methylene group (in the cyano-ester) active enough, i.e.,
increases the tendency for proton release, to react with ketones. Thus the
cyano-group is more strongly electron-attracting than the carbethoxy group.
Groups with a -I effect are usually known as negative groups. It is the
presence of a negative group that is the common feature in compounds
which tend to undergo condensation reactions, exhibit tautomerism, and
show increasedstrength as an acid and decreased strength as a base.

Many negative groups are characterised by the presence of multiple
bonds, e.g., N0 2 , CN, CO, etc., and it is these negative groups which give
rise to an active methylene group, and to tautomerism. On the other hand,
negative groups which do not contain multiple bonds, e.g., F, CI, Br, I,
•OCHj, do not give rise to an active methylene group, or to tautomerism.
All the negative groups, however, increase the strength of an acid and
decrease the strength of a base (examples of these phenomena will be found
in the text).

aP-Unsaturated acids may also be prepared by heating a-bromo-acids with
ethanolic potassium hydroxide, or better still, with potassium tert.-butoxide
(Cason et al., 1953); or by heating p-hydroxy-acids with aqueous sodium
hydroxide :

R-CIVCHBr-COijH + KOH ^^ ) R-CH:CH-C0 2 H + KBr + H a O

r-choh-ch 2 -co 2 h-^^->r-ch:ch-co 2 h + h 2 o

Py-Unsaturated acids may be prepared by heating an aldehyde with
sodium succinate and acetic anhydride at 100°. The product is a y-alkyl-
paraconic acid, and when this is heated at 150 , it eliminates carbon dioxide
to form the Py-unsaturated acid (Fittig et al., 1885) :

R-CHO I ] H * C °* m «*■<*»*>. r*-CH-CH-CQ 2 H "1
CH 2 -C0 2 Na IOO ° L OH CH 9 -C0 2 hJ
•CH— CH-

•H,0

i 2 v^ 2 i

R-CH— CH-CO»H

> r-ch:ch-ch 2 -co 2 h + co 2

*2

CO
This reaction is really an extension of the Perkin reaction (see p. 651).

282 ORGANIC CHEMISTRY

Unsaturated alcohols may be oxidised to the corresponding unsaturated
acid, provided the double bond is protected, e.g., acrylic acid from allyl
alcohol (see p. 252).

Oxidation of unsaturated aldehydes produces unsaturated acids. The
oxidation, however, cannot be carried out with the usual oxidising agents,
such as acid or alkaline permanganate, acid dichromate, etc., since these
will attack the double bond. A useful oxidising agent for unsaturated
aldehydes is ammoniacal silver nitrate (see p. 160) ; e.g., crotonaldehyde is
oxidised to crotonic acid :

ch 3 -ch:ch-cho + [Q] *™" > ch> .ch:CH-C0 1 H

AgNO, ° *

Unsaturated methyl ketones may be oxidised to unsaturated acids by
sodium hypochlorite (cf. the haloform reaction), e.g.,

r-ch:ch-co-ch 3 Na0cl > r.ch:ch-co 2 h

General reactions of the unsaturated acids. The esters of a(3-unsaturated
acids undergo the same addition reactions (including the Michael con-
densation) at the double bond as the a ^-unsaturated ketones. When
reduced by the Bouveault-Blanc method, <x(3-unsaturated esters are con-
verted into the corresponding saturated alcohol :

ch 3 -ch:ch-co 2 c 2 h 5 — -^— -> ch 3 -ch 2 -ch 2 -ch 2 oh

Lithium aluminium hydride, however, generally, reduces these esters to the
corresponding unsaturated alcohols {cf. p. 125).

Esters in which the double bond is further removed from the ester group
are reduced to the corresponding unsaturated alcohol (cf. acraldehyde,
above) ; e.g., butyl oleate is reduced to oleyl alcohol:

CH 3 -(CH 2 ) 7 -CH:CH-(CH a ) 7 -C0 2 C 1 H 9 Na/C ' H|0H >

ch 3 -(ch 2 ) 7 -ch:ch-(ch 2 ) 7 -ch 2 oh (82-84%)

All the unsaturated acids may be reduced catalytically to saturated acids.
a|3- and Py-Unsaturated acids add on halogen acid, the halogen atom
becoming attached to the unsaturated carbon atom which is further from
the carboxyl group :

ch 2 :ch-co 2 h + HC1 — > CH 2 C1-CH 2 -C0 2 H

CH 3 -CH:CH-CH 2 -C0 2 H + HBr — > CH 3 -CHBr-CH 2 -CH 2 -C0 2 H

This mode of addition (which may be contrary to Markownikoff's rule, as
in the case of acrylic acid) may be ascribed to the inductive effect of the
carboxyl group (cf. allyl alcohol, etc.). On the other hand, addition of
halogen acid to y8-unsaturated acids of the type CH 2 !CH'CH 2 *CH 2 'C0 2 H
takes place in accordance with Markownikoff's rule. This must be due to
the fact that the inductive effect of the carboxyl group ceases to be felt
beyond the (3-carbon atom.

When unsaturated acids are boiled with alkali, the double bond tends to
move so as to form the ap-unsaturated acid:

ch 3 -ch:ch-ch 2 -co 2 h -^V ch 3 -ch 2 -ch:ch-co 2 h

An interesting example of this migration of the double bond is the hydrolysis
of allyl cyanide with boiling alkali to produce crotonic acid. But-3-enoic
acid is probably formed first, and this then rearranges to crotonic acid:

ch 2 :ch-ch 2 -cn^V [CH 2 :cH-cH a -co a H] -^> ch 3 -ch:ch-co 2 h

UNSATURATED ALCOHOLS, ETHERS, CARBONYL COMPOUNDS AND ACIDS 283

The movement of a double bond to the a : p -position shows that the latter acid
is thermodynamically more stable, e.g., Linstead et al. (1934) showed that the
migration is reversible and that the equilibrium between the sodium salts of
vinylacetic acid and crotonic acid is 98 per cent, on the side of the latter :

CHs^H-CHj-CCV ===± CH 3 -CH=CH-C0 2 -

The mechanism of this shift is believed to be :

CH 2 =CH-CH,-CO,- ° " ; H L_ : CH,=CH-CH-CO,

^ CH,-CH=CH-CO,

This is supported by the fact that sodium vinylacetate exchanges H for D in
sodium hydroxide solution containing D a O (Ives et al., 1935); sodium crotonate
also exchanges, but much more slowly (Ives, 1938).

When unsaturated acids are fused with alkali, cleavage of the chain takes
place with the formation of two acids, one of which is always acetic acid.
This again indicates the migration of the double bond to the ap-position;
e.g., oleic acid gives acetic and palmitic acids:

ch 3 -(ch 2 ) 7 -ch:ch-(ch 2 ) 7 -co 2 h ^> [ch 3 -(ch 2 ) 14 -ch:ch-co 2 h]

KOH

> CH 3 -(CH 2 ) 14 -C0 2 H + CH 3 -C0 2 H

It is therefore obvious that fusion of an unsaturated acid with alkali cannot
be used to determine the position of the double bond. If, however, the
unsaturated acid is treated with cold dilute alkaline permanganate, or
hydrogen peroxide in acetic acid (or formic acid), the double bond is hydr-
oxylated, and the glycol formed under these conditions, i.e., without migra-
tion of the double bond, may now be oxidised in the usual way (see p. 73).
Alternatively, the position of the double bond may be ascertained by
ozonolysis (p. 74). In view of this shift of the double bond to the a : Im-
position, it therefore follows that the a : p-unsaturated acid is thermo-
dynamically more stable than any of the other isomeric unsaturated
acids.

ap-Unsaturated esters add on aliphatic diazo-compounds to form pyrazoline
derivatives; e.g., acrylic ester reacts with diazomethane to give pyrazoline-
5-carboxylic ester:

CH 2 :CH-C0 2 C 2 H 5 + CH 2 N 2 ^ CH 2 -CH-C0 2 C 2 H 5

HC. ,NH

N

ap-Unsaturated acids may be degraded by the Hofmann method (p. 206)
using a modified procedure. The a p-unsaturated acid amide in methanol
is treated with an alkaline solution of sodium hypochlorite, the urethan
produced being hydrolysed in acid solution to give the aldehyde in good
yield:

r-ch:ch-conh 2 NaOC " NaOH > r.ch:ch-nh-co 2 ch 3 -^> r-cho

CH a OH

In this case, two carbon atoms are eliminated from the acid. On the other
hand, Py- and y8-unsaturated acid amides eliminate one carbon atom to
produce the corresponding unsaturated primary amine, but in poor yield:

r-ch:ch-ch 2 -conh 2 — > r-ch:ch-ch 2 -nh 2

284 ORGANIC CHEMISTRY

Acrylic acid (prop-2-enoic acid), CH 2 :CH*C0 2 H, may be prepared:

(i) By the oxidation of allyl alcohol or acraldehyde (see p. 267).
(ii) By heating (J-hydroxypropionic acid with aqueous sodium hydroxide:

CH 2 OH-CH 2 -C0 2 H Na ° H > CH 2 :CH-C0 2 H + H 2
Alternatively, ethylene cyanohydrin may be heated with sulphuric acid:

CH 2 OH-CH 2 -CN H ' S0 ' > [CH 2 0H-CH 2 -C0 2 H] ~ H '° > CH 2 :CH-C0 2 H
Vinyl cyanide also gives acrylic acid on hydrolysis :

ch 2 :ch-cn Ha ° -> ch 2 :ch-co 2 h

It is therefore possible that the intermediate in the hydrolysis of ethylene
cyanohydrin may be vinyl cyanide or p-hydroxypropionic acid, or both.
Acrylic acid is prepared industrially with ethylene cyanohydrin (pre-
pared from ethylene oxide and hydrogen cyanide, p. 251) as the starting
material.

(iii) A new industrial preparation is by the interaction of acetylene,
carbon monoxide and water in the presence of nickel salts as catalyst :

CH-CH + CO + H 2 — >■ CH 2 :CH-C0 2 H

If an alcohol is used instead of water, the corresponding acrylic ester is
obtained, e.g.,

CH-CH + CO + CH3OH — > CH 2 :CH-C0 2 CH 3

Another industrial method is by the pyrolysis of an alkyl acetyl-lactate,
e.g.,

CH 3 -CH(0-CO-CH 3 )-C0 2 CH 3 — -> CH 2 =CH-C0 2 CH 3 + CH 3 -C0 2 H

Acrylic acid is a colourless liquid, b.p. 141°, which is miscible with water
in all proportions. On standing it slowly polymerises to a solid.

Methacrylic acid (2-methylprop-2-enoic acid), CH 2 ;C(CH 3 )*C0 2 H, exists
as colourless prisms, m.p. 15 . It may be prepared by removing a molecule
of hydrogen bromide from a-bromo«sobutyric acid:

CHo CHq

I I

CH 3 -CBrC0 2 H *" K °% CH 2 :C-C0 2 H

3 2 -HBr 2 2

Methyl methacrylate is very important commercially, since it polymerises
to polymethyl methacrylate under the influence of heat. This polymer is
tough, transparent, and can be moulded; one of its trade names is perspex.
One industrial method of preparing methyl methacrylate is as follows :

CH 3

CH3 V i^> CH, \^ CHa ° H > CH 2 :C-C0 2 CH 3 + NH 4 HS0 4

CH/ CH/ \CN H,so '

Methyl acrylate, which may be prepared by heating ethylene cyanohydrin
with methanol and sulphuric acid, also polymerises, but this polymer is
softer than that from the methacrylate.

Crotonic acid and teocrotonic acid both have the same structure

UNSATURATED ALCOHOLS, ETHERS, CARBONYL COMPOUNDS AND ACIDS 285

CH 3 'CHXH*COgH (but-2-enoic acid), but are geometrical isomers, crotonic
acid being the trans isomer, and isocrotonic the cis:

H— C— CH, H— C— CH 3

II II

H0 2 C-^C— H H— C— C0 2 H

crotonic acid t'socrotonic acid

The more stable form is crotonic acid, m.p. 72 . isoCrotonic acid, m.p.
15°, slowly changes into crotonic acid when heated at ioo°.

Crotonic acid (fraws-but-2-enoic acid) may be prepared by oxidising
crotonaldehyde with ammoniacal silver nitrate; by heating p-hydroxy-
butyric acid with sodium hydroxide; or by the Knoevenagel reaction
(see p. 280) ;

CH.-CHO + CH 2 (C0 2 H) 2 Pyridine > CH 3 -CH:CH-C0 2 H + C0 2 + H 2 (60%)

20°

isoCrotonic acid (rfs-but-2-enoic acid) may be prepared by the action of sodium
amalgam on (3-chloroisocrotonic ester which is obtained from acetoacetic ester by
reaction with phosphorus pentachloride :

pci, -Ha
CH 3 -CO-CH 2 -C0 2 C 2 H 5 --> CH 3 -CC1 2 -CH 2 -C0 2 C 2 H 5 ->

ch 3 -cci:ch-co 2 c 2 h 5 Na,Hg > ch 3 -ch:ch-co 2 c 2 h 5

The action of sodium amalgam is particularly interesting, since it normally
reduces a double bond in the a(J-position.

Angelic and tiglic acids are geometrical isomers with the structure
CH 3 -CHX(CH 3 )'C0 2 H (■z-methylbut-2-enoic acid) :

CH 3 — C— H

H G CHq

II
CH,-^— C0 2 H

II
CH 3 C C0 2 H

angelic acid, m.p. 45

tiglic acid, m.p. 64

Both acids are found in nature as esters.

Undecylenic acid (dec-g-ene-i-carboxylic acid), CH 2 !CH>(CH a ) 8 'C02H, may be
obtained by the destructive distillation of ricinoleic acid (n-hydroxyheptadec-8-
ene-i-carboxylic acid), which occurs as the glyceride ester in castor-oil. Heptanal
is the other product, 10 per cent, yield of each being obtained:

ch 3 -(ch 2 ) b -choh-ch 2 -ch:ch-(ch 2 ),-co 2 h — >

ch 3 '(ch 2 ) 5 -cho + ch 2 :ch-(ch 2 ) 8 -co a h

Undecylenic acid is a solid, m.p. 24-5°, and is said to be a preventive and a cure for
athlete's foot.

When ricinoleic acid is heated with sodium hydroxide in air, octan-2-ol and
sebacic acid are obtained:

NaOH

ch 3 -(ch 2 ) 6 -choh-ch 2 -ch:ch-(ch 2 ) 7 -co 2 h ^

air

CH 3 -(CH 2 ) 6 -CHOH-CH 3 + C0 2 H-(CH 2 ) 8 -C0 2 H

When ricinoleic acid (castor oil) is treated with concentrated sulphuric acid, it
gives a complex mixture consisting of the hydrogen sulphate of ricinoleic acid in
which the hydroxyl group is esterified, and a compound in which the sulphuric
acid has added to the double bond: esterification and addition do not occur
together in the same molecule of ricinoleic acid. The product, which is known
as Turkey-red oil, is used as a wetting agent.

Oleic acid (cis-heptadec-8-ene-i-carboxylic acid) occurs as the glyceryl
ester in oils and fats. It is a colourless oil, m.p. 16 , which is insoluble in

286 ORGANIC CHEMISTRY

water, but soluble in ethanol and ether. Catalytic reduction converts it
into stearic acid. Cold dilute alkaline permanganate, or hydrogen peroxide
in acetic acid, converts oleic acid into 9 : 10-dihydroxystearic acid, which,
on oxidation with, e.g., acid permanganate, gives nonoic acid and azelaic
acid:

ch 3 -(ch 2 ) 7 .ch:ch-(Ch 2 ) 7 -co 2 h J^>

CH 3 -(CH 2 ) 7 -CHOH-CHOH-(CH 2 ) 7 -C0 2 H

acid

KMno. ■*■ CH 3 -(CH a ) 7 -C0 2 H + C0 2 H-(CH 2 ) 7 -C0 2 H

This shows that the double bond in oleic acid is in the 9 : 10 position (car-
boxyl group being 1), and this is confirmed by ozonolysis.

Oleic acid is the cis form. The trans form is elaidic acid. m.p. 51 , which
may be obtained by the action of nitrous acid on oleic acid :

H-C-(CH 2 ) 7 -CH 3 HNOa CH 3 -(CH 2 ) 7 -C-H

H— C— (CH 2 ) 7 -C0 2 H H— C— (CH 2 ) 7 -C0 2 H

Linoleic acid (heptadeca-8 : ii-diene-i-carboxylic acid),
CH s -(CH 2 ) 4 -CH:CH-CH a -CH:CH-(CH 2 ) 7 -C0 2 H, b.p. 228714 mm., occurs as the
glyceryl ester in linseed oil, hemp oil, etc. ; it has been synthesised by Raphael
et al. (1950).

Linolenic acid (heptadeca-8 : n : 14-triene-i-carboxylic acid),
CH3-CH 2 -CH:CH-CH 2 -CH:CH-CH 2 'CH:CH-(CH 2 ) 7 -C0 2 H, is a liquid which occurs
as the glyceryl ester in, e.g., poppy-seed oil. Oils containing linoleic and linolenic
acids are drying oils (p. 263). Linolenic acid is the most abundant and wide-
spread triene acid in Nature; it has been synthesised by Weedon et al. (1956).

Erucic acid, m.p. 34°, is the cis form of heneicos-12-ene-i-carboxylic acid,
CH 3 -(CH 2 ) 7 *CHICH , (CH 2 ) 11 , C0 2 H. It occurs as the glyceryl ester in various
oils, e.g., rape oil, cod-liver oil, etc. It is converted into the trans isomer
brassidic acid, m.p. 61 -5°, by the action of nitrous acid (cf. oleic acid, above) :

H-C— (CH 2 ) 7 -CH 3 HNOj CH 3 -(CH 2 ) 7 — C-H

h4

-(CH 2 ) n -C0 2 H H— C— (CH 2 ) u -CO a H

Nei\omcacid(tricos-i4-ene-i-carboxylicacid),CH a '(CH. 2 ) 7 , Cii'.CH'(CH 2 ) 13 'CO i iI,
occurs in human brain-tissue and in fish oils. Both the cis and trans forms are
known, but it appears to be the cis acid, m.p. 43 , which occurs naturally.

Propiolic acid (propargylic acid, prop-2-ynoic acid), CH:OC0 2 H, is the simplest
acetylenic carboxylic acid. It is conveniently prepared by the action of dry
carbon dioxide on sodium acetylide:

CH^CNa -^-> CH:C-CO„Na d ''" te > CH-C-C0 2 H

HjSO,

It is a colourless liquid, m.p. 9°, which smells like acetic acid; it is reduced by
sodium amalgam to propionic acid; it adds on halogen acid to give the (3-halo-
geno-acrylic acid. It forms the silver or cuprous salt when treated with an
ammoniacal solution of silver nitrate or cuprous chloride, respectively. When
exposed to sunlight, it polymerises to trimesic acid (benzene 1:3: 5-tricarboxylic
acid; cf. acetylene) :

C0 2 H

(A

3 CH : C-C0 2 H >■ H02 cJI Jco 2 H

a fJ- Acetylenic acids are stronger acids than their ethylenic and saturated
analogues due to the acetylene bond exerting a strong electron-attracting effect
(cf. p. 96).

UNSATURATED ALCOHOLS, ETHERS, CARBONYL COMPOUNDS AND ACIDS 287

THE PRINCIPLE OF VINYLOGY

According to Fuson (1935), the principle of vinylogy may be described as
follows : If E 2 and E 2 represent non-metallic elements, then in the compound of
the type A— E 1 =E 2 or A — E^Ej, if a structural unit of the type / — C=C\ is

interposed between A and E x , the function of E 2 remains qualitatively un-
changed, but that of Ej may be usurped by the carbon atom attached to A. Thus

A— /C=C\„— E,=E.. or A— /C=C

•/C=Cy— E^E, or A— /C=C\„-

form vinylogous series. This may be illustrated by the compounds ethyl acetate
and ethyl crotonate. These are vinylogues (w being equal to 1) :

CH 3 — C=0 CH 3 — I CH=CH ;

A<

OC 2 H 5

A is equivalent to CH 3 and E 1 =E 2 to — C=0

OC 2 H 5

Ethyl acetate condenses with ethyl oxalate to form oxalacetic ester (this, of
course, is an example of the Claisen condensation) :

Na

C a H 5 OOC-COOC 2 H 5 + CH 3 -COOC 2 H 5 ^

ether

C 2 H 6 OOC-COCH 2 -COOC 2 H 5 + C 2 H 6 OH
In the same way, ethyl crotonate also condenses with ethyl oxalate :

c 2 h 5 ooc-cooc 2 h 5 + ch 3 -(ch==ch:-cooc 2 h 5 -^-->

ether

c 2 h 5 ooc-co-ch 2 {ch=chVcooc 2 h 5 + C 2 H 5 OH

Several other examples of the application of vinylogy will be found in the text.

KETENS (KETENES)

Ketens are compounds which are characterised by the presence of the
grouping >G=C=0. If the compound is of the type R'CHICO, it is
known as an aldoketen; and if R 2 CC:0, then a ketoketen.

The most general method of preparing a keten is by debrominating an
a-bromoacyl bromide with zinc, e.g., dimethylketen from a-bromo/sobutyryl
bromide:

(CH 3 ) 2 CBr-COBr + Zn — y (CH 3 ) 2 C:C:0 + ZnBr 2

The simplest member of the keten series is keten (ketene), CH 2 X:0.
Keten may be prepared by debrominating bromoacetyl bromide with zinc :

CH 2 Br-COBr + Zn — >- CH 2 '.CO + ZnBr 2

It is, however, usually prepared by the thermal decomposition of acetone
(ethyl acetate or acetic anhydride may also be used as the starting material) :

ch 3 -coch 3 — > ch 2 :c:o + ch 4

In practice, acetone vapour is passed over an electrically heated metal
filament at 700-750 ; the filament is made of chromel wire (an alloy of
80 per cent, nickel and 20 per cent, chromium). The yield of keten is high,
being usually between 90 and 95 per cent. An alternative procedure,

288 ORGANIC CHEMISTRY

which gives a much lower yield (25-29 per cent.), is to pass acetone vapour
through a long combustion tube filled with broken porcelain heated to
redness.

The thermal decomposition of acetone has been shown to be a free-radical
chain reaction. The exact details are not known with certainty; Rice and
Walters (1941) have suggested the following:

CH 3 -COCH 3 — > CH 3 > + CH 3 -CO — > 2 CH 3 - + CO

CH 3 -COCH 3 + CH 3 « — > CH 3 -COCH 2 - + CH 4

CH 3 -COCH 2 >- CH 2 :CIO + CH 3 -, etc.

Keten is prepared industrially by passing acetone vapour through a copper
tube at 700-850 , and by the decomposition of acetic acid in the presence of
phosphoric acid or ethyl phosphate as catalyst :

CH 3 -C0 2 H -^> CH 2 =C=0 + H,0

cat.

Keten is a colourless, poisonous, pungent gas, b.p. — 41 , which oxidises
in air to the unstable peroxide, the structure of which may be as shown.

£tt r=n Keten is a very reactive compound, and so is important as

I a I a synthetic reagent. It rapidly polymerises to diketen, and

k A because of this tendency to polymerise, keten is usually used

immediately when prepared.

The reactions of keten are generally those of an acid anhydride; it
acetylates most compounds with an active hydrogen atom (p. 198), pro-
vided the compound can be dissolved in some solvent inert to the action
of keten. The yield of acetyl derivative is almost 100 per cent, (based on
keten). The following reactions clearly show the acetylating property of
keten:

(i) When keten is passed into water, acetic acid is formed slowly:

CH 2 :CO + H 2 > CH 3 -C0 2 H

(ii) When keten is passed into glacial acetic acid, acetic anhydride is
formed :

CH 2 :CO + CH 3 -C0 2 H — > (CH 3 -CO) 2

By means of this reaction a mixed anhydride may be obtained, one radical
of which, of course, must be the acetyl group :

R-C0 2 H + CH 2 :CO — > R-COOCO-CH 3

(iii) Keten reacts with aliphatic or aromatic hydroxy-compounds :

CH 2 =CO -f- ROH — > CH 3 -C0 2 R

The reaction, however, is best carried out in the presence of a catalyst,
e.g., sulphuric acid.

It also acetylates enolisable compounds, e.g.,

CH ' — CO

CH 3 -COCH 2 -C0 2 C 2 H 5 H ' go > CH 3 -C=CH-C0 2 C 2 H 5 +

OCOCH 3

CH 2 =C-CH 2 -C0 2 C 2 H 5

OCOCH,

UNSATURATED ALCOHOLS, ETHERS, CARBONYL COMPOUNDS AND ACIDS 289

(iv) Keten reacts with ammonia to form acetamide, and with primary
or secondary amines to form iV-alkyl-acetamides:

CH 2 :CO + NH S — >■ CH 3 -CONH 2
CH 2 :CO + R-NH 2 — >■ CH 3 -CO-NH-R

Primary amino-groups are acetylated extremely readily, and hence it is
possible to acetylate this group selectively in compounds containing both
an amino and hydroxyl group, e.g., ^-aminophenol :

HO/ \nH 2 + CH 2 =CO — > HO/ ^NH-COCH 3

A particularly useful reaction of this type is the acetylation of amino-acids:

R-CH(NH 2 )-C0 2 H + CH 2 :CO — > R-CH(NH-COCH 3 )-C0 2 H

Amides are also acetylated by keten in the presence of sulphuric acid, but
at elevated temperatures (in the absence of sulphuric acid) the cyanide is
formed.

R-CO-NH a + CH 2 =CO — > R-CN + CH 3 -C0 2 H

(v) Thiols are acetylated (preferably in the presence of sulphuric acid as
catalyst) :

R-SH + CH 2 =CO — >■ R-S-CO-CH 3

(vi) Keten reacts with Grignard reagents to form methyl ketones:

RMgX + CH 2 =CO — > R-COCH 3

In addition to behaving as an acetylating reagent, keten behaves as an
unsaturated compound, the carbonyl group showing no reactivity, e.g.,
(i) Keten adds on bromine to form bromoacetyl bromide:

CH 2 :CO + Br 2 — > CH 2 Br-COBr

The halogeno-acetyl halide is also formed by the interaction of keten and
phosphorus pentahalides :

ch 2 :co + PC1 5 — > CH 2 C1-C0C1 + PC1 3

(ii) Keten adds on halogen acid to form the acetyl halide :

ch 2 :co + HX — > CH 3 -COX

Alkyl hahdes also react with the keten in the presence of charcoal at ioo°
to form acid chlorides:

RX + CH 2 =CO — > R-CH a -COX

This offers a means of stepping up a series.

(hi) A particularly interesting addition reaction of keten is that with
aldehydes to form p-lactones; e.g., with formaldehyde, $-propiolactone is
formed:

CH 2 :CO + H-CHO — > CH 2 -CH 2 -CO

1 '

Diketen is readily formed by passing keten into acetone cooled in solid
carbon dioxide (yield: 50-55 per cent.). Diketen is a pungent-smelling
liquid, b.p. 127°, which, when strongly heated (550-600 ), is depolymerised
to keten.

290 ORGANIC CHEMISTRY

The structure of diketen is still uncertain. Various structures have been
suggested, e.g.,

CH 2 — CO

CO — CH 2

cyc/obutane-i : 3-dione

(I)

CH 2

CO-

mono

—C-OH

— CH
-enol of I
(II)

CH„=C

1 1
CH 2 -C=0

vinylaceto-^-lactone

(III)

CH 3 — C=CH
O— CO

jS-crotonolactone
(IV)

CH 3 -

-C— CH=C=0

acetylketen
(V)

According to Whiffen and Thompson (1946), based on their work of
infra-red measurements, the structure of diketen is most likely III or
IV, or a mixture of both. These authors also point out that the evidence
seems to favour III. This formula is also favoured by Blomquist and
Baldwin (1948), who treated diketen with iV-bromosuccimide and showed
that the resulting products agree with structure III (and not with IV).
Structure III is also supported by mass spectra studies of Long et al. (1953).
Structure I has been excluded by work with isotope 14 C (Roberts d al.,
1949). It is interesting to note, however, that the structure of dimethyl-
keten dimer is 2:2:4: 4-tetramethylcycZobuta-i : 3-dione {i.e., it corre-
sponds to I), and the ring is planar (Robertson et al., 1956).

Diketen may be further polymerised to dehydroacetic acid (p. 227).

Diketen reacts with alcohols to form esters of acetoacetic acid, and this
reaction is now being used to prepare acetoacetic ester industrially (see
p. 224). It also reacts with primary amines to form AT-substituted aceto-
acetamides, e.g., with aniline:

CH 2 =C O + C 6 H 6 -NH 2 — > CH 3 -CO-CH 2 -CO-NH-C 6 H 5

CH 2 — CO

Diketen also undergoes the Friedel-Crafts reaction with benzene to form
benzoylacetone :

CH 2 =C O A, A1C1 ^ /\cO-CH !! -CO-CH3

CH 2 -CO \/- \^

QUESTIONS

1. Using acetone or glycerol as the starting material, show how you would prepare : —
(a) 3-methylbut-2-enoic acid, (6) /3 : /3-dimethylacrylic acid, (c) /3-hydroxypropionic
acid, (d) a : oe-dimethylsuccinic acid, (e) a : /3-dibromopropionic acid, (/) glyceraldehyde.

2. Name the following compounds and suggest one synthesis for each:

(a) CH 2 :CH-CH 2 -CH(CH 3 )-C0 2 H

(b) Me a C-CH 2 -C0 2 H (c) CH a -CH a -CO-CH 3

Me 2 C-CH 2 -C0 2 H CH(CO s H) 2

(d) Me 8 C(CH 2 -C0 8 H) 2 (e) CH 3 -CO-CH 2 -CHMe-CH 2 -0-CH s

(/) Me 2 C(CH a -CH 2 -C0 2 H) 2 .

3. How would you distinguish between allyl alcohol and propargyl alcohol?

4. Describe the preparation and the more important properties of: — (a) allyl alcohol,
(6) allyl halides, (c) crotyl alcohol, (d) propargyl alcohol, (e) methyl vinyl ether, (/)
divinyl ether, (g) crotonaldehyde, (h) propargylaldehyde, (i) methyl vinyl ketone,
(j) crotonic acid, (ft) propiolic acid.

UNSATURATED ALCOHOLS, ETHERS, CARBONYL COMPOUNDS AND ACIDS 29I

5. How may acraldehyde, mesityl oxide, phorone and acrylic acid be prepared?
Name the compounds and state the conditions under which they are formed when the
above substances, respectively, are treated with : — (a) Br 2 , (b) HBr, (c) HCN, Id) NaOH
(e) NH 3 , (/) H 2 , (g) NaHSO,, (h) NH 2 -OH, (i) CH 2 (C0 2 Et) 2 , (j) E.A.A., (ft) CH 2 N 2 .

6. Define and give examples of: — (a) hyperconjugation, (b) the allylic rearrangement,
(c) the 1 : 4 mechanism, (d) Michael condensation, (e) Knoevenagel reaction, (/) negative
groups, (g) geometrical isomerism, (h) vinylogy, (i) halochromic salt, (j) cyanoethylation.

7. Discuss the general methods of preparation and the more important properties of
the unsaturated monocarboxylic acids. Give an account of the methods that can be
used for determining the position of the double bond in an unsaturated acid.

8. Write an account of the preparation and properties of keten and diketen, including
in your answer a discussion of the structure of diketen.

9. Write an account of the reduction of a/S-unsaturated carbonyl compounds.

READING REFERENCES

Walsh, Remarks on the Strengths of Bonds, Trans. Faraday Soc, 1947, 4^. 60, 158, 342.
Bent, Distribution of Atomic s Character in Molecules and its Chemical Implications, /.

Chem. Educ, i960, 37, 616.
Organic Reactions, Wiley. Vol. V. (1949), Ch. 2. Cyanoethylation. Vol. X (1959),

Ch. 3. The Michael Reaction.
Unsaturated Acids, Ann. Reports, 1940, 37, 211.
Boxer and Linstead, The Knoevenagel Reaction, J.C.S., 1931, 740.
Fuson, The Principle of Vinylogy, Chem. Reviews, 1935, 16, 1.

Crawford, Hyperconjugation, Quart. Reviews (Chem. Soc.), 1949, 3, 226.

Baker, Hyperconjugation, Oxford Press (1952).

Rao, Is Hyperconjugation Necessary? Nature, i960, 187, 913.

Organic Reactions, Wiley. Vol. Ill (1946), Ch. 3. Preparation of Ketenes and

Ketene Dimers.
Advances in Organic Chemistry, Vol. 2 (i960). Lacey, Ketene in Organic Syntheses,

p. 217.
Johnson, Lecture on Some Applications of Acetylenic Compounds in Organic Synthesis,

Royal Institute of Chemistry (1948).
Gilman, Advanced Organic Chemistry, Wiley (1942, 2nd ed.).
(i) Vol. I, Ch. 7. Unsaturation and Conjugation.

(ii) Vol. I, Ch. 9. Catalytic Hydrogenation and Hydrogenolysis.
Raphael, Acetylenic Compounds in Organic Synthesis, Butterworth (1955).
Gunstone, Recent Developments in the Preparation of Natural Straight-Chain Fatty

Acids, Quart. Reviews (Chem. Soc), 1953, 7. T 75-
DeWolfe and Young, Substitution and Rearrangement Reactions of Allylic Compounds,

Chem. Reviews, 1956, 56, 753.
Friedlander and Robertson, The Crystal Structure of Dimethylketen Dimer, J.C.S.,

1956. 3 o8 3-
Evans, Oxidations by Manganese Dioxide in Neutral Media, Quart. Reviews (Chem. Soc),

1959, 13. 61.

CHAPTER XIII

NITROGEN COMPOUNDS

Some compounds containing nitrogen have been described in Chapter IX,
since these were regarded primarily as acid derivatives. It is now proposed
to deal with many other nitrogen compounds, most of which may be regarded
as alkyl nitrogen compounds.

Hydrocyanic acid, hydrogen cyanide (prussic acid), HCN, was discovered
by Scheele (1782), who obtained it from bitter almonds which contain the
glycoside amygdalin. Amygdalin, on hydrolysis with dilute acid, yields
hydrogen cyanide, benzaldehyde and glucose :

C 20 H 27 O u N + 2H a O -^> HCN + C 6 H 5 -CHO + 2C 6 H 12 6

Hydrogen cyanide also occurs in the leaves of certain plants, e.g., laurel.
It may be prepared by heating sodium cyanide with concentrated sulphuric
acid, the gas being dried over calcium chloride:

NaCN + H 2 S0 4 — > HCN + NaHS0 4 (93-97%)

Hydrogen cyanide is prepared industrially:

(i) By passing a mixture of ammonia, air and methane over a platinum-
rhodium gauze catalyst at 1000 :

2NH3 + 30 2 + 2CH 4 — > 2HCN + 6H 2

(ii) By passing a mixture of carbon monoxide and ammonia over an alu-
mina catalyst at 500-700° :

CO + NH 3 — > HCN + H 2

Hydrogen cyanide is a colourless, poisonous liquid, b.p. 26°. It is a very
weak acid, and is miscible in all proportions with water, ethanol and ether.
It hydrolyses slowly in aqueous solution, and more rapidly in the presence
of inorganic acids to form first formamide, and then ammonium formate:

HCN — V H-CO-NH 2 — V H-C0 2 NH 4
Hydrogen cyanide is reduced by nascent hydrogen to methylamine:

HCN + 4[H] Zn/HC1 > CH 3 -NH 2

It is a useful reagent for certain syntheses, e.g., cyanohydrins by combination
with carbonyl compounds (p. 151), and in the Gattermann aldehyde synthesis

(P- 6 47)-
Structure of hydrogen cyanide. Hydrogen cyanide is believed to exist in

two forms because it gives rise to two kinds of alkyl derivatives : the alkyl

cyanides, R*CN, which are derivatives of hydrogen cyanide, HCN ; and the

alkyl isocyanides, R-NC, which are derivatives of hydrogen isocyanide, HNC.

The structures of HCN and R-CN, respectively, were originally believed

to be H'C:N and R'CSN, but recent determinations of the dipole moments

of the alkyl cyanides indicate that they are resonance hybrids of structures

I and II, the contribution of II being about 57 per cent. :

R— C=N : <-> r— C=N :

(I) PI)

On the other hand, infra-red studies strongly favour I.

292

NITROGEN COMPOUNDS 293

The structures of HNC and R'NC proved more difficult to elucidate.
Originally, the formula of the alkyl j'socyanides was written R — N=C
(nitrogen with a valency of five and carbon, four). This, however, was
inconsistent with the chemical properties of *'«?cyanides, and was soon
considered unlikely. It is now known, according to the electronic theory
of valency, that four bonds between nitrogen and carbon is impossible on
theoretical grounds.

Normally, on addition to unsaturated compounds, the addendum adds to
each atom joined by the double bond; this does not occur with isocyanides,
the addendum adding only to the carbon atom :

r.NCS <i— R-NC -^-> R-NCBr 2

These facts led Nef (1891-1892) to suggest that isocyanides did not contain
quadrivalent carbon but bivalent carbon. He therefore wrote their formula
as R — N=C^, with two " free " valencies by means of which t'socyanides
add on the addendum to the carbon atom alone (as shown in the above
examples).

Study of the dipole moments and Raman spectra of tsocyanides indicated
that bivalent carbon did not represent the true state of affairs, and led to
the suggestion that the bond between the nitrogen and carbon atoms was
a triple bond. If we use the electronic formula of Nef 's structure, we can
introduce a triple bond between the nitrogen and carbon atoms by rearrang-
ing the lone pair of the nitrogen atom to act as a donor to the carbon atom,
i.e.,

R— N=C : — > R— N^C

Raman spectra of cyanides and isocyanides show the presence of the C=N
group. This is satisfied by the N=C group. On the other hand, more
recent work indicates that alkyl »socyanides are resonance hybrids of III
and IV:

R— N=C: -<-> R— N^C

(III) (IV)

Furthermore, the value of the C — N — C valency bond angle has been found
to be 180°, i.e., these three atoms form a linear arrangement, which shows
that IV makes a very large contribution to the actual state (see Ch. 11).

Thus, in view of the existence of alkyl cyanides and tsocyanides, hydrogen
cyanide is believed to be tautomeric, giving rise to the nitrile-isonitrile
diad system:

H— C=N ^ C^N— H

The position, however, cannot be regarded as being completely satis-
factory. It appears that only one form of acid is known, viz., hydrogen
cyanide; all its reaction indicate the structure H — C=N, e.g., its hydrolysis
to formic acid.

ALKYL CYANIDES

These are also known as nitrites or carbonitriles.

Nomenclature. This group of compounds is usually named either as
the alkyl cyanides {i.e., the alkyl derivatives of hydrogen cyanide), or as the
nitrite of the acid which is produced on hydrolysis, the suffix -ic of the trivial

294 ORGANIC CHEMISTRY

name being replaced by -onitrile. The following examples illustrate both
systems of nomenclature:

HCN hydrogen cyanide or formonitrile

CH 3 -CN methyl cyanide or acetonitrile

(CH 3 ) 2 CH-CN wopropyl cyanide or Mobutyronitrile

General methods of preparation, i. By the dehydration of acid amides
with phosphorus pentoxide :

R-CO-NH. _ P ;°; o > R-CN (g.)

Amides may also be converted into cyanides by heating with phosphorus
pentachloride or thionyl chloride. Dehydration of the amide to cyanide is
also very conveniently effected by heating with sulphamic acid (Kirsanov
etal., 1950).

R-CONH 3 + NH 2 -S0 3 H — > R-CN + NH 4 HS0 4

Stephens et al. (1955) have shown that ^>-toluenesulphonyl chloride in
pyridine readily dehydrates amides to cyanides (yield: 58-76 per cent).

High molecular weight acid amides are dehydrated to the corresponding
cyanide by heat alone (p. 205).

An acid amide may be converted into the cyanide by heating it with the amide-
boron trifluoride complex in the presence of a small amount of carboxylic acid
(Nieuwland, 1937), e i-<

CH 3 -CO-NH 2 -BF 3 + CH 3 -CO-NH 2 CH » C0 ' H > CH 3 -CN + CH 3 -CO a H + H 3 N-BF 3

Cyanides are prepared industrially by passing a mixture of carboxylic
acid and ammonia over alumina at 500 . This reaction probably occurs as
follows:

R-C0 2 H + NH 3 — > R-C0 2 NH 4 ^%-

H 2 + R-CONH 2 -^% H 2 + R-CN

2. By the dehydration of aldoximes with phosphorus pentoxide, or better,
with acetic anhydride:

R-CHIN-OH ' ch ' C0) - > r. CN (g .)

— H2O

3. The most convenient method is to heat an alkyl iodide with potassium
cyanide in aqueous ethanolic solution; a small amount of isocyanide is also
obtained:

RI + KCN — >■ R-CN + KI (g.)

This method is satisfactory only if R is a primary or secondary alkyl radical.
If it is a tertiary radical, very little cyanide is obtained, the tertiary iodide
being converted into the corresponding olefin (c/. p. 108). Tertiary alkyl
cyanides are best prepared by method 4 below.

Many cyanides, particularly the lower members, maybe readily prepared
by warming the potassium alkyl sulphate with potassium cyanide (some
j'socyanide is also obtained) :

RO-SCyOK + KCN — > R-CN + K 2 S0 4 (g.-v.g.)

The cyanide may be freed from s'socyanide by shaking with dilute hydro-
chloric acid, which hydrolyses the latter, but does not affect the former:

HC1

R-NC + 2H 2 -» R-NH 2 + H-C0 2 H

NITROGEN COMPOUNDS 295

Tosylates, i.e., ^>-toluenesulphonates (see p. 612), are also useful for cyanide
preparation:

^-CH 3 -C 6 H 4 S0 3 R + KCN — > R-CN + ^-CH 3 -C 6 H 4 S0 3 K

4. By the interaction of a Grignard reagent and cyanogen chloride:

RMgCl + C1CN — > R-CN + MgCl a

This is the best method of preparing tertiary alkyl cyanides.

Methyl cyanide may be formed by passing a mixture of acetylene and ammonia
over zirconia at 400-500 , the yield reaching as high as 99 per cent. (Amiel and
Nomine, 1947.)

General properties. The alkyl cyanides are stable neutral substances
with fairly pleasant smells, and are not as poisonous as hydrogen cyanide.
The lower members are liquids which are soluble in water (with which they
can form hydrogen bonds), but the solubility diminishes as the molecular
weight increases. They are all readily soluble in organic solvents.

Reactions. 1. The alkyl cyanides are hydrolysed by acids or alkalis to
the corresponding acid via the intermediate formation of an amide (see
also p. 204) :

R. C =N ^> [ R - C \ N "] — > R-CONH a -5£> R-CO a H + NH 3 (g.^.g.)

When a solution of alkyl cyanide in an alcohol is heated with concentrated
sulphuric acid or hydrochloric acid, the ester is obtained:

R-CN + R'OH + H a O — > R-CO a R' + NH 3

On the other hand, if dry hydrogen chloride is passed into the solution of
an alkyl cyanide in anhydrous alcohol, the imidic ester hydrochloride is
formed:

^NH-HC1
R-CN + R'OH + HC1 — > R-Cf

x OR'

When an alkyl cyanide is treated with dry hydrogen chloride, the imido-
chloride is formed :

^NH-HC1
R-C=N + 2HCI — > R-C< .

2. Alkyl cyanides combine with dry ammonia to form amidines:

^NH

R-CN + NH 3 — y R-C<

^NH a

3. Alkyl cyanides combine with acid anhydrides on heating to form
tertiary acid amides:

R-CN + (R-CO) 2 — ^ (R'CO) 3 N

4. Alkyl cyanides may be converted into aldehydes; e.g.,

R-CN BSA - Ha ) R-CHO

(ii) H.O

This, as we have seen (p. 148), is a general method for preparing aldehydes.

296 ORGANIC CHEMISTRY

5. Alkyl cyanides are reduced by nascent hydrogen (sodium and ethanol)
to give primary amines (the Mendius reaction, 1862) :

R-CN + 4 [H] Na/C,H '° H > R-CH 2 -NH 2

Some secondary amine, (R'CH 2 ) 2 NH, is also formed. Catalytic reduction of
an alkyl cyanide also produces a primary amine accompanied by the secondary
amine, but in this case the amount of the latter is greater than when the reduction
is carried out with nascent hydrogen. Von Braun and co-workers (1933)
suggested that the mechanism of the formation of secondary amine is :

R-C=N "' > R-CH:NH "' > R-CH 2 -NH 2

Ni

R-CH-.NH + R-CH 2 -NH 2 — > R-CH-NH-CH 2 -R — >

NH 2
NH 3 + R-CH:N-CH 2 -R — ^-> R-CH 2 -NH-CH 2 -R

Support for this mechanism is to be found in the fact that the formation of
secondary amine is prevented if the reduction is carried out with a Raney nickel
catalyst in the presence of ammonia.

Alternatively, the formation of secondary amine may be prevented by carry-
ing out the reduction in acetic anhydride solution, using the Adams' platinum
catalyst (p. 65). Under these conditions, the primary amine formed is
acetylated; this acetyl derivative cannot react with the imine, R'CHINH.
Reduction of alkyl cyanides with lithium aluminium hydride produces primary
amine only.

It should be noted that the above reactions 1 to 5 clearly indicate that
the alkyl group in alkyl cyanides is attached to the carbon atom of the CN
group.

6. Alkyl cyanides (except methyl cyanide) react with Grignard reagents
to form ketones:

R'
R-C-N + R'MgX — > R— C=N-MgX ^-> R-COR'

7. Alkyl cyanides undergo condensation reactions in the presence of
sodium, only the a-hydrogen atoms being involved. These condensations
are to be expected in view of the fact that the cyano-group is a very strong
negative group (p. 281). When the reaction is carried out in a solvent,
e.g., ether, two molecules of cyanide condense (cf. Thorpe's reaction, p. 471) :

NHCH 3

III
CH 3 -CH 2 -C=N + CH 3 -CH 2 -CN — + CH 3 -CH 2 -C— CH— CN

If, however, the reaction is carried out at 150 in the absence of a solvent,
three molecules condense to form a heterocyclic compound (a pyrimidine
derivative) :

c 2 h 5 -c— ch-ch 3 c 2 h 5 -c— c-ch 3

Hhcn ^-> 1 Unh,

I I

C 2 H B -C=N C 2 H 5 -C=N

On the other hand, alkyl cyanides with a-hydrogen atoms can condense
with esters if the reaction is carried out in ether solution in the presence of

NITROGEN COMPOUNDS 297

sodamide (Levine and Hauser, 1946) ; e.g., methyl cyanide condenses with
ethyl propionate to form propionylmethyl cyanide :

CH 3 -CN + NaNH 2 — > [ICH 2 -CN]Na + + NH 3

ch,-ch,.co,c,h. > CH3 . CH2 . C q. CH8 . cn + c 2 H 5 OH (40%)

ALKYL moCYANIDES

These are also known as isonitriles or carbylamines.

General methods of preparation. 1. By heating an alkyl iodide with
silver cyanide in aqueous ethanolic solution: a small amount of cyanide is
also formed:

RI + AgCN — y R-NC + Agl

2. By heating a mixture of a primary amine and chloroform with ethanolic
potassium hydroxide:

R-NH 2 + CHCI3 + 3KOH — > R-NC + 3KC1 + 3H a O

It has been argued that by analogy with this reaction, replacement of a
primary amine by ammonia would lead to the formation of hydrogen isocyanide :

H— NH 2 + CHC1 3 + 3KOH — > H— NC + KC1 + 3H a O

The mechanism of the isocyanide reaction is not certain. Robinson (1961)
has suggested that the reaction proceeds via the intermediate formation of di-
chloromethylene, which is produced from the chloroform in alkaline solution

(p. 117):

R-NH 2 CC1 2 > R-NHyCCl 2 : —> R-NH-CCL. ^^> R-N=CC1 ^> R-N=C

General properties. The alkyl isocyanides are poisonous, most unpleasant-
smelling liquids, with lower boiling points than the isomeric cyanides.
They are not very soluble in water, the nitrogen atom not having a lone
pair of electrons available for hydrogen bonding.

Reactions. 1. Alkyl isocyanides are hydrolysed to an amine and formic
acid by dilute acids, but are not hydrolysed by alkalis:

R-NC + 2H 2 -^> R-NH 2 + H-C0 2 H

2. Alkyl wocyanides are reduced to secondary amines either by nascent
hydrogen or by catalytic reduction:

R-NC + 4[H] — >■ R-NH-CH 3

Both reactions 1 and 2 clearly indicate that the alkyl group in alkyl
isocyanides is attached to the nitrogen atom of the CN group [cf. cyanides,
above) .

3. Alkyl isocyanides add on halogen to form alkyliminocarbonyl halides,
sulphur to form alkyl isothiocyanates, and are readily oxidised by mercuric
oxide to alkyl isocyanates :

R-NC4-X 2 — ^R-NCX 2
R-NC + S — ^ R-NCS
R-NC + HgO — > R-NCO + Hg

4. When alkyl j'socyanides are heated for a long time, they rearrange to
the cyanide :

R-NC — > R-CN

298 ORGANIC CHEMISTRY

Cyanogen (ethanedinitrile), C 2 N 2 , may be prepared by heating the cyanides of
mercury or silver:

Hg(CN 2 ) >. C 2 N 2 + Hg

It may also be prepared by adding sodium cyanide to copper sulphate solution
and heating. The cupric cyanide first formed is unstable, decomposing into
cuprous cyanide and cyanogen :

2CuS0 4 + 4 NaCN — > 2CuCN + C 2 N 2 + 2Na 2 S0 4

When heated with phosphorus pentoxide, oxamide is dehydrated to cyanogen :

CO-NH, Pj0( CN

CONHjj ~ 2H ' CN

Cyanogen is a poisonous, colourless gas, b.p. — 21 , very soluble in water. It
burns with a violet flame to give carbon dioxide and nitrogen :

-> 2CO a + N 2

Its aqueous solution slowly decomposes to give a brown precipitate of azulmic
acid, C 4 H B ON 5 , and the solution contains ammonium oxalate, hydrogen cyanide,
formic acid and urea. When cyanogen is heated at 400°, it polymerises to
paracyanogen, which, at 8oo°, regenerates cyanogen. The structure of cyanogen
is probably best represented as a resonance hybrid, I being the most important:

: n=c— c=n : -<— > : n=c=c=n : <-> : 'n=c=c=n :
(i) (ii) (in)

Cyanogen chloride, C1CN, may be prepared by passing chlorine into a cooled
solution of sodium cyanide :

NaCN + Cl 2 y C1CN + NaCl

It is a poisonous, colourless gas, b.p. 12°, which is very soluble in water, ethanol
and ether. It readily polymerises to cyanuric chloride, C1 3 C 3 N 3 . This is a
crystalline solid, m.p. 154 , which is decomposed by heating with water under
pressure at 125 to give cyanuric acid:

C1 3 C 3 N S + 3H a O > H 3 3 C 3 N 3 + 3HCI

Cyanogen bromide, BrCN, may be prepared by adding sodium cyanide to well-
cooled bromine water:

NaCN + Br 2 ^ BrCN + NaBr (73-85%)

It is a colourless, crystalline solid, m.p. 52°, which polymerises to cyanuric
bromide, Br 3 C 3 N 3 . Cyanogen bromide is a useful reagent for converting tertiary
amines into secondary amines (see p. 318).

Cyanic acid, HOCN. Urea, on dry distillation, gives cyanuric acid:

3 CO(NH 2 ) 2 — > H 3 OAN 3 + 3NH3

The yield is improved by heating urea with zinc chloride. Cyanuric acid
is a colourless, crystalline solid, not very soluble in water, and is strongly
acid, reacting as a mono-, di- and tribasic acid. It has a cyclic structure
(a triazine derivative) and is believed to be tautomeric {amido-imidol triad
system) :

O OH

H C

0=Cx ,0=0 " F ~~ HO— Cv x,C— OH

NH N

NITROGEN COMPOUNDS 299

On the other hand, X-ray analysis of cyanuric acid in the solid state indicates
that the acid is best represented as a resonance hybrid of the carbonyl
form (amido-form). Furthermore, absorption spectra studies in neutral
solution show that the same forms are present as in the solid state (Koltz
and Askounis, 1947).

When cyanuric acid is heated, it does not melt but decomposes into
cyanic acid vapour which, when condensed below 0°, gives a colourless
condensate:

H3O3C3N3— > 3 HOCN

Cyanic acid is a colourless, volatile, strongly acid liquid which, above o°,
readily polymerises to cyanuric acid and cyamelide, which is a white solid
isomeric with cyanuric acid. The structure of cyamelide is believed to be

HN=c/ \>=NH

I I

C

II
NH

Aqueous solutions of cyanic acid rapidly hydrolyse to give carbon dioxide
and ammonia:

HOCN + H 2 — > C0 2 + NH 3

Cyanic acid was believed to be a tautomeric mixture of cyanic acid, IV, and
tsocyanic acid, V (amido-imidol triad system) :

H— O— CseN ;=t 0=C=N— H

(IV) (V)

Infra-red studies, however, have indicated that cyanic acid is V; only
derivatives of V are known, e.g., alkyl isocyanates, R-NCO.

When a solution of potassium i'socyanate (usually called potassium
cyanate) and ammonium sulphate is evaporated to dryness, urea and
potassium sulphate are obtained, the urea being formed by the molecular
rearrangement of the intermediate product ammonium wocyanate:

NH 4 NCO — > CO(NH 2 ) 2

Cyanic acid reacts with alcohols to form urethans; excess of cyanic acid
converts urethans into allophanates, which are esters of the unknown acid,
allophanic acid, NH 2 -CONH-C0 2 H :

ROH + HNCO — > NH 2 -C0 2 R HNC ° > NH 2 -CONH-C0 2 R

According to Close et al. (1953), the formation of urethan is, if it is truly the
intermediate step, not the important one in the usual preparation of allophanates.
These authors suggest that allophanate formation occurs via a concerted attack
of two molecules of cyanic acid to form a chelate intermediate.

R— O /j^NH R— 0/ \NH

1 L — > ii

H>0fc>=0 H x X>=0

NH

300 ORGANIC CHEMISTRY

iV-substituted urethans are formed by the reaction between alkyl or aryl
tsocyanates and an alcohol. These are well-defined crystalline solids, and
so are used to characterise alcohols, phenyl and i-naphthyl isocyanates
being the wocyanates used for this purpose :

C 6 H 5 -NCO + ROH > C 6 H 6 -NH-C0 2 R

Phenyl tsocyanate is also used to characterise primary and secondary
amines, with which it forms substituted ureas (see p. 315).

Phenyl wocyanate is usually prepared by the action of carbonyl chloride
on aniline:

C,H B -NH 2 + C0C1 2 — > HC1 + C 2 H 5 -NH-C0C1 —V HC1 + C 6 H 6 -NCO

This is a general method for preparing wocyanates; another general method
is the Curtius reaction (p. 209).

Di-j'socyanates, e.g., hexamethylene di-isocyanate, OCN-(CH 2 ) 6 -NCO, are used
to prepare polyurethan plastics by reaction with di- and polyhydroxyalcohols,
e.g., tetramethylene glycol:

HO-(CH 2 ) 4 -OH + OCN-(CH 2 )„-NCO + HO-(CH 2 ) 4 -OH + . . . >

HO-(CH 2 ) 4 -0-CO-NH-(CH 2 ) e -NH-CO-0-(CH 2 ) 4 -0- . . .

Fulminic acid. HO-N=^C, may be prepared by decomposition of its salts
with acid. Mercury fulminate, [(CNO) 2 Hg] 2 H 2 0, which is used as a detonator,
may be prepared by dissolving mercury in nitric acid, and then adding ethanol.
Free fulminic acid is unstable. It is soluble in ether, the ethereal solution also
being unstable, the acid readily polymerising.

Cyanamide, NH 2 -CN, may be prepared by the action of ammonia on cyanogen
chloride :

C1CN + NH, — y NH 2 -CN + HC1

It may also be prepared by heating urea with thionyl chloride:

CO(NH 2 ) 2 + SOCl 2 — y NH 2 -CN + SO a + 2HCI

It is also readily prepared by the action of water and carbon dioxide on
calcium cyanamide (see below).

Cyanamide is a colourless, crystalline solid, m.p. 42°, very soluble in water,
ethanol and ether. It is converted into guanidine (q.v.) by ammonia, and
into thiourea (q.v.) by hydrogen sulphide. When cyanamide is melted it forms
the dimer dicyanodiamide (NH a ) 2 C!N'CN, and the trimer melamine, which is a
cyclic compound (and tautomeric) :

NH. NH

c X

n/ ^n HN/\ra

NH 2 <L i-NHj"'" HN=L yC=NH

N NH

Melamine is used for making melamine-formaldehyde plastics.

Cyanamide itself is a tautomeric compound (amidine system), and Raman
spectra investigations have shown that two forms are present in equilibrium
in the solid or fused state, or in solution :

H 2 N— - C^N ^±: HN=C=NH

Cyanamide forms salts, the most important of which is the calcium salt.
This is manufactured by heating calcium carbide mixed with 10 per cent, its
weight of calcium chloride in a stream of nitrogen at 800 :

CaC 2 + N 2 — > CaN'CN + C

NITROGEN COMPOUNDS 301

The calcium cyanamide-carbon mixture is used as a fertiliser; it is hydrolysed
in the soil to cyanamide, which is then hydrolysed to urea, which, in turn,
is converted into ammonium carbonate by bacteria in the soil:

CaN-CN + 2H 2 — > NH a »CN + Ca(OH) a

NH 2 -CN + H 2 > CO(NH a ) 2

CO(NH 2 ) a + 2H a O > (NH 4 ) 2 CO $

Calcium cyanamide is also used to prepare urea industrially.

NITROPARAFFINS

Nomenclature. The nitroparaffins are named as the nitro-derivative
of the corresponding paraffin, the positions of the nitro-groups being in-
dicated by numbers, e.g.,

CH 3 -N0 2 nitromethane

N0 2

,-CHh

CHa'CH-CHg 2-nitropropane

N0 2 N0 2

CH 3 -CH-CH 2 -CH'CH 2 -CH 3 2 : 4-dinitrohexane

The nitroparaffins, the structure of which is R*N0 2 , are isomeric with
the alkyl nitrites, RO-NO. The evidence that may be adduced for these
respective formulae is to be found in the study of the reactions of these two
groups of compounds. The reactions of the nitrites have already been
described (p. 196), and those of the nitroparaffins are described below. It
is, however, worth while at this stage to mention the reaction which most
clearly indicates their respective structures, viz., reduction. When alkyl
nitrites are reduced by nascent hydrogen, an alcohol and ammonia or
hydroxylamine are formed. This shows that the alkyl group in nitrites
is attached to an oxygen atom :

R— ONO — -> ROH + NH 3 + H 2

On the other hand, when nitroparaffins are reduced, a primary amine is
formed. This shows that the alkyl group is attached to the nitrogen atom,
since the structure of a primary amine it known to be R — NH 2 from its
method of preparation (see below) :

R.N0 2 -5U. R-NH 2 + 2H 2

^0
The structure of the nitro-compounds was originally written R — tf y ,

*0
but work on the dipole moments indicates that they are resonance hybrids:

R— Nf <-> R— IK or R— N< _ <-> R— Nl

XO ^O x O ^O

For most purposes, however, the original formula R — Nn or the non-
committal formula R*N0 2 , is satisfactory. ^0

302

ORGANIC CHEMISTRY

From the M.O. point of view, the nitro-group is conjugated (Fig. i), and
delocalisation of bonds increases its stability (the two oxygen atoms are
equivalent; cf. the carboxylate ion, p. 184).

■N

R — tf

Fig. 13. 1.

General methods of preparation. 1. By heating an alkyl halide with
silver nitrite in aqueous ethanolic solution :

RX + AgONO — > R-N0 2 + RONO + AgX

This method is only useful for the preparation of primary nitroparamns.
With sec.-halides the yield is about 15 per cent., and with tert.-halides 0-5
per cent. ; the yield of nitrite increases from primary to tertiary halides
(Kornblum et ah, 1954, 1955)- Kornblum et al. (1956) have now described
a simple new synthesis of primary and secondary nitro-compounds. These
authors have shown that, contrary to general opinion, sodium nitrite reacts
with alkyl halides to give good yields of nitroparafhn (55-62 per cent.) together
with alkyl nitrites (25-33 P er cent.). The success of the reaction depends on
the use of dimethylformamide as solvent, and the addition of urea increases
the solubility of sodium nitrite. Alkyl bromides and iodides are most
satisfactory; the chlorides react too slowly to be useful.

2. Nitromethane may be prepared by boiling an aqueous solution of sodium
nitrite with a halogeno-acetic acid, e.g.,

CHXl-C0 2 Na + NaN0 2 — -> NaCl + [CH 2 (N0 2 )-C0 2 H] — >

CH 3 -N0 2 + C0 2 (35-38%)

This method has no value for preparing higher nitroparaffms (Treibs et ah,

1954)-

The ready decarboxylation of the intermediate nitroacetic acid may be
explained on the assumption that the electron-attracting nitro-group facilitates
the loss of carbon dioxide (cf. p. 177) :

6§

>N— CH,

<

o

3. Until fairly recently, the preparation of nitro-compounds by direct
nitration was of academic interest only, but mainly from the work of
Hass and his co-workers direct nitration is now an important industrial
process. Two techniques have been evolved: liquid-phase nitration and
vapour-phase nitration.

In liquid-phase nitration, the hydrocarbon is heated with concentrated
nitric acid under pressure at 140°. Nitration under these conditions is
always slow, and a large amount. of polynitro-compounds is produced. It

NITROGEN COMPOUNDS 303

is interesting to note that a mixture of nitric and sulphuric acids is not
suitable for nitrating paraffins (c/. aromatic hydrocarbons, p. 553).

In vapour-phase nitration, the hydrocarbon is heated with nitric acid
(or with oxides of nitrogen) at 150-475 ° ; each hydrocarbon has its optimum
temperature, e.g.,

CH 3 -CH 2 -CH 3 ^%- CH 3 -CH 2 -CH 2 -N0 2 + CH 3 -CH(N0 2 )-CH 3 +

4 °°° CH 3 -CH 2 -N0 2 + CHa-NOj

Vapour-phase nitration is more satisfactory than liquid-phase nitration.

Hass and Shechter (1947) have formulated general rules of vapour-phase
nitration of paraffins (and cyctoparamns).

(i) Polynitro-compounds are formed only from paraffins of fairly high molecular
weight.

(ii) Any hydrogen atom in the hydrocarbon is capable of replacement by a
rdtro.-group, and the ease of replacement is tertiary hydrogen > secondary >
primary. As the temperature rises, however, the ease of replacement tends to
become equal.

(iii) Any alkyl group present in the paraffin can be replaced by a nitro-group,
i.e., chain fission takes place; e.g., isopentane yields nine nitroparaffins. The
fission reaction increases as the temperature rises.

(iv) Oxidation always accompanies nitration, resulting in the formation of
nitro-compounds and a mixture of acids, aldehydes, ketones, alcohols, nitrites,
nitroso-compounds, nitro-olefins, polymers, carbon monoxide and carbon dioxide.
Catalysts such as copper, iron, platinum oxide, etc., accelerate oxidation rather
than nitration.

4. Another recent method for preparing nitro-compounds is the hydrolysis
of a-nitro-olefins with water, acid or alkali; e.g., 2-methyl-i-nitroprop-i-
ene gives acetone and nitromethane in almost quantitative yield (Levy and
Scaife, 1947):

(CH 3 ) 2 C:CH-N0 2 + H 2 0— =► (CH 3 ) 2 CO + CH 3 -N0 2

5. Kornblum et al. (1956) have introduced a practical synthesis of tert.-
nitro-compounds by the oxidation of tertf.-carbinamines with potassium
permanganate:

R 3 ONH 2 KMn °'> R 3 ON0 2 (70-80%)

General properties. The nitroparaffins are colourless (when pure),
pleasant-smelling liquids which are sparingly soluble in water. Most of
them can be distilled at atmospheric pressure.

The nitroparaffins in commercial use at the moment are the lowest four
members of the series: nitromethane, nitroethane and 1- and 2-nitro-
propanes. These are prepared by the vapour-phase nitration of propane,
and are used as solvents for oils, fats, cellulose esters, resins and dyes.

Reactions. 1. Nitroparaffins are reduced in acid solution to primary
amines:

R-N0 2 + 6[H] metel/acid > R-NH 2 + 2 H 2

Catalytic reduction also produces a primary amine, the yield with a Raney
nickel catalyst being 90-100 per cent. When the reduction is carried out
in neutral solution, e.g., with zinc dust and ammonium chloride solution,
nitro-compounds are converted into hydroxylamine derivatives:

R-N0 2 + 4 [H] Zn/NH ' C '> R-NH-OH + H 2

304 ORGANIC CHEMISTRY

When stannous chloride and hydrochloric acid are used as the reducing
agent, nitro-compounds are converted into a mixture of hydroxylamine
derivative and oxime:

R-CH 2 -N0 2 ^^>R-CH 2 -NH-OH + R-CHIN-OH

Since hydroxylamine can give rise to two types of derivatives, e.g., R-NH-OH
and NH 2 *OR, it is necessary to distinguish one from the other. A common
method is to name the former as the JV-alkylhydroxylamine, and the latter
as the O-alkylhydroxylamine, the capital letters N and 0, respectively,
indicating where the alkyl group is attached in the molecule.

2. Primary nitro-compounds are hydrolysed by boiling hydrochloric
acid or by 85 per cent, sulphuric acid to a carboxylic acid and hydroxyl-
amine:

R-CH 2 -N0 2 + H 2 ^> R-C0 2 H + NH 2 -OH

This reaction is now used for manufacturing hydroxylamine, and may
become a commercial source of propionic acid (using i-nitropropane).

Secondary nitro-compounds are hydrolysed by boiling hydrochloric acid
to ketones and nitrous oxide:

2R 2 CH-N0 2 -^> 2R 2 C0 + N 2 + H 2

Tertiary nitro-compounds are generally unaffected by hydrochloric acid.

3. Primary and secondary nitro-compounds, i.e., those containing a-
hydrogen atoms, are acidic in character due to tautomerism (the nitro-group
is a very strong negative group and contains multiple bonds) :

aQ /OH

R-CH 2 -N< ^=± R-CH=N<

(I) (U)

The nitro-form I is often called the pseudo-acid form; II is known as the
aa'-form or nitronic acid. This is an example of a triad system, the nitro-
acinitro system; the equilibrium is almost completely on the left, and this
may be due to the nitro-form being stabilised by resonance (of the nitro-
group).

These nitronic acids do not dissolve in aqueous sodium carbonate, but
do in aqueous sodium hydroxide, which disturbs the nitro-acinitro equili-
brium by removing the latter form. Thus nitro-compounds behave as acids
in the presence of strong alkalis, but not in their absence ; hence they are said
to be pseudo-acids :

R 2 CH-N0 2 ^= R 2 CIN0 2 H -^> R 2 C:N0 2 Na

These sodium compounds are true salts, i.e., they exist as ions : [R 2 C!N0 2 ]~Na + .
When the sodium salt is acidified at low temperature, there is not always
an immediate separation of oily drops (of the nitro-form). On standing,
however, the acidified solution slowly deposits oily drops due to the slow
tautomeric change of the nitronic acid into the nitro-compound. Further-
more, when the sodium salt is carefully neutralised with hydrochloric acid,
the resulting solution has a conductivity which is greater than that calculated
for the sodium chloride content. Hence, in addition to the sodium and
chloride ions, there must be present other ions which must be those from
the nitronic acid. Thus the following changes probably take place:

[R 2 GN0 2 ]-Na + + HC1 — > Na+Cl- + [R 2 CN0 2 ]"H + — >

R 2 CIN0 2 H =^= R 2 CH-N0 2

NITROGEN COMPOUNDS 305

This is evidence for the existence of the nitronic acid ; actually Hantzsch and
Schultze (1896) isolated both forms of phenylnitromethane, QHg'CHa'NOg,
thus confirming the existence of the nitro-acinitro tautomeric system.

When the sodium salt solution of the nitronic acid is acidified with 50 per
cent, sulphuric acid at room temperature, an aldehyde (from a primary
nitro-compound) and a ketone (from a secondary nitro-compound) is
obtained, e.g., (R' is either an alkyl radical or a hydrogen atom) :

2RR'C:NO a Na+2H 2 S0 4 — > 2RR'CO+N 2 0+2NaHS0 4 +H a O (85%)

When treated with stannous chloride and hydrochloric acid, the sodium salt
of the nitronic acid is reduced to the aldoxime or ketoxime:

RR'GN0 2 Na Sna,/HC1 > RR'GN-OH

These oximes are readily converted into the parent carbonyl compound
by steam distillation or by direct hydrolysis with acid.

4. Primary and secondary nitro-compounds are readily halogenated in
alkaline solution in the a-position only:

R 2 CN0 2 Na + Br 2 -^> R 2 CBr-N0 2 + NaBr

Primary nitro-compounds can form the mono- and dibromo-derivatives,
whereas secondary form only the monobromo-derivative : nitromethane
is exceptional in that it can form the tribromo-derivative. Chloropicrin
(p. 117) is manufactured by the reaction between nitromethane, chlorine
and sodium hydroxide.

When liquid or gaseous nitro-compounds are treated with halogen in the
absence of alkali, indiscriminate substitution takes place resulting in the
formation of the a-, |3-, y-, halogeno-compounds, e.g.,

CH 3 -CH 2 -N0 2 -?±+ CH 3 -CHC1-N0 2 + CH 2 C1-CH 2 -N0 2

5. Nitro-compounds react with nitrous acid, the product formed depending
on the nature of the alkyl group.

Primary nitro-compounds form nitrolic acids; these are crystalline
substances which dissolve in sodium hydroxide to give red solutions:

K

H

+
H

=NOH ^NOH

N0 2

R— C< + H 2

\N0 2

Secondary nitro-compounds form pseudonitroles (ifi-nitroles) ; these are
colourless crystalline substances which dissolve in sodium hydroxide to
give blue solutions (the blue colour is probably due to the presence of the
nitroso-group ; see below) :

/ H + NO —NO /NO

R 2 C( — > R 2 C( + H 2

NN0 2

"Nno,

Tertiary nitro-compounds do not react with nitrous acid since they have no
a-hydrogen atom.

These reactions with nitrous acid are the basis of the " red, white and
blue " test for the nature of monohydric alcohols (Victor Meyer et al., 1874).
The alcohol under investigation is treated as follows:

ROH-^>RBr^^R-N0 2

306 ORGANIC CHEMISTRY

The nitrocompound is now treated with an alkaline solution of sodium
nitrite, acidified with hydrochloric acid, and finally made alkaline with
sodium hydroxide. The development of a red colour indicates a primary
alcohol, blue colour a secondary, and the solution remaining colourless
(" white "), a tertiary alcohol. This test is now only historically important.
6. Owing to the presence of active a-hydrogen atoms, primary and
secondary nitro-compounds undergo condensation with aldehydes; e.g., (i)
nitro-methane condenses with benzaldehyde in the presence of ethanolic
potassium hydroxide to form a-nitrostryrene:

C 6 H s -CHO + CH 3 -N0 2 -^> C 6 H b -CH:CH-N0 2 + H 2

(ii) Nitroethane condenses with formaldehyde in the presence of aqueous
potassium hydrogen carbonate to form the bishydroxymethyl compound,
2-methyl-2-nitropropane-i : 3-diol (cf. p. 165).

N0 2
khco I /CH 2 OH

CH 3 -CH 2 -N0 2 + 2HCHO '-> CH 3 -C(

i - \CH 2 OH

These condensations are similar to the aldol condensations, the active
methylene or methyne group being produced by the adjacent nitro-group.

The sodium salts of the nitronic acids condense with a(3-unsaturated
ketones (p. 279) ; e.g., with mesityl oxide nitromethane forms 4 : 4-dimethyl-
5-nitropentan-2-one :

(CH 3 ) 2 C:CH-COCH 3 + CH 2 :N0 2 Na — >

(CH 3 ) 2 C— CH 2 -COCH 3 > (CH 3 ) 2 C— CH 2 -COCH 3

CH:N0 2 Na CH 2 -N0 2

Primary and secondary nitro-compounds also undergo the Mannich
reaction (1917). This is the condensation between formaldehyde, ammonia
or a primary or secondary amine (preferably as the hydrochloride), and a
compound containing at least one active hydrogen atom. In this reaction,
the active hydrogen atom is replaced by an aminomethyl group or substituted
aminomethyl group:

N0 2

R 2 CH-N0 2 + H-CHO + NH 4 C1 — > R 2 C— CH 2 -NH 2 -HC1 + H 2

The Mannich reaction offers a means of preparing a large variety of com-
pounds, e.g., nitro-amines, diamines, etc.

NITROSO-PARAFFINS

The nitroso-paraffins contain a nitroso-group, — N=0, directly attached to a
carbon atom. They are named as the nitroso-derivatives of the corresponding
paraffin, e.g.,

(CH 3 ) 3 C # NO 2-methyl-2-nitrosopropane

General methods of preparation. 1. By the addition of " nitrous fumes " to
olefins. The compound formed depends on the nature of the olefin; usually a
mixture is obtained, e.g., Michael and Carlson (1940) treated isobutene with

NITROGEN COMPOUNDS 307

dinitrogen tetroxide under various conditions, and believed that the following
compounds were formed (see also p. 71) :

(CH 3 )X-

O-NO, NO

r(CH 3 ;

-CH,

O-NO NO

nitrosate nitrosite

(CH 3 ) 2 C CH 2 (CH 3 ) 2 CCH-N0 2

NO, NO a

dinitro-compound nitro-compound

2. By the addition of nitrosyl chloride or bromide to olefins, whereby olefin
nitrosohalides are formed (p. 71 ) :

CHj CH 2 2 molecules TCH, CH,

CH 2 =CH 2 + NOC1 — > \ I y

NO CI

|-CH 2 -CH 2 -|
Lno CI J s

Thorne (1956) has shown that, in general, structures which increase the
availability of electrons at the double bond favour the formation of nitroso-
chlorides and nitrosates. Thus it was found that a carboxyl group fairly close
to the double bond prevents formation of adducts, e.g., 4-phenylbut-3-enoic and
5-phenylpent-4-enoic acid do not form nitrosochlorides or nitrosates; oleic acid,
however, forms a nitrosochloride. Furthermore, the position of a double bond
in an olefin determines whether an adduct is formed, e.g., hex-i-ene does not
form a nitrosochloride whereas hex-2-ene does.

3. By the action of nitrous acid on certain types of compounds, e.g., secondary
nitro-parafnns (see p. 305).

4. By the oxidation of primary amines containing a tertiary alkyl group with
e.g., Caro's acid (peroxy(mono)sulphuric acid) :

R 3 C-NH 2 + 2[0] H ' S0 "> R 3 C-NO + H 2

On the other hand, Emmons (1957) has prepared primary, secondary, and
tertiary nitroso compounds by oxidation of amines with neutralised peracetic
acid in methylene dichloride (yield: 33-80 per cent.).

5. Gowenlock et al. (1956) have prepared nitrosoparaffins (C t — C 4 ) by the
pyrolysis of alkyl nitrites, and have proposed the following mechanism (e.g., for
ethyl nitrite) :

CH 3 -CH 2 -0-NO > NO + CH 3 -CH 2 -0- >

NO + CH 3 - + CH 2 >■ CH 3 -NO + CH 2

General properties. Nitroso-compounds, which are usually blue or green
liquids, tend to associate to give colourless solids (smelling like camphor) which
are dimers. These bimolecular solids regenerate the monomer when fused or
when dissolved in solution. Chilton et al. (1955) and Gowenlock et al. (1955) have
examined the absorption spectrum of the two solid forms of nitrosomethane,
and have concluded that they are geometrical isomers (N.B., a double-bond is

CH 3 \ /CH 3 CH 3 \ *0

;n=n( >n=n(

O* X> O^ \CH 3

cis trans

necessary for this molecule to exhibit geometrical isomerism) .

Nitroso-compounds exist as such only when the nitroso-group is attached to a
carbon atom not joined to hydrogen, i.e., to a tertiary carbon atom. If the
nitroso-group is attached to a primary or secondary carbon atom, the nitroso-
compound is generally unstable, tending to rearrange to the oxime :

r 2 ch— no — > r 2 c:n-oh

blue and colourless and

unstable stable

308 ORGANIC CHEMISTRY

This is known as the nitroso-oximino triad system. This system is potentially
tautomeric, but no example is known of the isomerisation of the oxime directly
to the nitroso-compound. On the other hand, when an aldoxime is treated with
bromine, the bromonitroso compound is formed; this, on fusion, is converted
into the isomeric bromo-oxime :

R-CH:N-OH Br ' > R-Cr-NO fU5C > R-CBr:N-OH
\Br

It has also been found that the bimolecular nitroso-chlorides rearrange, if possible,
to the oxime when treated with sodium hydroxide :

"CH a Cl

NaOH CH 2 C1

* 2 1

ch:n-oh

The so-called isonitroso-compounds (oximino-compounds) are actually the
oximes, e.g., isonitrosoacetone, CH S *COCHIN'OH, formed by the action of
nitrous acid oh acetone, is really the half oxime of methylglyoxal (q.v.).

Nitroso-compounds may be oxidised to the nitro-compound by nitric acid, and
reduced to the primary amine by, e.g., tin and hydrochloric acid:

R 3 ONO a < HN °' R3ONO i^> R 3 ONH 2

MONOAMINES

Amines are derivatives of ammonia in which one or more hydrogen
atoms have been replaced by alkyl groups. The amines are classified as
primary, secondary or tertiary amines according as one, two or three
hydrogen atoms in the ammonia molecule have been replaced by alkyl
groups. Thus the general formulae of primary, secondary and tertiary
amines may be written R*NH 2 , R a NH and R 3 N, and each is characterised
by the presence of the amino-group — NH 2 , the imino-group ^NH, and the
tertiary nitrogen atom ^N — , respectively.

In addition to the amines the tetra-alkyl derivatives of ammonium
hydroxide, [R 4 N] + OH~, are known; these are called the quaternary am-
monium hydroxides.

Nomenclature. The suffix of the series is amine, and each member is
named according to the alkyl groups attached to the nitrogen atom, e.g.,

CH 3 'NH 2 methylamine

(CH 3 ) 2 NH dimethylamine

(CH S ) 3 N trimethylamine

(C 2 H 5 )2N'CH(CH3) g diethyh'sopropylamine

The amines are said to be " simple " when all the alkyl groups are the same,
and " mixed " when the alkyl groups are different.

The method of naming a quaternary ammonium hydroxide is illustrated
by the following examples:

(CH 3 ) 4 NOH tetramethylammonium hydroxide

(CH 3 ) 3 N'CgH 5 }OH ethyltrimethylammonium hydroxide

General methods of preparation. Many methods are available for pre-
paring amines, but it is instructive to consider them in the following groups:

Methods for the three classes of amines. 1. Hofmann's method (1850).
This is by means of ammonolysis, an alkyl halide and an ethanolic solution
of ammonia being heated in a sealed tube at ioo°. A mixture of all three

NITROGEN COMPOUNDS 309

classes of amines is obtained, together with some quaternary ammonium
compound (see p. 314 for the structure of amine salts) :

RX + NH 3 — y R-NH 2 + HX — > R-NH 2 -HX
R-NHjj-HX + NH 3 ^=±: R-NH 2 + NH 4 X
R-NH 2 + RX — > R a NH + HX — > R 2 NH-HX
R 2 NH + RX — > R 3 N + HX — > R 3 N-HX

R 3 N + RX — >■ [R 4 N] + X-

In many cases a good yield of primary amine may be obtained by using
a large excess of ammonia, and a good yield of tertiary amine by using alkyl
halide in slight excess required by the equation:

3RX + NH 3 — > R S N + 3HX

Carrying out the reaction in liquid ammonia (in excess) gives a better yield
of primary amine than by using aqueous or ethanolic ammonia; e.g., Watt
and Otto (1947) found that ammonolysis of ethyl iodide by liquid ammonia
in excess at o° was complete in less than 15 minutes, and gave 46 per cent,
ethylamine, 31 per cent, diethylamine and 17 per cent, triethylamine.

The order of reactivity of the alkyl halides in ammonolysis is alkyl
iodide >bromide>chloride, and the method is limited to primary alkyl
halides and the secondary halide, isopropyl bromide. All other secondary
halides and tertiary halides eliminate a molecule of halogen acid to form
the olefin and no amine when heated with ethanolic ammonia. Amines of
the type R 3 C-NH 2 are best prepared by means of a Grignard reagent (p. 357).

2. A mixture of the three types of amines may be prepared by the am-
monolysis of alcohols; the alcohol and ammonia are heated under pressure
in the presence of a catalyst, e.g., copper chromite or alumina:

ROH + NH 3 — > R-NH 2 + H 2
R-NH 2 + ROH — >■ R 2 NH + H 2
R 2 NH + ROH — > R 3 N + H 2

The primary amine may be obtained as the main product by using a large
excess of ammonia (cf. method 1).

Separation of amine mixtures. When the mixture contains the three
amine salts and the quaternary salt, it is distilled with potassium hydroxide
solution. The three amines distil, leaving the quaternary salt unchanged
in solution, e.g.,

R 2 NH-HX + KOH — > R 2 NH + KX + H 2

The distillate of mixed amines may now be separated into the individual
amines as follows :

Fractional distillation. This is the most satisfactory method and is now
used industrially, its success being due to the high efficiency of industrial
fractionation apparatus.

Hinsberg's method (1890). The mixture of amines is treated with an
aromatic sulphonyl chloride. Benzenesulphonyl chloride, C e H 5 -S0 2 Cl,
was used originally, but has now been replaced by ^-toluenesulphonyl
chloride, CH 3 -C 6 H 4 *S0 2 C1. After treatment with this acid chloride, the
solution is made alkaline with potassium hydroxide. Primary amines form
the alkyl sulphonamide, e.g., ethylamine forms ethyl ^-toluenesulphonamide:

CH 3 -C 6 H 4 -SO a Cl + C 2 H 5 -NH 2 — > CH 3 -C 6 H 4 -S0 8 -NH-C 8 H 5 + HC1

310 ORGANIC CHEMISTRY

These sulphonamides are soluble in potassium hydroxide solution, possibly
due to the formation of the soluble potassium salt of the enol form of the
sulphonamide :

OH OH 0K +

CH 3 -C 6 H 4 -S— N— C 2 H 5 ^ CH 3 -C e H 4 -S=N-C 2 H 5 — t >CH 3 -C 6 H 4 -S=N-C 2 H 5

Secondary amines form dialkyl sulphonamides, which are insoluble in
potassium hydroxide solution because these sulphonamides are incapable
of tautomerism (there is no a-hydrogen atom adjacent to the negative
group); e.g., diethylamine forms diethyl />-toluenesulphonamide :

CH 3 -C 6 H 4 -S0 2 C1 + (C 2 H 6 ) 2 NH — -> CH 3 -C 8 H 4 -S0 2 -N(C 2 H 6 ) 2 + HC1

Tertiary amines do not react with ^>-toluenesulphonyl chloride.

The alkaline solution is distilled, and the tertiary amine thereupon distils
off. The residual liquid is filtered: the filtrate contains the primary amine
derivative which, on acidification, gives the alkyl sulphonamide ; the residual
solid is the dialkyl sulphonamide. The amines are regenerated from the
sulphonamides by refluxing with 70 per cent, sulphuric acid or 25 per cent,
hydrochloric acid.

Preparation of primary amines. 1. By the reduction of nitro-compounds
with metal and acid, or with hydrogen and a nickel catalyst :

R-N0 2 + 3H 2 — -> R-NH 2 + 2 H 2 (ex.)

This method is becoming increasingly important since more aliphatic nitro-
compounds are becoming available and is particularly useful for preparing
amino-alcohols from the starting materials obtained by the condensation
between formaldehyde and nitro-compounds (see p. 306).

2. (i) By the reduction of alkyl cyanides :

R-CN + 4 [H] Na/QH '° H > R-CH 2 -NH 2

This method is particularly useful for the preparation of high molecular
weight amines, since the cyanides are readily prepared from the long-chain
fatty acids.

(ii) By the reduction of oximes with sodium and ethanol, or catalytically :

R-CH:N-OH + 4[H] — > R-CH 2 -NH 2 + H 2 (g.)
R 2 C:N-OH + 4 [H] — > R 2 CH-NH 2 + H 2 (g.)

(iii) By the reduction of amides with sodium and ethanol, or catalytically :

R-CO-NH, + 4 [H] — ^ R-CH 2 -NH 2 + H 2 (f.g.-g.)

In all cases there are also obtained varying amounts of secondary amine as
by-products, which increase in quantity when the reduction is carried out
catalytically (see alkyl cyanides, p. 296). Reeve et al. (1956), however, have
found that Raney cobalt (and Raney nickel) generally gives a primary
amine in high purity by the reduction of oximes. On the other hand, nitro-
compounds, cyanides, oximes and amides may be reduced by lithium alu-
minium hydride to primary amines unaccompanied by secondary amines.

3. By passing a mixture of aldehyde or ketone and a large excess of
ammonia and hydrogen under pressure (20-150 atm.) over Raney nickel

NITROGEN COMPOUNDS 311

at 40-150°. The reaction may also be carried out at 3 atmospheres in the
presence of excess ammonium chloride and Adams' platinum catalyst :

R-CO-R' + NH 3 + H a — > RR'CH-NH 2 + H 2

Small amounts of secondary and tertiary amines are obtained as by-products.
This process of introducing alkyl groups into ammonia or primary or second-
ary amines by means of an aldehyde or ketone in the presence of a reducing
agent is known as reductive alkylation.

4. By the Leuckart reaction (p. 156) :

R-CO-R' + 2H-C0 2 NH 4 — >

RR'CH-NH-CHO + 2H 2 + C0 2 + NH 3 -^%- RR'CH-NH 2

5. By the hydrolysis of alkyl j'socyanates with boiling alkali (Wurtz,
1849):

R-NCO + 2KOH — > R-NH 2 + K 2 C0 3

This method is not very useful in practice owing to the inaccessibility of the
isocyanates. It is, however, historically important because Wurtz dis-
covered amines by this reaction. He at first thought he had obtained
ammonia, but subsequently found the gas was inflammable.

6. By Hofmann's degradation method (p. 206) :

R-CO-NH 2 + Br 2 + 4KOH > R-NH 2 + 2KBr + K 2 C0 3 + 2H 2

This is generally the most convenient method of preparing primary amines.

7. By the Curtius reaction (p. 209) :

R .CO-N 3 CH -° H > R-NCO -^% R-NH-C0 2 CH 3 -^> R«NH 2

This method of preparing primary amines works well for all members of the
series.

8. By the Lossen rearrangement (p. 207) :

R-CO-NH-OH -^-^ H a O + R-NCO — -> R-NH 2
This has very little importance as a practical method.

9. The Schmidt reaction (1923). This reaction is carried out by 'shaking
a solution of a carboxylic acid in concentrated sulphuric acid with a chloro-
form solution of hydrazoic acid (cf. p. 178).

R-C0 2 H + HN 3 H ' SO * > R-NH 2 + C0 2 (v.g.)

This method generally gives better yields of primary amines than the
Hofmann or Curtius reaction, but it is somewhat dangerous because of the
explosive and poisonous character of the hydrazoic acid.

10. Gabriel's phthalimide synthesis (1887). In this method phthalimide
is converted, by means of ethanolic potassium hydroxide, into its salt
potassiophthalimide, which, on heating with an alkyl halide, gives the
2V-alkylphthalimide. This is then hydrolysed to phthalic acid and a primary
amine by heating with 20 per cent, hydrochloric acid under pressure, or by
remixing with potassium hydroxide solution:

(Qco£ + R ' NH -

3 I 2 ORGANIC CHEMISTRY

When hydrolysis is difficult, the alkylphthalimide can be treated with
hydrazine to give the amine (Ing, 1926) :

\^ U H 2 N ^CO-NH

Gabriel's synthesis is a very useful method since it gives a pure primary amine.

1 1 . Decarboxylation of amino-acids. This is carried out by distilling the amino-
acid with barium hydroxide; e.g., glycine gives methylamine :

CH 2 (NH 2 )-C0 2 H Ba '° H> ' -> CH 3 -NH 2 + C0 2

Decarboxylation of amino-acids may also be effected by bacteria, especially
putrefying bacteria (many of which occur in the intestines).

12 . By means of a Grignard reagent and chloramine. The product is pure :

RMgX + C1NH 2 — > R-NH 2 + MgXCl

Alternatively, a primary amine may be prepared by interaction of a Grignard
reagent and O-methylhydroxylamine (see p. 357).

13. A good method for preparing primary amines containing a tert.-alkyl
group is to add a te^.-alcohol (or an alkene) to acetic acid in which is dis-
solved sodium cyanide, and then add sulphuric acid (Ritter et ah, 1948) :

R 3 COH H ' SO ' > R3OOSO3H — -> R 3 ON=CH-OS0 3 H

-^> H 2 S0 4 + R 3 ONH-CHO — ^*— > R 3 ONH 2 (40%)

The iV-alkylformamide produced is hydrolysed with sodium hydroxide.

Preparation of secondary amines. 1. By the reduction of an alkyl iso-
cyanide :

R-NC + 4[H] — > R-NH-CH 3

The amine produced always contains a methyl group as one radical; the
method is of academic interest only.

2. By heating a primary amine with the calculated quantity of alkyl halide :

R-NH 2 + R'X — y R-NH-R' + HX

3. Aniline is heated with an allkyl halide, and the product, dialkylaniline,
is treated with nitrous acid; the ^-nitroso-dialkylaniline so formed is then
boiled with sodium hydroxide solution. A pure secondary amine and
^>-nitrosophenol are produced :

NH 2 NR 2 NR a OH

/% RX /\ HNO, /\ Na0H /%

II I — HI I H I H +R 2 NH

NO NO

This is one of the best methods of preparing pure secondary amines.

4. By the catalytic reduction of a Schiff 's base, which is formed by inter-
action of a primary amine and an aldehyde (Henze and Humphreys, 1942).

R-NH 2 + R'-CHO — > H 8 + R-NICH-R' RaneyM > R-NH-CH.-R'

Hi

Secondary amines prepared this way usually contain an aromatic group
as one of the radicals (R is generally the phenyl radical, C„H 5 -).
5. By the hydrolysis of a dialkyl cyanamide with acid or alkali:

CaN-CN Na ° H > Na 2 N-CN — -> R 2 N-CN -^> R 2 NH + CO a + NH 3

NITROGEN COMPOUNDS 313

Preparation of tertiary amines, i. The best method of preparation is
to heat an ethanolic solution of ammonia with alkyl halide which is used
in slight excess required by the equation:

3RX + NH 8 — > R 3 N + 2HX

When the reaction mixture is made alkaline and distilled, the tertiary
amine is obtained. The residual quaternary compound may be distilled
in vacuo to yield more tertiary amine (see quaternary compounds, p. 318).

2. By the reduction of a carbonyl compound (in excess) in the presence
of hydrogen, ammonia or a primary or secondary amine, with Raney nickel
as catalyst (cf. primary amines, method 3), e.g.,

R 2 NH + R'-CHO + H 2 -^-> R 2 N-CH 2 -R' + H 2

Preparation of quaternary compounds. There is only one satisfactory
method of preparing quaternary compounds, viz., by heating ammonia
with a very large excess of alkyl halide; a primary, secondary or tertiary
amine may be used instead of ammonia. The starting materials will
depend on the nature of the desired quaternary compound, e.g., ethyltri-
methylammonium iodide may be prepared by heating trimethylamine with
ethyl iodide :

(CH 3 ) S N + C 2 H 5 I — > [(CH 3 ) 3 N-C 2 H 5 ] + I-

General properties of the amines. The lower members are gases, soluble
in water (with which they can form hydrogen bonds). These are followed
by members which are liquids, and finally by members which are solids.
The solubility in water decreases as their molecular weight increases. All
the volatile members have a powerful fishy smell, and are combustible.

The reactions of an amine depend very largely on the class of that amine.

Reactions given by all three classes of amines. 1. All the amines are basic,
and they are stronger bases than ammonia.

RNH 2 + H 2 ^ RNH 3 + + OH~

One factor that controls the basicity of amines will be the availability of the
lone pair for protonation. At first sight, it would appear that as the number
of alkyl groups increases, then because of their + I effect, the lone pair will
become more available, thereby increasing the basicity of the amine. This,
however, is not the case, as can be seen from the following order of basicity:

NH 3 < Me 3 N < MeNH 2 < Me 2 NH

The reason for this order is not clear. It has been suggested that a steric
factor operates. Addition of the proton increases crowding and so sets up
strain, which will be greatest in the tertiary amine, and consequently the
stability of this molecule is decreased, i.e., its basicity is reduced. How-
ever, when we consider that the three bonding pairs and the lone pair all
occupy sp a orbitals, and that a lone pair causes crowding of bonding pairs
(p. 29), it would seem that protonation should relieve crowding by trans-
forming the lone pair into a bonding pair.

All the 'onium ions carry a positive charge and so are solvated. The
greater the ion, the less will be the solvation and so the less stabilised is the
ion. It seems possible that in going from Me 2 NH to Me 3 N, although the
availability of the lone pair is increased, the ion has now grown sufficiently
in size at this stage to become less solvated and so the stabilisation lost due
to this decrease in solvation is greater than the inductive effect making the
lone pair available for protonation.

314 ORGANIC CHEMISTRY

2. All the amines combine with acids to form salts, e.g., methylamine
combines with hydrochloric acid to form methylammonium chloride:

CH 3 -NH 2 + HC1 — > [CH 3 -NH 3 ] + C1-

The nitrogen is quadricovalent unielectrovalent in amine salts; but to
show their relationship to the amines, their salts are often written as, e.g.,
CH 3 -NH 2 -HC1, methylamine hydrochloride; [(C 2 H s ) 2 NH] 2 'H 2 S0 4 , diethyl-
amine sulphate.

Amine salts of certain complex acids, particularly chloroplatinic acid,
are used for the determination of the molecular weights of amines (see

P- 5).

3. All the amines combine with alkyl hahdes to form alkyl-substituted
ammonium hahdes with more alkyl groups than the amine used.

4. When an amine salt is heated at high temperature, a molecule of alkyl
hahde is eliminated (i.e., reverse of reaction 3, above) :

HC1 HC1

R 3 N-HC1 — > RC1 + R 2 NH > RC1 + R-NH 2 > RC1 + NH 4 C1

If the tertiary amine is a mixed amine, it is the smallest alkyl group that
is eliminated first :

(C 2 H 6 ) 2 N-CH 3 -^-> (C 2 H 5 ) 2 NH + CH 3 C1

This is the basis of estimating the methylimino-group, ^N-CH 3 , in natural
substances such as alkaloids.

When the methyl ether of a high boiling hydroxy compound is heated
with concentrated hydriodic acid at 100°, methyl iodide is formed (cf.
p. 142) :

ArOCH 3 + HI -^->- ArOH + CH 3 I

This is the Zeisel method for the quantitative estimation of methoxyl groups

(P- 637)-

When the temperature is raised to 150°, iV-methyl groups are converted
into methyl iodide:

Ar-NH-CH 3 + HI -^> Ar-NH 2 + CH 3 I

This is the Herzig-Meyer method for the quantitative estimation of methyl-
imino-groups. Thus methoxyl and methyhmino-groups can be estimated
separately when both are present in the same compound.

Reactions given by primary and secondary amines. 1. Primary and
secondary amines react with acid chlorides or anhydrides to form the iV-alkyl
acid amide, e.g.,

R-NH 2 + (CH 3 -CO) 2 — > CH 3 -CO-NH-R + CH 3 -CO a H

These monoacetyl derivatives are easily prepared and, since they are usually
well-defined crystalline solids, are used to characterise primary amines.

It is difficult to prepare the diacetyl derivative ; excess of the acetylating
agent and high temperature, however, often yields the diacetyl derivative :

R-NH-COCH 3 + (CH 3 -CO) 2 ► R-N(CO-CH 3 ) 2 + CH 3 -C0 2 H

Secondary amines, obviously, can form only the monoacyl derivative:

R 2 NH + CH 3 -COCl ► CH 3 -CONR 2 + HC1

The acetylated amines are neutral substances which do not form salts with
inorganic acids. It is the presence of the acetyl group, a negative group,
which decreases the basic strength of the amine (see p. 569).

NITROGEN COMPOUNDS 315

Sulphonyl chlorides also react with primary and secondary amines to
form iV-alkyl sulphonamides (see the Hinsberg separation, above).

2. Halogens, in the presence of alkali, react with primary and secondary
amines to form halogeno-amines. Primary amines form the mono- or
dihalogeno-derivative according to the amount of halogen used:

R. N H 2 -i™% R-NHX + R-NX 2

R 2 NH X ' /Na ° H > R 2 NX

3. Primary amines react with nitrosyl chloride to form an alkyl chloride:
secondary amines are converted into a nitrosamine :

R-NH 2 + NOC1 — > RC1 + N 2 + H 2
R 2 NH + NOC1 — >- R 2 N-NO + HC1

4. Primary and secondary amines form sodium salts when heated with
sodium :

R-NH 2 + Na —>■ [R-NH]"Na+ + £H 2
R 2 NH + Na — > [R 2 N]-Na + + £H 2

5. Primary amines react in two stages with Grignard reagents, the first
stage taking place at room temperature, the second only at high temperature :

R-NH 2 + CH 3 -Mg-I — > CH 4 + R-NH-Mgl
R'NH-Mgl + CH 3 -Mg-I — > CH 4 + R-N(MgI) 2

Secondary amines, since they contain only one active hydrogen atom, can
only react with one molecule of a Grignard reagent :

R 2 NH + CH 3 -Mg-I — > CH 4 + R 2 N-MgI

6. Primary and secondary amines form substituted ureas with phenyl
wocyanate, and may be characterised by these derivatives :

R-NH 2 + C„H 5 -NCO — >■ R-NH-CO-NH-C 6 H 5
R 2 NH + C 8 H 5 -NCO — > R 2 N-CO-NH-C 6 H 5

7. Primary and secondary amines can participate in the Mannich reaction
(p. 306), e.g.,

C 6 H 5 -CO-CH 3 + H-CHO + R-NH 2 -HC1 — >

C g H 5 -CO-CH 2 -CH 2 -NHR-HCl + H 2

Reactions given by primary amines, i. Primary amines form wocyanides
when heated with chloroform and ethanolic potassium hydroxide:

R-NH 2 + CHC1 3 + 3KOH — > R-NC + 3 KC1 + 3H 2

This reaction is used as a test for primary amines (or for chloroform).

2. When warmed with carbon disulphide, primary amines form a dithio-
carbamic acid, which is decomposed by mercuric chloride to the alkyl iso-
thiocyanate:

R-NH 2 + CS 2 — > S=C( -i^> R-NCS + HgS + 2HCI

X SH

This is known as the Hofmann mustard oil reaction, and may be used as a test
for primary amines.

3. Primary amines combine with aromatic aldehydes to form Schiff bases:

C 6 H B -CHO + R-NH 2 — y C 6 H 5 -CH:NR + H 2 Q

316 ORGANIC CHEMISTRY

4. Primary amines react with nitrous acid with the evolution of nitrogen.
The equation is usually written :

R-NH 2 + HNO a — > ROH + N a + H 2

The reaction, however, is far more complicated than this equation indicates.
According to Whitmore (1941), methylamine does not form any methanol
at all when treated with nitrous acid; dimethyl ether is produced. Ethyl-
amine gives 60 per cent, of ethanol, and ^-propylamine 7 per cent, of n-
propanol, 32 per cent, of wopropanol and 28 per cent, of propylene. Linne-
mann (1868) had claimed to have obtained methanol in 50 per cent, yield
from methylamine. Austin (i960) has examined the deamination of methyl-
amine in detail and showed (using no excess of inorganic aad) that the follow-
ing compounds are formed: methanol, methyl nitriten(major product),
methyl chloride, nitromethane, and methylnitrolic acid (N0 2 *CH=NOH).
Austin has shown that nitrous acid, as such, is'not a "participant in the initial
interaction with the amine; the effective nitrosating reagent is dinitrogen
trioxide (or NOX; X=C1, Br, I); thus:

R-NH 2 + N 2 3 *" > R-NH 8 -NO — >
R-NH-NO — > R-N=NOH — > R-N 2 + + OH" — >■ R+ + N 2
The products from ^-propylamine may thus be explained as follows :

Cri3*CH2 , CH 2

CH 3 -CHOH-CH 3 -^1 CH 3 -CH-CH 3

OH-

CHo'CH^^CHo

CH 3 -CH 2 -CH 2 OH

Whatever are the products formed by the action of nitrous acid on a
primary amine, nitrogen is always evolved. Thus this reaction may be
used as a test for primary amines, since none of the other classes of amines
liberates nitrogen.

Primary amines may be converted into the corresponding alcohols in good
yields by means of the von Braun method (see also p. 762) :

C,H.'COCl PCI.

R-CH 2 -NH 2 -^ > R-CH 2 -NH-COC 6 H 5 ->

NaOH

C 6 H 5 -CN + POClj + R-CH 2 C1 y R-CH 2 OH

White (1954) has introduced a new method for the deamination of aliphatic
amines :

R'-COCl N.O. heat

r.NH 2 > R-NH-COR' ^— > R-N(NO)-CO-R' y

CH a *CO a Na

N 2 + R-OCOR' Na ° H > ROH (v.g.)

5. Oxidation of primary amines. The products obtained depend on the oxidis-
ing agent used, and on the nature of the alkyl group, e.g., (i) with potassium
permanganate :

toi H,o

-!^-> R-CH:NH — --> R-CHO + NH 3

aldimine

[O] H.O

> R,C:NH — — > R 2 CO + NHj

ketimine
R 3 ONH 2 — ^L> R 3 ON0 8

NITROGEN COMPOUNDS 317

(ii) with Caro's acid (H 2 S0 5 ) :

roi /OH

R-CH 2 -NH 2 ■ ■ > R-CHj— NH-OH + R-CH.'N-OH + R-qC

^-N-OH
N-alkyl aldoxime hydroxamic

hydroxylamine acid

R 2 CH-NH 2 [0] > R 2 C:NOH
R 8 C-NH 2 [ ° ] > R 3 C-NO

Reactions given by secondary amines, i. Secondary amines react with
nitrous acid to form insoluble oily nitrosamines; nitrogen is not evolved:

R 2 NH + HN0 2 — > R 2 N-NO + H a

Nitrosamines are yellow neutral oils which are steam-volatile. When
warmed with a crystal of phenol and a few drops of concentrated sulphuric
acid, nitrosamines form a green solution which, when made alkaline with
aqueous sodium hydroxide, turns deep blue. This procedure may be used
as a test for secondary amines; it is known as Liebermann's nitroso reaction.
Nitrosamines are readily hydrolysed to the amine by boiling with dilute
hydrochloric acid:

HC1

R 2 N-NO + H 2 >■ R 2 NH + HN0 2

Peroxytrifluoroacetic acid converts secondary nitrosamines into nitramines
(Emmons, 1954):

R 2 N-NO ***"> R 2 N-N0 2 (73-95%)

2. When warmed with carbon disulphide, secondary amines form a
dithiocarbamic acid, which is not decomposed by mercuric chloride to the
alkyl isothiocyanate (cf. primary amines) :

/NR 2
S=C=S + R 2 NH — > S=C(

\SH

3. Secondary amines may be oxidised, the product depending on the oxidising
agent used, e.g.,

„ KMnO. H.SO.

R a NH V R 2 N— NR 2 R 2 NH -^!> R 2 N-OH

tetra-alkylhydrazine

Reactions given by tertiary amines. 1. Tertiary amines dissolve in cold
nitrous acid to form the nitrite salt, R 3 N-HN0 2 or [R 3 NH] + NOjf . When
this solution is warmed, the nitrite decomposes to form a nitrosamine and
alcohol:

[R 8 NH] + NO a - — > R 2 N-NO + ROH

2. Tertiary amines are not affected by potassium permanganate, but are
oxidised to the amine oxide by Caro's acid or by hydrogen peroxide:

R3N + [O] — > R 3 N -> O

These amine oxides aire basic and exist as [R 3 N'OH] + OH- in solution. When
the solution is evaporated to dryness, the dihydrate, R 3 N0.2H 2 0, is obtained as
crystals. These, when carefully heated in vacuo, are converted into the
anhydrous amine oxide, R 3 NO. Amine oxides form addition compounds with
gaseous hydrogen halide, e.g., [R 3 N-0H]+C1-, and with alkyl halides, e.e
[R,N-OR]+Cl-. 1.6,

3l8 ORGANIC CHEMISTRY

3. Tertiary amines react with cyanogen bromide to form a dialkyl cyan-
affttdc ' ■

R 3 N + BrCN — -> [R 3 N-GN] + Br" — > R 2 N-CN + RBr

The dialkyl cyanamides are readily hydrolysed by acid or alkali to a
secondary amine:

R 2 N-CN -^> R 2 N-C0 2 H — -> R 2 NH + C0 2

This method therefore offers a means of converting a tertiary amine into
a secondary, but is mainly used to open the ring of a cyclic amine containing
a tertiary nitrogen atom (see p. 762).

QUATERNARY AMMONIUM COMPOUNDS

The quaternary ammonium salts are white crystalline solids, soluble in
water, and completely dissociated in solution (the nitrogen is quadricovalent
unielectrovalent). When a quaternary ammonium halide is heated in
vacuo, it gives the tertiary amine :

[R 4 N] + X- — > R 3 N + RX (g.-v.g.)

When a quaternary ammonium halide is treated with moist silver oxide,
the quaternary ammonium hydroxide is produced:

[R 4 N]+X- + ' AgOH ' — > [R 4 N] + OH- + AgX

This change may also be effected by treating the quaternary ammonium
halide with a methanolic solution of potassium hydroxide. Potassium
halide, which is precipitated, is removed by filtration, and the solvent then
evaporated from the filtrate. It is important to note that the quaternary
ammonium hydroxide is not liberated by aqueous potassium hydroxide.

The quaternary ammonium hydroxides are white deliquescent crystalline
solids which are as strongly basic as sodium and potassium hydroxides
(all exist as ions in the solid state). Their solutions absorb carbon dioxide,
and will liberate ammonia from its salts.

The thermal decomposition of quaternary ammonium hydroxides is a
very important reaction. Only tetramethylammonium hydroxide decom-
poses to give an alcohol:

[(CH 3 ) 4 N] + OH- — > (CH 3 ) 3 N + CH 3 OH

All other quaternary ammonium hydroxides give an olefin and water, e.g.,

[(C 2 H 6 ) 4 N] + OH- — ► (C 2 H 6 ) 3 N + C 2 H 4 + H 2

If the quaternary ammonium hydroxide contains different alkyl groups of
which one is methyl, then the methyl group is always retained by the
nitrogen, and an olefin and water are formed. Furthermore, if the ethyl
group is one of the radicals, then ethylene is formed preferentially to any
other olefin (Hofmann's rule). Water is always eliminated by the com-
bination of the hydroxyl ion with a ^-hydrogen atom of one of the alkyl
groups (preferably ethyl) :

[(CH 3 ) 2 (C 3 H 7 )N-CH 2 -CH 3 ]+OH- — -> (CH 3 ) 2 N-C 3 H 7 + C 2 H 4 + H 2

If there is no p-hydrogen in the quaternary hydroxide, e.g., tetramethyl-
ammonium hydroxide, then no olefin is formed.

The above reaction is the basis of the Hofmann exhaustive methylation
method (1851). In certain cases it is used to prepare unsaturated compounds

NITROGEN COMPOUNDS 319

of known structure, e.g., benzene (see p. 506), but it is generally used to
ascertain the nature of the carbon skeleton of cyclic compounds containing
a nitrogen atom in the ring (see p. 762).

According to Ingold et al. (1927, 1933) the reaction proceeds by a bimolecular
elimination (E2) mechanism:

HO-^ H— CH 2 — CH a -Q& 3 — -> H 2 + CH 2 =CH 2 + NE S

The reason for the preferential elimination of ethylene has been explained as
follows (Ingold et al., 1927). The positive charge on the nitrogen atom produces
an inductive effect which causes positive charges to be produced on neighbour-
ing carbon atoms. This weakens the C — H bonds sufficiently for a fi-proton to
be eliminated by the E2 mechanism.

H

I + Je^-X OH-
CH 3 ->-CH^-CH 2 ->-N-<-CH 2 — CH 2 — H >-

At \u CH 3 -CH 2 -CH 2 -N(CH 3 ) 2 + CH 2 =CH 2 + H a O

CH 3 C±i 3

Since the terminal methyl group of the w-propyl radical is electron-releasing,
the positive charge induced on the p-carbon atom (of the propyl group) is
partially neutralised. This " tightens " the bonding of these fj-hydrogen
atoms, and so a less tightly bound (S-hydrogen atom in the ethyi radical is
eliminated preferentially with the formation of ethylene.

As can be seen from the foregoing account, the formation of ethylene (and, in
general, the least branched olefin) is attributed to the polar factor. Brown
et al. (1956), however, have concluded from their work that the Hofmann type
of elimination must be attributed to the large steric requirements of the group
undergoing elimination and not to the positive charge on the 'onium group
(cf. Saytzeff's rule, p. in).

Another interesting application of the method of exhaustive methylation
is the preparation of methyl esters of acids which are difficult to esterify by the
usual methods. The acid is converted into the quaternary methylammonium
salt by titration with tetramethylammonium hydroxide in methanolic solution;
the salt, on heating at 200-300 , is decomposed into the methyl ester and tri-
methylamine (Prelog and Piantanida, 1936) :

[(CH 3 ) 4 N]+[OCO-R]- > R-C0 2 CH 3 + (CH 3 ) 3 N (v.g.)

Methylamine, dimethylamine and trimethylamine may be prepared by any of
the general methods, but are conveniently prepared by special methods.

Methylamine: By heating ammonium chloride with two equivalents of
formaldehyde (in formalin soln) ; the yield is 45-51 per cent, based on the
ammonium chloride:

2H-CH0 + NH 4 C1 > CH 3 -NH 2 -HC1 + H-C0 2 H

It is prepared industrially by passing a mixture of methanol and ammonia
over a catalyst, and separating the mixture by fractional distillation (see
p. 309).

Methylamine is a gas, b.p. —7-6°, which is used as a refrigerant.

Dimethylamine : By heating ammonium chloride with about four equiva-
lents of formaldehyde (in formalin solution) :

NH 4 C1 + 4H-CHO — =>- (CH 3 ) 2 NH-HC1 + 2H-C0 2 H

It is a gas, b.p. 7 .

Trimethylamine : By heating a solid mixture of ammonium chloride and
paraformaldehyde :

2NH 4 C1 -f gH-CHO — > 2(CH 3 ) 3 N-HC1 + 3CO a (89%)

320 ORGANIC CHEMISTRY

It is a gas, b.p. 3-5°. It occurs in sugar residues, and is used as a source of
methyl chloride:

(CH 3 ) 3 N + 4HCI — — — ► 3CH3CI + NH 4 C1

pressure

N-methylation of primary and secondary amines can readily be effected
by formaldehyde (formalin solution) ; the mechanism may be :

^NH + CH 2 — > ^N-CH 2 OH -^5L> >J-CH 3 + H 2

The 2[H] is produced according to the following reactions:

2CH 2 + H 2 — > CH3OH + H-C0 2 H

H-C0 2 H — > C0 2 + 2[H]

Carbon dioxide is always evolved in these reactions.

The long-chain amines, prepared by the catalytic reduction of alkyl
cyanides (from fatty acids and ammonia), are used as antioxidants, sterilising
agents and flotation agents.

DIAMINES

Diamines may be prepared by methods similar to those for the mono-
amines, using polymethylene halides instead of alkyl halides.

Ethylenediamine. NH 2 'CH 2 -CH 2 -NH 2 , may be prepared by heating, under
pressure, ethylene bromide with a large excess of ammonia:

CH 2 Br-CH 2 Br + 2NH3 — > NH 2 -CH 2 -CH 2 -NH 2 + 2HBr

It is a colourless liquid, b.p. 118 , soluble in water. It forms piperazine
when its hydrochloride is heated :

hc1-nh 2 -ch 2 -ch 2 -nh 2 -hc1 /ch 2 -ch 2 \

+ — > hci-hn( ;nh-hci + 2 nhxi

hc1-nh 2 -ch 2 -ch 2 -nh 2 -hc1 x:h 2 -ch/

Ethylenediamine forms chelate compounds with many metals, e.g., cobalt.

Putrescine, tetramethylenediamine, NH 2 *(CH 2 ) 4 'NH 2 , is formed by the
putrefaction of proteins (in flesh). It may be prepared as follows:

CH 2 Br-CH 2 Br — \ NOCH.-CH.-CN Na/c ' H ' OH > NH 2 -(CH 2 ) 4 -NH 2

It is a poisonous solid, m.p. 27°, with a disagreeable odour, and is soluble in
water. When its hydrochloride is heated, pyrrolidine is formed:

CH 2 -CH 2 -NH 2 -HC1 CH 2 — CH 2 \

I — > I )NH-HC1 + NH 4 C1

CH S -CH 2 -NH 2 -HC1 CH 2 — CH/

Cadaverine, pentamethylenediamine, NH 2 -(CH 2 ) 5 *NH 2 , is formed by the
putrefaction of proteins (in flesh). It may be prepared by an analogous
method to that used for putrescine :

Br-(CH 2 ) 3 -Br — \ NO(CH 2 ) 3 -CN Na/C,H '° H > NH 2 -(CH 2 ) 6 -NH 2

It is, however, prepared more conveniently by heating pentamethylene
bromide with an excess of ammonia:

Br-(CH 2 ) 5 -Br + 2NH3 — > NH 2 -(CH 2 ) 5 -NH 2 + aHBr

The starting material pentamethylene bromide is readily obtained from
piperidine (see p. 762).

NITROGEN COMPOUNDS 321

Cadaverine is a poisonous, syrupy, fuming liquid, b.p. 178-180 , with a
disagreeable odour, and is soluble in water. When its hydrochloride is
heated, piperidine is formed:

/CH 2 -CH 2 -NH 2 -HC1 /CH 2 — CH 2 \

CH 2 ( > CH 2 ( )NH-HC1 + NH 4 C1

\CH 2 -CH 2 -NH 2 -HC1 X CH 2 — CH/

Hexamethylenediamine is manufactured by the catalytic hydrogenation
of adiponitrile prepared from adipic acid :

C0 2 H-(CH 2 ) 4 -C0 2 H — ^-> CN-(CH 2 ) 4 -CN — - t — >

2 v z/4 * catalyst v "* catalyst

NH 2 -(CH 2 ) 6 -NH 2

It is a colourless crystalline solid, m.p. 39 , and is used in the manufacture
of nylon (p. 379).

Spermine, a deliquescent crystalline solid which is isolated from human sperm,
is atetramine: NH 2 -(CH 2 ) 3 -NH-(CH 2 ) 4 -NH-(CH 2 ) 3 -NH 2 .

Diamines with two amino-groups attached to the same carbon atom are
unknown (cf. the group C(OH) 2 , p. 168). On the other hand, JV-substituted
derivatives have been prepared, e.g., tetraethylmethylenediamine :

(C 2 H 5 ) 2 NH + CH a O + NH(C 2 H 5 ) 2 > (C 2 H 5 ) 2 N-CH 2 -N(C 2 H S ) 2 + H 2

UNSATURATED AMINES

The simplest unsaturated amine would be vinylamine (ethenylamine) if it
existed. AH attempts to prepare it result in the formation of the cyclic compound
ethyleneimine (aziridine), and a good method is to treat 2-aminoethanol with
sulphuric acid and then to heat the product with aqueous sodium hydroxide.

NH

H.SO. + NaOH / \

NH 2 -CH 2 -CH 2 OH -> NH 3 -CH 2 -CH 2 OS0 3 - >CH 2 — CH 2

Thus this compound is a three-membered heterocyclic ring. Ethyleneimine is a
syrupy liquid, b.p. 56 , with a strong ammoniacal odour; it is miscible with
water, and is strongly basic. It combines with sulphurous acid to form taurine
(2-aminoethanesulphonic acid), which occurs in human bile:

CH 2 — CH 2 + H 2 S0 3 — > NH 2 -CH 2 -CH 2 -S0 3 H or NH 3 -CH 2 -CH 2 -S0 3 -

\/
NH

Neurine (trimethylvinylammonium hydroxide), [(CHjJjN'CHXHJ+OH - , is
found in the brain. It may be prepared :

(i) By boiling choline with barium hydroxide solution :

[(CH 3 ) 3 N-CH 2 -CH 2 OH]+OH- ^"N [(CH 3 ) 3 N-CH:CH 2 ]+OH- + H 2

(ii) By heating trimethylamine and ethylene bromide (one molecule of each),
and subsequently heating the product with silver oxide in water :

(CH 3 ) 3 N + BrCH a -CH 2 Br >■ [(CH 3 ) 3 N-CH 2 -CH 2 Br]+Br- " Ag0H ' I >

[(ch 3 ) 3 n-ch:ch 2 ]+oh-

(iii) A very interesting synthesis is by the interaction of acetylene and tri-
methylamine under pressure, in the presence of water at 60° :

C 2 H 2 + (CH 3 ) 3 N + H a O— -> [(ch 3 ) 3 n-ch:ch 2 ]+oh-

Neurine is a very poisonous syrupy liquid.

Allylamine (2-propenylamine) , CH 2 ;CH-CH 2 'NH 2 , may be prepared by heating
allyl iodide with ammonia:

CH 2 :CH-CH 2 I + NH 3 y CH 2 :CH-CH 2 -NH 2 + HI

M

322 ORGANIC CHEMISTRY

It is more conveniently prepared by boiling allyl wothiocyanate with dilute
hydrochloric acid:

CH 2 :CH-CH 2 -NCS + H 2 HC ' > CH 2 :CH-CH 2 -NH 2 + COS (70-73%)

It is a colourless liquid, b.p. 53°, with an ammoniacal smell, and is miscible with
water.

AMINOALCOHOLS

The simplest aminoalcohol is 2-aminoethanol, NH 2 -CH 2 -CH 2 OH (cholamine,
ethanolamine, 2-hydroxyethylamine) . It occurs in kephalins (p. 263), and is best
prepared by the action of ethylene oxide on excess of ammonia.
O

CH 2 — CH a + NH 3 y NH 2 -CH 2 -CH 2 OH

It is a viscous liquid, b.p. 171°, miscible with water, and is strongly basic.

Choline (-z-hydroxyethyltrimethylammonium hydroxide) occurs in lecithins
(p. 263) ; it is best prepared by the action of ethylene oxide on trimethylamine in
aqueous solution:

A

CH 2 — CH 2 + (CH 3 ) 3 N + H 2 y [(CH 3 ) 3 N-CH 2 -CH 2 OH] + OH-

It is a colourless viscous liquid, soluble in water, and is strongly basic. It forms
neurine when boiled with barium hydroxide solution (see above). It is present
in the vitamin B complex; it is a growth factor in chicks.

AMINO-ACIDS

Amino-acids are derivatives of the carboxylic acids in which a hydrogen
atom in the carbon chain has been replaced by an amino-group. The amino-
group may occupy the a- or p- or y- . . . position ; there may also be two
or more amino-groups present in the chain.

The three basic classes of foods are: proteins, fats and carbohydrates.
The proteins are nitrogenous substances which occur in most cells of the
animal body; they also occur in plants. When hydrolysed by strong
inorganic acids or by enzymes, proteins yield a mixture of amino-acids, all
of which are a-amino-acids. The number of amino-acids so far obtained
appears to be about twenty-five, of which about ten are essential, i.e., a
deficiency in any one growth in young animals, and may even cause death.

The amino-acids are classified in several ways; Table VI shows a convenient
classification; the letters g, I and e which follow the name of the acid indicate that
the acid is respectively of general occurrence, lesser occurrence, and essential (in
man).

In this book we shall consider in detail only the simplest amino-acid,
glycine (aminoacetic acid, glycocoll), CH 2 (NH 2 )-C0 2 H. This acid is found
in many proteins, and occurs in certain animal excretions, usually in com-
bination, e.g., hippuric acid (in horses' urine), C e H 5 *CO\NH-CH 2 -C0 2 H.
Glycine may be readily prepared by the action of concentrated ammonium
hydroxide solution on chloroacetic acid:

CH 2 C1-C0 2 H + 2NH 3 — y CH 2 (NH 2 )-C0 2 H + NH 4 C1 (64-65%)
It may also be prepared pure by Gabriel's phthalimide synthesis (p. 311):

|| lcg/NK + C1-CH 2 -C0 2 H — > KC1 + || X |g^N-CH 2 -C0 2 H

HCl /Vfl w

y II feo H + CH 2 (NH 2 )-C0 2 H

NITROGEN COMPOUNDS
TABLE VI

323

Name

Formula

Neutral Amino-acids (one amino-group and one carboxyl group)

1. Glycine (g)

2. Alanine (g)

3. Valine (g, e)

4. Leucine (g, e)

5. isoLeucine (g, e)

6. Norleucine (7)

CH 2 (NH 2 )-C0 2 H

CH 3 -CH(NH 2 )-C0 2 H

(CH 3 ) 2 CH-CH(NH 2 )-C0 2 H

(CH 3 ) 2 CH-CH 2 -CH(NH 2 )-C0 2 H

(C 2 H 5 )(CH 3 )CH-CH(NH 2 )-C0 2 H

CH 3 - (CH 2 ) 3 -CH(NH 2 ) -C0 2 H

7. Phenylalanine (g, e)

/ %CH 2 -CH(NH 2 )-C0 2 H

8. Tyrosine {g)

HO^ ^CH 2 -CH(NH 2 )-C0 2 H

9. Serine (g)

10. Cysteine (g)

11. Cystine (g)

12. Threonine (g, e)

13. Methionine (g, e)

HOCH 2 "-CH(NH a )-C0 2 H
HS-CH a -CH(NH 2 )-C0 2 H
(— S-CH 2 -CH(NH 2 )-C0 2 H) 2
CH 3 -CHOH-CH(NH 2 )-C0 2 H
CH 3 -S-CH 2 -CH 2 -CH(NH 2 )-C0 2 H
I

14. Di-iodotyrosine or iodogorgic
acid (/)

HO^ ^CH 2 -CH(NH 2 )-C0 2 H

I =

I I_

HO^ /— °— \ ^— CH 2 -CH(NH 2 )-C0 2 P

f^ r

Br

15. Thyroxine {I)

16. Dibromotyrosine (I)

HO^ %— CH 2 -CH(NH 2 )-CO a H

17. Tryptophan (g, e)

j^j j|CH 2 -CH(NH 2 )-C0 2 H

NH
CH 2 — CH 2

18. Proline (g)

CH 2 CH-CO a H

\ /

NH
HOCH— CH 2

19. Hydroxyproline (I)

CH 2 CH-C0 2 H

\ /

NH

Acidic Amino-acids (one 2

imino-group and two carboxyl groups)

20. Aspartic acid (g)

21. Glutamic acid (g)

22. j8-Hydroxyglutamic acid (/) '

C0 2 H-CH 2 -CH(NH 2 )-C0 2 H

C0 2 H-CH 2 -CH 2 -CH(NH 2 )-C0 2 H

CO a H-CH 2 -CHOH-CH(NH 2 )-C0 2 H

Basic Amino-acids (two

imino-groups and one carboxyl group)

23. Ornithine *

NH 2 -CH 2 -CH a -CH 2 -CH(NH 2 )-C0 2 H

NH 2

24. Arginine {g, e)

25. Lysine {g, e)

HN=C— NH-CH 2 -CH 2 -CH 2 -CH(NH 2 )-C0 2 H
NH 2 -CH 2 -CH 2 -CH 2 -CH 2 -CH(NH 2 )-CO a H
, CH 2 -CH(NH 2 )-C0 2 H

26. Histidine (g, e)

HN N

W

* Ornithine is probably not present in proteins, but is formed by the hydrolysis of
arginine.

1 Occurrence in proteins uncertain.

3 2 4 ORGANIC CHEMISTRY

Glycine exists as white prisms which melt, with decomposition, at 289-
292 . It has a sweet taste, and is soluble in water but insoluble in ethanol
and ether. Since it contains an amino-group and a carboxyl group, it
combines the properties of a base and an acid, i.e., it is amphoteric. The
following reactions are typical of all a-amino-acids.

Reactions characteristic of the amino-group. 1. Glycine forms salts with

strong inorganic acids, e.g., C1{H 3 N-CH 2 -C0 2 H

2. Glycine reacts with acetyl chloride or acetic anhydride to give the
acetyl derivative:

H 2 N-CH 2 -C0 2 H + (CH 3 -CO) 2 — >

CH 3 -CONH-CH 2 -C0 2 H + CH 3 -C0 2 H

Similarly, with benzoyl chloride, it forms benzoylglycine (hippuric acid; see
above).

These acylated derivatives are acidic, the basic character of the amino-
group being effectively eliminated by the presence of the negative group
attached to the nitrogen (see p. 569).

3. When glycine is treated with nitrous acid, nitrogen is evolved and
glycollic acid is formed :

CH 2 (NH 2 )-C0 2 H + HN0 2 — > CH 2 OH-C0 2 H + N 2 + H 2

4. Nitrosyl chloride (or bromide) reacts with glycine to form chloro- (or
bromo-) acetic acid:

CH 2 (NH 2 )-C0 2 H + NOC1 — > CH 2 C1-C0 2 H + N 2 + H 2

Reactions characteristic of the carboxyl group. 1. Glycine may be
esterified by an alcohol in the presence of inorganic acid, e.g.,

CH 2 (NH 2 )-C0 2 H + C 2 H 6 OH -^> HC1-H 2 N-CH 2 -C0 2 C 2 H 5 + H 2

The ester is liberated from its hydrochloride by alkali.

2. Glycine forms metallic salts when its aqueous solution is warmed
with a metallic oxide or hydroxide. These salts are chelate compounds,
e.g., the copper salt (deep blue needles) is:

0=C — O H 2 N-CH,

CH 2 -NH 2 OC=0

3. Amino-acids may be reduced to aminoalcohols by lithium aluminium
hydride (Vogel et al., 1952).

R-CH(NH 2 )-C0 2 H LiA ' H ' > R-CH(NH 2 )-CH 2 OH

Reactions due to both the amino- and carboxyl groups. 1. When measured
in aqueous solution, the dipole moment of glycine is found to have a large
value. To account for this large value it has been suggested that glycine
exists, in solution, as an inner salt :

H 2 N-CH 2 -C0 2 H + H 2 ^ H 3 N-CH 2 -C0 2 - + H 2

Such a double charged ion is known, in addition to an inner salt, as a zwitter-
ion, ampholyte, or a dipolar ion. This dipolar ion structure also accounts
for the absence of acidic and basic properties of an amino-acid (the carboxyl

NITROGEN COMPOUNDS 325

and amino-groups of the same molecule neutralising each other to form a
salt). The properties of crystalline glycine, e.g., its high melting point
and its insolubility in hydrocarbon solvents, also indicate that it exists as
the inner salt in the solid state.

Owing to its amphoteric character, glycine cannot be titrated directly with
alkali. When formalin solution is added to glycine, methyleneglycine is
formed (reaction probably more complex) :

H 2 N-CH 2 -C0 2 H + H-CHO — >■ CH 2 :N-CH 2 -C0 2 H + H 2

This is a strong acid (the basic character of the amino-group being now
suppressed), and can be titrated with sodium hydroxide. This method is
known as the Sorensen titration.

2. When heated, glycine forms diketopiperazine; glycine esters give a
better yield:

ch 2 -co oc,h 5 h nh /ch 2 — co\

1 ' -> nh( ;nh + 2C 2 h 5 o

HN |H C 2 H 6 0| 0C-CH 2 x CO— CH 2

Betaines. These are the trialkyl derivatives of glycine, which exist as
+ —
dipolar ions of formula R 3 N-C0 2 . Betaine itself is the trimethyl derivative,
and may be prepared by heating glycine with methyl iodide in methanolic
solution :

H 3 N-CH 2 -C0 2 + 3CH3I — -> (CH 3 ) 3 N-CH 2 -C0 2 + 3 HI

It is more conveniently prepared by warming an aqueous solution of chloro-
acetic acid with trimethylamine :

(CH 3 ) 3 N + CH 2 C1-C0 2 H — > (CH 3 ) 3 N-CH 2 -C0 2 + HC1

Betaine is a solid, m.p. 300° (with decomposition). It occurs in nature,
especially in plant juices. It behaves as a base, e.g., with hydrochloric acid

it forms the stable crystalline hydrochloride, CI (CH 3 ) 3 N*CH 2 , C0 2 H.

ALIPHATIC DIAZO-COMPOUNDS

The aliphatic diazo-compounds are characterised by the presence of the
group ^CN 2 .

Diazomethane, CH 2 N 2 , may be prepared in various ways; the first is of
historical importance, and the others are very convenient methods.

Method of Von Pechmann (1894). Methylamine is treated with
ethyl chloroformate to give iV-methylurethan which, on treatment
with nitrous acid in ethereal solution, forms iV-methyl-JV-nitroso-
urethan. This, on warming with methanolic potassium hydroxide,
decomposes into diazomethane, which is collected on cooled ether :

CH 3 -NH 2 + C1-C0 2 C 2 H 5 — >■ CH 3 -NH-C0 2 C 2 H 5 HN °' >

CH 3 -N(NO)-C0 2 C 2 H 5 K ° H > CH 2 N 2 + C0 2 + C 2 H B OH

Method of McKay (1948). Methylamine hydrochloride and nitro-
guanidine are allowed to react in potassium hydroxide solution, the
product iV-methyl-JV'-nitroguanidine treated with nitrous acid, and the

326 ORGANIC CHEMISTRY

A^-methyl-iV-nitroso-iV'-nitroguanidine so produced is then warmed with
potassium hydroxide:

NH

CH 3 -NH 2 -HC1 + NH 2 -ONH-N0 2 + KOH — >

NH

HNO,

NH 3 + KC1 + CH 3 -NH-C-NH-NO,

NH

CH 3 -N(NO)-ONH-N0 2 K ° H > CH 2 N 2

Method of Backer et al. (1951). The nitroso-derivative of ^>-toluene-
iV-methylsulphonamide is distilled with ethanolic potassium hydroxide:

CH 3 -C 6 H 4 -S0 2 N(CH 3 )-NO + KOH — > CH 2 N 2

+ CH 3 -C 6 H 4 -S0 3 K + H 2 (80-90%)

Method of Miiller et al. (1955) . Nitrous oxide is passed into an ethereal
solution of methyl-lithium, and the precipitate decomposed in ether
with aqueous potassium hydroxide :

N 2 + CH 3 Li — > CH 3 -N=N-OLi — > LiOH + CH 2 N 2

^U CH 4 + CHN 2 Li -^% LiOH + CH=N-NH — -> CH 2 N 2

Diazomethane is a yellow, poisonous gas; liquid diazomethane, b.p.
— 24 , is explosive. The gas is soluble in ether, and since the ethereal
solution is fairly safe to handle, reactions with diazomethane are usually
carried out in ethereal solution. Diazomethane is neutral; it is reduced by
sodium amalgam to methylhydrazine CH 3 'NH-NH 2 .

Reactions. 1. Diazomethane is widely used as a methylating agent for
hydroxyl groups. In these reactions nitrogen is always liberated.

It reacts with halogen acid to form a methyl halide, e.g.,

CH 2 N 2 + HC1 — >■ CH3CI + N 2

Diazomethane methylates acidic hydroxyl groups very readily : carboxylic
acids, sulphonic acids, phenols, and enols, e.g.,

R-C0 2 H + CH 2 N 2 — > R-C0 2 CH 3 + N 2

Alcohols can also be methylated by diazomethane in the presence of a suitable
catalyst, e.g., an aluminium alkoxide.

\ + — CH N

ROH + Al(OR') 3 =^= yO— Al(OR') 3 — *-^-
W

R\ + -
N 2 + )0— Al(OR') 3 ^=±: R-0-CH 3 + Al(OR') 3

H 3 C/

Alcohols may, however, be attacked directly by certain diazo-compounds, e.g.,
diphenyldiazomethane :

(C 6 H 5 ) 2 CN 2 + C 2 H 5 OH — ^ (C 6 H 5 ) 2 CH-0-C 2 H 5

NITROGEN COMPOUNDS 327

Diazomethane also reacts with aldehydes, converting them into methyl
ketones ; but in some cases the ethylene oxide derivative is formed, particu-
larly when R is a negative group.

__ R-CO-CH,
R-CHO + CH 2 N 2

R-CH— CH,
\ / 2
O

Ketones also react in a similar manner, but only in the presence of water
as catalyst; e.g., acetone forms both the higher ketone and the ethylene
oxide derivative:

O

CH N / \

CH 3 -CO-CH 3 ^— > CH 3 -COCH 2 -CH 3 + (CH 3 ) 2 C— CH 2

Certain cyclic ketones undergo ring expansion when treated with diazo-
methane (p. 479).

2. Diazomethane adds on to ethylenic compounds to form pyrazoline
derivatives (cf. ap-unsaturated esters, p. 283); with ethylene, pyrazoline is
formed :

CH 2 CH 2 — CH

|| + CH 2 N 2 ^| ||

CH 2 CH 2 N

NH

Diazomethane also adds on to acetylenic compounds, in this case to form
Pyrazole derivatives; with acetylene, pyrazole is formed:

CH CH— CH

+ CH 2 N 2 -^|| ||
H CH ij

NH

3. Diazomethane is used in the Arndt-Eistert synthesis (1935). This is
a means of converting an acid (aliphatic, aromatic, alicyclic, or heterocyclic)
into the next higher homologue. The acid chloride is treated with diazo-
methane (2 molecules), and the resulting diazoketone is warmed with water in
the presence of silver oxide as catalyst :

R-COC1 + 2CH 2 N 2 — > R-COCHN 2 + CH 3 C1 + N 2

R-CO-CHN, + H 2 -^> R-CH 2 -C0 2 H + N 2 (50-80%)

This rearrangement of the diazoketone is usually referred to as the Wolff
rearrangement (1912).

Diazoketones react with alcohols and ammonia in the presence of silver
oxide as catalyst to form esters and amides, respectively:

R'OTT NTT

R-CH 2 -C0 2 R' < R-COCHN 2 V R-CH 2 -CONH 2

In the absence of a catalyst and in the presence of water and formic acid,
diazoketones form hydroxymethyl ketones:

R-CO-CHN 2 + H 2 H "°°' H > R-COCH 2 OH + N 2

H+

328 ORGANIC CHEMISTRY

The mechanism of the Wolff rearrangement is uncertain; a possibility is one
involving a 1,2-shift:

R R _ OH

o=c— ch-Q\t=n: ~ n * > o=c— CH — >- 0=C— CHR -^--> o=c— chr

i I

•I OH

I

0=C=CHR 0=C— CH 2 R

aldoketen acid

Now ketens react with water to form acids, with alcohols to form esters, and
with ammonia to form amides (p. 288). This mechanism via a keten is supported
by the fact that ketens have actually been isolated during this rearrangement
{e.g., by Staudinger et al., 1916). Further support is given by the experiments
of Huggett et al. (1942) who showed by means of isotope 13 C that the CO group
of the diazoketone is converted into the acid group:

C 6 H 6 MgBr + CO a — > C 6 H 6 -C0 2 H (0 S0C '' > C 6 H 5 -CO-CHN 2 -^2>

(ii) CHjNj HjO

* CuCr.O, *

C 6 H 6 -CH 2 -C0 2 H ^-VC 6 H 5 -C0 2 H + C0 2

The intramolecular character of the rearrangement has also been demonstrated
by Lane et al. (1941), who used an optically active R group and showed that the
configuration was retained during the rearrangement.

4. A particularly interesting reaction is the formation of the p-lactam
of iV-phenyl-p-alanine when diazomethane reacts with phenyl wocyanate
(Sheehan and Izzo, 1948) :

C 6 H 5 -N CO

C 6 H 5 -NCO + 2CH 2 N 2 — > | I +2N 2

CH 2 CH 2

Structure of diazomethane. Curtius (1889), working with diazoacetic

' ester (see below), proposed the ring structure ^C. || for the ^CN 2 group.

/N \N

Thus diazomethane would be CH» II. Angeli (1907) suggested that the

structure of diazomethane was linear, and proposed the formula CH 2 =N=N.
This contains a quinquecovalent nitrogen atom, and therefore, if it is going
to be accepted, requires modification. Three modifications are possible:

CH 2 =N=N I, CH 2 — N=N :, and CH 2 — N=N .\ Chemical evidence, how-
ever, is insufficient to show whether one of these formulae or the cyclic formula
of Curtius is correct. Measurement of the dipole moment of diazomethane
shows that its value is small; each linear structure should have a large
dipole moment. It would therefore appear that the linear formula is un-
tenable. On the other hand, Boersch (1935) showed by electron diffraction
studies that the diazomethane molecule is linear; this is supported by infra-
red spectra measurements. If we assume that diazomethane is a resonance
hybrid of all three linear structures, we can then account for the small dipole
moment, i.e., diazomethane is best represented as a resonance hybrid:

CH 2 =N=N I <-► CH.— N^N '. ++ CH.— *N=N :

NITROGEN COMPOUNDS 329

The first two forms make the greatest contribution, but Huisgen (1955)
believes that the following two resonance structures also contribute to some
extent :

ch 2 — n=n: ■<-> cH a =N— n:

The interesting point here is that the first of these two structures could
behave as an electrophilic reagent at the terminal nitrogen atom. There is
some evidence to support this, e.g., diazoacetophenone reacts with aqueous
potassium cyanide solution as follows (Wolff, 1902) :

— . . + (i) KON
PhCO-CH— N=N: > PhCO-CH = N-NH-CN

(li)H+

Diazoacetic ester (ethyl diazoacetate) , CHN 2 -C0 2 C 2 H,;, may be readily
prepared by treating a cooled solution of the hydrochloride of ethyl glycine

ester with cold sodium nitrite solution:

rn n ^~~ tf* C1-H 3 N-CH 2 -C0 2 C 2 H 5 + NaN0 2 — ►

^ 2 <g^ c^y CHN 2 -C0 2 C 2 H 5 + NaCl + 2 H 2 (85%)

It is a yellow oil, b.p. i4i°/720 mm., insoluble

I3 ' 2 ' in water, soluble in ethanol and ether.

The reactions of diazoacetic ester are similar to those of diazomethane.

It is reduced by zinc dust and acetic acid to ammonia and glycine. When

boiled with dilute halogen acid, it eliminates nitrogen to form glycollic

ester:

CHN 2 -C0 2 C 2 H 5 + H 2 > CH 2 OH-C0 2 C 2 H 5 + N 2

When, however, diazoacetic ester is warmed with concentrated halogen acid,
ethyl halogeno-acetate is formed, e.g.,

CHN 2 -C0 2 C 2 H 5 + HC1 > CH 2 C1-C0 2 C 2 H 5 + N 2

Diazoacetic ester reacts with compounds containing an active hydrogen
atom, e.g., it forms acetylglycoUic ester with acetic acid, and the ethyl
ether of glycollic ester with ethanol:

CH 3 -C0 2 H + CHN 2 -C0 2 C 2 H 5 — > CH 3 -CO-0-CH 2 -C0 2 C 2 H 5 + N 2
C 2 H 5 OH + CHN 2 -C0 2 C 2 H 5 — > CH 2 (OC 2 H 6 )-C0 2 C 2 H 6 + N 2

It reacts with ethylenic compounds to form pyrazoline derivatives, e.g.,
with ethylene it forms pyrazohne-3-carboxylic ester:

CH 2 CH 2 — C-C0 2 C 2 H 5

|| +CHN 2 -C0 2 C 2 H 5 — > I I

CH 2 CH 2/ N

NH

With acetylenic compounds it forms pyrazole derivatives, e.g., with acetyl-
ene, it gives pyrazole-3-carboxylic ester :

CH CH— OC0 2 C 2 H 5

+ CHN 2 -C0 2 C 2 H 5 — > || ||
H CH N

\ /
NH

33° ORGANIC CHEMISTRY

QUESTIONS

i. Draw up an analytical table to show how you would distinguish between aqueous
solutions of AcNH 2 -HCl, MeNH 2 -HCl, Me 2 NH-HCl, Me 3 N-HCl, Me 4 NCl.

2. Write out the structures and names of the isomeric amines of formula C 4 H U N.

3. Discuss the problem of hydrogen cyanide as a tautomeric mixture, including in
your answer an account of the structures of the alkyl cyanides and isocyanides.

4. Describe the methods for preparing MeCN. Name the compounds and state the
conditions under which they are formed when MeCN is treated with: (a) NaOH,

(b) HC1, (c) ROH + HC1, (d) NH 3 , (e) Ac 2 0, (/) SnCl 2 , (g) H, (h) RMgBr, (i) Na,
(j) AcOEt.

5. Write an account of the preparation and properties of cyanic acid and its related
compounds.

6. How may EtNO a be prepared? Name the compounds and state the conditions
under which they are formed when MeN0 2 is treated with: — (a) H, (6) HC1, (c) NaOH,
(d) NaOH followed by H 2 S0 4 , («) NaOH followed by SnCL/HCl, (/) Br,, (g) UNO.,
(h) H-CHO, (i) C 6 H 5 -CHO, (j) CH 2 :CH-CO-CH 3 .

7. Suggest as many methods as you can: — (a) for separating the three classes of
amines, (6) for distinguishing between the three classes of nitro-compounds.

8. Discuss the various methods whereby: — (a) a primary amine may be converted
into a secondary, and vice-versa, (6) a secondary amine into a tertiary, and vice-versa,

(c) EtOH into M-PrOH, and vice-versa.

9. Write an account of the general methods for preparing amines. Name the com-
pounds and state the conditions under which they are formed when Et-NH 2 , Et 2 NH
Et 3 N, respectively, are treated with:— (a) H 2 0, (6) H 2 S0 4 , (c) Mel, (d) AcCl, (e) Br 2 ,
(/) NOCl, (g) Na, (h) MeMgl, (i) C„H 6 -NCO, (j) CS 2 , (k) C 6 H s -CHO, (I) HN0 2 , (m) KMn0 4 ,
(n) H 2 SOj, (o) BrCN.

10. Write an account of the preparation and properties of the quaternary ammonium
hydroxides.

11. Describe the preparation and properties of: — (a) ethylenediamine, (6) putrescine,
(c) cadaverine, (d) aziridine, (e) neurine, (/) allylamine, (g) cholamine, (h) choline.

12. Suggest one synthesis for each of the following :— (a) CH 3 -CH(NH 2 )-CH 2 -NH 2 ,
(6) Me 2 C(NH 2 )-CH 2 OH, (c) NH 2 -(CH 2 ) 3 -CHOH-CH 3 , (d) Me 2 N-NH 2 , (e) Me 2 N-OH,
(/) spermine.

13. Describe the preparation and properties of glycine. What are betaines?

14. Describe the preparation of: — (a) CH 2 N 2 , (6) CHN 2 -C0 2 Et. Discuss their
structures. Name the compounds and state the conditions under which they are
formed when CH 2 N 2 and CHN 2 -C0 2 Et, respectively, are treated with: — (a) HBr, (b)
AcOH, (c) Et-S0 3 H, (d) EtOH, (e) Et-NH 2 , (/) AcNH 2 , (g) CH 3 -CHO, (h) Me 2 CO,
(j) C 2 H 4 , (j) C 2 H 2 , (k) Et0 2 C-CH = CH-C0 2 Et.

15. Define and give examples of: — (a) tautomerism, (6) pseudo-acids, (c) Mannich
reaction, (d) Schmidt's reaction, (e) Gabriel's synthesis, (/) Zeisel method, (g) Herzig-
Meyer method, (h) Liebermann's nitroso-reaction, (i) Hofmann's exhaustive methylation
method, (j) Sorensen titration, (k) Arndt-Eistert synthesis, (I) reductive alkylation.

READING REFERENCES
Sidgwick, The Organic Chemistry of Nitrogen (New Edition by Taylor and Baker,

I 937). Oxford Press.
Mowry, The Preparation of Nitriles, Chem. Reviews, 1948, 42, 189.
Levy and Rose, The Aliphatic Nitro-Compounds, Quart. Reviews (Chem. Soc.), 1948,

1. 358.
Hass and Shechter, The Vapour-Phase Nitration of Saturated Hydrocarbons, Ind.

Eng. Chem., 1947, 39, 817.
Austin, The Action of Nitrous Acid on Aliphatic Amines, Nature, i960, 188, 1086.
Organic Reactions, Wiley. Vol. I (1942), Ch. 10. The Mannich Reaction.
ibid., Vol. I (1942), Ch. 2. The Arndt-Eistert Synthesis.
ibid.. Vol. Ill (1946). Ch - 8 - The Schmidt Reaction.

ibid., Vol. IV (1948), Ch. 3. The Preparation of Amines by Reductive Alkylation.
ibid., Vol. VII (1953), Ch. 3. Carbon-Carbon Alkylations with Amines and

Ammonium salts, Ch. 6. The Nitrosation of Aliphatic Carbon Atoms.
ibid., Vol. VII (1954), Ch. 8. The Reaction of Diazomethane and its Derivatives

with Aldehydes and Ketones.
ibid., Vol. XI (i960), Ch. 5. Olefins from Amines: The Hofmann Elimination

Reaction and Amine Oxide Pyrolysis.
Smith, Aliphatic Diazo-Compounds, Chem. Reviews, 1938, 23, 193.
Huisgen, Altes und Neues iiber aliphatische Diazoverbindungen, Angew. Chem., 1955,

67, 439-
Thome, The Formation of Nitrosochlorides and Nitrosates, J.C.S., 1956, 4271.

NITROGEN COMPOUNDS 331

Gowenlock and Liittke, Structure and Properties of C-Nitroso-compounds, Quart.

Reviews (Chem. Soc), 1958, 12, 32L
Kornblum et al., The Mechanism of the Reaction of Silver Nitrite with Alkyl Halides,

/. Amer. Chem. Soc, 1955, 77, 6269.
Kornblum et al, A New Method for the Synthesis of Aliphatic Nitro Compounds, ibid.,

!956, 78, 1497.
Finar, Organic Chemistry, Vol. II, Longmans, Green (1956). Ch. XIII. Amino- Acids

and Proteins.

CHAPTER XIV

ALIPHATIC COMPOUNDS OF
SULPHUR, PHOSPHORUS, ARSENIC AND SILICON

SULPHUR COMPOUNDS

Mercaptans or thioalcohols, RSH. These compounds occur in petroleum
and give rise to " sour petrol ".

Nomenclature. One method is to name them as alkyl mercaptans, the
— SH group being known as the mercapto or sulph-hydryl group. On the
other hand, according to the I.U.P.A.C. system of nomenclature, the — SH
group is known as the thiol group, and the suffix of the series is thiol. This
method of naming the mercaptans arises from the fact that they are the
sulphur analogues of the alcohols, the usual procedure of showing that an
oxygen atom has been replaced by a sulphur atom being indicated by the
prefix thio.

CHj-SH methyl mercaptan methanethiol

C 2 H 6 *SH ethyl mercaptan ethanethiol

(CH 3 ) 2 CH > CH 2 -SH isobutyl mercaptan 2-methylpropanethiol

General methods of preparation. One method is to heat an alkyl halide
with potassium hydrogen sulphide in ethanolic solution :

RX + KSH y RSH + KX {g.)

Alternatively, a potassium alkyl sulphate may be distilled with potassium
hydrogen sulphide (the yield of thioalcohol is variable) :

RO-S0 2 -OK + KSH — ->■ RSH + K 2 S0 4

Tosylates may also be used instead of the alkyl sulphates (cf. the alkyl
cyanides, p. 295).

Another method depends on the direct replacement of oxygen in an alcohol
by sulphur. This may be carried out by heating the alcohol with phosphorus
pentasulphide :

5R0H + P 2 S 5 ^5RSH + P 2 6 (p.)

A more satisfactory method of replacing the oxygen by sulphur is to pass
a mixture of alcohol vapour and hydrogen sulphide over a thoria catalyst
at 400°:

ROH + H 2 S T "°' > RSH + H 2 (f.-g.)

The best method for preparing thioalcohols is to decompose an S-alkyliso-
thiouronium salt with alkali; these salts may be prepared by interaction
of an alkyl halide (preferably the bromide or iodide) and thiourea:

/NH 2 ^NH N OH

RBr + S=< — >R— S— C( -HBr > RSH (80-90%)

X NH 2 \NH 2

On the other hand, Frank and Smith (1946) have found that the S-alkyh'so-
thiouronium salt may be prepared by heating an alcohol and thiourea with
48 per cent, hydrobromic acid (for 7 hours) :

^NH
ROH + S=C(NH 2 ) 2 + HBr — y R— S— C< -HBr + H 2

X NH 2

332

ALIPHATIC COMPOUNDS OF SULPHUR, PHOSPHORUS, ARSENIC, SILICON 333

The yields are as good as those when the alkyl bromide is used, and so
eliminate the step of converting the alcohol into the alkyl bromide.

General properties. The thiols, except methanethiol, which is a gas,
are colourless volatile liquids with disagreeable smells. Their boiling points
are lower than those of the corresponding alcohols; this is probably due
to the fact that they are very little associated, hydrogen bonding not readily
taking place between hydrogen and sulphur. The thiols are also less soluble
in water than the corresponding alcohols, again, no doubt, due to their
inability to form hydrogen bonds with water. They are more strongly
acidic than the alcohols; this is to be expected, since the thiols are the alkyl
derivatives of hydrogen sulphide, which is a stronger acid than water, of
which the alcohols may be regarded as alkyl derivatives.

Reactions. The thioalcohols resemble the alcohols in many ways (sulphur
and oxygen occur in the same periodic group) ; the main difference is their
behaviour towards oxidising agents.

1. Thiols form mercaptides with the evolution of hydrogen when treated
with alkali metals:

2RSH + 2Na — ■» 2RSNa + H 2

These alkali mercaptides are salts: [RS]~Na + , and are decomposed by water:

RSNa + H 2 > RSH + NaOH

This reaction is reversible for the low-molecular-weight thiols, since these
dissolve in aqueous alkali, e.g.,

C 2 H 5 SH + NaOH > C 2 H 8 SNa + H 2

2. Thiols precipitate mercaptides when treated with an aqueous solution
of the salt of a heavy metal, e.g.,

2RSH + (CH 3 -C0 2 ) 2 Pb y (RS) 2 Pb + 2CH 3 -C0 2 H

Thiols also attack mercuric oxide in aqueous solution to form the mercury
mercaptide:

2RSH + HgO > (RS) 2 Hg + H 2

It is this reaction which is the origin of the name mercaptans (mercurius,
mercury; captans, seizing). These heavy metals mercaptides are covalent
compounds.

An aqueous solution of cupric chloride or an alkaline solution of sodium
plumbite containing sulphur (" doctor solution ") converts thiols into
disulphides:

2RSH + 2CuCl 2 > R— S— S— R + 2CuCl + 2HCI

2RSH + Na 2 Pb0 2 + S > R— S— S— R + PbS + 2NaOH

3. Thiols react with carboxylic acids, preferably in the presence of
inorganic acid, to form a Molester:

H+

R-C0 2 H + R'SH =^ R-COSR' + H 2

4. Thiols may be oxidised, the nature of the product depending on the
oxidising agent used. Mild oxidation with, e.g., air, hydrogen peroxide,
cupric chloride or sodium hypochlorite results in the formation of dialkyl-
disulphides:

2RSH + H 2 2 > R— S— S— R + 2H 2

Dialkyl-disulphides may also be prepared by the action of iodine on sodium
mercaptides :

2RSNa + I 3 > RjSj + 2NaI

334 ORGANIC CHEMISTRY

Field et al. (1958) have shown that lead tetra-acetate oxidises thiols to di-
sulphides :

2RSH + (AcO) 4 Pb > R 2 S 2 + (AcO) 2 Pb + aAcOH

These disulphides have an unpleasant smell (not as unpleasant as that of thiols),
and their formation by the oxidation of thiols is known as " sweetening " (p. 58).
Allyl disulphide, (CH 2 !CH-CH 2 — ) 2 S 2 , occurs in garlic. Disulphides are reduced
to thiols by lithium aluminium hydride.

When oxidised with vigorous oxidising agents, e.g., nitric acid, thiols
are converted into sulphonic acids :

RSH + 3[0] HN °' > R-S0 3 H

5. Thiols readily combine with aldehydes and ketones, in the presence
of hydrochloric acid, to form mercaptals and mercafttols respectively; e.g.,
ethanethiol forms diethylmethyl mercaptal with acetaldehyde, and di-
ethyldimethyl mercaptal with acetone (see also sulphones, below) :

CH 3 -CHO + 2C 2 H 5 SH — '->CH 3 -CH(SC 8 H 6 ) a + H 2

(CH 3 ) 2 CO + 2C 2 H 5 SH y (CH 3 ) 2 C(SC 2 H 5 ) 2 + H 2

Refluxing mercaptals (and mercaptols) with ethanol in the presence of freshly
prepared Raney nickel replaces the thiol group by hydrogen (Wolfram et al.,

1944) :

>C(SR) 2 + 4 [H] ► ^CH 2 + 2RSH

Thus a carbonyl group can be converted into a methylene group (cf. the
Clemmensen and Wolff-Kishner reductions).

The aldehyde group may be " protected " in acid solution by conversion
into a mercaptal, and can be regenerated by treatment of the latter with
mercuric chloride in the presence of cadmium carbonate (cf. acetals, p. 161).

Thioethers or alkyl sulphides, R 2 S. These are the sulphur analogues of
the ethers, from which they differ considerably in a number of ways.

General methods of preparation. 1. By heating potassium sulphide with
an alkyl halide or a potassium alkyl sulphate :

2RX + K 2 S ► R 2 S + 2KX (f.g.-g.)

2ROS0 2 -OK + K 2 S > R 2 S + 2K 2 S0 4 (g.)

Tosylates may also be used instead of the alkyl sulphates.

2. Heating an ether with phosphorus pentasulphide :

5 R 2 + P 2 S 5 > 5 R 2 S + P 2 5 (p.)

3. By heating an alkyl halide with a sodium mercaptide (cf. Williamson's
synthesis, p. 141) :

RX + R'SNa — -> R— S— R' + NaX (g.-v.g.)

4. By passing a thiol over a mixture of alumina and zinc sulphide at
300 :

2RSH >- R 2 S + H 2 S

5. By the addition of a thiol to an olefin in the presence of peroxides; in
the absence of the latter very little reaction occurs (Kharasch et al., 1939) :

R-CH=CH 2 + R'SH > R-CH 2 -CH 2 -SR'

ALIPHATIC COMPOUNDS OF SULPHUR, PHOSPHORUS, ARSENIC, SILICON 335
This reaction probably occurs by a free-radical mechanism :
(R-C0 2 ) 2 — > 2R- + 2 co 2

R- + R'SH >- RH + R'S-

R"-CH=CH 2 + R'S-— >R''-CH-CH 2 SR'-^L>R''-CH 2 -CH 2 SR' + R'S-; etc.

General properties and reactions. The thioethers are unpleasant-smelling
oils, insoluble in water but soluble in organic solvents. Chemically they
are comparatively inert. They may be oxidised to sulphoxides which, on
further oxidation, are converted into sulphones: e.g., ethyl sulphide, on
oxidation with hydrogen peroxide in glacial acetic acid, gives first diethyl
sulphoxide and then diethyl sulphone:

(C 2 H 5 ) 2 S ^% (C 2 H 5 ) 2 S=0 i^> (C 2 H 5 ) 2 S^

Other oxidising agents which bring about the same changes are potassium
permanganate, nitric acid, perbenzoic acid, etc.

Originally, sulphur was thought to be quadrivalent in sulphoxides and sexa-
valent in sulphones. Then, to conform with the " octet " theory, the valencies
were changed to 3 and 4, respectively, the oxygen atoms now being linked by
co-ordinate bonds. More recently, however, bond length measurements of the
S — O bond in sulphoxides and sulphones indicate that these bonds are almost all
double bonds. M.O. calculations have also shown that the S — O bond is largely
double in character and that the 3d orbitals of the sulphur atom are involved in
the formation of this bond (Moffitt, 1950).

The valency shell of Sulphur is (3s) 2 (3£) 4 . Also the nature of hybridisation
used is decided by the number of a-electrons and lone pairs in the atom in the
bonded state (p. 28). Unless one wishes to consider multiple bonds as " bent "
bonds, electrons in re-bonds (double or triple) are not counted when deciding the
type of hybridisation. In sulphides, each R group supplies one electron, and
since the S atom has 6 valency electrons, the bond orbitals are sp 3 , two being
occupied by bonding pairs, and two by lone pairs, i.e., R 2 S. In sulphonium salts,
the halide ion takes its bonding pair, and the R group uses one of the lone pairs
of the sulphur atom, i.e., R 3 S + X~. In sulphoxides, R 2 S=0, the orbitals are
still sp 3 , but here the R groups are joined by a-bonds, the oxygen atom by a a
and it,,-,, bond (one of the sulphur electrons has been promoted to 3^), and a

lone pair remains in the fourth sp 3 orbital. In sulphones, R2S.X . there are

four a-bonds and two ic i - r bonds (one to each oxygen atom) .

The alkyl sulphides form various addition products, e.g., with bromine
the alkyl sulphide dibromide is formed, R 2 SBr 2 . Alkyl sulphides also com-
bine with a molecule of alkyl halide to form sulphonium salts, in which
the sulphur is tercovalent unielectrovalent, e.g., the formula of triethyl-

sulphonium iodide is (C 2 H 6 ) 3 S:}I. When a sulphonium salt is heated, it
decomposes into alkyl sulphide and alkyl halide (cf. quaternary ammonium
salts, p. 318):

[R 3 S] + I- — ^ R 2 S + RI

When they are treated with moist silver oxide, the sulphonium hydroxide
is formed :

[R 3 S] + I- + " AgOH " — ^ [R 3 S] + OH- + Agl

Sulphonium hydroxides are strongly basic, and on heating form alkyl
sulphide and olefin, e.g.,

[(C 2 H 5 ) 3 S] + OH- -^ (C 2 H 5 ) 2 S + C 2 H 4 + H 2

33^ ORGANIC CHEMISTRY

This is believed to occur by an E2 mechanism [cf. p. 319) :

HO- H— CH 2 — CH-P§(C 2 H 6 ) 2 > H a O + CH 2 =CH 2 + (C 2 H 5 ) 2 S

This reaction can also occur by the Ei mechanism provided the alkyl groups are
of a suitable type, e.g., tf-butyldimethylsulphonium chloride in alkaline solution
decomposes by the Ei mechanism:

rv+

Me,C— SMe 2 > Me 2 S + Me 3 C+ — > Me a C=CH 2 + H+

Challenger et al. (1948) have isolated dimethyl-(3-carboxyethylsulphonium
+ —

chloride, [(CH 3 ) 2 S-CH 2 -CH 2 -C0 2 H]C1, from natural sources.

Mustard gas, 2 : 2'-dichlorodiethyl sulphide, bis{2-chloroethyl) sulphide,
(C1CH 2 , CH 2 -) 2 S, may be prepared by the action of sulphur monochloride
on ethylene:

2 C 2 H 4 + S 2 C1 2 > (C1CH 2 -CH 2 — ) 2 S + S

It may also be prepared, in a purer state, by heating ethylene chlorohydrin
with sodium sulphide, and treating the product with hydrochloric acid :

2H0CH 2 -CH 2 C1 + Na 2 S > 2NaCl + (HOCH 2 -CH 2 — ) 2 S

-> (C1CH 2 -CH 2 — ) 2 S

Mustard gas is an oily liquid, b.p. 215-217 , with a mustard-like smell. It
is almost insoluble in water, but soluble in most organic solvents. It is a
poison, and a vesicant.

Thiocyanic acid, isothiocyanic acid and their derivatives. Thiocyanic
acid appears to be a tautomeric substance, the equilibrium mixture of thio-
cyanic acid, HSCN, and i'sothiocyanic acid, HNCS. Spectroscopic studies of
thiocyanic acid indicate the structure HNCS (Beard et al., 1947) :

H— S— C=N ^ S=C=N— H

Salts and esters of both forms are known [cf. cyanic acid, p. 299).

Thiocyanic acid may be prepared by heating a mixture of potassium thio-
cyanate and potassium hydrogen sulphate :

KSCN + KHS0 4 >- HSCN + K 2 S0 4

Thiocyanic acid distils over as a colourless liquid, m.p. 5°. It is soluble in water,
ethanol and ether in all proportions. Its dilute aqueous solutions are fairly
stable, the concentrated solutions decomposing to form carbonyl sulphide and
ammonia :

HSCN + H 2 y COS + NH S

Alkyl thiocyanates, R-SCN, may be prepared by heating potassium
thiocyanate with an alkyl halide (or tosyl ester, p. 612) :

RI + KSCN — > R-SCN + KI

They may also be prepared by the action of cyanogen chloride on a lead
mercaptide :

(RS) 2 Pb + 2CICN > 2R-SCN + PbCl 2

The alkyl thiocyanates are fairly stable volatile oils, with a slight odour of
garlic. They are oxidised to sulphonic acids by concentrated nitric acid,
and reduced to thiols by, e.g., zinc and sulphuric acid; both of these re-

ALIPHATIC COMPOUNDS OF SULPHUR, PHOSPHORUS, ARSENIC, SILICON 337

actions show that the alkyl radical in alkyl thiocyanates is directly attached
to sulphur:

R.SO3H «-=!*- R-SCN Zn/H ' S °' > RSH
They are converted into sulphonyl chlorides by chlorine water.

R-SCN + Cl 2 -^> R-S0 2 C1 + C1CN

When heated at 180 , alkyl thiocyanates rearrange to the alkyl wothio-
cyanate :

R-SCN -^-> R-NCS

Alkyl isothiocyanates or mustard oils, R-NCS, may be prepared by heating
alkyl thiocyanates at 180 , but a much more satisfactory method is the
Hofmann mustard-oil reaction (1868), which is carried out by heating a
mixture of a primary amine, carbon disulphide and mercuric chloride.
The mechanism is uncertain; it may be via the formation of a dithio-
carbamic acid salt (see p. 390 for further details) :

/NH-R „ n
R-NH 2 + S=C=S -^ S=C( g ' > R-NCS + HgS + 2 HC1

X SH

Another convenient method is to add an aqueous solution of a primary
amine to carbon disulphide in sodium hydroxide solution, and then ethyl
chloroformate :

R-NH 2 + CS 2 + NaOH > R-NH-CS 2 Na + H 2 C1 ' C °' C ' H '>

NaCl + R-NH-CS 2 C0 2 C 2 H 8 > R-NCS + COS + C 2 H 5 OH (60-70%)

The alkyl wothiocyanates are liquids with a powerful mustard smell;
they are lachrymatory and vesicatory. They are hydrolysed to primary
amines when heated with hydrochloric acid, and reduced to primary amines
and thioformaldehyde by, e.g., zinc and sulphuric acid; both of these
reactions show that the alkyl radical in alkyl isothiocyanates is directly
attached to nitrogen:

R-NCS + 2 H 2 y R-NH 2 + C0 2 +H 2 S

R-NCS + 4 [H] Zn/H ' S °' > R-NH 2 + H-CHS

■ Allyl isothiocyanate (allyl mustard oil), CH 2 XH-CH 2 -NCS, occurs in
mustard seed as the glucoside sinigrin, which, on hydrolysis by acid or by
the enzyme myrosin (which is found in mustard seeds), gives allyl j'sothio-
cyanate, glucose and potassium hydrogen sulphate. Allyl wothiocyanate
is a colourless oil, b.p. 151 , and is the substance which gives mustard its
characteristic odour and taste. It is lachrymatory and vesicatory, and is
a convenient starting material for the preparation of allylamine.

Gmelin et al. (1955) have isolated CH 3 -S-(CH 2 ) 4 -NCS and CH 3 -S-(CH 2 ) 3 -NCS
from natural sources.

Thiocyanogen, ( — SCN) 2 , may be prepared by treating lead thiocyanate with
bromine in ethereal solution at 0°.

Pb(SCN) 2 + Br 2 > (— SCN) 2 + PbBr 2

Thiocyanogen is a gas and resembles the halogens in that it adds on to double
bonds (thiocyanation) .

338 ORGANIC CHEMISTRY

Alkyl sulphoxides, R 2 S=0, may be prepared by oxidising alkyl sulphides with
the theoretical amount of hydrogen peroxide (in acetic acid) or with dilute nitric
acid:

R 2 S + [O] > R a S=0

The state of oxidation depends largely on the conditions, and usually a
mixture of sulphoxide and sulphone is obtained. Chromium trioxide in acetic
acid appears to be specific for oxidation to sulphoxide (Knoll, 1926), and
Edwards et al. (1954) have shown that saturated (and unsaturated) aliphatic
sulphides may be oxidised to sulphoxides by chromium trioxide in pyridine or with
manganese dioxide in light petroleum (yield: 49-74 per cent.). Alkyl sulphides
are also readily oxidised by liquid dinitrogen tetroxide to sulphoxides, and there
is no further oxidation to sulphone (yield: 90 per cent.; Addison et al., I95 6 )-

The sulphoxides are odourless, relatively unstable solids, soluble in water,
ethanol and ether, and are feebly basic, e.g., they form salts with hydrochloric acid.
The structure of these salts is uncertain; it may be [R 2 S — OH] + Cl~. Sulphoxides
are reduced to sulphides by zinc and acetic acid :

R 2 SO + 2[H] Zn/CH -' C0,H > R 2 S + H a O

Dimethyl sulphoxide may be used as an oxidising agent for certain com-
pounds, e.g., phenacyl bromide (see p. 661).

^°
Alkyl sulphones, R 2 SX , may be prepared by oxidising alkyl sulphides with

hydrogen peroxide in excess (in acetic acid) or with concentrated nitric acid :

R 2 S + 2[0] — > R 2 SO a

They are colourless, odourless, very stable solids, soluble in water; they are very
resistant to reduction, but some sulphones (and sulphoxides) are reduced to
sulphide by lithium aluminium hydride. Many sulphones produce sulphinic
acids when fused with potassium hydroxide at 200 , e.g.,

(C 2 H 5 ) 2 S0 2 + KOH > C 2 H 4 + C 2 H 5 -S0 2 K + H a O (60%)

This reaction may be used to prepare aliphatic sulphinic acids.

A very important sulphone is sulphonal [2 : 2-bis(ethylsulphonyl) -propane],
which may be prepared by the oxidation of dimethyldiethyl mercaptol with
potassium permanganate:

(CH 3 ) 2 C(SC 2 H 5 ) 2 + 4 [0] > (CH 3 ) 2 C(SCyC 2 H 5 ) 2

Sulphonal is a colourless solid, m.p. 126 , stable to acids and alkalis. It has been
used as an hypnotic.

Sulphur dioxide adds on to unsaturated compounds to form long linear
polymers known as poly sulphones, the valency of the sulphur changing from
4 to 6 :

O

I II
-C— S-

1 U

The reaction takes place in the liquid phase at room temperature under the
influence of light, or in the presence of a catalyst such as silver nitrate or peracetic
acid. The properties of the polysulphones depend on the nature of the olefin
used; all polysulphones, however, are thermoplastic, resinous substances which
are insoluble in water, acids and most organic solvents.

Sulphonic acids, R-S0 3 H. The aliphatic sulphonic acids are named
either as alkylsulphonic acids or as alkanesulphonic acids; in the latter case
the sulphonic acid group is considered as a substituent group, e.g.,

ALIPHATIC COMPOUNDS OF SULPHUR, PHOSPHORUS, ARSENIC, SILICON 339

CH 3 'SO s H methylsulphonic acid or methanesulphonic acid

CH 3 'CH 2 'CH 2 , S0 3 H M-propylsulphonic acid or propane-i-sulphonic acid
(CH 3 ) 2 CH'S0 3 H s'sopropylsulphonic acid or propane-2-sulphonic acid

General methods of preparation, i. By the action of fuming sulphuric
acid or chlorosulphonic acid on a paraffin. The sulphonic acid group enters
the chain, probably at the second carbon atom, e.g.,

CH 3 -(CH 2 ) 4 -CH 3 + S0 3 — ^^CH 3 -(CH 2 ) 3 -CH(S0 3 H)-CH 3

Sulphuric acid itself has no action ; it is the free sulphur trioxide that is the
sulphonating reagent.

2. By the action of sulphuryl chloride on a hydrocarbon in the presence
of light and a catalyst, e.g., pyridine, at 40-60 . The sulphonyl chloride
is obtained, often in high yield; the sulphonyl chloride group appears to
enter mainly at the second carbon atom, but varying amounts of product
with this group at the first carbon are also obtained:

RH + S0 2 C1 2 > R-S0 2 C1 + HC1

3. By the oxidation of a lead mercaptide with concentrated nitric acid,
the lead sulphonate produced being converted into the free acid by hydrogen
sulphide:

(RS) 2 Pb + 6[0] V (R-SO s ) 2 Pb -^> 2R-S0 3 H {g.)

In practice it is usually more convenient to use a thiol or a thiocyanate as
the starting material, e.g.,

RSH + 3 [0] ^% R-S0 3 H (g.-v.g.)

This method is most satisfactory, in that the position of the sulphonic acid
group is known with certainty (the position of the mercapto or thiocyanate
group being determined by the nature of the alkyl halide used in the pre-
paration of the thiol or thiocyanate).

4. The Strecker reaction (1868). This reaction is carried out by heating
an alkyl halide with sodium sulphite; the sodium salt of the sulphonic acid
is produced:

RX + Na 2 S0 3 > R-S0 3 Na + NaX (g.-v.g.)

5. Sodium hydrogen sulphite adds on to olefins in the presence of peroxides
to form sulphonic acids (Kharasch et al., 1939).

R-CH=CH 2 + NaHS0 3 > R-CH 2 -CH 2 -S0 3 Na

6. By the oxidation of S-alkyk'sothiouronium salts (p. 390).

General properties and reactions. The sulphonic acids are generally
thick liquids, soluble in water. They are isomeric with the alkyl hydrogen
sulphites:

O O

R— S— OH R— O— S— OH

A

alkylsulphonic acid alkyl hydrogen sulphite

The alkyl hydrogen sulphites are esters readily hydrolysed by alkali; the
sulphonic acids are not hydrolysed, but form salts with alkali. Many of
the reactions of the sulphonic acids and their methods of preparation show
that in these acids the sulphur is directly attached to the alkyl group.

34° ORGANIC CHEMISTRY

The sulphonic acids are strong acids, forming salts with metallic hydroxides
or carbonates; the lead and barium salts are very soluble in water. They
form the acid chloride, the sulphonyl chloride, when treated with phosphorus
pentachloride (see p. 610) :

R-S0 3 H + PC1 5 — > R-S0 2 C1 + HC1 + POCl 3

These sulphonyl chlorides are only very slowly hydrolysed by water (c/.
acyl chlorides) ; they react readily with concentrated aqueous ammonia to
form sulphonamides :

R-S0 2 C1 + 2NH 3 > R-S0 2 -NH 2 + NH 4 C1

Sulphonyl chlorides also react with alcohols to form esters :

R-S0 2 C1 + R'OH > R-S0 3 R' + HC1

Alkyl sulphonates cannot be produced by heating a mixture of sulphonic
acid and alcohol; no esterification takes place. Sulphonyl chlorides are re-
duced by lithium aluminium hydride to thiols (Marvel et al., 1950).

The sulphonic acids undergo many double-decomposition reactions (see
the aromatic sulphonic acids, p. 608). There are, however, two important
differences between aliphatic and aromatic sulphonic acids. In the former,
the sulphonic acid group (a) is not eliminated by heating with hydrochloric
acid, and (6) is hardly replaced, if at all, by hydroxyl when fused with alkali.

Thioaldehydes and thioketones. These readily polymerise to the trimer and the
isolation of the monomer is difficult, and impossible in some cases, e.g., thio-
formaldehyde.

Thioaldehydes and thioketones may be prepared by the action of hydrogen
sulphide on an aldehyde or ketone in the presence of hydrochloric acid, e.g.,
(R' = H or alkyl) :

R-COR' + H 2 S -^i> H 2 + [R-CS-R'J P °' ymenSeS > [R-CS-R^

These trimers are cyclic compounds (1:3: 5-trithians) , and many of those of the
thioaldehydes have been isolated in two forms. Baumann and Fromm (1891)
suggested that the two forms were geometrical isomers, the cis form being the one
with the three hydrogen atoms all on the same side of the plane :

\¥ A \¥ A

C S R C S H

i i

cis trans

Schonberg and Barakat (1947), however, believe that the ring is not planar, but
puckered, and that the two forms are related to each other as " chair " and
" boat " forms (cf. eycZohexane, p. 488). Hassell and Viervoll (1947) have shown
by electron diffraction studies that the ring is puckered and that it is of the
" chair " type. If this is so, then the existence of the two forms can be explained
by cis-trans isomerism.

Thioacids, R-COSH, may be prepared by the action of phosphorus penta-
sulphide on a carboxylic acid:

5 R-CO a H + P 2 S 6 > 5 R-COSH + P 2 O s

The thioacids have a most disagreeable odour, and slowly decompose in air.
They have lower boiling points and are less soluble in water than the correspond-
ing oxygen compounds; they are soluble in most organic solvents.
In most of their reactions, the thioacids and their salts behave as if they

ALIPHATIC COMPOUNDS OF SULPHUR, PHOSPHORUS, ARSENIC, SILICON 341

contained the mercapto-group, but in a few reactions they behave as if they
contained a hydroxyl group. This may be accounted for by tautomerism for
the acid and resonance for the salts :

^O /O-H

thioacid R — C/ ^=i R — C<

\S— H ^S

salts R— C/ _ <-> R— C<.

The existence of the acid as a tautomeric mixture is supported by the preparation
of both the O- and 5-esters.

Nomenclature. The methods of nomenclature are illustrated by the following
example :

CH 3 -COSH CH 3 -CS-OH

(I) (II)

According to the trivial system of nomenclature, both I and II are named as thio-
acetic acid. According to the I.U.P.A.C. rules, the suffix -oic of the corresponding
oxygen acid is changed to thioic. Thus both I and II are named either as ethane-
thioic acid or methanecarbothioic acid. On the other hand, the suffix -thiolic is
used if it is certain that the oxygen of the hydroxyl group is replaced by sulphur,
and the suffix thionic if it is the oxygen of the carbonyl group. Thus I is ethane-
thiolic acid or methanecarbothioUc acid, and II is ethanethionic or methane-
carbothionic acid.

The most characteristic reaction of the thioacids is their extreme readiness to
acylate alcohols and amines:

R-COSH + R'OH y R-COOR' + H 2 S

R-COSH + R'-NHj ^ R-CONH-R' + H a S

Ditbioacids, R-CS 2 H, may be prepared by the action of a Grignard reagent on
carbon disulphide :

RMgX + cf > R— cf — -> R-CS 2 H

^S \S-MgX

Nomenclature. The methods of nomenclature are illustrated by the following
example :

CH 3 -CS 2 H is named as dithioacetic acid, ethanethionthiolic acid or methane-
carbodithioic acid.

A very important dithioacid is dithiocarbonic acid, HOCS 2 H. The free acid is
unknown, but many of its derivatives have been prepared, e.g., potassium xanthate
may be prepared by the reaction between potassium hydroxide, ethanol and
carbon disulphide :

KOH + CS 2 + C 2 H 6 OH >■ C 2 H 6 OCS 2 K + H 2

The name xanthate is derived from the property of these compounds, giving a
yellow precipitate of copper xanthate with copper salt (Greek: xanthos, yellow).

Vulcanisation of rubber. This process is carried out by heating crude rubber
with 4-5 per cent, sulphur and certain organic compounds which accelerate the
reaction between the rubber and sulphur. These organic compounds are known
as accelerators, and all contain sulphur or nitrogen, or both. Vulcanising rubber
causes the rubber to lose its stickiness, makes it no longer sensitive to temper-
ature changes, causes it to retain its elasticity over a wide temperature range, and
increases its tensile strength. The function of the sulphur appears to be to cross-
link the long hydrocarbon chains in crude rubber.

PHOSPHORUS COMPOUNDS

Alkyl-phospbines. All three classes of phosphines are known, the tertiary
phosphines being the commonest; the quaternary phosphonium compounds
are also known.

342 ORGANIC CHEMISTRY

General methods of preparation, i. Primary and secondary phosphines
are produced when phosphonium iodide is heated with alkyl halide in the
presence of zinc oxide, e.g., ethyl and diethylphosphine :

2PHJ + 2C 2 H 6 I + ZnO ► 2C 2 H 6 -PH 2 -HI + Znl 2 + H 2

PHJ + 2C 2 H B I + ZnO > (C 2 H 5 ) 2 PH-HI + Znl 2 + H 2

If the reaction is carried out in the absence of zinc oxide, the tertiary
phosphines and quaternary phosphonium compounds are produced, e.g.,
triethylphosphine and tetraethylphosphonium iodide :

PH 4 I + 3 C 2 H 5 I > (C 2 H 5 ) 3 P-HI + 3 HI

(C 2 H S ) 3 P + C 2 H 5 I ► [C 2 H 5 ) 4 P] + I-

Secondary phosphines may be prepared by heating a primary phosphine
with the calculated amount of alkyl halide :

R-PH 2 + R'l ^RR'PH-HI

Tertiary phosphines may be prepared from secondary in a similar manner:

R 2 PH + R'l > R 2 R'P-HI

2. A mixture of tertiary phosphine and quaternary phosphonium com-
pound is produced when phosphonium iodide is heated with one of the lower
alcohols or lower ethers, e.g.,

PHJ + CH 3 OH ► PH 3 + CH 3 I + H 2

PHJ + 3 CH 3 I > (CH 3 ) 3 P-HI + 3HI (v.p.)

(CH 3 ) 3 P + CH 3 I > [(CH 3 ) 4 P] + I- (v.p.)

A small amount of tertiary phosphine is produced when a metallic phosphide
is heated with alkyl halide, e.g.,

Ca 3 P 2 + 6C 2 H 5 Br > 2(C 2 H 5 ) 3 P + 3CaBr 2

A very small amount of quaternary phosphonium compound is produced
when phosphorus is heated with an alkyl halide or an alcohol.

3. Tertiary phosphines are prepared most conveniently by the action of
excess Grignard reagent on phosphorus trihalide:

3 R-MgBr + PCl 3 ^R 3 P + 3 MgClBr (g.-v.g.)

General properties and reactions. Except for methylphosphine (a gas),
all the alkyl-phosphines are colourless, unpleasant-smelling liquids. They
resemble the corresponding nitrogen compounds in many ways, but differ
in being less basic and more easily oxidised, e.g.,

O

C 2 H 5 -PH 2 + 3 [0] 2^V C 2 H 5 -P(OH) 2

ethylphosphonic acid

o

HNO, 1 1

(C 2 H 5 ) 2 PH + 2[0] V (C 2 H 5 ) 2 P-OH

diethylphosphonic acid

(C 2 H 6 ) 3 P + [O] ^V (C 2 H 5 ) 3 P=0

triethylphosphonoxide

ALIPHATIC COMPOUNDS OF SULPHUR, PHOSPHORUS, ARSENIC, SILICON 343

In naming organic acids and oxides derived from phosphorus, arsenic
or antimony, the syllable -on denotes the quinquevalent and -in the tervalent
state of the central atom (British Chemical A bstracts, 1948).

When quaternary phosphonium halides are treated with moist silver
oxide, the quaternary phosphonium hydroxide is produced :

[(C 2 H 6 ) 4 P] + I- + ' AgOH ' — -> [(C 2 H 5 ) 4 P] + OH- + Agl

The quaternary phosphonium hydroxides are strongly basic, comparable
in strength with the quaternary ammonium hydroxides and sodium hydr-
oxide. When heated they form the trialkyl-phosphonoxide and a hydro-
carbon (c/. R 4 NOH, p. 318).

[R 4 P] + OH- > R 3 P=0 + RH

Phosphorus differs from nitrogen in being able to expand its valency shell to
form a decet; compounds of the type R 3 PX 2 have been prepared (cf. phosphorus
pentahalides). Arsenic also forms compounds of this type. Wittig et al. (1948)
have prepared pentaphenylphosphorus, (C 6 H 5 ) 5 P.

Triphenylphosphine (which is now obtainable commercially) is used to
prepare triphenylphosphine methylene, which is used in the Wittig reaction
(p. 64):

C,H t Li

(C 6 H 6 ) 3 P + CH 3 Br — > (C 6 H S ) 3 PCH 3 + Br- > (C 6 H 5 ) 3 P = CH 2

ARSENIC COMPOUNDS

Alkyl-arsines. There are primary, secondary and tertiary arsines, and
quaternary arsonium compounds. Primary arsines may be prepared by
the action of dialkyl-mercury on arsenic trichloride and reducing the product,
alkyl-dichloroarsine, with zinc and sulphuric acid :

AsCl 3 + R 2 Hg — > RHgCl + R-AsCl 2 Zn/H;S0 ' -> R-AsH 2

Alternatively, they may be prepared by reducing an alkyl-arsonic acid
with zinc and sulphuric acid:

,-, . „ -r-r Zn/H,SO. „ . „

R-As0 3 H — > R-AsH 2

Secondary arsines may be prepared by the reduction of dialkyl-chloroarsine:

Zn/HjSO,

RjAsCl >■ R 2 AsH

Tertiary arsines may be prepared by the action of excess Grignard reagent
on arsenic tribromide:

3R-MgX + AsBr 3 — > R 3 As + 3MgClX

Quaternary arsonium hahdes may be prepared by the addition of alky]
halide to a tertiary atsine:

R 3 As + RT — > [R„R'As] + I-

General properties and reactions. Except for methylarsine (a gas), the
alkyl-arsines are colourless poisonous liquids with a garlic smell. They

344 ORGANIC CHEMISTRY

have practically no basic properties, and do not form salts with inorganic
acids. They are readily oxidised when exposed to air:

O

CH 3 -AsH 2 -^-> CH 3 -As— (OH) a
methylarsonic acid

o

(CH 3 ) 2 AsH -^-> (CH 3 ) 2 As-(OH)

dimethylarsonic acid

(CH 3 ) 3 As -^-MCH 3 ) 3 As=0

trimethylarsenoxide

Quaternary arsonium halides form the corresponding hydroxide when
treated with moist silver oxide. These hydroxides are strongly basic and,
on heating, decompose into the trialkyl-arsonoxide and hydrocarbon {cf.
R 4 POH) :

[R 4 As] + OH- — > R 3 AS=0 + RH

Cacodyl oxide (dimethylarsinoxide) , (CH 3 ) 2 As— O — As(CH 3 ) 2 . Cadet (1760)
distilled a mixture of equal parts of arsenious oxide and potassium acetate,
and obtained a vile-smelling, spontaneously inflammable oil, which sub-
sequently became known as Cadet's liquid :

As 4 6 + 8CH 3 -C0 2 K — - -> 2[(CH 3 ) 2 As— ] 2 + 4K 2 C0 3 + 4 C0 2 (17%)

This reaction was later investigated by Bunsen (1837-1843), who showed
that Cadet's liquid was a mixture of cacodyl oxide and cacodyl, (CH 3 ) 2 As —
As(CH 3 ) 2 . It was Berzelius who proposed the name cacodyl (Greek : kakodes,
stinking).

A better method for preparing cacodyl oxide is to pass a mixture of the
vapours of arsenious oxide and acetic acid over an alkali metal acetate
catalyst at 300-400 (yield: 66 per cent., Fuson and Shive, 1947).

Cacodyl oxide is an extremely poisonous liquid, b.p. 150°, insoluble in
ethanol and ether. It is not spontaneously inflammable when pure; its
inflammability is due to the presence of cacodyl. Cacodyl oxide is feebly
basic, reacting with hydrochloric acid to form cacodyl chloride (dimethyl-
chloroarsine) , b.p. 109 :

[(CH^aAs— ] 2 + 2HCI > 2(CH 3 ) 2 AsCl + H 2

Pure cacodyl oxide may be prepared by heating cacodyl chloride with
potassium hydroxide:

a^H^aAsCl + 2KOH > [(CH 3 ) 2 As] 2 + 2KCI + H 2

Cacodyl (tetramethyldiarsine) , [(CH 3 ) 2 As — ] 2 , may be prepared by heating
cacodyl chloride with zinc in an atmosphere of carbon dioxide (cf. Wurtz
reaction) :

2(CH 3 ) 2 AsCl + Zn > (CH^aAs— As(CH 3 ) 2 + ZnCl 2

It is a colourless poisonous liquid, b.p. 170°, spontaneously inflammable
in air, giving carbon dioxide, water and arsenious oxide.

Cacodylic acid (dimethylarsonic acid), (CHgJgAsO^OH), is formed when
cacodyl oxide is oxidised with moist mercuric oxide :

[(CH 3 ) 2 As— ] 2 .+ 2HgO + H 2 ► 2(CH 3 ) 2 AsO(OH) + 2 Hg

ALIPHATIC COMPOUNDS OF SULPHUR, PHOSPHORUS, ARSENIC, SILICON 345

It is a crystalline odourless substance, m.p. 200°, soluble in water, the
aqueous solution being acid to phenolphthalein. Cacodylic acid reacts
with hydrogen sulphide to form cacodyl sulphide, [(CH 3 ) 2 As— ] 2 S.

Lewisite (dichloro-$-chlorovinylarsine), ClCH!CH>AsCl 2 , may be prepared by
passing acetylene into arsenic trichloride in the presence of anhydrous aluminium
chloride :

CH-CH + AsCl 3 > cich:ch-asci 2

Continued action of acetylene produces bis(|3-chlorovinyl)chloroarsine and finally
tris((3-chlorovinyl)arsine :

cich:ch-asci 2 ^"S (cich:ch— ) 2 a s ci c ' h *> (cich:ch— ) 3 as

Lewisite is a liquid, b.p. 190° (with decomposition), and is a powerful vesicant.

Antimony compounds. Antimony forms only tertiary stibines and quaternary
stibonium compounds. Tertiary stibines may be prepared by the action of a
Grignard reagent on antimony trichloride, e.g., trimethylstibine :

3 CH 3 -MgI + SbCl 3 > (CH 3 ) 3 Sb + 3 MgClI

Trimethylstibine slowly adds on methyl iodide to form tetramethylstibonium
iodide, [(CH 3 ) 4 Sb] + I~, which, on treatment with moist silver oxide, forms tetra-
methylstibonium hydroxide, f(CH 3 ) 4 Sb] + OH~. The quaternary stibonium com-
pounds closely resemble the corresponding arsenic compounds.

Bismuth compounds. The organo-bismuth compounds resemble the organo-
compounds of mercury, lead and tin in reactivity, i.e., they are true organo-
metallic compounds (p. 348). Only tertiary bismuthines are known, and these
are best prepared by the action of a Grignard reagent on bismuth trichloride :

3 R-MgX + BiCl 3 > R 3 Bi + 3 MgClX

Tertiary bismuthines do not add on a molecule of alkyl halide; generally, they
react to form an alkyl-halogeno-bismuthine, e.g., trimethylbismuthine, when
heated with methyl iodide at 200 , forms methyl-di-iodo-bismuthine and ethane :

(CH 3 ) 3 Bi + 2 CH 3 I _f^l>CH 3 BiI 2 + 2C 2 H,

Pentaphenylbismuth, (C 6 H 5 ) 6 Bi, has now been prepared by Wittig et al. (1952) ;
it is an unstable solid.

SILICON COMPOUNDS

The organo compounds of silicon have come into prominence in recent
years due to the discovery that resins could be prepared from them.

Alkyl-silanes. These are silicon hydrides (silanes) in which one or more
hydrogen atoms have been replaced by an alkyl group, e.g., C 2 H 6 SiH 3 ,
ethylsUane; (CH 3 ) 4 Si, tetramethylsilane.

General methods of preparation. 1. By the action of a dialkyl-zinc on
silicon tetrachloride (Friedel and Crafts, 1863) :

2SiCl 4 + 2(C 2 H 5 ) 2 Zn > 2C 2 H 5 SiCl 3 + ZnCl a

ethyltrichlorosilane

SiCl 4 + 2(CH 3 ) 2 Zn > (CH 3 ) 4 Si + 2ZnCl 2

2. By the action of a Grignard reagent on silicon tetrachloride (Kipping,
1904):

SiCl 4 + 2CH 3 -MgI > (CH 3 ) 2 SiCl 2 + 2MgClI

dimethyldichlorosilane

SiCl 4 + 4CH 3 -MgI > (CH 3 ) 4 Si + 4MgClI

3. By heating a mixture of silicon tetrachloride and alkyl bromide with
sodium (cf. Wurtz reaction) :

SiCl 4 + 4C 2 H 6 Br + 8Na > (C 2 H 6 ) 4 Si + 4NaCl + 4NaBr

346 ORGANIC CHEMISTRY

4. By passing alkyl halide vapour over a mixture of silicon and copper
at 350 (Patnode and Rochow, 1945) ; e.g., methyl chloride gives a mixture
of methyltrichlorosilane, dimethyldichlorosilane, trimethylchlorosilane and
methyldichlorosilane, of which the first two predominate. This reaction
has been shown to take place via the formation of methylcopper, which
then undergoes homolytic fission to give a methyl free radical (Hurd and
Rochow, 1945):

2Cu + CH3CI > Cu— CH 3 + CuCl

Cu— CH 3 > Cu + CH 3 -

CuCl + Si > =SiCl + Cu

ssSiCl + 3CH 3 y (CH 3 ) 3 SiCl

5. By passing a mixture of alkyl hahde vapour and silicon tetrachloride
over an aluminium or zinc catalyst at 400 . The same mixture is obtained
as in method 4; and probably the mechanism is the same.

Methods 1, 4 and 5 are used on a large scale for preparing the intermediates
required for the preparation of silicones (see below).

General properties and reactions. The alkylsilanes are colourless oils,
stable in air, and are not affected by acids or alkalis. The alkyl groups in
tetra-alkylsilanes may be chlorinated, and these chlorinated compounds
undergo many of the reactions characteristic of the alkyl halides, e.g.,

(C 2 H 8 ) 4 Si + Cl 2 > (C 2 H B ) 3 Si-C 2 H 4 Cl + HC1

(C 2 H 5 ) 3 Si-C 2 H 4 Cl + RONa > (C 2 H 5 ) 3 Si-C 2 H 4 -OR + NaCl

Alkyl-halogenosilanes may be prepared by any of the above methods
(1-5). They are usually colourless oily liquids that fume in the air due to
their ready hydrolysis to silanols.

Silanols. Hydrolysis of trialkyl-chlorosilanes, dialkyldichlorosilanes and
alkyl-trichlorosilanes gives the silanols, R 3 SiOH, silanediols, R 2 Si(OH) 2 ,
and silanetriols, RSi(OH) 3 , respectively.

Trialkyl silanols are colourless stable liquids, many of which have a camphor-
like odour. On the other hand, the silanediols and silanetriols are unstable,
eliminating water to form polymers. Hydrolysis is carried out by dissolving the
alkyl-chlorosilane in a solvent, and mixing with another solvent containing water.
Under these conditions hydrolysis is slow : rapid hydrolysis leads to gel formation.
The hydroxy-compounds are polymerised by heating.

Hydrolysis of dimethyldichlorosilane produces cyclic polymers containing 3-9

yO— Si(CH 3 ) 2
oxygen atoms in the ring, e.g., the trimer, (CH 3 ) 2 Sk ^o ; linear

\0— Si(CH s ) a
polymers, HOSi(CH 3 ) 2 '[OSi(CH 3 ) 2 ] n -OH, are also obtained, and the value of n
may be increased by treating with sulphuric acid, resulting in higher polymers
which are viscous liquids or soft plastics. Hydrolysis of methyltrichlorosilane
results in the formation of amorphous powders or hard brittle solids. These
polymers consist of linear, cyclic and cross-linked structures, depending on the
conditions of the hydrolysis; e.g., the end product of the dehydration process is
cross-linked :

I I

O O

•-0— Si(CH 3 )— O— Si(CH 3 )— O— Si(CH 3 )— O—

O

— O— Si(CH 3 )— O— Si(CH 3 )— O— Si(CH 3 )— O—

o o

ALIPHATIC COMPOUNDS OF SULPHUR, PHOSPHORUS, ARSENIC, SILICON 347

Thus by controlling the conditions of hydrolysis, and using mixtures of inter-
mediates, polymers with different properties can be obtained. These polymers
are known as the silicones. Silicones may be roughly classified as follows :

(i) Silicone fluids : used in high-temperature baths and diffusion pumps,
as hydraulic fluids, and form water-repellent surfaces,
(ii) Silicone rubbers : used as electrical insulators.

(iii) Silicone greases : used as lubricants at high and low temperatures,
(iv) Silicone resins : used as electrical insulators.

QUESTIONS

1. How may EtSH be prepared? Name the compounds and state the conditions
under which they are formed when EtSH is treated with: — (a) Na, (6) KOH, (c)
(AcO) 2 Pb, (d) HgO, (e) Na 2 PbO a , (/) AcOH, (g) H a 2 , (A) NaOCl, (*) HN0 3 , (1) CH,-CHO,
(/) Me 2 CO.

2. Describe the preparation and properties of: — (a) Et 2 S, (6) (ClCH a -CH a — ) 2 S, (c)
Me-SCN, (d) CH 2 :CH-CH a -NCS, (*) Et a SO, (/) Me a SO a , (g) Me 2 C(S0 2 Et) 2 .

3. Discuss the structures of the sulphoxides, sulphones, sulphonium salts, trimeric
thioaldehydes and thioacids.

4. Suggest one synthesis for each of the following: — (a) Me-NH-CS-OEt, (6)
Me-NH-CS-NH-Et, (c) Et-NH-CS 2 Me. What would each of these give on vigorous
hydrolysis?

5. Compare and contrast the methods of preparation and the properties of the
onium compounds of N, S, P and As.

6. How may Et-SO s H be prepared? How may this be distinguished from EtOSO a H?
How may Et-SO s H be converted into:— (a) Et-SO,Cl, (b) Et-S0 3 Na, (c) (Et-S0 3 ) 8 Pb,
(d) Et-SO a Et, (e) Et-SO a -NH 2 , (/) Et-SO a -NH-Et?

7. Compare and contrast the methods of preparation and the properties of the
phosphines and arsines.

8. Write an account of the preparation and properties of the cacodyl compounds,
and discuss the part played by these compounds in the history of organic chemistry.

9. Write an account of the preparation and properties of the tetra-alkyl-silanes and
the alkyl-chlorosilanes, and discuss the use of organo-silicon compounds in industry.

READING REFERENCES

Gilman, Advanced Organic Chemistry, Wiley (1942, 2nd ed.). Vol. I, Ch. 10. Organic

Sulphur Compounds.
Schonberg and Barakat, The Stereochemistry of Trimeric Thioaldehydes, J.C.S.,

1947. 6 93-

Snell, Surface Active Agents, Ind. Eng. Chem., 1943, 35, 107.

Ann. Reports, 1930, 27, 143. The Decomposition of the Quaternary Compounds

of Nitrogen and Phosphorus.
ibid., 1945, 42, 103. Phosphorylation.

ibid., 1948, 45, 198. Recent Work on the Reactions of Organic Sulphur Compounds.
Newton Friend's Inorganic Chemistry :

(i) Vol. XI, Part I (1928), pp. 246-291. Silicon Compounds.

(ii) Vol. XI, Part II (1930), pp. 3-61. Arsenic Compounds.

(iii) Vol. XI, Part III (1936), pp. 3-46. Phosphorus Compounds.
Gilman and Yale, Organo Bismuth Compounds, Chem. Reviews, 1942, 30, 281.
Emblem and Sos, Silicon Organic Compounds, Chem. and Ind., 1946, 51, 450.
Burkhard et al., The Present State of Organo-Silicon Chemistry, Chem. Reviews, 1947,

4 1 . 97- •
Hardy and Megson, The Chemistry of Silicon Polymers, Quart. Reviews {Chem. Soc),

1948, 2, 25.

Jeffes, Some Aspects of the Industrial Development of Silicones, Chem. and Ind.,

1954, 498.
Atherton, Some Aspects of the Organic Chemistry of Derivatives of Phosphorus

Oxyacids, Quart. Reviews (Chem. Soc), 1949, 3, 146.
Abrahams, The Stereochemistry of Sub-group VIB of the Periodic Table, ibid., 1956,

10, 407.
MacDiarmid, Silyl Compounds, ibid., 1956, 10, 208.
Wittenberg and Gilman, Organosilylmetallic Compounds, Quart. Reviews {Chem. Soc),

1959, 13. "6.

CHAPTER XV

ORGANO-METALLIC COMPOUNDS

Generally speaking, organo-metallic compounds are those organic com-
pounds in which a metal is directly joined to carhon (see, however, alkali
metals, p. 361). The most widely studied and the most useful are the
magnesium compounds, but in recent years the organo-compounds of sodium
and lithium have been studied in great detail; the lithium compounds in
particular are becoming increasingly important in synthetic work.

THE GRIGNARD REAGENTS

The alkyl-magnesium halides, R — Mg — X, or Grignard reagents, intro-
duced by Grignard in 1900, are extremely valuable in laboratory organic
syntheses, and recently are being used on a large scale (cf. silicon compounds,
P- 345)- A Grignard reagent is generally prepared by reaction between mag-
nesium (1 atom) and alkyl halide (1 molecule) in dry, alcohol-free ether.

RX + Mg > RMgX (v.g.-ex.)

The mechanism of the formation of a Grignard reagent is still not clear.
A free radical mechanism has been suggested (Gomberg and Bachmann, 1927;
cf. triphenylmethyl bromide, p. 702). The reaction starts by the formation of a
trace of magnesium halide (Wurtz reaction) and then proceeds as follows :

2RX + Mg y R— R + MgX 2

Mg + MgX 2 =^±: 2-MgX
RX + -MgX — > MgX 2 + R-
R- + -MgX > RMgX

The ethereal solution of the Grignard reagent is generally used in all
reactions. Other solvents besides ether may be used, e.g., tertiary amines,
tetrahydrofuran and the dimethyl ether of ethylene glycol. Tetrahydro-
furan is being used increasingly as a solvent, since it has been found that it
increases the reactivity of organic halides towards magnesium. Dimethyl
sulphate also reacts readily with magnesium in tetrahydrofuran to give
CHgMg-SO^Hg (Normant et al., 1957). The reactions of this new reagent
appear to be similar to those of CH 3 MgX.

The ease with which an alkyl* halide forms a Grignard reagent depends
on a number of factors. It has been found that for a given alkyl radical
the ease of formation is alkyl iodide >bromide>chloride. It has also been
found that the formation of a Grignard reagent becomes increasingly
difficult as the number of carbon atoms in the alkyl group increases, i.e.,
the ease of formation is CH 3 X>C 2 H 5 X>C 3 H 7 X> • • •. Since tertiary
alkyl iodides readily eliminate hydrogen iodide with the formation of an
olefin, tertiary alkyl chlorides are used.

Normant (1953, 1957) has shown that Grignard reagents can be prepared
from vinyl and aryl halides if tetrahydrofuran is used as solvent.

Structu