CD

OU1 58482

OSMANIA UNIVERSITY LIBRARY

Call No. 5 L TH"5'"1 Accession No. Gv 10 ,<-* G

A <0 O
Author

Title O x o\ OUJ> Xc - * ^a v ^ v ^^ r v >
This book should be returned on or before the date last marked below.

ORGANIC ELECTRODE
PROCESSES

MILTON J. ALLEN

Director of the Physical Research Laboratories

Ciba Pharmaceutical Products Inc.

Summit, New Jersey

London: Chapman & Hall Ltd.

REINHOLD PUBLISHING CORPORATION
NEW YORK

FIRST PUBLISHED 1958

LONDON AND TONBR1DGE

To
JODY

whose tenderness and thoughtfulness
illuminated the period of this endeavour

FOREWORD

IN INORGANIC CHEMISTRY the electrochemical method is both well
understood and widely used, but in organic chemistry the reverse
is true. This is a strange anomaly, the reasons for which are obscure,
but whatever they may be there is no doubt of the desire of many
organic chemists to know more about the principles and practice of
organic electrochemistry. As Dr. Allen points out in his preface, this
book is his response to many requests for information and instruction.
The enthusiasm which he shows for the subject will do much to advance
interest in the field. It will do much also to dispel a certain mystery
which surrounds the subject, and which seems, in part, to originate in
a shyness which many organic chemists have towards subjects of a
physical chemical nature.

The author is well equipped for the task, having spent the greater
portion of his research career in investigations of experimental electro-
organic chemistry, and in this book he carefully balances the sifting of
knowledge with the application of principles to practice.

The future of the subject may merit speculation. There seems at
the present time to be a growing interest in radical reactions at the
anode, with all that that may imply for modern technology. On the other
hand, the steady inflation which is occurring in the cost of conventional
reducing metals may well initiate a resurgence in the importance of
cathodic reduction. In any event, Dr. Allen's book seems to have made
a timely appearance to inspire thought, interest, and experiment.

CHRISTOPHOR L. WILSON

University of Notre Dame,
Indiana, U.S.A.
March 1957.

AUTHOR'S PREFACE

DURING the last few years I have been privileged to deliver lectures
here and abroad on the various aspects of electro-organic chemistry.
Many members 'of my audiences have suggested that if there were
available a comprehensive monograph which would enable the novice
to obtain a complete picture of the techniques involved, and of the
results one might anticipate when undertaking an organic electro-
chemical investigation, they would be much more desirous of attempting
to utilize the technique. Although a number of excellent books have
been written on the subject, each suffers from inadequacy in one
or more of the necessary aspects of this science. Further, some
of the better books have been out of print for some time. Therefore,
in view of the steadily increasing interest in organic electrochemistry,
I have attempted to present as complete a picture as possible of the
various phases with which the potential investigator need be acquainted.
There are many subjects, especially those involving the theoretical
aspects, which could have been presented in a more elaborate form, but
I felt this complexity might discourage the reader. If I have erred in
omission of facts which may be deemed vital by some readers I sincerely
beg forgiveness, and take refuge in the word " oversight."

In closing I wish to express my sincere appreciation to Dr. A. H.
Corwin, whose initial stimulus led me into this most fascinating field ;
to Drs. J. Bockris and A. Hickling for their moral support; to Drs. J.
Marsh and R. Mizzoni for their helpful suggestions; to Mr. R. Wolf
and Miss R. Cole for their artistic and photographic contributions ; and
last, but not least, to three wonderful girls, Misses P. McGovern, V.
Powell, and J. Siragusa, whose efforts helped bring this manuscript to
its final form.

MILTON J. ALLEN
Summit, NJ.
U.S.A.
March 1957.

IX

CONTENTS

CHAPTER PAGE

INTRODUCTION xiii

I. GENERAL CONSIDERATIONS i

II. INSTRUMENTATION AND TECHNIQUES OF ELEC-

TROLYSIS 20

III. CATHODIC REDUCTION 47

(1) MECHANISM OF CATHODIC REDUCTION 47

(2) REDUCTION OF NITRO COMPOUNDS 49

(3) REDUCTION OF IMINES AND IMIDIC ESTERS 55

IV. CATHODIC REDUCTION OF CARBONYL TYPE COM-

POUNDS 58

(1) REDUCTION OF ALDEHYDES AND KETONES 58

(2) REDUCTION OF CARBOXYLIC ACIDS 68

(3) REDUCTION OF ESTERS 72

(4) REDUCTION OF AMIDES AND IMIDES 73

V. MISCELLANEOUS CATHODIC REACTIONS 84

(1) ADDITION OF HYDROGEN TO UNSATURATED COM-

POUNDS 84

(2) REDUCTION OF ALKALOIDS 87

(3) REDUCTION OF SULPHUR, SULPHONYL CHLORIDE,

AND ARSENIC-CONTAINING COMPOUNDS 88

(4) DEHALOGENATION 91

(5) DESULPHONATION 92

VI. ANODIC OXIDATION 95

(i) MECHANISM OF ANODIC OXIDATION 95

, (2) OXIDATION OF FATTY ACIDS AND THEIR SALTS 96

(a) Mechanism of the Kolbe Reaction 97

(b) Monocarboxylic Acids 102

(c) Mixtures of Carboxylic Acids 104

CONTENTS Xi

CHAPTER PAGE

(d) Branched-chain Acids and Half-Esters 105

(e) Unsaturated Acids 106
(/) Aromatic Acids 107
(g) Substituted Acids 107
(h) Dicarboxylic Acids 109
(i) Acetoxylation, Alkoxylation, and Alkylation in

VII. ANODIC OXIDATION OF ALIPHATIC, HETERO-

CYCLIC AND AROMATIC COMPOUNDS 116

(1) OXIDATION OF ALIPHATIC COMPOUNDS 116

(a) Alcohols 116

(b) Aldehydes 118

(c) Ketones 119

(d) Sugars 119

(e) Carboxylic Acids 120
(/) Alicyclic Compounds 121
(g) Miscellaneous Aliphatic Anodic Oxidations 123

(2) OXIDATION OF HETEROCYCLIC COMPOUNDS 124

(3) OXIDATION OF AROMATIC COMPOUNDS 125

(a) Aldehydes and Ketones 132

(b) Carboxylic Acids 133

(c) Miscellaneous Aromatic Oxidations 135

VIII. ANODIC SUBSTITUTION 142

(1) ANODIC HALOGENATION 143

(a) Chlorination 144

(b) Bromination 151

(c) lodination 155

(d) Fluorination 159

(2) ANODIC THIOCYANATION 161

(3) ANODIC NITRATION 162

(4) ANODIC ALKOXYLATION 163

INDEX 168

INTRODUCTION

For out of olde fildes, as men seith,
Cometh al this newe corn fro yeer to yere
And out of old bokes, in good feith,
Cometh al this newe science that men lere.

CHAUCER.

T

'ODAY in the field of electrochemical processes we discuss subjects
_ such as overpotential, effects of medium, temperature, concen-
tration, etc., on the progress of an electrolysis as if they were new
conceptions. But the literature indicates that investigators of the pre-
vious century were well acquainted with these phenomena, although
perhaps not to the point of creating definite terms for their observations.
Therefore, our present-day knowledge in the field of electrolytic
processes is a result, in part, of rediscovery and elaboration.

The fact that electricity is a potential oxidizing or reducing agent
was realized very early in the nineteenth century by Rheinold and
Erman (i) who electrolysed dilute aqueous solutions of alcohol.
Grutthus made the interesting observations that indigo-white, upon
electrolysis in a basic medium, gave a blue precipitate at the anode.
This precipitate disappeared upon reversal of the current. Liidersdorff
in 1830 was the first to undertake a detailed study of the products
obtained using different electrodes in the oxidation of alcohol. It was
Faraday who first realized the potentialities of using electric currents
to bring about the synthesis or degradation of a compound. However,
in spite of Faraday's extensive researches, it was not until the time of
Kolbe's important discoveries that the principles of electrolytic
oxidation and reduction were established. In 1845, by electrolytic
means, he completely replaced the chlorine with hydrogen in
chloromethylsulphonic acid. He also observed that trichloromethyl-
sulphonic acid was completely decomposed in an aqueous solution at a
platinum electrode. Concerning this he stated: " This acid, which is
stable to the strongest oxidizing media, breaks down with little difficulty
at a platinum anode." Kolbe now turned his attention to the anodic

xiii

xiv INTRODUCTION

oxidation of organic substances from which resulted the synthesis of
hydrocarbons by electrolysis of salts of monobasic aliphatic acids.
These discoveries led to extended studies on the effects of electrolysis
on aromatic hydrocarbons and their derivatives. Toward the closing
years of the last century, investigators such as Gattermann, Haber, Lob,
Tafel and others made tremendous contributions to the knowledge of
the quantitative effects of electrolytic oxidation and reduction on
organic compounds.

Fritz Haber laid a sound foundation for the methodical study
of electro-organic processes by introducing the concept of controlled
electrode potential. In 1898 there appeared a paper by Haber (2) on
the stepwise reduction of nitrobenzene. He showed, for example, that
by variation of the cathodic potential he could obtain either azoxy-
benzene or hydrazobenzene. In his discussion he stated that: " The
electric current up to this time has been regarded in organic electro-
chemistry as a means of reaction whose effects are determined through
current density, current duration and occasionally through the material
of the electrode. This view is incomplete, for oxidation and reduction
processes depend mainly on the potential of the electrode at which they
take place. The current density, current duration and electrode
material are important only in so far as they determine the electrode
potential and its changes in the process of electrolysis. " This fact seems
to have been overlooked to a great extent by subsequent workers in the
field, with the result that only empirical information has been forth-
coming. However, of late there has been renewed interest in controlled
potential electrolysis, and it is now possible to anticipate more
quantitative information as a result of the rediscovery of Fritz Haber's
work.

REFERENCES

(1) Gmelim Handbook, English translation, Vol. I, 1848; Vol. VIII, 1853;
Moser, Die elektrolytischen Prozesse der organishe Chemie> 1010, pp i-n

(2) Haber, Z. Elektrochem., 1898, 4, 506.

CHAPTER I

GENERAL CONSIDERATIONS

(1) POLARIZATION

THE passage of an electric current tends to displace the potentials
of the electrodes in the system concerned. As a result, during an
electrolytic oxidation or reduction the electrodes are not in a state of
equilibrium. Under such conditions the electrodes are considered to
be polarized. Therefore, any pair of electrodes which comprise a cell,
and are connected to an outside source of voltage so as to permit the
passage of current, are considered to be polarized. Substances which
tend to re-establish the equilibrium state are referred to as depolarizers.
Thus if an electrolytic oxidation or reduction is performed on a com-
pound which is susceptible to oxidation or reduction, this substance
will tend to return the electrode to its state of equilibrium.

(2) REVERSIBLE ELECTRODE PROCESSES

Our prime concern will be with organic electrode processes which
are mainly irreversible, therefore we shall touch only lightly on rever-
sible processes.

As reduction involves the increase in negative charge or the gain of
electrons, and oxidation the decrease in negative charge or the loss of
electrons, one can create a system wherein there is a state of equilibrium
between the oxidized and the reduced forms represented by

Red. ^n e + Ox.

where Red. and Ox. represent the reduced and the oxidized form and n
the number of electrons involved in this change of state.

In order to have a reversible oxidation-reduction process it is neces-
sary that both species be present in the system, and also that minute
changes in potential result in the change of one form to another,
depending on the extent of the potential change. In other words, if

ORGANIC ELECTRODE PROCESSES

the potential is made more positive, then oxidation with loss of electrons
should tend to compensate for this positive change in potential. A
more negative potential should be compensated for by addition of
electrons to the oxidized form. Therefore, the equilibrium potential
for any reversible system as indicated by the above equation can be
given by the following expression :

F - F RT , ( X

A> ^ In p .

nF (Red.)

where E is the standard oxidation-reduction potential, Ox. and Red.
the concentrations of oxidized and reduced forms, and n the number of
electrons involved in the transformation.

(3) IRREVERSIBLE ELECTRODE PROCESSES

As mentioned previously, most processes involving the electrolytic
oxidation or reduction of organic compounds are irreversible. There
are, however, some noteworthy exceptions such as quinone ^ hydro-
quinone, leuco base ^ coloured form of the triphenylmethane dyes,
etc. However, these systems are rare and in general we must be
content with empirical data. An irreversible system does not give
a definite potential which varies according to the amount of oxidized
and reduced form in the system, as is obtained by application of the
thermodynamic expression for the reversible system. As the potential
of the systems depends more on the nature of the medium, its degree
of acidity or alkalinity, the electrode and its history, than on the
concentrations of oxidized and reduced forms, it is not possible to
apply the simple principles which hold for reversible electrode processes.
It must be remembered that there is a distinct difference between
chemical reversibility and thermodynamic reversibility. A ketone ^
secondary alcohol may be reversible chemically, but it certainly does not
generally give a definite thermodynamically reversible potential. That
does not mean to imply that once the ketone is electrolytically reduced
to the alcohol it cannot be again oxidized electrolytically back to the
ketone. Such is possible, but it requires totally different conditions of
potential, medium, pH and electrode material, to achieve this trans-
formation.

If we may take licence from Fritz Haber and assume from his state-
ment that an oxidation or reduction process depends mainly on the
potential of the electrode at which it occurs, then there is a focal point

2

GENERAL CONSIDERATIONS

for a beginning. However, it must be remembered that without con-
sideration of the medium in which the electrolysis is performed, acidity
or alkalinity of this medium, concentration of depolarizer, electrode
material, etc., electrode potential becomes an abstract and meaningless
term. Therefore, at this point, electrode potential and each of the
various factors which control and contribute to this potential at the
electrode surface will be discussed.

(4) ELECTRODE POTENTIAL

If a ketone is reduced with tin and hydrochloric acid one usually
obtains an alcohol. If the synthetic chemist thinks further about the
matter he realizes that a two electron change has taken place with the
addition of hydrogen to form the alcohol. He may not extend his
thoughts to the fact that this particular reaction has a definite E.M.F.
Therefore, if this reaction were carried out electrolytically at the same
E.M.F. in a suitable medium the electrode would contribute the two
electrons and the medium the hydrogen necessary for the reduction.

We may refer to this E.M.F. of the reaction as the electrode potential
or as a reference potential v. a standard reference electrode (e.g. saturated
calomel electrode). There are two methods by which one can attain
the desired electrode potential. The first is in the use of electrode
material of proper overpotential. By overpotential is meant the
difference between potential at which gas evolution is first observed,
and the theoretical reversible potential of the electrode in the same
solution (i). The higher the hydrogen overpotential of the electrode
generally the more powerful its reducing ability, or the greater the energy
content of the hydrogen liberated. The oxidizing ability and the
energy content of the oxygen liberated at the anode is similarly depen-
dent on the oxygen overpotential. We may therefore regard the poten-
tial as being a measure of the potential energy of the hydrogen or
oxygen liberated.

It has been shown by LeBlanc (2) that aqueous solutions of acids
and bases, in general, have a " decomposition potential " in the vicinity
of 1-7 volts at a smooth platinum electrode (Table I: i).

The term " decomposition potential " refers to that potential at
which appreciable current begins to flow through the solution, accom-
panied by hydrogen evolution at the cathode and oxygen evolution at
the anode. The reason given by LeBlanc for the " decomposition
potential " being the same in different solutions is that the hydrolysis

ORGANIC ELECTRODE PROCESSES

TABLE i : i
Decomposition Potentiah of Acids and Bases in Aqueous Solutions

Acids

Volts

Bases

Volts

Nitric acid
Sulphuric acid
Phosphoric acid
Trichloracetic acid .
Perchloric acid

1-69
1-67
170
r66
1-65

Ammonium hydroxide
Sodium hydroxide .
Trimethylammonium hydroxide
Diethylammonium hydroxide .
Potassium hydroxide

1-74
1-69
1-74
1-62
1-67

of water is common to all cases, with the discharge of hydrogen ions
at the cathode and hydroxyl ions at the anode, the hydroxyl ions forming
oxygen according to the following scheme :

2OH~

2-OH

* 2'OH

H 2 O +

2C

i/2O a

The fact that aqueous solutions of halogen acids have lower decom-
position potentials (Table i : 2) is explained by the fact that the discharge
potential of the halogen is lower than that required for hydroxyl ion
discharge.

TABLE 1:2
Decomposition Potentials of Aqueous Halogen Acid Solutions

Hydriodic acid
Hydrobromic acid
Hydrochloric acid ,

0-52 volts
0-94 volts
1-31 volts

As previously mentioned, the decomposition potential of aqueous
acid or base solution is about 1-7 volts. This is only true if smooth
platinum electrodes are used. Other electrodes in the same medium
will give different decomposition potentials.

An indication of the relative ability of a metal to reduce or oxidize
can be obtained from the Table 1:3.

From the table it can be seen that mercury, zinc, lead and tin should
be the most desirable electrodes to use for difficult reductions, and
nickel, jplatinum and copper for compounds easily reduced.

ThisTis in general agreement with observations that have been made
in the past. The reduction of nitro compounds in acid media at a high
overpotential electrode such as lead, mercury, tin or zinc yields the

GENERAL CONSIDERATIONS

TABLE 1:3
Cathodic and Anodic Electrode Overpotentials

Hydrogen
Overpotentials
(In N.H 2 S0 4 )

Oxygen
Overpotentials
(In N.KOH)

Palladium
Gold .
Iron
Smooth Platin
Silver .
Nickel .
Copper
Cadmium
Tin
Lead .
Zinc
Mercury

um

o-oo

0'02
0'08

O'og
0-15

0'2I
0.23
0-48

0'53
0*64
0-70
0-78

0'43
0'53
0-25

Q'45
0-41
0-06

0*43

0-31

amine ; a low overpotential metal such as nickel or platinum yields inter-
mediate reduction products such as phenylhydroxylamines or azoxy-
and hydrazo- compounds derived from phenylhydroxylamines (3).

The electrodes used for electrolytic oxidations are somewhat more
limited because it is difficult to obtain a stable anode potential with
many electrodes in the presence of a depolarizer. The potential
generally rises rapidly from the low value, at which the anode dissolves,
to the high value for passivity and oxygen evolution. However, since
platinum and sometimes gold are nearly always passive it is possible to
obtain graded potentials with these metals.

It is difficult to give definite rules concerning the efficiency of an
electrode for anodic oxidation. If the oxidative process is due to the
presence of oxygen in an active form it might be anticipated that a high
anode potential would indicate a more effective oxidizing power. This
does not seem to hold in all instances, for, as will be seen later, there are
processes which appear to be independent of the anode potential. In
these cases it is possible that the effective oxidizing agent may be some-
thing other than oxygen, perhaps hydrogen peroxide. The prime
prerequisite for a metal to be used as an anode is that it be essentially
passive in the electrolytes used. In the absence of halogens, metals
such as platinum or gold are usually not attacked and therefore are
suitable. These may on occasion be used with chloride providing the

ORGANIC ELECTRODE PROCESSES

amount is low. This also holds for anodes of cobalt and nickel provided
high current densities are used. At still higher current densities iron
can be utilized. However, once this metal has become passive it is
then possible to use the electrode at low current densities. Lead can
also be used in acid solutions of moderate concentration. In a sulphuric
acid medium lead first forms a sulphate coating which is converted
into an oxide coating. With an alkaline medium it is necessary to use
metals whose oxides are insoluble. Lead even behaves as an insoluble
electrode at high current densities in this medium, although at low
current densities it is not passive and will go into solution. In this type
medium platinum is as suitable as it is in an acid medium.

The second method, whereby the desired electrode potential may
be attained at the working electrode, is to utilize a high hydrogen over-
potential metal (mercury) for reductions, or a high oxygen overpotential
metal (platinum) for oxidations, and to control the applied voltage
across the cell so as to obtain the desired electrode potential. The
potential at the working electrode is measured against that of a standard
reference electrode with a vacuum tube voltmeter. This will be dis-
cussed further in Chapter II.

Often by utilizing the same electrode, the same medium and the
same temperature, it is possible to obtain different products from the
same starting material by merely changing the electrode potential.
This fact has been demonstrated by Haber (4) in his studies on the
reduction of nitrobenzene, in which it was shown that the first step is
the formation of nitrosobenzene, followed by the hydroxylamine and
finally aniline :

C 6 H 5 N0 2 ^ C 6 H 6 NO C 6 H 5 NHOH C 6 H 5 NH 2

Each of these steps is controlled by the potential at the cathode surface.
It has also been demonstrated that/>-aminoacetophenone can be reduced
at a mercury cathode to either the hydrol or the pinacol, depending upon
the potential at the mercury cathode (5), thus :

-NH 2 C 6 H 4 COCH 3 ?^- />-NH 2 C 6 H 4 CH(OH)CH 3

h / OH

2e / I

2/>-NH 2 C 6 H 4 COCH 3 -~~> I />-NH 2 C 6 H 4 C

\ CH 3 / 2

GENERAL CONSIDERATIONS

The stepwise reduction of 9-(o-iodophenyl) acridine (6) is another
example of the use of controlled potential to obtain desired partially
reduced compounds. This reduction proceeds according to the
following scheme :

2H+

26

The fact that the electrode potential is probably the most important
factor in determining the reaction occurring during electrolytic reduction
is not difficult to comprehend. It is known, for instance, that in a
reduction at a constant current density the initial potential will
increase as the reduction proceeds, as illustrated by the reduction of
J>-methoxyacetophenone (Table 1 : 4) (7). This means that if one

TABLE 1:4

Reduction of p-Methoxyacetophenone to its Pinacol in an Aqueous-
Methanolic Potassium Acetate Medium

Time (min.)

Ref Pot.
v. S.C.E.

c.d.amp.lcm*

o

2*40

0-288

2

2-46

3

2-48

5

2-50

7

2-58

9

2-60

1 1

2-79

13

2-86

1

i4'6

3-00

0-288

desires to attack the more easily reduced centre in a compound
having two reducible centres, under constant current density
conditions, one should choose an electrode of proper over-
potential to prevent possible reduction of both centres. In other

7

ORGANIC ELECTRODE PROCESSES

words electrolysis at a specific potential is equivalent to using one
reducing or oxidizing reagent, whereas constant current density
electrolysis is comparable to using a number of reducing or oxidizing
reagents in the same reaction vessel. The maximum potential obtain-
able would of course depend on the electrode material used.

It is not meant to imply by the above that constant current density
electrolysis is of no value. If one has a compound with one or more
reducible centres and a complete reduction is desired, it is usually of
no consequence what the electrode potential is, so long as it is greater
than the minimum potential required for the desired reaction. In these
cases constant current density electrolysis will give satisfactory results
with possible sacrifice in current efficiency.

(5) VOLTAMMETRIC CURVES

If an increasing potential is applied on a cathode in a medium which
does not contain a depolarizer, there will be a relatively small flow of
current (#), which can possibly be attributed to diffusion of hydrogen
from the electrode, or the presence of a trace of reducible material in
solution. When the potential reaches a certain voltage (b) the evolution
of hydrogen gas bubbles will be observed from the cathode, and the
current will increase rapidly (A in Fig. 1:1).

CURRENT DENSITY
FIG. I : i. Reference Potentials in Cathodic Reductions.

In the presence of a depolarizer which reduces relatively slowly as
many organic compounds do, a voltammetric curve such as B will be

8

GENERAL CONSIDERATIONS

obtained. A compound which is rapidly reduced will give a curve as
illustrated by C. From curves B and C it can be seen that by con-
trolling the potential at a particular point, one can control the reducing
power of the electrode. It is not enough to set the current density at a
particular value, for, as it was shown previously, removal of depolarizer
due to its reaction with hydrogen will cause the potential to increase,
with resultant hydrogen evolution. However, the current density can
be regulated so as to maintain a constant electrode potential.

(6) CATALYTIC EFFECT OF ELECTRODES

In general the overpotential of the electrode is a suitable indication
of its reducing ability. There are, however, occasions in which the
nature of the electrode material exerts a pronounced catalytic effect on
the reaction. In these instances a low overpotential electrode may be
as effective as, or perhaps more effective than a higher overpotential
electrode. As an example we might consider the cathodic reduction of
nitric acid to ammonia. If a relatively high overpotential cathode such
as amalgamated lead is used, a small percentage of ammonia is obtained
together with a large amount of hydroxylamine, whereas spongy copper,
which has a low overpotential, will give very high yields of ammonia (8).

Similarly, when ketones are reduced at cathodes of relatively high
overpotentials, the pinacol is generally obtained. If a cadmium
electrode, which has a relatively lower potential, is used, the more
highly reduced form, the hydrocarbon, is obtained (9). The ability
of high overpotential electrodes to reduce resistant compounds can also
be hindered by the presence of catalytic quantities of other metals such
as copper, platinum or silver in the medium or electrode. Therefore,
it is essential that the electrode be absolutely pure. This is the reason
mercury is usually suitable for reductions and platinum for oxidations,
since these metals can be obtained in an extremely pure state.

In some instances the catalytic effect of the anode metal may over-
shadow the consideration of potential. In the oxidation of iodate in
alkaline medium at a smooth platinum electrode, increase in electrode
potential results in increased yield of periodate. However, use of
a lead dioxide electrode results in a still better yield, even though
the anode potential of this electrode is somewhat lower (10). Therefore,
it is difficult to say in this case that the potential is the governing factor
in the process. Methyl alcohol, when subjected to anodic oxidation in
a dilute sulphuric acid medium at a smooth platinum anode, yields

ORGANIC ELECTRODE PROCESSES

formaldehyde with an 80 per cent efficiency. When platinized platinum
or lead dioxide anodes are used, the decrease in yield of formaldehyde
is accompanied by increased quantities of the more highly oxidized
products, formic acid and carbon dioxide (n). In many oxidations
better results can be obtained with platinized platinum or lead dioxide.
This is especially so in the oxidation of benzene derivatives.

(7) ALLOY ELECTRODES

As has been suggested previously, the use of certain electrodes,
though of lower potential, may reduce a compound to a greater degree
than that of a higher overpotential electrode. It has also been found
that alloy electrodes may be more efficient in certain instances.

The reduction efficiency of nitrobenzene in an alcoholic-alkaline
solution is increased from 58 to 72 per cent by the addition of 12 per
cent iron to a nickel cathode (12). The yield of pinacol from the
electrolytic reduction of acetone is tremendously improved by the use
of a lead-tin or lead-copper electrode (13).

(8) CATALYTIC EFFECTS OF ADDED SUBSTANCES

In addition to the catalytic effect of a particular electrode material
or an alloy electrode on the reaction, the addition of various salts to
the electrolyte may have an effect on the products obtained from an
electrolytic reduction. The addition of salts of lead and mercury to
low overpotential electrodes, such as nickel, will increase the efficiency
of the electrode. Also, certain metals may be deposited on the cathode,
react with the depolarizer, and then be redeposited from solution.
The reduction of nitro-compounds to their hydrazo-derivatives in a
basic medium is facilitated by zinc, lead or tin hydroxide dissolved in
the catholyte (14). The lead probably causes an increase in the over-
potential of the cathode, but with the zinc or tin salts the effects may be
partly due to the chemical action of the metal (15). Indigo is reduced
to indigo-white at a zinc cathode. This will also occur at other cathodes
in the presence of zinc salts (16). The same reduction can be accom-
plished chemically with zinc. Similarly, in the synthesis of methyl
anthranilate from the methyl ester of o-nitrobenzoic acid, a 40 per cent
yield i& obtained using a lead cathode. Addition of a catalytic amount
of stannous chloride to the electrolysis medium will increase the yield
to 90 per cent (17).

Electrolytic reductions performed at a mercury cathode in an

10

GENERAL CONSIDERATIONS

alkaline medium, or in the presence of an alkali salt, may be a result of
chemical reduction by the amalgam. Benzoic acid in an alkaline medium
can be reduced to the tetrahydro compound (18). The same reaction
can be accomplished chemically with sodium amalgam. Lob and Moore
(19), in their experiments on the controlled potential reduction of nitro-
benzene in an alkaline medium, reported that the addition of copper
powder increased the efficiency for the formation of aniline from 33-2
to 52-6 per cent. In this case it is doubtful if chemical action can
account for the reduction, even though phenylhydroxylamine can be
reduced to aniline with copper powder. Some German patents (20)
claim the addition of powdered metals, particularly copper when used
with a copper cathode, have facilitated the reduction of nitro-compounds
to their respective amines. Thus it has been possible to achieve the
difficult reduction of some azoxy compounds such as o-azoxychloro-
benzene to their respective amines.

Another type of catalytic effect which can, in some cases, produce
increased reduction efficiency is obtained by the addition of ions capable
of existing in two stages of oxidation; e.g. titanium, vanadium,
chromium, iron and cerium. These are sometimes referred to as
" hydrogen carriers. " As an example Ti+ 4 is reduced at the cathode to
Ti+ 3 , which, being a strong reducing agent, reacts with the substance
present in the solution and in so doing is reoxidized to Ti+ 4 . The
resultant ions are once more reduced cathodically, and the process goes
continuously.

An acid solution of chlorate is reduced much more effectively at a
platinum cathode if a small quantity of vanadic acid is present to
facilitate conversion to the chloride. The reduction of nitro compounds
to their respective amines is often facilitated by the addition of a tin
salt (21). The addition of antimonic or arsenious oxides to the sulphuric
acid medium used in the reduction of N,N-dimethylphenylacetamide
results in a considerable increase in yield of the phenyl ethyl amine
(22). The yield at a lead cathode increased from 41 to 80 per cent as
a result of using the oxide catalysts. The addition of iodine has been
reported to increase the speed of reduction of certain nitro compounds
to the amine (23).

There have been some reports on anti-catalytic effects of certain
substances. Ordinarily, use of a lead cathode will result in an excellent
yield of aniline from nitrobenzene. If it is desired to stop the reduction
at />-aminophenol, the efficiency of the lead cathode can be advan-

ii

ORGANIC ELECTRODE PROCESSES

tageously diminished by the addition of arsenic or bismuth powder or
copper sulphate to the catholyte (24).

Many electrolytic oxidations are also facilitated by the presence of
ions capable of existing in two valence states, and the action of these
" oxygen carriers " is comparable to that of the " hydrogen carriers "
previously mentioned. Potassium permanganate has been used as a
catalyst for the manufacture of saccharin by electrolysis of toluene
sulphonamide (25). The oxidation of anthracene to anthraquinone in
an acetone-sulphuric acid solution at a platinum anode takes place with
a 55 per cent efficiency. The addition of a cerium, chromium or
manganese salt results in an increase to an 80 per cent efficiency (26).
Similarly acetic acid has been obtained by oxidation of paraldehyde in
a dilute sulphuric acid medium, in the presence of salts of cerium,
manganese, molybdenum, or vanadium (27).

The effectiveness of various catalysts varies. In the oxidation of
naphthalene to naphthoquinone the catalytic effect increases in the
following order: manganous sulphate, dichromate and ferrocyanide,
potassium chlorate, vanadic acid, eerie nitrate (28). There have been
reported instances wherein the use of different catalysts has resulted in
different oxidation products. If a chromium salt is used as a catalyst
in the oxidation of toluene in a sulphuric acid medium, benzoic acid is
obtained. If a manganous salt is used the product obtained is benzal-
dehyde (29).

It has been reported that fluorides in the electrolyte have the effect
of increasing the overpotential at platinum, gold, silver and carbon
anodes (30). Therefore, their use has a pronounced effect on the
reaction process. Thus Skirrow (31) reportedly obtained a more
efficient oxidation of naphthalene and benzene in an aqueous hydro-
fluoric acid solution than in a medium containing a comparable quantity
of sulphuric acid. On occasion, perchlorates have been found to have
an effect similar to that of fluorides.

(9) NATURE OF MEDIUM

The products obtained as a result of anodic or cathodic electrolysis
are influenced to a marked degree by the nature of the electrolyte used.
In the bimolecular reduction of />-dimethylaminoacetophenone an acid
medium favours the high melting form, whereas a basic medium
favours the lower melting pinacol (32). It has also been reported that
when benzophenone is reduced in an aqueous-acidic medium the

12

GENERAL CONSIDERATIONS

pinacone, a rearrangement product of the benzophenone pinacol, is
obtained. In a basic medium the pinacol is obtained (33).

In an acid solution, acetaldehyde, when oxidized yields acetic acid ;
in an alkaline medium formic acid is obtained (34). Aniline in a dilute
acid medium is oxidized to aniline black or quinone, whereas in an
alkaline medium, azobenzene is obtained (35). The use of a con-
centrated sulphuric acid medium results in a formation of />-amino-
phenol.

In general there are a limited number of solvents that may be used
for anodic or cathodic processes. In most cases water alone cannot be
used because of the low solubility of the organic compound in this
medium. However, in many instances it is possible to obtain satisfactory
concentration of depolarizer by mixing with other solvents. Electrolytic
oxidations can generally be performed in aqueous solutions of methanol
or acetic acid, absolute methanol, glacial acetic acid, and mixtures in
varying proportions of acetic acid and ethylene dichloride or pyridine.
Reductions of compounds not too soluble in water can be performed in
such media as methanol, ethanol, acetic acid, acetic acid-benzene
mixtures, dimethylformamide and dioxane. It should be remembered
that if it is possible to have some water in the solvent system the
resistance of the needed solution will be diminished, effecting a saving
in the power required to accomplish the reaction and in the power
dissipated as heat. If the latter is excessive, it may in some cases cause
undesirable chemical side reactions. If it is necessary to use a non-
aqueous medium for an oxidation or reduction, it is possible, by
utilizing proper membranes to separate the anolyte from the catholyte,
to use either an aqueous catholyte or anolyte respectively. This aspect
will be discussed in Chapter II.

There are many electrolytes which can be used in an aqueous
medium. In the nonaqueous media mentioned above salts such as
lithium chloride, lithium nitrate, potassium acetate have been found
most suitable. Acids, such as sulphuric, are generally as satisfactory
in nonaqueous media as they are in an aqueous medium. In addition
to potassium hydroxide there are organic bases, such as tetramethyl-
ammonium hydroxide, which are suitable for basic electrolysis in non-
aqueous media. McKee and Brockman have developed a novel method
for increasing the solubility of organic compounds (36). They obtain
the desired solubility by " salting-in " the organic compound with
3aturated solutions of sodium />-toluene-sulphonate or sodium cymene-

13

ORGANIC ELECTRODE PROCESSES

sulphonate. In addition to these saturated salt solutions being good
solvents they are also excellent conductors.

Wetting agents, such as the " aerosols," have on occasion been found
satisfactory for solubilizing the organic compound in an aqueous
medium. The use of nonaqueous solvents may be avoided by utilizing
a vigorously agitated suspension of the organic compound. Although
satisfactory results can be obtained in many instances, the efficiency of
this method is considerably lower than that employing a solution of the
substance.

(10) CONCENTRATION OF DEPOLARIZER

In general the concentration of the depolarizer does not affect the
nature of the products obtained. However, one would anticipate that
the formation of bimolecularly reduced compounds would be favoured
by high concentration, as the chance of collision between the reactive
specie would be increased. The reaction rate is affected by increasing
the concentration of depolarizer. In this manner a higher current
density can be maintained with good process efficiency. In a reversible
process the " limiting current density " at which one attains 100 per
cent efficiency is directly proportional to the concentration of the
depolarizer. For an irreversible process this can only be considered as
an approximation, a guide in deciding on the concentration of depolarizer
to use for a given current density, in order to try to approach a maximum
efficiency.

In order to further understand the evaluation of " limiting current
density " for a reaction at a particular concentration, it might be
desirable to briefly describe the so-called " diffusion layer " which exists
at the electrode surface. The concentration of depolarizer which is
being oxidized or reduced at the electrode surface differs from that in the
bulk of the medium. There will be, therefore, a concentration gradient
in passing from the electrode to the electrolyte. The so-called
" diffusion layer, " then, is the surface layer in which most of the
concentration change occurs.

As mentioned previously, it is possible to obtain an approximate
" lirAiting current density " for irreversible reactions if certain assump-
tions\re made. The first is that the thickness of the diffusion layer is
0*05 cm. thick (37), bearing in mind that this will decrease with increased
temperature and agitation. The second assumption is that the diffusion
coefficient of the depolarizer is 0-6 to 0-8 cm. 2 day- 1 , which is a value

H

GENERAL CONSIDERATIONS

obtained for nonelectrolytes capable of reversible oxidation or reduc-
tion, e.g. quinone ^ hydroquinone. With this information we can
utilize the equation developed by Nernst (38) from Pick's Law

_ ADcn
~ 0-895 i

where 8 is the thickness of the diffusion layer, A the area of the electrode,
D the diffusion coefficient, i the current, c the concentration of de-
polarizer, and n the difference in charge between the oxidized and
reduced form. Substitution of 0-05 cm. for 8 and expression of i\A
per sq. cm. as the current density (c.d.) we arrive at the following
equation :

c.d. (amp./cm. 2 ) = 0-0223 DCn ( 2O c -)

with concentration (c) given as gram-moles/litre. This " limiting
current density " will increase at approximately 2-5 per cent per degree.
Agitation also has a pronounced effect on the limiting current. Usually
the greater the agitation rate the greater the limiting current.

The optimum " limiting current density " is only for a specific
concentration of depolarizer, and therefore, to maintain maximum
efficiency, the depolarizer should be replenished as it reacts, or the
efficiency will diminish. Performing the reduction at a controlled
potential will accomplish essentially the same purpose (39). As the
depolarizer is reduced the current drops until a plateau is reached,
which indicates essentially a completed reaction.

An illustration of the use of the concept of " limiting current density "
in preparative studies can be found in a recent report on the reduction
of N,N-dimethylaminoethyl tetrachlorophthalimide to the respective
isoindoline. A comparison was also made between the optimum current
efficiency obtained using constant current density and that obtained at a
controlled electrode potential. With the former a 45 per cent current
efficiency was obtained, whereas the use of controlled potential increased
the efficiency to 77 per cent (40).

Another method which can be utilized to improve the efficiency of
the process is to increase the area of the electrode either by increasing
its size or by utilization of gauze electrodes.

(ii) AGITATION

A necessary factor in maintaining a high concentration of depolarizer
at the electrode surface is adequate stirring of the electrolyte. The

ORGANIC ELECTRODE PROCESSES

more efficient the agitation the higher is the " limiting current density "
which can be used for maximum reduction or oxidation efficiency with
depolarizers which react rapidly. For more slowly reactive species the
rate of agitation will not have as pronounced effect on the maximum
current for optimum efficiency. However, some agitation is necessary
here in order to bring unreacted depolarizer to the electrode surface.

(12) TEMPERATURE

The temperature of the catholyte or anolyte during a cathodic
reduction or anodic oxidation has an influence on a number of factors
controlling the process. Firstly, an increase in temperature will decrease
the overpotential of the electrode. This decrease will vary somewhat
with the electrode material, but generally this variation is of the order
of 0-02 to 0-03 volt for each 10 C. rise. Thus electrodes such as
platinum, silver, nickel and copper, whose hydrogen overpotential
values are in the vicinity of 0*15 volt at 20 C. and may be effective for
certain reductions, will have essentially no action and a neglible over-
potential when used at 100 C. With a high overpotential electrode,
such as mercury, even at elevated temperatures the overpotential will
be high enough to accomplish many difficult reductions. This tempera-
ture effect on overpotential has been utilized to stop the course of a
reaction at an intermediate point. Thus if ethanol is subjected to anodic
oxidation in dilute sulphuric acid at elevated temperatures a satisfactory
yield of acetaldehyde is obtained. However, if the same reaction is
performed at 30 degrees, the more highly oxidized product, acetic acid,
is obtained (41). These results may be due partly to the increase of
overpotential with the decrease in temperature, and partly to the lower
volatility of the acetaldehyde at this temperature which permits its
further oxidation. A second factor to consider is that at elevated
temperatures it is possible to use effectively higher current densities.
This will enable attainment of shorter electrolysis periods with better
yields, especially in instances in which prolonged exposure of the
reaction product will cause its destruction (42). Since the rate at which
the unreacted specie will come in contact with the electrode surface
increases as the temperature rises, the efficiency of a slow process may
be considerably improved by temperature elevation. For example,
the efficiency of oxidation of an alkaline formate solution at an iron
anode is increased twofold when the temperature of the medium is
maintained at 75 C. instead of at 16 C. (43.) We must not overlook

16

GENERAL CONSIDERATIONS

the possibility of an adverse effect on the reaction being directly due
to an increased electrolysis temperature. The reduction of a ketone to a
pinacol proceeds quite satisfactorily at the lower temperatures when
using an acidic medium. However, at elevated temperatures there is
an excellent chance that the desired pinacol will rearrange to the
pinacone. Another possibility which must be recognized is that the
starting material may be unstable at elevated temperatures. Thus if a
phthalimide were subjected to cathodic reduction in an acid medium
under such conditions, the reduction would be preceded by hydrolysis,
with the net result being a negligible yield of the isoindoline.

(13) SUPERIMPOSITION OF ALTERNATING CURRENTS

There is little that can be said concerning this aspect at the present
time, except that at one time it was thought that superimposing an
alternating electric current on a direct current electrolysis would tend
to lower the overpotential of the electrode (44). Reitlinger claimed
that by utilizing an alternating current in this way ethyl alcohol could
be oxidized to acetaldehyde without being oxidized further to acetic
acid. He also reported that />-benzaldehyde sulphonic acid would be
obtained from />-toluenesulphonic acid in this manner (45). An
attempt was made to improve the yield of terephthalic acid obtained by
anodic oxidation of p-toluic acid by superimposing an alternating current.
Normally a great deal of carbon dioxide and water is obtained as the
side product. This modification in the process did not have the desired
effect (46).

(14) INFLUENCE OF TIME ON THE ELECTROLYTIC PROCESS

If the electrolysis is performed under conditions in which the
electrode potential is controlled at a predetermined value, then we need
not fear too prolonged exposure of the product to the current. It has
been pointed out previously that, in general, the current will drop until
a plateau is reached, at which time the electrolysis is essentially com-
pleted.

In using constant current density conditions the electrolysis time
should be kept at a minimum when the products of electrolysis are
chemically unstable, or when further oxidation or reduction may occur.
The latter may be due to an increase in electrode potential at the working
electrode with decreasing depolarizer concentration. In addition we
must not overlook the possibility that prolonged electrolysis may

ORGANIC ELECTRODE PROCESESS

result in partial destruction of the reaction product due to its migration
to the opposite electrode. Generally it is best when exploring a new
reaction to allow it to proceed for its theoretical time (T), as computed
from the following equation based on Faraday's Law :

_ MW 96,500 n
VIM -=- ^7

Where MW is the molecular weight of the depolarizer, g the weight of
reacting substance in grams; n the number of electrons involved in the
process, and i the current in amperes. The results obtained then can
be used as a guide to future efforts. Since the potential of the electrode
increases with electrolysis time, it may be, on occasion, that a prolonged
period of electrolysis will result in an increased efficiency. Also, cases
where the electrolysis is performed using a suspension, prolonged
electrolysis may be necessary to attain satisfactory yields because of
low efficiency.

REFERENCES

(1) For a complete discussion see Bockris, J. Amer, electrochem . Soc., 1951,
9t I53C.

(2) LeBlanc, Z. physikal Chem., 1891, 8, 299.

(3) Elbs, Chem.-Ztg , 1893, 17, 209; Z. Elektrochem., 1896, 2, 472; 1901, 7,
589.

(4) Haber, Z. Elektrochem., 1898, 4, 506; Haber and Schmidt, Z. physikal
Chem., 1900, 32, 271.

(5) Allen and Corwin, J. Amer. chem. Soc., 1950, 72, 114.

(6) Lingane, Swain and Fields, J. Amer. chem. Soc., 1943, 65, 1348.

(7) Allen and Fearn, unpublished data.

(8) Tafel, Z. anorg. Chem., 1902, 31, 289.

(9) Tafel, Ber., 1909, 42, 3146; Tafel and Schepss, ibid, 1911, 44, 2148;
Z. Elektrochem. 1911, 17, 972; Schepss, Ber., 1913, 46, 2564; Swann,
Trans. Amer. electrochem. Soc., 1932, 62, 177.

(10) Muller, Z. Elektrochem., 1904, 10, 50.

(n) Elbs and Brunner, Z. Elektrochem. , 1900, 6, 604

(12) Glasstone, Trans. Faraday Soc., 1924, 19, 574.

(13) D.R.P., 252759, 1911; 306304, 306523, 324919, 324920, 1918.

(14) B.P., 15750, 1915-

(15) Buchner, Z. Elektrochem., 1903, 9, 646.

(16) Binz, Z. Elektrochem. 1898, 5, 103; 1899, 6, 261.

(17) I^ishin, Babievskii and Izgaryshev, Doklady Akad. Nauk, U.S.S.R.,
1955, 104, 744.

(18) Mettler, Ber., 1906, 39, 2933; Baur and Mttller, Z. Elektrochem., 1928,
34, 98; Somlo, ibid, 1929, 35, 264, 769.

(19) L6b and Moore, Z. physikal. Chem., 1904, 47, 418.

18

GENERAL CONSIDERATIONS

(20) D.R.P., 130742, I3I44>

(21) D.R.P., 116942, 1899; 121835, 122046, 1900.

(22) Kindler, Ber., 1923, 56, 2063.

(23) Kirkhgof, Khim. Farm. Prom., 1933, 326.

(24) D.R.P., 295841, 1915-

(25) Tafel and Stern, Ber., 1900, 33, 2224; Tafel and Emmert, Z. physikal
Chem.y 1906, 54, 433; Zerbes, Z. Elektrochem., 1912, 18, 619.

(26) Marie, Compt. rend., 1905, 140, 1248.

(27) McKee and Brockman, Trans. Amer. electrochem. Soc. y 1932, 62, 203.

(28) Ahrens, Z. Elektrochem., 1896, 3, 99; Ogura, Mem. Coll. Set. Kyoto, 1929,
i2A, 339.

(29) U.S.P., 867575, 1907.

(30) Isgarischev and Stepanov, Z. Elektrochem., 1924, 30, 138.

(31) Skirrow, Z. anorg. Chem., 1903, 33, 25.

(32) Allen, y. chem. Soc., 1951, 1598.

(33) Swann, Trans. Amer. elektrochem. Soc., 1933, 64, 313; Eng. Exp. Sta.
Bull. Univ. Illinois, No. 236.

(34) Coehn, Z. Elektrochem., 1901, 7, 681; Heimrod and Levene, Ber., 1908,
41,4443.

(35) Moser, Elektrolytischen Prozesse, 1910, 68; GoppelsrSder, Ding. Polyt.J.,
1876, 221, 75; 1877, 223, 317, 634; Compt. rend., 1875, 81, 944; 1875,
82, 331, 1199; Coguillion, ibid., 1875, 81, 408; 1875, 82 228; B.P.,
573, 1902.

(36) McKee and Brockman, Trans. Amer. elektrochem. Soc., 1921, 39, 441.

(37) Glasstone and Reynolds, Trans. Faraday Soc., 1933, 29, 399; Glasstone,
Trans. Amer. elektrochem. Soc., 1931, 59, 277.

(38) Nernst, Z. physikal. Chem., 1904, 47, 52.

(39) Lingane, Swain and Fields, loc. cit., ref. 6; Allen, Anal. Chem., 1950,
22, 804; y. Amer. chem. Soc., 1950, 72, 3797; ibid., 1951, 73, 3503;

y. org. Chem., 1950, 15, 435; J. chem. Soc., 1951, 1598; Allen and
Corwin, y. Amer. chem. Soc., 1950, 72, 114, 117; Levine and Allen,
J. chem. Soc., 1952, 74* 254; Allen, Fearn and Levine, ibid., 1952, 2220;
Allen and Steinman,^. Amer. chem. Soc., 1952, 74, 3932.

(40) Allen and Ocampo, J. Amer. electrochem. Soc., 1956, 103, 452.

(41) Schlotter, Diss., Munich, 1902; Askenasy, Leiser and Grunstein, Z
Electrochem., 1909, 15, 846; Arnstein, Trans. Amer. elektrochem. Soc.,
1924, 45, 129 Discussion', D.R.P., 274032, 1911.

(42) Elbs and Illig, Z. Elektrochem., 1898, 5, in.

(43) Fritz-Foerster, Elektkochemie, 1923, 816.

(44) Rothmund, Ann. Physik, 1904, 15, 193; Bennewitz, Z. physikal Chem.
1910, 72, 216; Glasstone and Reynolds, Trans. Faraday Soc., 1932
28, 582; ibid., 1933, 29, 399.

(45) Reitlinger, Z. Elektrochem., 1914, 20, 261.

(46) Allmand and Puttick, Trans. Faraday. Soc., 1927, 23, 641.

CHAPTER II

INSTRUMENTATION AND TECHNIQUES
OF ELECTROLYSIS

INSTRUMENTATION

(i) CONSTANT CURRENT AND MANUAL-CONTROLLED
POTENTIAL DEVICES

THE equipment required for electrolytic oxidations or reductions
can be constructed from readily available components. The complexity
of the equipment will depend to a great extent on the desired type of
control. If a constant current electrolysis assembly is desired, the
equipment will consist of a source of direct current, a voltmeter, ammeter
and the necessary rheostats to control the output voltage.

It would of course be an ideal situation if it were possible to obtain
the direct current from the mains. This, unfortunately, is not possible
in most instances, and therefore it is necessary to depend a great deal
on batteries, a rectifier unit, or a generator. Generally, batteries are
not suitable because they are short-lived, dry cells needing replacement
and wet cells recharging. Therefore it is best to obtain a rectifier unit
which is reasonable in cost for the low current and voltage outputs
required in small scale laboratory preparations.

In obtaining a rectifier one should give consideration to the " ripple
voltage." It is absolutely essential that this voltage does not exceed the
desired range of reference potential if the unit is also to be used for
controlled potential electrolysis at the working electrode. Simply stated,
if a po^ver supply has a maximum output of 30 volts and 10 amperes
and one wishes to control at maximum output to 0*03 volt, the
ripple voltage cannot exceed o-i per cent. Of course with this same
unit narrower limits can be attained when operating at a lower output.
As the power fleeds increase, the cost of " low-ripple " rectification

20

INSTRUMENTATION AND TECHNIQUES OF ELECTROLYSIS

becomes quite expensive and it is at this point that consideration should
be given to a direct current generator.

There are occasions where it will be desired to investigate the
possibilities of controlled potential electrolysis, either because the
molecule under study has more than one reactive oxidative or reductive
centre and selectivity is desired and/or one desires to obtain the
maximum process efficiency. As mentioned in the previous chapter,
selectivity may also be obtained by utilizing the electrode of desired
overpotential characteristics and conducting the electrolysis at constant
current density, with the possibility of adversely affecting the current
efficiency of the process. In this instance a mercury or lead cathode
may be used for reductions, or platinum for oxidations, and the applied
voltage controlled so as to maintain the potential at the working electrode
surface constant v. a standard reference electrode. This reference
potential can be measured by a vacuum tube voltmeter, a pH meter
with a voltage scale or a potentiometer connected between the reference
electrode and the working electrode. The essential prerequisite of this
measuring circuit is that it does not draw an appreciable current from
the system, as this would result in an ambiguous reference potential
value. Such a simple circuit as described for manual control of the
reference potential is illustrated in Fig. n: i. It however suffers from

REF. ELECTRODE

FIG. II : i . Simple Electrolysis Circuit.

the disadvantage of requiring constant supervision and thus, if con-
siderable work is contemplated, an automatic device should be utilized
for controlled potential electrolysis.

(2) AUTOMATIC CONTROLLED POTENTIAL DEVICES

In the past, many instruments for controlled potential electrolysis
have been described (i). Hickling's potentiostat, since superseded by
more suitable designs, was a major contribution to the field of controlled

21

ORGANIC ELECTRODE PROCESSES

potential electrolysis, for it was the first such instrument which achieved
automatic control at a predetermined potential. Prior to its conception,
it was only possible to change or maintain the potential by altering either
the applied voltage across the cell or the current density. The main
disadvantage of Hickling's early instrument was the fact that a maximum
of only 0-3 ampere was obtainable; this imposed a limitation on its use
in preparative organic electrochemistry. The same disadvantage was
present in many subsequent designs. However, with some instru-
ments it is possible by modification of the described circuit to obtain
currents larger than 5 amperes (2).

The limitations imposed by prior instruments were taken into
consideration when a Redoxotrol utilizing the electronically con-
trolled amplidyne generator as a power source and control unit was
constructed (3).

Basically, the amplidyne is an externally driven d.c. generator.
However, its unique use of a short circuit and compensating winding
creates a precise electrical balance so that the smallest signal will control
a relatively large power output (4). Essentially this instrument will give
automatically controlled impressed potential to secure constant cathode
or anode potentials with respect to a reference electrode. Its method of
operation is illustrated by reference to the schematic diagram shown in
Fig. H : 2.

The desired electrode potential is obtained by opposing the E.M.F.
of the reference electrode (EscE) with an E.M.F. from the reference
supply (ERS). The error signal (EES) equals zero when the (RS) equals
(EscE). When the (EscE) changes, due to a change in the amount of
depolarizer present in solution, there will be a flow of voltage across the
error signal resistor (ES) which in turn will be transmitted to the
amplifier, resulting in a change in applied voltage output by the
generator. This instrument can deliver 20 amperes and 75 volts. The
control sensitivity is 0-005 v lt an d response is almost instantaneous
with no hunting. Generally if one uses an aqueous or predominantly
aqueous system for electrolytic oxidations or reductions, a reference
potential range of 2 volts is sufficient. As soon as hydrogen evolution
takes glace at the cathode during reductions, or oxygen evolution
at the anode during oxidations, this potential becomes meaningless
and it is just as satisfactory to operate at constant current density.
However, as is the case with most organic compounds, their solubility
in an aqueous medium is limited, and therefore it is often quite

22

INSTRUMENTATION AND TECHNIQUES OF ELECTROLYSIS

GENERATOR

X

VAC.TUBE
VOLTMETER

FIG. ii : 2. Operational Schematic of Redoxotrol.

necessary to use a mixed aqueous-nonaqueous medium such as aqueous
alcohol or aqueous acetic acid with a suitable electrolyte such as a mineral
acid, inorganic or organic bases, or salts such as potassium acetate or
lithium chloride. Often in more difficult cases it is necessary to use
alcohol, acetic acid, or even a mixture of ethylene dichloride and acetic
acid with a suitable electrolyte. Therefore, as we go further away from
the conductivity of an aqueous system, the reference potential attainable
at the working electrode will be much greater than in an aqueous system
before gas evolution occurs. Hence an available reference potential
range of 6 volts was incorporated in the design of this instrument.

Since the original publication, the circuit has been modified (Fig.
11:3) so as to eliminate the small error obtained when the current-
carrying line was used also to carry the signal. A more precise
balancing circuit was designed, batteries eliminated from the reference
voltage supply, and replaced by a stable voltage supply, and necessary
switching incorporated to enable use of the instrument for electrolytic
oxidations. The assembled Redoxotrol Model II is shown in Fig. n: 4.

23

ORGANIC ELECTRODE PROCESSES

FIG. II : 4. Redoxotrol Model II.

A second instrument has been constructed which embodies many of
the features of the original instrument, with the added advantage of
being able to deliver higher currents and voltages. The basic system

24

INSTRUMENTATION AND TECHNIQUES OF ELECTROLYSIS

is essentially a closed loop control system consisting of a controlled
generator, a d.c. amplifier, a reference potential supply, and a sampling
circuit. Since the standard d.c. generators have too long a response
time the amplidyne generator was used in this instrument also. This
amplidyne has a rated output of 24 amperes at 125 volts, and the two
control fields have a resistance of 900 ohms each. The response time
is approximately one-tenth that of a conventional generator. A power
change of o.i watt per field will produce an output change of 600 watts ;
this represents a power gain of 6,000. A 44o-volt, 6o-cycle, 3 -phase,
5 h.p. induction motor drives the generator.

A cascaded voltage regulator tube system provides a well regulated
reference voltage, stable over long periods. The reference potential is
variable from o to 6 volts through a lo-turn, 2OO-ohm Helipot and is
indicated on the reference potential meter, whose accuracy is 0-5 per
cent. In order to make it possible to obtain a voltametric curve for a
preliminary evaluation of the electrochemical behaviour of a new
compound, the reference voltage potentiometer is connected to a syn-
chronous motor which drives at a constant rate of 0-3 volt per minute.

Because certain oxidations require reference potentials which are
less positive than a reference electrode, such as the standard calomel
electrode, a " voltage-subtractor " circuit was included in the design of
the reference voltage supply. By proper switching, these opposing
potentials can be set on the reference voltage meter (Mi) (Fig. 11: 5)
and is variable from o to 0-5 volt.

The stability and gain are taken care of by use of a balanced two-stage
amplifier which consists of two 6AU6 (VT3, VT4) pentodes for the
input stages and two 6L6 (VTi , VT2) for the output stages. The control
fields of the amplidyne form the plate loads for the 6L6 tubes. The
amplifier is balanced (balanced condition indicated by zero output) by
means of a potentiometer (Pi), in the plate circuit of the output tubes.
This balanced condition is extremely critical, and in order to compensate
for any drift that may occur during operation, a self-balancing circuit
is incorporated. This connects the input of the amplifier to the output
through a 10 M? resistor. Thus any unbalanced condition is trans-
mitted through the closed loop in such a phase as to cancel. In the
design of any d.c. amplifier the problem of coupling together stages
that have a common plate supply is always present. It was found
desirable to use a common plate supply and to couple the stages
together without a reduction in gain. This was accomplished

25

ORGANIC ELECTRODE PROCESSES

PI*'

AMPLIDYNI
BALANCE

Z50V

FIELD !
N2 U <

I

>

>R1
>

.

y

VTI/JL\ /j

7> R5

FIG. ii : 5 . Schematic Circuit Diagram of Model III Redoxotrol.

26

INSTRUMENTATION AND TECHNIQUES OF ELECTROLYSIS

Components

Ri iK, 25 W, W.W. Ci, C2 .02 [if, 600 V d.c

R2 350 i), 25 W, W.W. C 3 o.i |if,6ooVd.c

R3 .68 Ma, i W, Comp. C4 i .5 (if, 600 V d.c

R4 .68 MO., J W, Comp C$ 2.0 (if, 600 V d.c.

R 5 100 K, i W, Comp. VTi, VT 2 6L6

R6 100 K, i W, Comp. VT 3 , VT 4 6AU6

R7 10 K, 5 W, W.W. VTs, VT6, VT7 OC 3 , VR-ios

R8 2.2 K, i W, Comp. VTg 6SC7

Rg 47 K, iW, Comp. Mi 0-1.5, 3.0, 7.5 V d.c.

Rio 47 K, | W, Comp. M2 o-i, 10, 50 Amps.

RII .47 Ma, J W, Comp M3 0-30, 75, 150 V d.c.

Ri2 .1 Mft, i W, Comp. Rec. 2 Pen 100 MV Recorder.

Ri3 820 O, 4 W, Comp. Same range as M2 and M3

Ri4 .27 MO., \ W, Comp Mot.i 5 H.P., 440 V, 6o~, 3^

Ri5 10 Ma, \ W, Comp. Gen.i 3 KW, 24 A, 125 V Ampli-

Pi 10 K, 4 W, 3 Turn dyne

P2 o.i Ma, 2 W, Comp. Mot. 2 115 V, 6o~, i<f> Synch.

P 3 i .o MO., 2 W, Comp. Motor, 15 RP-HR., 3 W

?4 200 O, 4 W, 10 Turn

P5 i oo Q, 4 W, 3 Turn

P6 10 Q, 4 W, 3 Turn

S-i SPDT

NOTES :

1. 440-Volt line was stepped down to 115 volts and stabilized with a
constant voltage transformer.

2. A timer is provided to shut off all power at the end of a timed run.

3. One 250-volt, 2OO-mA power supply provides power for the VTVM
and amplifier.

4. A separate VR tube supply provides the regulated reference voltage.

5. Separate switches are provided for selecting the applied voltage and
cell current ranges.

6. A switch connects Mi across P6 for setting the voltage added and
between P$ (centre tap) and the -f of the reference voltage for setting the
voltage subtracted.

7 . Fields of amplidyne are interchanged for oxidations

8. P6 is always set to zero for oxidations

ORGANIC ELECTRODE PROCESSES

through the use of voltage regulator tube coupling VTs (003) and VT6
(OC3). A single 25O-volt plate supply is used. As stability is often
difficult to achieve in a closed loop system, two stabilizing circuits have
been incorporated into the design. The stability control (P$) is connected
in series with a condenser (C3), the electrolysis cell, and the input
resistor (Ri4) of the amplifier. The circuit eliminates very rapid
changes which could cause instability in operation. To prevent hunting
of the amplidyne an " anti-hunt " circuit comprising a resistor (Pz) and
a capacitor (5) was connected between the two stages.

A gain of approximately 1-5 X io 7 is obtained from the amplifier.
This, added to the gain of the amplidyne, gives an over-all system gain
of 9 X io 10 which is sufficient to maintain the electrode potential within
maximum limits of i io millivolts over the entire voltage span.

The desired electrode potential is obtained by placing the standard
reference electrode against the surface of the cathode for a reduction
and the anode for an oxidation. The E.M.F. of the reference electrode
is opposed by the reference potential developed across (P4). This
reference potential is read on voltmeter (Mi) as previously described.
The off-balance between the E.M.F. of the standard reference electrode
and the reference potential is developed across resistor (Ri4). This
off-balance is referred to as the error signal, and is the input to the
amplifier. When the error signal is zero the amplifier is balanced and
the control fields of the amplidyne conduct equal and opposite currents,
resulting in zero output from the amplidyne. An error signal will
develop any time there is a difference between the electrode potential
and the reference potential. The error signal is amplified and drives
the control fields of the amplidyne in opposite directions. In turn, the
control fields will cause the amplidyne to increase or decrease in output
whichever is necessary to produce an electrode potential equal to the
reference potential.

In addition, the instrument also incorporates a Leeds and Northrup
two-pen Speedomax recorder which simultaneously plots cell current
and applied voltage v. time. A one-tube (VTg-GSCy) vacuum tube
voltmeter circuit monitors the error signal. The completed Redoxotrol
Model III is shown in Fig. n : 6.

An apparatus for automatic control of cathode potential published
by Palmer and Vogel (5) is of interest, firstly, because all components
are obtainable in Great Britain, and, secondly, it will deliver up to io
amperes and 20 volts, making it suitable for many organic reductions.

28

INSTRUMENTATION AND TECHNIQUES OF ELECTROLYSIS

FIG. II : 6. Redoxotrol Model HI.

It appears that with very little modification this same instrument can
also be used for electrolytic oxidations. The schematic drawing of their
assembled apparatus is shown in Fig. n : 7.

The operation of the instrument depends on the detection of a d.c.
" error- volt age " between the cathode and the standard reference
electrode. The parts of the instrument include a d.c. output unit
operating from the a.c. supply line which is used to supply the necessary
current and voltage to the electrolysis cell; a motor driven " Variac "
transformer to control the a.c. input to the d.c. output unit ; a potentio-

29

ORGANIC ELECTRODE PROCESSES

3

2
I

I

$
^
s
2

I

2

fe

^

3

O

INSTRUMENTATION AND TECHNIQUES OF ELECTROLYSIS

meter that can be set to the desired potential drop between the cathode
and the potential of the reference electrode ; and an amplifier unit which
operates at i millivolt d.c. and is connected to a reversible motor
controlling the " Variac " transformer. The cathode reference potential
is measured by a vacuum tube voltmeter. A voltage-stabilized power
supply is also utilized to supply the necessary power for the amplifier
unit and the vacuum tube voltmeter.

(3) MEMBRANES

As there are a number of materials available which can be used as
membranes to separate the catholyte from the anolyte, it is usually no
great problem to find one suitable for the experimental conditions. The
ideal membrane should have the following characteristics : firstly, it will
offer little resistance to the passage of current ; secondly, it will offer
a maximum resistance to the passage of the depolarizer ; thirdly, it will
not clog during prolonged electrolysis; and, lastly, it will be able to
withstand the medium and elevated temperatures, if necessary, required
by the experiment. Unfortunately, the " ideal " membrane is not as
yet available, but there are many suitable substitutes which the
author has found satisfactory. These will be given together with a
discussion of their limitations.

Cellophane sausage casing, which can be obtained in tube form, is
very suitable for electrolysis when used in an aqueous acid or neutral
medium. It can also be used with a 50 per cent methanol or ethanol
medium. However, if the alcoholic concentration is increased beyond
this, the membrane will deteriorate. This also occurs in a basic medium
and at temperatures over 55 C. Regardless of these limitations, this
type of membrane is extremely useful at times. In addition it has the
advantage of offering little resistance to the passage of current and high
resistance to the passage of the depolarizer. An added advantage is
that it is probably the most inexpensive membrane material available,
and therefore can be discarded after each use.

Another type of membrane material which is probably used to a
greater degree than any other is some form of porous porcelain. The
writer has found the most desirable to be " Alundum " (6). This can
be obtained in various sizes and shapes. Of the number of mixes
available, the RA 1143 mix has been found most satisfactory. Its
advantages are that it can withstand almost any medium, at any
temperature, and its resistance to the passage of an electric current is

ORGANIC ELECTRODE PROCESSES

quite low. Because it does not offer as great a resistance to the passage
of the organic depolarizer as does cellophane, there is always danger
of loss of organic material. For experimental purposes this is not a
serious consequence, if electrolysis periods are not longer than eight
hours. Beyond this there is a tendency for the membrane to clog,
with a resultant increase in electrical resistance. These membranes
are best cleaned by first washing well with water, or any other suitable
solvent, then soaking overnight in a 50 per cent solution of nitric acid.
This step can be accelerated by heating the medium to 100 C. and
permitting the membrane 'to soak therein for one hour. After exposure
to the nitric acid medium the membrane is washed exhaustively with
water. Treatment of the membrane in this manner makes it possible
to use it quite a number of times before it need be discarded.

Other materials which have been found as suitable as the porous
porcelain type membrane are fine mesh sintered glass disks, and glass
cloth. The former has the disadvantage of requiring a permanent
mounting in the cell, and the latter, being nonrigid, requires a special
type cell, as illustrated by Fig. n: 13.

The membrane materials discussed previously are generally suitable
for most purposes. Another type of membrane material is the " ion-
exchange " resin membrane sold by one organization under the trade
name of " Amberplex " (7). This membrane has the electrical con-
ductivity of a o-i N potassium chloride solution and is extremely
resistant to the diffusion of the organic depolarizer. Within the writer's
experience no medium has been found which could not be used with
this membrane. However, one disadvantage is that it cannot be
used for electrolysis at temperatures over 70 C. This membrane
shows its tremendous advantage under extremely difficult conditions.
We may take as an example a situation in which the organic de-
polarizer requires a nonaqueous medium. In such an instance it
would be necessary, if an " Alundum " membrane were used, to
employ essentially the same type medium in the other compart-
ment. This would naturally result in a relatively high resistance
between the anode and cathode contributed to by the poor conductivity
of the medium. Using an " ion-exchange resin " type membrane it is
possible, due to the nature of the membrane material, to use an aqueous
medium in the nonworking compartment. Thus a sterol can be
dissolved in a mixture of ethylene dichloride and acetic acid with an
acid such as sulphuric acid, and the whole placed either in the anode or

INSTRUMENTATION AND TECHNIQUES OF ELECTROLYSIS

cathode compartment, depending on the type of reaction desired. In
the other compartment can then be placed an aqueous solution of the
electrolyte without concern to clogging of the membrane or appreciable
diffusion of the aqueous solution, with resultant precipitation of the sterol.

The only prerequisite for using these membranes is that a new
membrane must be properly prepared prior to its use, by soaking for at
least two hours in the experimental medium. Thereafter it may be
used any number of times with the same medium without removing it
from its mounting in the cell. A cell designed for use with this as well
as other types of non-rigid membranes will be illustrated in another
section of this chapter.

There are occasions, of course, when a membrane is unnecessary.
If, in an oxidation, the product formed is not readily reduced, a low
overpotential cathode such as platinum may be used, and a membrane
is thus of no consequence. Also, in reductions, wherein oxidation of
the reduced product is not liable to oxidation, a low overpotential anode,
such as iron or stainless steel, may be used. To further diminish the
possibility of reaction at the non- working electrode this electrode should
be of a considerably smaller area than the working electrode.

(4) ELECTROLYSIS CELLS

The cells used for electrolysis can be as varied as one's imagination
will permit. There are a few general types which will be described
merely to serve as a guide.

The simplest type can be constructed from a beaker. It will consist
of an anode, cathode, membrane, a reference electrode if a controlled
potential electrolysis is to be performed, and an agitator. A cooling
bath is placed around the beaker to maintain the desired electrolysis
temperature. Such a cell is schematically described in Fig. n : 8.

If a solid cathode is used it is not necessary to utilize a side-arm on
the beaker. It is merely necessary to extend a length of the solid cathode
material through the rubber stopper and make necessary wire contact
in this manner. With the larger cells of this design, that is, over a
125 ml. capacity, it is preferable to use a paddle-type stirrer to obtain
uniform agitation. The cell design illustrated in Fig. n : 9 has been
found most suitable for quantities of 25 to 50 ml. This design
can also be utilized for performing electrolysis at reflux temperature
by merely inserting an additional side-arm to which is attached a
reflux condenser.

33

INSTRUMENTATION AND TECHNIQUES OF ELECTROLYSIS

FIG. 11:9. Micro-Cell. B. Magnetic stirrer; C. Cathode; D.
Reference electrode ; E. Cathode contact ; H. Nitrogen gas (outlet not
shown); F. Anode contact.

35

ORGANIC ELECTRODE PROCESSES

-I

Fig. II : 10. Resin Flask Electrolysis Vessel. A. Solid anode for
oxidation, or solid cathode when cell used for reductions; B. Solid
cathode for oxidations or solid anode for reductions; C. Alundum
membrane; D. Salt bridge with fibre contact (see Fig. n: 16); E.
Reference electrode ; F. Paddle stirrer; G. Condenser; H. Thermo-
meter; /. Resin-flask; /. Cover, with multi-standard taper openings.

When this cell is used with a mercury cathode, A is eliminated and
a tube similar to Fig. n : I2-J is used to make electrical contact between
cathode and power supply. The salt bridge D is replaced by the type
described in Fig. n: I2-G or Fig. n: 16.

The cell illustrated in Fig. n: n contains an internal cooling coil
and is suitable for reactions in which considerable heat is generated.

36

INSTRUMENTATION AND TECHNIQUES OF ELECTROLYSIS

E

FIG. ii : ii Internally Cooled Electrolysis Cell. A. Paddle type
stirrer; C. Cathode; D. Reference electrode; E. Cathode contact;
F. Anode contact; G. anode (membrane surrounding anode is not
shown); H. Nitrogen inlets; / Cooling coil.

A cell designed to handle larger quantities of material, approximately
3-5 litres, is illustrated in Fig. ii: 12. The capacity can be increased
further by modification of the given dimensions. As illustrated,
the cell is used for reductions with a mercury cathode, but this too can
be modified so as to use other cathodes or anodes, if the larger chamber
is to be used for oxidations.

37

ORGANIC ELECTRODE PROCESSES

FIG. ii : 12. Large Scale Electrolysis Cell. A. Glass container
8i" i.d. X 8J" high; B. Alundum membrane ^ H wide X 6" long X
6" deep, J" wall, mix RA 1143 (Norton Company); C. Membrane
support " diameter rod; D. Stirrer 16" long X \* diameter; E.
Mercury seal type bearing with JS joint; F. Nitrogen inlet tubes to
cathode and anode chambers ; G. Salt bridge, top portion 2" diameter
X 4" high narrowing to A* X 12" long, at end of which is sealed a
Jt male joint, f " up from the bottom of this joint is a -^ hole, over
this is a female collar in length; H. Support for salt bridge,
I j joint extended 1-5" ; /. Reference electrode ; /. Contact to mercury
cathode, consisting of tube iV' o.d. 14" long, at bottom of which is
sealed a piece of platinum wire. The tube is partly filled with mercury
and connected to the negative side of power supply for reductions.
This tube passes through a rubber stoppered J$ joint ; K. Platinum
atlode 4" X 4"; L. Perspex cover; M. Cooling chamber n" X n"
X 7i" high o.d. thick Perspex; O. Water inlet to cooling chamber
V i.d.; P. Water outlets, \" i.d.; N. Cathode

INSTRUMENTATION AND TECHNIQUES OF ELECTROLYSIS

The type of cell illustrated in Fig. n: 13 has been found satisfactory
if an ion-exchange resin type membrane, or any other sheet type
membrane, such as cellophane, to is be used. This type can
be constructed of two elbows manufactured of Pyrex industrial
pipe which is available in many diameters. The membrane is
placed between two teflon or neoprene gaskets and held between the
two ends of the pipe by the flanges. This cell can be immersed in a
bath for cooling, or heated by wrapping with tape heaters.

FIG. n : 13. Cell for Non-rigid Type Membranes. A. Stirrer;
B. Thermometer ; C. Working electrode ; D. Non -working electrode ;
E. Gasketed membrane ; F. Reference electrode ; G. Salt bridge of the
type illustrated by Fig. II : 16.

Although the cells previously described are constructed of glass, it
is not meant that the electrolysis cells cannot be constructed of other
materials. In fact, any rigid material can be used, so long as it does
not react with the electrolysis medium, or serve as an electrical conduc-
tor. Thus, cells can be constructed of glass-lined or resin-coated steel,
hard rubber, polyethylene, etc. As mentioned, it is not usually desirable
to construct the cell of a conductor. However, on occasion this may
have a distinct advantage. For example, if a cathodic process required a
lead cathode, it might be advantageous to have the electrolysis cell
constructed of this same material, which would result in the availability
of a greater electrode area.

39

ORGANIC ELECTRODE PROCESSES

In general, the cooling coils should be constructed of a chemically
and electrically inert material, such as glass. Here too, it may be
possible to have the electrodes serve a dual purpose. Thus an oxidation
performed at a platinum anode can utilize a platinum-plated coil,
through which water is passed for cooling the anode chamber. In the
cathode chamber one can use a stainless steel coil for the dual purpose
of cathode and cooling coil. Of course, in such a system one must be
extremely careful that the conductivity of the coolant is considerably
lower than that of the electrolysis medium, otherwise the usual conse-
quence of a short circuit will result. In general practice, however, it is
usually possible to cool at least the external chamber with a cooling
bath or coil and, therefore, it would only be necessary to use a cooling-
coil-electrode in the internal chamber if it were isolated from the
cooling bath itself, as in cells illustrated in Figs. 11: 8-12.

(5) ELECTRODE MATERIALS

Many of the metals generally used for electrodes require no extensive
purification prior to use, as they are readily available in pure form.
However, after being used once, they require some form of cleaning.

Mercury, reagent grade, can be used directly for the electrolytic
reaction. If a batch process is in progress it is generally not necessary
to change the electrode for each electrolysis. However, in the
laboratory, where the same cell may be used for various types of
electrolysis, it is necessary to use fresh mercury for each procedure.
In our laboratories the mercury is purified in the following manner:
A suction flask is filled to approximately one-third its height with the
contaminated mercury. The surface scum is removed by allowing
water to flow through the top of the flask for half an hour. The water
is decanted and to the flask is added a 20 per cent solution of nitric acid
in water, equivalent in volume to the mercury. The flask is closed with
a rubber stopper through which is passed a i mm. bore capillary tube,
which extends to the bottom of the flask. The side-arm of the flask is
connected to a vacuum line and air is pulled through the capillary for
approximately twelve hours. At the termination of this period, the
supernatant liquid is decanted from the mercury, the mercury is
washed well with distilled water, and then filtered through a mercury
filter df the type based on the gold adhesion principle (8). Use of this
method results in a product of exceptional purity.

Lead, although essentially quite pure, has a surface layer of grease

40

INSTRUMENTATION AND TECHNIQUES OF ELECTROLYSIS

and other contaminants which must be removed prior to use. After
the lead is washed well with a suitable grease solvent, it is placed centrally
in a beaker, the internal periphery of which is coated with aluminium
foil. The positive lead from a d.c. power supply is connected to the
lead electrode and the negative lead to the aluminium electrode. Using
a medium consisting of a 20 per cent solution of sulphuric acid, the lead
electrode is subjected to anodic electrolysis for ten minutes at a current
density of 0-02 amp./cm. 2 . The leads are then reversed and the
electrolysis continued for an additional ten minutes. Finally the leads
are reversed a second time, and after a ten-minute electrolysis, the lead
electrode is removed, washed well with distilled water, hot ethanol and
air dried (9). Prior to use as a cathode the lead electrode is subjected
to a ten minute pre-electrolysis in the experimental background medium.
If a number of experiments are to be performed with essentially the
same medium, it is not usually necessary to repurify the electrode prior
to use, providing of course that the electrode is kept dry and protected
from air. In instances in which an oxidation process requires a lead
oxide anode, the treated lead electrode need not be further subjected
to pre-electrolysis as an anode.

Generally, other electrode materials such as copper, cadmium,
platinum, tin, zinc, and nickel require only a degreasing and pre-
electrolysis in the medium to be used for the electrolytic experiment.
At all times, the metals obtained for electrolytic studies should be of
the purest quality available. In addition to the use of sheet stock for
solid electrodes, one may find it desirable to use gauze electrodes in
order to increase the electrode area within a given space. This effect
can also be achieved to a lesser extent by pleating the solid electrode.
In addition to the solid electrodes, it might on occasion be desirable
to plate the desired metal on a sheet of the same metal, or another
metal, to obtain an exceptionally pure electrode surface. Plated elec-
trodes such as platinum black on copper (10), lead on lead, lead on copper,
and copper on copper have been used for electrolytic processes (i i).

There is in many instances a need for alloy electrodes, either from
the point of view of obtaining a catalytic plus a high overpotential
effect, or simply for obtaining a physically workable electrode. Thus a
vertical electrode of the high overpotential characteristics of mercury
can be obtained by the amalgamation of lead with mercury. This is
accomplished by using a lead sheet as the supporting material and
coating with mercury, by either allowing the lead to stand in a solution

4 1

ORGANIC ELECTRODE PROCESSES

of mercuric chloride (12), or by simply rubbing mercury on the lead
surface (13). Amalgamated nickel and zinc have also been prepared by
allowing sheets of these metals to be exposed to a solution of mercuric
chloride (14). Other alloy electrodes such as phosphor bronze and
Monel metal have also been used as electrodes (15).

Platinum, lead oxide, iron, and nickel have been used as anodes for
electrolytic oxidation. Platinum, because of its inertness, permits one
to obtain graded potentials. However, caution must be observed in
using this electrode in a halogen medium, as there is some attack on the
metal. This can be overconie by using a platinum electrode containing
a small amount of iridium. Iron and nickel should only be used as
anodes when the electrolysis is performed in an alkaline medium.

(6) REFERENCE ELECTRODES

In general, any half-cell of known constant value can be used as a
reference electrode. However, the prime prerequisite of such an
electrode is that its potential shall not be altered to any significant
degree by the passage of small currents, as would be the case even were
a high-resistance voltmeter to be used to measure the potential between
the working electrode and the reference electrode. Thus the only
satisfactory reference electrodes are essentially the saturated calomel
half-cell and the silver-silver chloride half-cell. These electrodes are
easily prepared and their potentials remain constant almost indefinitely.

The saturated calomel electrode is probably the more commonly
used electrode, of the two mentioned. It consists of a solution that is
saturated with respect to both potassium chloride as well as mercurous
chloride. This is in direct contact with a mercury pool, resulting in
the reaction :

Hg 2 Cl 2 + ae ^i 2Hg + 2Cr

The chloride ion activity is constant at a given temperature due to the
presence of an excess of potassium chloride. Since there also is present
an excess of mercurous chloride, the mercurous ion (which is responsible
for the potential of the cell) is also kept constant. Therefore passage
of small currents in either direction does not affect the half-cell potential.

The potential of the saturated calomel electrode (S.C.E.) by
definition is the E.M.F. of the following cell:

* - H 2 (i atm.) 1 H + (a = i) || KCl(s), Hg 2 Cl 2 (s) | H +
which is expressed as :

= + 0-242 7-6 X io~ 4 (t 25)
42

INSTRUMENTATION AND TECHNIQUES OF ELECTROLYSIS

Thus at 25 C. the saturated calomel electrode has a potential of + 0-242
volt v. the normal hydrogen electrode (N.H.E.). It will be noted from
the equation that this electrode has a large temperature coefficient.
In most electrolytic processes, a precision of 10 millivolt is sufficient,
thus this is of no consequence.

A number of instrument manu-
facturers can supply saturated
calomel electrodes of various
designs. As most of these usually
have a high resistance they can
only be used with a vacuum tube
voltmeter type measuring circuit.
When this type electrode is not
available, a saturated calomel elec-
trode of the type seen in Fig.
ii : 14 can be constructed.

The bottom of the vessel is
filled to approximately 3 mm. with
mercury, over which is placed an
equivalent amount of mercurous
chloride and potassium chloride.
The vessel is then filled with
saturated potassium chloride.
Electrical connection is made
through a platinum wire sealed in
a glass tube and extending through
the stopper used to seal the vessel.
Contact between the calomel cell
and the working electrode is made
by placing a rubber " policeman "
on the end of the side-arm tube,
through the bottom of which has

been threaded a number of asbestos fibres. The resistance of this
electrode is quite low and, therefore, a high resistance voltmeter can be
used to measure the potential between the reference electrode and the
working electrode.

The silver-silver chloride electrode can also be used as a reference
electrode. The potential of this electrode is + 0-197 volt a.N.H.E.
or 0-045 volt v. S.C.E. This electrode can also be purchased from

43

FIG. ii : 14. Saturated
Calomel Electrode.

ORGANIC ELECTRODE PROCESSES

various manufacturers, or can be constructed as shown in Fig. ii: 15,
from a glass tube, one end of which is sealed by a rubber " policeman,"
pierced in a number of places at the tip, with asbestos fibres, as illustrated
in the fabrication of the calomel electrode. The tube is then filled with
a solution saturated with respect to both silver
chloride and potassium chloride. Through the
rubber stoppered closure is passed a silver wire.

In measuring the true potential of the working
electrode with respect to the reference electrode,
we are Confronted with the unavoidable difficulty
of an IR drop between the tip of the reference
electrode and the surface of the working electrode.
Therefore what is observed is related to the true
potential by the equation :

-Bobs. = factual +77?

Thus in reductions we appear to work at potentials
more negative and in oxidations more positive than
the actual potentials at the working electrodes.
This in no way interferes with obtaining the desired
results, except in so far as it has a small effect on
the efficiency of the reaction. Since in organic
electrolytic processes an actual 100 per cent
efficiency is generally not attainable, this slight
effect is of no consequence.

In order to minimize this IR drop the contact
point on the reference electrode should be as close
to the electrode surface as possible. This is ex-
tremely important when working with nonaqueous
or partially aqueous systems. With a mercury cath-
ode there is no great problem with the reference
electrodes described, except in cases wherein a salt
bridge is necessary, due to the height of the
electrolysis vessel. With mercury it is possible to
obtain satisfactory contact with the sleeve type salt bridge (Fig. n: 12).
However, with solid electrodes it is extremely difficult to obtain good
contadt between the solid surface of the commercially available reference
electrode and the working electrode. For this reason the salt bridges
illustrated in Fig. n: 16 have proved most suitable.

44

FIG. ii : 15.
Silver-Silver Chlo-
ride Electrode.

INSTRUMENTATION AND TECHNIQUES OF ELECTROLYSIS

\/

\/

-GLASS TUBE-

RUBBER POLICEMAN PERFORATED

WITH ASBESTOS FIBERS

-WORKING-

11 ELECTRODE SURFACE

SOLID ELECTRODE

FIG. 11:16.

MERCURY ELECTRODE

Salt Bridges.

The multi-perforated rubber " policeman " offers the advantage of
flexibility and softness, which makes it possible, by exertion of the proper
amount of pressure, to achieve good contact with the working electrode.
With solid electrodes it is possible to minimize the IR drop still further
by placing the reference electrode in the region of minimal current flux.
Thus, if the anode were opposite to the cathode, and the reference
electrode contact made on the reverse side of the working electrode, it
would be outside the region of high potential gradient.

GENERAL TECHNIQUE

It is extremely difficult to set forth a general technique which will
enable one to obtain the desired results immediately. Usually, from
a knowledge of the substance one wishes to oxidize or reduce, it is
possible to develop the proper electrolysis medium. It is always most
desirable to have the depolarizer in solution. This enables attainment
of maximum efficiency. The next step is to obtain an insight into the
reactivity of the oxidizable or reducible groups in the molecule. This
can be accomplished by obtaining a voltametric curve (Fig. i: i), or a
polarogram when possible, either at a dropping mercury cathode or a

45

ORGANIC ELECTRODE PROCESSES

rotating platinum anode (16). With the knowledge gained from these
curves it is possible to decide on the potential and/or the electrode
material to be used. If the oxidation or reduction goes in a stepwise
manner it will be necessary to investigate various potentials and the
products obtained at each, until the proper conditions which will yield
the desired results are found.

REFERENCES

(1) Hickling, Trans. Faraday Soc., 1942, 38, 27; Caldwell, Parker and Diehl,
2nd. Eng. Chem., Anal. Ed., 1944, 16, 532; Lingane, ibid., 1945, i?> 332;
Diehl: Electrochemical Analysis with Graded Cathode Potential Control,
G. F. Smith Chemical Co., Columbus, Ohio, 1948; Allen, Anal. Chem.,
1950, 22, 804; Lamphere and Rogers, ibid. t 1950, 22, 463; Wehner and
Hindman, J. Amer. chem. Soc., 1950, 72, 3911; Milner and Whitten,
Analyst, 1952, 77, n; Urabe and Yasukochi, J. Electrochem. Soc. Japan,
1954, 22, 525; Schwarz, Chem. Ing. Techn., 1956, 28, 423.

(2) Lingane, loc. cit., ref. i ; Lingane and Jones, Anal. Chem., 1950, 22, 1169.

(3) Allen, loc. cit., ref. i.

(4) Miller, Electronics, 1941, 14, 27, 105.

(5) Palmer and Vogel, Analyst, 1953, 78; 1928.

(6) Norton Company, Worcester, Mass., U.S.A.

(7) Rhom & Haas Co., Philadelphia, Pa., U.S.A.

(8) Bethlehem Apparatus Co., Bethlehem, Pa., U.S.A.

(9) See Tafel, Ber., 1900, 33, 2209.

(10) Itomi, Mem. Coll. Sci., Kyoto, 1930, I3A, 311.

(u) Fichter and Stein, Helv. Chim. Acta, 1929, 12, 821; Bruckner and

Karalos, Ann., 1935, 518, 226; Swann and Tucker, Trans. Amer. electro-

chem. Soc., 1939, 75, 411.

(12) Oroshnik and Spoeni, y. Amer. chem. Soc., 1941, 63, 3338.

(13) Creighton, U.S.P. 1990582, 1935.

(14) Bougault, Cattelain and Chabrier, Compt. rend., 1939, 208, 193; Shima,
Mem. Coll. Sci., Kyoto, 1930, 13 A, 315.

(15) McKee and Brockman, Trans. Amer. electrochem. Soc., 1932, 62, 203;
McKee and Gerastopolou, ibid. t 1935, 68, 329.

(16) Kolthoff and Lingane, Polarography, Interscience, New York, 1952;
Lord and Rogers, Anal. Chem., 1954, 26, 284; Allen and Powell, Trans.
Faraday Soc., 1954, 50, 1244; Wawzonek, Anal. Chem., 1956, 28, 638,

CHAPTER III

CATHODIC REDUCTION

(i) MECHANISM OF CATHODIC REDUCTION

As previously mentioned, Fritz Haber pointed out that it was the
cathodic potential which was probably the most important factor in
determining the outcome of an electrolysis. Quite often it is possible,
by maintaining constant conditions, i.e. electrode material, medium,
and temperature, to obtain different products by merely varying the
electrode potential.

A number of theories have been proposed to describe the potential
acquired by the electrode. The most widely accepted one states
that the higher the cathode potential, the higher the potential energy
of the hydrogen atoms liberated at the electrode surface. Therefore,
for a given electrode, it is possible to obtain different energies for the
discharge hydrogen atoms by merely changing the cathodic potential.
Generally, then, the reduction efficiency of a cathode depends on the
cathode potential. The higher the potential, the higher the efficiency.
Thus it would be expected that a ketone or aldehyde would be reduced
to the hydrol at a lower cathodic potential than that required to induce
free-radical formation, with subsequent bimolecular reduction to the
pinacol (i), or as has been amply demonstrated by Haber and others
(2, 3), nitrobenzene can be reduced in a stepwise manner by merely
changing the cathodic potential at which the reaction was being
performed.

Another way of looking at the situation is to take advantage of the
experimental results of Eyring, et al. (4), in which it was demonstrated
that a low overpotential metal forms a strong metal-hydrogen bond,
and a high overpotential metal forms a relatively weak metal-hydrogen
bond. We may extend this idea to protons in general. Thus the
reduction of a ketone to the hydrol, at a mercury cathode and at a
relatively low potential, would proceed as follows :

47

ORGANIC ELECTRODE PROCESSES

O

II
R,C

_ e

!0|
R..-C

_e

101

H

R,C-

H 4

M

R 2 CH(OH)

The charged mesomeric form is attracted to the cathode, where the
oxygen atom, which has a pair of unshared electrons, accepts a hydrogen
atom as indicated. Simultaneously, at the relatively lower cathodic
potential, because of the strength of the metal-proton bond, the
charged molecule is held at the cathode a sufficient length of time to
undergo a two-electron change, with resultant hydrol formation. At
the higher cathodic potential we have the following :

_e

10
RoC-

R 2 C-

M

H

l.oi

RoC

R 2 C

OH on

i i

R 2 C CR 8

wherein the charged form, because of the weak metal-proton bond,
remains only long enough for a one-electron transfer to the electron
deficient carbon, with the resultant formation of a highly reactive free
radical, which combines with another to form the pinacol (5). This
free radical mechanism is further substantiated by the fact that mixed
pinacols can be obtained by the electrolytic reduction at a mercury
cathode, of a mixture of />-dimethylaminoacetophenone and/>-methoxy-
acetophenone (5).

The first theory involving the potential energy of the hydrogen
liberated, does not adequately explain the reactivities of organic
compounds. The reactivity of the charged form with the electrode
appears to fill in the picture to the point where it is possible to visualize
so-called free-radical formation.

CATHODIC REDUCTION

We might also rationalize elimination of a hydroxyl group as being a
replacement reaction, similar in nature to an electrolytic dehalogenation.
It has been found that phthalimides (I) in an acidic medium undergo
a two-electron change to the respective hydrol (II) at a lower cathodic
potential than that required for complete reduction to the isoindoline
(III), thus:

If the hydrol (II), or the phthalimide (I), is reduced at the higher
potential the isoindoline (III) is obtained (7). An interesting observa-
tion that has been made is that the potential required for formation of the
isoindoline is essentially the same as that required for the partially
reduced phthalimidine. The latter can be isolated in good yield if the
electrolysis is interrupted before completion. This would tend to indi-
cate that the reduction of the first carbonyl group may possibly be the
rate-determining step in the reduction of the phthalimide to the
isoindoline.

(2) REDUCTION OF NITRO COMPOUNDS

Nitro compounds have been studied quite extensively, and there-
fore there is probably more information available on the electrochemical
reactivity of these compounds than on other types. Haber and others
(2, 3), in their experiments on the reduction of nitrobenzene, found that
this compound, in an acidic or basic medium, is first reduced to nitro-
benzene, then to phenylhydroxylamine, and finally to aniline, depending
on the electrode potential used in the experiment :

C 6 H 5 N0 2 + 2H + 2e ---- > C 6 H 5 NO + H 2 O
C 6 H 5 NO + 2H + 2e - -- C 6 H B NHOH
C 6 H 5 NHOH + 2 H + 2 e ---- > C 6 H 6 NH 2 -f- H 2 O

As the electrode potential for the reduction of nitrosobenzene in an
alkaline medium is 0-48 volt, it is a far better depolarizer than nitro-
benzene, which requires an electrode potential of 0-78 volt to initiate
the cathodic process. Thus the consequence in the overall scheme is

ORGANIC ELECTRODE PROCESSES

that the nitrosobenzene is rapidly reduced to the phenylhydroxylamine,
and thence to aniline and/or other side reaction products, depending
on the experimental conditions. Haber verified the formation of nitroso-
benzene as an intermediate by performing a reduction on nitrobenzene
in an alcoholic-alkaline medium, at an electrode potential of 0-98
volt, in the presence of a-napthol and hydroxylamine. As a result, as
soon as the nitrosoamine formed it reacted to give the azo dye, 4-(phenyl-
azo)-i-naphthol:

Cathodic process C 6 H 5 NO 2 + 2e + zH ----- C 6 H 6 NO + H 2 O

N = N
Chemical process

C 6 H 6 NO + NH 2 OH

OH

The formation of nitrosobenzene in an acidic medium was verified by
performing the electrolysis in the presence of oc-naphthylamine and
hydroxylamine, which resulted in the formation of 4-(phenyl-azo)-i-
naphthylamine.

Although nitrosobenzene is a good depolarizer and is easily reduced,
it can be obtained with satisfactory yields by performing the electrolysis
in a cell without a membrane, using a magnesium sulphate medium,
with a nickel cathode and a lead anode (8). The phenylhydroxylamine,
as formed at the cathode, is oxidized to nitrosobenzene at the anode,
thus:

Cathodic process C 6 H 5 NO 2 + 4 H + 46 --- * C 6 H 6 NHOH + H,O
Anodic process C 6 H 5 NHOH -f O ----- > C 6 H 5 NO + H 2 O + ze

The presence of the intermediate, phenylhydroxylamine, in the
reduction of nitrobenzene, was proved by reduction of the nitrobenzene
at a platinum cathode, in an alcoholic ammonium hydroxide solution
containing sufficient ammonium chloride to increase the conductivity.
In addition to the above major product, a smaller quantity of azoxy-
benzene is also obtained (2). The formation of phenylhydroxylamine as
an intermediate in an acid medium, was substantiated by Gatterman (9),
who reduced a mixture of nitrobenzene and benzaldehyde in a concen-
trated sulphuric or acetic acid medium, and obtained benzylidine-
hydroxyl-amine in the following manner:

5

CATHODIC REDUCTION
C 6 H 5 N0 2 ~f 4 H + 4 e > C 6 H 5 NHOH + H 2 O

O

t
C 6 H 5 NHOH + C 6 H 5 CHO C 6 H 5 N: CHC 6 H 5 + H 2 O

The side reactions, which may take place between the intermediates
obtained as a result of the electrolytic reduction of nitrobenzene, have
been investigated by Bamberger and his associates (10). They found
that the following three processes, that is the formation of azoxybenzene
(i), azobenzene (II) and a mixture of azoxybenzene and aniline (III),
occur predominantly in an alkaline medium, and also, to a very much
smaller extent, in an acid or neutral solution.

O

t

I. C 6 H 5 NHOH + C 6 H 6 NO > C 6 H 5 N: NC 6 H 5 + H 2 O

II. 2 C 6 H 5 NHOH > C 6 H 5 N:NC 6 H 6 { 2H 2

O

t

III. 3 C 6 H 5 NHOH > C 6 H 5 N: NC 6 H 5 + C 6 H 5 NH 2 + 2 H 2 O

The above are given in their respective rate orders, with the first being
the fastest.

To obtain the maximum amount of the azoxybenzene (I) the reaction
should be carried out in a slightly basic medium, and the electrode
potential controlled so that this compound cannot be further reduced
(2). Another way in which this can be accomplished is to use a suspen-
sion of the nitrobenzene in a dilute aqueous alkaline solution, with a
low overpotential electrode, such as nickel. The nitrobenzene is a
better depolarizer than the azoxybenzene, and therefore, if the tempera-
ture is not high, the azoxy-compound will precipitate out as it is formed,
and will not be further reduced.

If the alkalinity of the medium is not kept low and the potential is
permitted to approach a high value, one will obtain the hydrazo com-
pound, thus :

C 6 H 5 N: NC 6 H 5 + aH + ae > C 6 H 5 NH.NHC 6 H 5

To obtain the best yields of the hydrazobenzene, this reaction should
be performed in a basic aqueous-alcohol solution at an elevated tempera-
ture with a lead, tin, or zinc cathode. To prepare the hydrazobenzene
from nitrobenzene directly it is necessary to carry out the reaction in
two stages. In the first stage a suspension of the nitrobenzene is

51

ORGANIC ELECTRODE PROCESSES

electrolysed in a dilute alkaline solution, using an iron-gauze cathode
and moderate temperatures. As soon as the nitrobenzene has been
reduced, a high overpotential cathode is inserted and the temperature
increased; thereupon further reduction will occur (n). Utilization of
more drastic conditions initially would, of course, have resulted primarily
in the formation of aniline.

Benzidine (I), which is quite important to the dye industry, can be
prepared by reduction of azoxy- or azo-benzene in a weakly acidic
medium, to the hydrazobenzene, with subsequent rearrangement
according to the following scheme :

ze + aH
C 6 H 5 N : NOC 6 H 5 > C 6 H 5 N : NC 6 H 5

_.

v_y ...
\

H 2 N< >-< >NH

2H

Rearrangement C 6 H 5 NH . NHC 6 H 5

NH 2
II

If the rearrangement of hydrazobenzene is performed at temperatures
of 20 to 30 C., only small quantities of the biphenyl (II) will be
obtained (12).

The ease with which the hydroxylamine is reduced to aniline depends
on the cathode potential. If a high overpotential metal, such as mercury,
lead, or zinc, is used in an acidic or basic medium, then aniline is obtained
as the main product. It is reduced so rapidly that it does not have much
opportunity to undergo any side reactions.

Aminophenols, a side reaction product, can generally be obtained by
the reduction of a nitrobenzene in a sulphuric acid medium at a low
overpotential electrode, such as platinum (13). Thus, nitrobenzene is
reduced to phenylhydroxylamine, and then rearranged:

CeH 5 N0 2 4e + 4H ^ C 6 H 6 NHOH __**L> p-HOC 6 H 5 NH 2

Of course, if this same reduction were carried out with a high over-
potential electrode, the phenylhydroxylamine would not exist as such

52

CATHODIC REDUCTION

in the medium for a sufficient period to undergo rearrangement to the
amino phenol. Were a hydrochloric acid medium to be used instead of
the sulphuric acid, a mixture of ortho and parachloroanilines would be
obtained, probably with a chloroamine as the intermediate (14).

C 6 H 5 NHOH > C 6 H 5 NHC1 > NH 2 C 6 H 4 C1 (o and p)

When ethanol is used in the sulphuric acid electrolyte, the />-amino-
phenol is esterified to yield />-phenetidine (NH 2 C 6 H 4 OC 2 H 5 ). The
aminophenol can be simultaneously sulphonated if the nitrobenzene is
reduced in a fuming sulphuric acid medium, with the resultant forma-
tion of 2-hydroxy-5-amino benzenesulphonic acid (15). If the para posi-
tion of the nitrobenzene is occupied by a Cl or COOH it is replaced
by an OH to yield />-aminophenol ; if there is an OH or NH 2 in
the para position the nitro group undergoes the usual reductive changes
without any rearrangement (16). Nitrobenzenes containing N(CH 3 ) 2 ,
COCH 3 , OCH 3 , or CH 3 in the ortho position, when reduced to
the respective hydroxylamine, rearrange so rapidly to the ^-amino-
phenol, or a derivative, in a concentrated acid medium, that even the
use of a high overpotential electrode does not interfere (17).

Of course, it is possible to obtain the phenyl hydroxylamine without
the risk of rearrangement by utilizing a low overpotential electrode such
as platinum, and a medium close to neutral, preferably buffered with
either ammonium chloride-ammonia or sodium acetate-acetic acid (2).
This latter condition is obviously necessary, as the electrolyte in the
vicinity of the electrode becomes more acidic due to the liberation of
hydrogen ions at the cathode, and would result in the formation of
/>-aminophenol.

Azobenzene can be obtained in good yield by performing the
electrolysis on an alkaline-alcoholic solution containing a high concen-
tration of hydrazobenzene and nitrobenzene. This reaction requires a
low overpotential electrode and reflux temperature (18):

3 C 6 H 5 .NH.NHC 6 H 5 + 2 C 6 H 5 .NO 2

3 C 6 H 5 N: NC 6 H 5 + C 6 H 5 NO: NC 6 H 5 + 3 H 2 O

The azobenzene and azoxybenzene formed can only be reduced to
hydrazobenzene when using a high overpotential electrode, and
preferably with the aid of added catalysts.

As hydrazobenzene is very susceptible to air oxidation, azobenzene

53

ORGANIC ELECTRODE PROCESSES

is readily obtained by passing a stream of oxygen through a solution of
the above-mentioned compound (19):

2C 6 H B NH.NHC 6 H 5 + O 2 > 2C 6 H 5 N: NC 6 H 5 + 2H 2 O

Substituted nitro compounds generally undergo the same type of
reduction as the unsubstituted compounds. However, there are
instances where other types of reactions will take place. For example,
when a methyl group is in the meta position, the respective nitroso- and
hydroxylamine derivatives react chemically in a basic medium to give
the azoxy-compound. Wheii the methyl group is ortho or para, the
hydroxylamine and nitroso do not readily react, and as a result, the
reduction continues on to the amine (20). One finds very similar
characteristics amongst other derivatives, such as w-nitrophenol and
w-nitroaniline. The nitrophenol and nitroaniline will yield, on reduc-
tion, the azoxy compounds ; whereas the ortho and para substituted
compound will give the amine (21).

A number of poly-nitro compounds have been reduced electrolyti-
cally to the expected products (22). -Thus, i,5-dinitronaphthalene
yields the 1,5 -naphthalene- diamine when the reduction is performed in a
sulphuric acid medium at a lead cathode (23). In a similar manner
2,6-dinitrotoluene is first reduced to 2-amino-6-nitrotoluene, and
finally to 2,6-diaminotoluene.

\N0 2 6e + 6H / \NH 2 6e

N0 2 CH 3 NO 2 CH 3 NH 2 CH 3

However, it must be remembered that the presence of more than one
nitro group may lead to complex mixtures.

From the previous discussions we can present the following reactions
obtained by direct electrolytic reduction, as represented by con-
tinuous lines, and the secondary chemical reactions, as represented
by dotted lines :

C 6 H 5 N0 2 -- C 6 H 5 NO --- > C 6 H 5 NHOH -- > C 6 H 6 NH 2

O

T

C e H 6 N:NC 6 H 6
C 6 H 6 N: NC 8 H 5 ^.^r> C 6 H 6 NH.NHC 6 H 5

54

CATHODIC REDUCTION

In addition, the following chemical transformations occur in an
acidic medium :

conc - + P -HOC 6 H 6 NH 2

H 2 S0 4
C,H 5 NHOH 4r^T > p-HOC 6 H 5 NH 2

SO a H
H2S 4 - p-C 2 H 6 OC 6 H 5 NH 2

+ C 2 H 5 OH
conc. HC1

P-C1C 6 H 5 NH 2 and o-C!C 6 H 5 NH 2

The electrolytic reduction of aliphatic nitro compounds has not been
investigated as extensively as those previously discussed. Generally
they are first reduced to the hydroxylamine, and then to the amine.
Pierron (25) found that nitromethane was reduced to methylhydroxyl-
amine, using a nickel cathode, a medium consisting of a 10 to 15 per
cent solution of sulphuric acid in alcohol, and a temperature of 20 C.
Using a higher temperature, he was able to cause the reduction to go
to the amine. In a strongly acidic medium the nitromethane was
reduced to the formaldehyde oxime [CH 2 :NOH], which hydrolysed
to hydroxylamine and formaldehyde. He also found that under the
conditions described above, nitroethane could be reduced to ethyl-
hydroxylamine, and nitropropane to propylhydroxylamine. When
chloropicrin [CC1 3 .NO 2 ] was reduced in a weakly acidic medium at a
platinum cathode the dechlorinated methylhydroxylamine was obtained.
If a lead cathode was used the reaction yielded methylamine (26).
When nitrosamines or nitramines are reduced in a dilute acid medium,
hydrazines are obtained (27).

R 2 N: NO + 4 H + 4 e > R 2 NNH 2

RNHNO 2 + 6H -f 6e > RNHNH 2

(3) REDUCTION OF IMINES AND IMIDIC ESTERS

Imines and Imidic esters can readily be reduced to their respective
amines in either an acidic or basic medium, using a relatively high
cathode potential metal, such as lead or mercury. Law (28) reported
on the reduction of a number of substituted benzalanilines to their
respective benzylanilines, thus :

55

= NC 6 H 5 > R<!< >CH 2 NHC 6 H 5

ORGANIC ELECTRODE PROCESSES
26
2H

Imidic esters, such as ethyl acetimidate, have been reduced to
ethylamine, using a aN sulphuric acid medium and maintaining the
temperature at o C. (29) :

NH

/ 4H 4- 4e
CH 3 C > CH 3 CH 2 NH 2 + C 2 H 5 OH

OC 2 H 5

This same worker also accomplished the reduction of some sub-
stituted ethyl benzimidates to their respective amines with satisfactory
yields being obtained.

REFERENCES

(1) Allen and Corwin,^ Amer. chem. Soc., 1950, 72, 114.

(2) Haber, Z. Elektrochem., 1898, 4, 506; 1899, 5, 77; Haber and Schmidt,
Z. physikal. Chem., 1900, 32, 271.

(3) D.R.P., 21131, 1882; 116942, 1899; 123813, 127815, 130742, 131404,
1901; 168273, 1905; Elbs, et al.,y. pr. Chem., 1891, 43, 39; Z. Elektro-
chem., 1903, 9, 416, 428; Haussermann, Chem. Ztg., 1893, I 7 J 29, 2O 9J
Ber., 1901, 34, 3769; Lob, et al., Z. Elektrochem., 1896, 2, 472; Fichter
and Sulzberger, Ber., 1904, 37, 878; Brand, Ber., 1905, 38, 3076; Brand
and Stohr, ibid., 1909, 42, 2482; Otin, Z. Elektrochem., 1910, 16, 487;
Snowdon, jf. phys. Chem., 1911, 15, 797; Frederiksen, ibid., 1915, 19*
696 ; Trans. Amer. electrochem. Soc., 1915, 28, 345 ; Fierz and Wissenbach,
Helv. Chim. Ada., 1920, 3, 305; Brown and Warner, Trans. Amer.
electrochem. Soc., 1922, 41, 225; J. phys. Chem., 1923, 27, 455, 652;
U.S.P., 1536419, 1922; Doolittle, Trans. Amer. electrochem. Soc., 1924,
45 575 Norris and Cummings, Ind. Eng. Chem., 1925, 17, 305; Itomi,
Mem. Coll. Set. Kyoto, 1929, i2A, 343; Kirkhgof, Khim. Farm. Prom.,
I 933> 326; Holleck and Exner, Z. Elektrochem., 1952, 56, 46; Holleck
and Schmidt, ibid., 1955, 59 56; Sekine and Sugino, J. electrochem. Soc.
Japan, 1953, 21, 383; Holleck and Schmidt, Naturwissenschaften, 1954,
41, 87; Bergman and James, Trans. Faraday Soc., 1954, 50, 60.

(4) Eyring, Glasstone and Laidler, J. chem. Phys., 1939, 7* 1053.

(5) Allen, Proc. C.I.T.C.E., Butterworth's Sci. Pub., London, 1955, p. 481.

(6) Levine and Allen, J. chem. Soc., 1952, 254; Waters, Ann. Repts., 1952,

49> 135.

(7) Sakurai, Bull. chem. Soc. Japan, 1930, 5, 184; 1932, 7, 155; Allen and
Ocampo, J. Amer. electrochem. Soc., 1956, 103, 452.

(8) D.RP., 192519, 1905-

(9) Gatterman, Ber., 1896, 29, 3037, 3040.

(10) Bamberger, et al., Ber., 1897, 30, 2278; 1900, 33, 274.

(n) D.R.P., 100233, 100234, 1898; 116467, 1899; 189312, 1904.

56

CATHODIC REDUCTION

(12) D.R.P., 116871, 1899; Lob, Z. Elektrochem., 1901, 7, 337; 1902, 8, 778;
U.S.P., 1544357, 1923-

(13) Gatterman, et al., Ber., 1893, 26, 1844; 1894, 27, 1927; 1896, 29, 3034;
D.R.P., 79865, 80978, 81625, 82445, 1894; Elbs, Z. Elektrochem., 1896,
2, 471 ; Lob, ibid., 1897, 3, 46; Ber., 1896, 29, 1894; Brand, J.pr. Chem.,
1903, 6 7> 145; Friedlander, Ber., 1905, 38, 2837; U.S.P., 1501472, 1920;
1536419, 1922; McDaniell, et al., Trans. Amer. electrochem. Soc., 1921,
39, 441; Caeser, ibid., 1927, 52, 95; Brigham and Lukens, ibid., 1932,
61, 281; B.P., 254204, 1925.

(14) Lob, Ber., 1896, 29, 1894; Z. Elektrochem., 1896, 2, 46.

(15) Gatterman, Ber., 1893, 26, 1847; 1894, 27, 1929; Noyes and Clement,
z'faW., 1893, 26, 990.

(16) Noyes and Dorrance, ibid. t 1895, 28, 2349.

(17) Pinnow, J. r. C/zew., 1901, 63, 352; 1902, 65, 579; Brand, ibid., 1903,
67* 145-

(18) Lob, Z. Elektrochem., 1896, 2, 529; 1897, 3, 471 ; 1898, 4, 428; 1899, 5,
456; 1900, 7, 333; Ber., 1898, 31, 2201; Elbs, et al., Z. Elektrochem.,
1898, 5, 108, 113; 1902, 8, 789; 1903, 9, 416, 428; Rhode, Z. Elektro-
chem., 1900, 7, 322, 328, 338; Wohlfahrt, J. pr. Chem., 1902, 65, 295;
Brand and Stohr, Ber., 1906, 39, 4058; Inoue, J. Chem. Ind. Japan,
1921, 24, 121 ; McKee and Brockman, Trans. Amer. electrochem. Soc.,
1932, 62, 203.

(19) Elbs and Kopp, Z. Elektrochem., 1898, 5, 108; D.R.P., 141535, 1902.

(20) Elbs and Schwarz, Z. Elektrochem., 1900, 7, 144; 1904, 10, 579.

(21) Elbs, et al., Z. Elektrochem., 1898, 5, 108, 113; 1900, 7, 141; Rhode
ibid., 1900, 7, 328, 338; Brown and Warner, Trans. Amer. electrochem.,
Soc., 1922, 41, 225; Berezovskii and Varkov, J. Gen. Chem. (U.S.S.R.),

1953, 2 3 97, ioo.

(22) D.R.P., 79406, 1894; 105501, 1898; 116942, 1899; Moller, Z. Elektro-
chem., 1901, 7, 799; 1904, 10, 199, 201; Brand, Ber., 1905, 38, 4014;
Hofer and Jakob, ibid., 1908, 41, 3195; Brand and Zoller, ibid., 1907,
4> 3330; Brand and Eisenmenger, J. Prakt. Chem., 1913, 87 [27], 494;
Bradt and Linford, Trans. Amer. electrochem. Soc., 1936, 69, 353;
Phillips and Lowy, ibid., 1937, 71, 501 ; McGookin, Swift and Tittensor,
y. Soc. chem. Ind., 1940, 59, 92; Kikuchi and Vtsunomiya, J. chem. Soc.
Japan, Ind. Chem. Sect., 1951, 54, 534; Holleck and Schmidt, Natur-
zvissenschaften, 1954, 41, 870; Bergman and James, Trans. Faraday Soc.,

1954, 5> 60; Holleck and Schmidt, Z. Elektrochem., 1955, 59, 56.

(23) Moller, loc. cit., ref. 22.

(24) Brand and Zoller; Hofer and Jakob, loc. cit., ref. 22.

(25) Pierron, Bull. Soc. Chim., 1899, 21 [3], 780; Z. Elektrochem., 1900, 7. 1900.

(26) Frank, Diss., Giessen, 1903.

(27) Ahrens, Z. Elektrochem., 1896, 2, 578; Ahrens and Sollman, Chem.
Ztscher., 1903, 2, 414; Backer,^, chem. Soc., 1912, 101, 592; Rec. Trav,
Chim. Pays-Bas et Belg. (Amsterd.) 1912, 31, 15, 172, 190; Bircher,
Ingersoll, Armendt and Cook, J. Amer. chem. Soc., 1925, 47, 391; Yu
Chang and Sah, Sri. Rep. Nat. Tsing Hua Univ., 1933, 2, 157; Wells,
Babcock and France,^. Amer. chem. Soc., 1936, 58, 2630.

(28) Law, J. Chem. Soc., 1912, 101, 154; See also Knusden, Ber., 1909, 42
3998; Brand, ibid. 9 1909, 42, 3461; Brand and Hoing, Z. Elektrochem.,
1912, 18,747; D.R.P., I43I97, 1902; 148054, 1903.

(29) D.R.P., 360529, 1920; Wenker, J. Amer. chem. Soc., 1935, 57, 772.

57

CHAPTER IV

CATHODIC REDUCTION OF CARBONYL
TYPE COMPOUNDS

s electrochemical reactivity of the .carbonyl group which is
JL common to aldehydes, ketones, acids, esters, amides, and imides,
has been investigated almost as extensively as the nitro group. That
this group is not as easily reduced as the nitro group is demonstrated
by the fact that iw-nitrobenzaldehyde can be reduced to azoxybenzalde-
hyde in an alcoholic-sulphuric acid medium, at a copper cathode, with-
out disturbing the aldehyde portion of the molecule (i).

O

\
NO 2 N=N

\ \ r

OHC <^~S > OHC <(^J)> /__V-CHO

Generally the cathodic reduction of the carbonyl group requires a
relatively high overpotential, such as that obtained with a mercury or
lead electrode. However, depending on the type of reduction desired,
it may be necessary, at times, to depend on a catalytic electrode which
will yield a more highly reduced form, even though the overpotential
of the electrode is lower than that of mercury or lead.

The mechanism as proposed for the reduction of aldehydes and
ketones (p. 47) may very well apply to other carbonyl compounds. The
formation of dimers is not too common in types other than aldehydes
and ketones, possibly because the energy required for the formation of
the free radical is beyond experimental capabilities at the present time.

(i) REDUCTION OF ALDEHYDES AND KETONES

The reduction of aldehydes can be accomplished a little more
readily than the reduction of ketones. Usually it requires a somewhat

58

CATHODIC REDUCTION OF CARBONYL TYPE COMPOUNDS

lower electrode potential to reduce the carbonyl group to the alcohol
than to the pinacol. In order to attain the more highly reduced form,
a hydrocarbon, it is necessary in most instances to utilize a so-
called catalytic electrode. For this purpose a cadmium electrode
has been found quite satisfactory. Thus the reduction of an
aldehyde or ketone may result in the formation of three different
compounds, depending on the conditions of the electrolysis and the
electrodes used (2).

2e/mole | le/mole | 4e/mole

I j

RoCHOH R 2 C. .CR 2 RCH 2

,11 f

I OH HO _J

ze/mole

The process wherein the pinacol is obtained is independent of the
other processes and cannot be obtained by cathodic reduction from the
alcohol. The complete reduction may involve the alcohol as an
intermediate. This assumption is based on studies performed with
N-substituted tetrachlorophthalimides, wherein it was possible, in an
aqueous-acetic sulphuric acid medium, to obtain the hydroxyphthali-
mide by using a lead cathode and maintaining the potential v. a standard
calomel electrode at 0-68 volt. This compound could then be further
reduced to the isoindoline at a reference potential of i-io volt (3).
The reaction sequence is :

Aliphatic aldehydes are generally reduced to the corresponding
alcohols, although in some cases it is possible to obtain the glycol (4).
Formaldehyde and acetaldehyde in a basic medium, at copper or silver
cathode, give excellent yields of the respective alcohols. In an acidic
medium there is a tendency to obtain the hydrocarbon using the
above electrodes. Using cadmium, it is also possible to obtain good
yields of the hydrocarbon (5). If a high overpotential cathode, such

59

ORGANIC ELECTRODE PROCESSES

as mercury, is used with an acidic medium, 2,3-butanediol results
from the electrolysis of acetaldehyde. Thus acetaldehyde can, under
proper conditions, yield three different products.

CH,CHO

CH^

le + i H /mole
(acidic medium)

-I-CH CH 3

26 +2H

(basic medium)
r Cu or Ag
,CH 2 OH CH :

4e + 4 H
(acidic medium)
,Cd
,CH, CH 3 CI

OH OH

Isobutyraldehyde has been reduced electrolytically under varied
conditions with the formation of isobutane, isobutene, and iso-
butyl alcohol (6). It was found that a yield of 60 per cent butyl alcohol
was obtained at a mercury cathode. With this cathode no hydrocarbon
was obtained. However, at a lead cathode, a 40 per cent yield of the
hydrocarbon resulted. The temperature of the medium also played a
part in the results, for it was found by the same investigators that, at
room temperature, the alcohol was the main product. Elevated
temperatures resulted in an increase in hydrocarbon formation.

The reduction of glucose and mannose to mannitol has been
achieved (7) at a graphite cathode in a dilute sulphuric acid medium,
and at an amalgamated lead electrode in a basic medium (7).

H H OHH 2e H H OHH

HOCH a - C C C C CHO > HOCH 2 C C C C CH 2 OH

OH OHH OH OH OHH OH

Mannose is also reduced to mannitol at an amalgamated lead
electrode in an aqueous sodium sulphate medium.

According to the patent literature (8), acetylene, when passed into
a 5 to 10 per cent solution of sodium sulphate, sulphuric acid, or
orthophosphoric acid containing a mercury salt and reduced at a mercury
or lead cathode, yields ethyl alcohol :

+ 26

CH ! CH + H 2 O > CH 3 CHO > CH 3 CH 2 OH

<K + 2H

Conditions of electrolysis, e.g. temperature, degree of acidity, and the
concentration of unreduced aldehyde must be carefully controlled;

60

CATHODIC REDUCTION OF CARBONYL TYPE COMPOUNDS

otherwise, such side products as butanediol, ethyl ether, and butyl
alcohol will result.

Glyoxylic acid, when reduced at an iron, nickel, or silver electrode
in a neutral or alkaline medium, yields tartaric acid (I), whereas at a
lead, cadmium, or mercury cathode, the more highly reduced form,
succinic acid (II) is obtained (9):

zCHO.COOH

HOCH-COOH CH 2 COOH

+ 2e -f 2H

HOCH-COOH CH 2 COOH

I II

Aliphatic ketones are generally somewhat more susceptible to cathodic
reduction. However, the use of a basic medium is undesirable, because
of the ease with which these compounds undergo self-condensation.
Although a quantity of condensation products is obtained, it is possible
to obtain the pinacol and isopropyl alcohol, using a lead cathode and a
4 per cent aqueous sodium hydroxide medium for the reduction of
acetone. Using a lead cathode and an acid medium results in the
formation of both the pinacol and isopropyl alcohol without any conden-
sation products. With mercury, no pinacol is obtained; but isopropyl
alcohol and a small quantity of propane are isolated (10). So far as
can be determined, no electrolytic method has been found to give
satisfactory yields of the pinacol, although yields of isopropyl alcohol
as high as 40 per cent have been reported (n).

If the reduction is allowed to proceed in the absence of air at a
lead cathode, lead di-isopropyl [Pb(C 3 H 7 ) 2 ] and lead tetra-isopropyl
[Pb(C 3 H 7 ) 4 ] are obtained (12). This parallels the results obtained
chemically, when a sulphuric acid solution of acetone is treated with
lead-sodium (13).

Methyl ethyl ketone, when reduced at a lead cathode in a sodium
bicarbonate medium, yields secondary butyl alcohol (14). Using a
lead cathode and a 5 to 10 per cent sulphuric acid medium, the pre-
dominant product is methyl ethyl pinacol, with a small quantity of
secondary butyl alcohol (15). Other aliphatic ketones usually respond
to cathodic reduction in the same manner, with the product being
either the alcohol, pinacol, or hydrocarbon, or a mixture of these
end-products.

Aromatic aldehydes and ketones generally are similar in their
electrolytic behaviour to the aliphatic aldehydes and ketones. How-

61

ORGANIC ELECTRODE PROCESSES

ever, they can form pinacols more readily. The products obtained,
of course, depend on the conditions of electrolysis. For the preparation
of pinacols, in an acidic medium, the temperature generally must be
kept below 40 C. ; otherwise, rearrangement to the pinacone may
occur. The use of a basic medium must also be avoided when there is
the possibility of condensation taking place, as for example with amino
ketones, such as />-aminopropiophenone, and cycloparaffin ketones,
such as cyclopentanone (16). It has been found in many cases, that
aromatic aldehydes will form hydrobenzoins, and in some cases hydro-
carbons more readily than will ketones (17).

Benzaldehyde, upon reduction in a medium of alcohol and acid or
base, at cathodes of nickel, platinum, iron, or copper, yields benzyl alcohol.
Under certain conditions it has been reported (18) that in an acidic
medium it is possible to obtain benzyl alcohol, stilbene, and benzene.

The same investigator reports that in an alkaline medium the
following hydrobenzoins were obtained in 10 to 95 per cent yields.

Benzaldehyde > hydro- and isohydrobenzoin

Cuminaldehyde > hydro- and isohydrocuminoin

Anisaldehyde > hydro- and isohydroanisoin

Vanillin > hydrovanilloin

Salicylaldehyde > hydrosalicyloin

The reduction of a mixture of benzaldehyde, ethyl iodide, and
triethylamine at a copper cathode, to yield predominantly the ethyl
ether of ethyl phenyl carbinol (19), is also of interest.

C 2 H 5

C-OC 2 H 5
H

Halogenated benzaldehydes, such as ortho, meta, and para chloro-
benzaldehyde, have also been reduced electrolytically to the hydro-
benzoin, the stilbene, the benzyl alcohol and the toluene (20).

When m-nitrobenzaldehyde is reduced in an alcoholic-acid medium
at a copper cathode, azoxybenzaldehyde is obtained in the following
manner:

/NO = N
+6e+6H ,-^ I

CHO > OHC<f > J-*

^=/ <^J>CHO + 3 H,0

62

CATHODIC REDUCTION OF CARBONYL TYPE COMPOUNDS

At a higher overpotential electrode, such as lead, the m, w'-bis (w-
aminobenzalamino)-hydrobenzoin was obtained with excellent yield
(21). This product was obtained as a result of combined electrolytic
and chemical reactions in the following manner:

2 N H 2 N

J- 6e + 6H ,U

(0 \ ;>CHO -- > <f >CHO + 2H 2

\ - / \r ?./

w-aminobenzaldehyde
H 2 N H 2 N H 2 N

H <CHO + H a O

m- (w-aminobenzalamino)
benzaldehyde

>- v 2e + 2H

H /CHO

NH 2

H

OH OH
m, w'-bis (m-aminobenzalamino)-hydrobenzoin

The reduction of />-dimethylaminobenzaldehyde in an aqueous-
alcoholic potassium hydroxide medium at a reference potential v.
S.C.E. of 1-9 volts, with a mercury cathode, results in a 97 per cent
yield of the bimolecular compound (22). The high melting form of
the 4,4'-6w-dimethylaminohydrobenzoin precipitated out during the
reaction, and was obtained in a 39 per cent yield. The lower melting
form was obtained from the mother liquor in a 58 per cent yield:

OH OH

H H

The reduction of />-acetamidobenzaldehyde to 4,4'-Wi-/>-acetamido-
hydrobenzoin was accomplished using a mercury cathode at a reference

63

ORGANIC ELECTRODE PROCESSES

potential of 1-55 volts v. S.C.E. in an aqueous-alcoholic potassium
acetate medium at reflux temperature (23).

OH OH

HN/~CHO 2e + * H

2 | N==:/

C^O C=,0 H H

CH 3 CH 3 CH 3

/>-Hydroxybenzaldehyde, when reduced in an aqueous-alkaline
medium at a reference potential of 2-0 volts v. S.C.E., using a mercury
cathode, resulted in a 95 per cent yield of the desired hydrobenzoin
(24). A 93 per cent yield of p-hydroxybenzyl alcohol was obtained,
using a sodium bicarbonate medium and a mercury cathode (25). At
a cadmium cathode in an acidic medium, ^-cresol was obtained (26).

A 63 per cent yield of a mixture of hydrofuroin and isohydrofuroin
has been obtained (27) from the reduction of furfural with a lead cathode
in a potassium dihydrogen phosphate buffer, in the following manner :

CH - CH CH - CH HC = CH

26 + 2H

>0 0<

CH =CH CHO CH =C CH(OH)-(HO)HC C -CH

Aromatic ketones generally can also be reduced with satisfactory
yields to either the alcohol, the pinacol, or the fully reduced hydro-
carbon (28). Usually the alcohol is obtained with a lower overpotential
cathode, whereas the pinacol requires a higher overpotential electrode.
As with the aromatic aldehydes and the aliphatic ketones, aromatic
ketones are more readily reduced completely if a catalytic type electrode
is used.

It has been reported (29) that benzophenone, when subjected to
catHbdic reduction in an aqueous-acidic medium at a high overpotential
electrode, will give satisfactory yields of the rearranged benzpinacol,
benzpinacone, thus:

CATHODIC REDUCTION OF CARBONYL TYPE COMPOUNDS

OH OH

If this reduction is performed in an acetic-sulphuric acid medium the
resultant pinacone will be obtained, even at relatively low overpotential
electrodes, such as iron, nickel, and copper. Thus the effect of the
medium on the reaction can be seen, in an increase in the over potentials
of those electrodes (30) so that they were comparable in activity to a
high overpotential electrode in an aqueous medium.

The pinacols of />-aminoacetophenone and ^-aminopropiophenone
can be obtained by reduction of the respective ketones in an aqueous
hydrochloric acid medium, at a mercury cathode, using a cathodic
reference potential of 1-5 volts v. S.C.E. (31).

It might be of interest to point out that this is one of the instances
wherein the desired compound, 2,3-6*s-(^-aminophenyl)-2,3-butanediol,
could only be obtained using an electrolytic procedure. The usual
chemical methods gave only the carbinol.

The reduction of the ^-aminoacetophenone at a reference potential
v. S.C.E. of i -i volts in the same medium yields the carbinol. The
pinacol of ^-aminoacetophenone can also be obtained using a tin
cathode (32). At a cathodic reference potential of 2-2 volts v.
S.C.E. using a mercury cathode, />-hydroxyacetophenone, upon
electrolysis in an aqueous sodium hydroxide medium, gave a
77 per cent yield of the respective pinacol. The pinacol, 2,3-61$-
(]8-pyridyl)-2,3-butanediol, can also be obtained, by reduction of
j8-acetyl pyridine in aqueous hydrochloric acid using a mercury cathode
under controlled potential conditions (22) :

O K P. 65 F

ORGANIC ELECTRODE PROCESSES

OH OH

/

\CH 8 + 2e + 2H - r8v

Utilization of controlled potential conditions at a mercury cathode
has also resulted in the selective reduction of p-carboxyacetophenone
and ^-dimethylaminoacetophenone to their respective pinacols (33). A
number of Mannich bases have also successfully yielded the desired
pinacols under similar conditions (34). The Mannich base, />-acet-
amido-co-dimethyl-aminopropiophenone in an aqueous hydrochloric
acid medium, yielded 3,4-61$ -^-acetamidophenyl- 1 ,6-to-dimethyl-
aminohexane-3,4-diol :

+ 2e + 2H

- 1-7 v.

CH 2

CH,

N(CH,) 2

OH

OH

CH, CH,

CH 2 CH 2

N(CH 8 ) 2 N(CH 3 ) 2

Of interest is the preparation of the unsymmetrical pinacol, 2-p-
dimethylaminophenyl-3-/)-methoxyphenylbutane-2,3-diol from a mix-
ture of the ketones, using a mercury cathode (35).

(CH 9 ) 2 N

/

+ CH,

+ 26 + 2H

OH

(CH S ) 2 N<(^>C

CH,

OH

CH

CH 8

This compound was obtained using either an aqueous-ethanol

66

CATHODIC REDUCTION OF CARBONYL TYPE COMPOUNDS

hydrochloric acid medium at a reference potential of ro6 volts v.
S.C.E., an aqueous-ethanolic-potassium acetate medium at a reference
potential of 1-97 volts v. S.C.E., or an aqueous-ethanolic-potassium
hydroxide medium at a reference potential of 1*86 volts v. S.C.E.
The latter two media were kept at reflux temperature during the reaction.
The yield of unsymmetrical pinacoi was 30 to 37 per cent, depending
upon the medium used.

The reduction of a number of oc-aminoketones, in an aqueous
sulphuric acid medium, at various electrodes and different temperatures,
has been reported (36). The reduction of i-methyl-2-ethyl-3-piperi-
done (I) at 60 C. with a cadmium electrode, yielded i-methyl-2-
propylpyrrolidine (II) and i-methyl-2-ethyl-3-hydroxypiperidine (III).
At a temperature of 20 C. only N-methyl-heptylamine (IV) could be
isolated :

O

26 +2H

N C 2 H 5
CH 3

O

CH 2 C 2 H 5

NH

I
CH 3

r

l_

OH

N
CH 3

H 2

H 2 O

N C 3 H 7

I
CH 3

2e 4-2H

OH

4H

2e

N C 2 H B

CH 8
III

CH 3 NHC 7 H 15

IV

N C 8 H 7

CH 3
II

With a lead cathode at 60 C. N-methylheptylamine (IV) and i -methyl-
2-propylpyrrolidine (II) were obtained. The same cathode at 20 C.
yielded only the N-methylheptylamine (IV). A copper cathode at

67 F 2

ORGANIC ELECTRODE PROCESSES

60 C. yielded i-methyl-2-propyl-2-pyrroline (V) and i-methyl-2-
propylpyrrolidine (II). This same electrolysis, when performed at
20 C., gave a quantitative recovery of the starting material. Similar
results were obtained when a tin electrode was used, with electrolysis
temperatures of 20 C. and 60 C.

The results indicate that it is possible to obtain a reductive-
rearrangement of the Clemmensen type, by electrolysis. The fact that
no apparent reaction took place with a tin electrode might indicate that
this type reaction is not dependent only on the potential of the electrode,
but possibly on the catalytic effects of the electrode material. Tin is
a higher overpotential electrode than copper or cadmium but lower
than lead. It is interesting to observe that reduction at 20 C. gave only
N-methylheptylamine. At this temperature the overpotential of the
electrode would be higher, and thus one might expect the more highly
reduced form. At the lead cathode at 60 C., the N-methylheptylamine
and i-methyl-2-propylpyrrolidine are obtained, with the former com-
pound predominating. Increasing the potential of this already high
potential electrode by performing theL reaction at 20 C. yielded only
the N-methylheptylamine.

The electrolytic method has also been applied to the synthesis of
medium-size ring compounds containing a nitrogen atom (37). This is
achieved by reduction of bicyclic or tricyclic a-aminoketones at 60 C.
in an aqueous sulphuric acid medium, using a lead cathode. As an
example, 7-keto-i-azabicyclo 6.4.0 dodecane, upon reduction under
these conditions, yields y-hydroxyazacyclodecane, thus:

O OH

N

(2) REDUCTION OF CARBOXYLIC ACIDS

The two available possibilities for the reduction of the carbonyl
group of a carboxylic acid result in the formation of an aldehyde which
can be further reduced to the alcohol, depending on the conditions of
electrolysis :

68

CATHODIC REDUCTION OF CARBONYL TYPE COMPOUNDS

RC

o

OH

2H + 26

OH

RC H
OH

-H 2

O

II
RCH

2e

RCH 2 OH

Generally, when the reduction is performed in an acidic medium at
a low overpotential electrode such as platinum, nickel, or iron, no reaction
takes place. However, at a higher overpotential electrode, such as lead,
the primary alcohol will be obtained.

The monobasic acids do not usually behave in this manner, although
the lowest members of the aliphatic series, that is, formic, phenylacetic,
oxalic, and glyoxylic acid, do respond to electrolytic reduction. Formic
acid, when dissolved in a 10 per cent sulphuric acid medium and sub-
jected to cathodic electrolysis in a divided cell at a lead electrode,
especially in the presence of a " hydrogen carrier " such as eerie
sulphate, has been reported by Ellis and McElroy to yield either
formaldehyde at a low current density (low electrode potential) or
methanol at a high current density (high electrode potential) (38):

HCOOH

-f 26 + 2H

low c.d.

HCHO

+ 2e -f 2H
high c.d.

CH.OH

However, it should be mentioned that a subsequent investigator (39)
was unable to duplicate the results of Ellis and McElroy.

Phenylacetic acid can also be reduced to the corresponding alcohol
(40). However, the ease of reduction of this acid when substituents are
present on the benzene ring will, to a great extent, possibly be determined
by the inductive effect of these substituents. In an acid medium at a
mercury cathode, />-hydroxyphenyl acetic acid is not reduced to the
corresponding alcohol, whereas j>-dimethylaminophenyl acetic acid will
give the desired carbinol (41).

Oxalic acid, upon reduction in an aqueous sulphuric acid medium
at a mercury or lead cathode, will yield glyoxylic acid (I) when the
reaction is performed at temperatures below 10 C., and glycolic acid
(II) at temperatures over 40 C. (42).

ORGANIC ELECTRODE PROCESSES
COOH CHO CH 2 OH

| 2H + 2C | 2H -f 26 |

COOH > COOH > COOH

I II

CH(OH)COOH

CH(OH)COOH
III

As glyoxylic acid will, upon reduction in an acid medium, yield tartaric
acid (III) (43) in a manner similar to that in which aldehydes and
ketones will yield bimolecular products, it is possible that glyoxylic acid
as such may not be intermediate to glycollic acid. It is, however,
entirely possible that if the glyoxylic acid were present in low concen-
tration, the opportunity for collision of the activated intermediate (the
free radical form) with another of its kind would be minimized, and thus
further reduction to the alcohol would be anticipated. This is further
indicated by experiments in which the carbinol was obtained from
/>-aminoacetophenone, when the latter was present in small quantities
in the medium and subjected to electrolysis conditions which would
have normally yielded the pinacol at a higher ketone concentration (44).
Thus it is entirely possible, due primarily to the concentration of the
depolarizer, to obtain either of two products.

Of interest is the reduction of carbonic acid, as the carbonate, at
mercury, amalgamated lead, amalgamated zinc, or zinc electrodes, to
give almost quantitative yields of formic acid (45). The reduction of
carbon monoxide to formaldehyde has also been claimed (46).

Aromatic acids are readily reduced to the alcohol in an acidic medium
at a high overpotential electrode, such as lead or mercury (47). This
method has been successfully applied to the preparation of substituted
benzyl alcohols (48), which in many instances are difficult to obtain by
purely chemical means, whereas the respective benzoic acids are readily
obtainable. Some of the reductions achieved are as follows :

o-,m-, and ^-chlorobenzoic acid > o-, m-, and />-chlorobenzyl alcohol

o-, and m-nitrobenzoic acid > o-, and m-aminobenzyl alcohol

m-bromobenzoic acid > m-bromobenzyl alcohol

3,5-dit|como-2-hydroxybenzoic acid > 3,5-dibrorro-2-hydroxybenzyl alcohol

3 -hydroxy benzoic acid > 3 -hydroxy benzyl alcohol

5-chloro-2-hydroxybenzoic acid > 5 -chloro-2-hydroxy benzyl alcohol

Reductions of this type are usually performed in an aqueous-ethanol

70

CATHODIC REDUCTION OF CARBONYL TYPE COMPOUNDS

sulphuric acid medium, and in order to preclude ester formation, the
temperature is maintained below 30 C. If it is possible to eliminate
the ethanol from the medium, higher temperatures may be used. How-
ever, caution must be used with the substituted benzoic acids, as there is
always present the possibility of decarboxylation at elevated temperature
(49). The aldehyde is only obtained in the presence of a sodium salt
and boric acid at a mercury cathode (50), thus salicylic acid under these
conditions yields salicylaldehyde :

OH OH

COOH

26 + 2H

Y-CHO

The reduction of the carboxyl group does not usually take place in a
basic medium, as indicated by the fact that />-carboxyacetophenone,
when subjected to cathodic reduction at a mercury cathode, at a reference
potential of 2-0 volts v. S.C.E. in an aqueous sodium hydroxide
medium, yields the respective pinacol (51).

Another interesting example is the reduction of the potassium salt
of 2-carboxybenzaldehyde at a mercury cathode to yield hydrodiphthalyl
(III), and a small quantity of phthalide (IV) (52). It is probable that
the initial reaction is electrolytic in nature, with the reduction of the
aldehyde group to the pinacol (I) and 2-carboxybenzyl alcohol (II).
These then lactonize to yield the isolated products, thus :

H

~CH 2 OH
COOH

IV

ORGANIC ELECTRODE PROCESSES

Aromatic dicarboxylic acids are reduced to the corresponding
phthalide, or dihydric alcohol, depending on the position on the ring of
the carboxyl groups. Phthalic acid, when reduced at a lead cathode in
a neutral or weak'y acidic medium, will give phthalide (53), with
phthalic anhydride being intermediate in the reaction, thus:

The phthalide can also be prepared by reduction of phthalic
anhydride.

Isophthalic acid upon reduction at a lead cathode gives w y o/-
dihydroxy-w-xylene (54), thus :

HOOC (HO)H 2 C

~ + 8H + 8e , -/~~\CH 2 (OH) + 2 H 2 O

The carboxyl group of /?-keto acids is not attacked but the ketone
group is reduced to either carbinol or methylene compound.

(3) REDUCTION OF ESTERS

In some cases the esters of aliphatic acids respond to cathodic
reduction. There is, however, an indication that the nature of the acid
portion of the molecule influences the ease of reduction. The ethyl
esters of acetic, phenylacetic, and cyanoacetic acids are not reduced,
whereas those of fumaric, tartaric, and succinic acids are reduced slowly.
The esters of oxalic, malonic, acetoacetic, oxalacetic, benzoic, halo-
benzoic, and phthalic acids are reduced quite readily at a high over-
potential electrode in an acidic medium.

The products obtained from the reduction of acetoacetic acid, and
its derivatives, result in a number of compounds in which either or both
carbonyl groups are acted upon electrolytically. Generally, the final
product is the hydrocarbon. Using a lead cathode, an aqueous-alcohol
sulphuric acid medium and elevated temperatures results in the
formation of butane from the ethyl ester of acetoacetic acid (56). The
reduction of butylacetoacetic acid at a lead cathode in an aqueous-
alcohol sulphuric acid medium yields n-octane, thus :

72

CATHODIC REDUCTION OF CARBONYL TYPE COMPOUNDS
O

CH 3 C CHC OC 2 H 5 -> CH 3 (CH 2 ) a C 4 H 9

I

O C 4 H 9

This indicates that an isomerization has occurred during the course of
the reduction; otherwise, 3-methyl heptane, would have been obtained.
Isobutylacetoacetic ester yields octane, and not the expected product,
2,4-dimethylhexane (57).

In the aromatic series the corresponding alcohol or ether is generally
obtained. The initial reaction is the cathodic reduction of the carbonyl
group to the fully reduced form, to yield the ether, which may then
cleave, due to the acidity of the medium, to give the alcohol (58):

C 6 H 5 COOC 2 H 5 + 4 H + 4e > C 6 H 5 CH 2 OC 2 H 5 + H 2 O

C 6 H 6 CH 2 OC 2 H 5 + H 2 H + y C 6 H 6 CH 2 OH + C 2 H 5 OH

(4) REDUCTION OF AMIDES AND IMIDES

The carbonyl groups of both amides and imides can generally be
fully reduced at a high overpotential electrode in an acidic medium.
The ease with which this group is reduced in amides and thio-amides,
is to a great extent dependent upon the nature of the substituent on the
nitrogen atom (59). It has been found that the greater the number of
methyl groups attached to the nitrogen of oc-phenylacetamide the more
rapidly does the electrochemical reduction occur. Thus the reduction
of a-phenylacetamide will yield less than i per cent of j3-phenylethyl-
amine, whereas N-methyl-oc-phenylacetamide yields 80 per cent of
N-methyl-/J-phenylethylamine, and N,N-dimethyl-a-phenylacet-
amide yields 92 per cent of the corresponding amine. It was also
found that a lead cathode gives a better yield of amine in the reduction
of N,N-dimethyl-a-phenylacetamide, than a zinc-amalgam electrode.
Addition of a small quantity of arsenious or antimonic oxides increased
the yield at a lead cathode still further (60).

In addition to the effect of replacement of the N-hydrogens by
methyl groups on the ease of reduction of amides, other workers (61)
found that replacement by phenyl groups also had a desirable effect.
Their findings further indicated that substitution of one or more phenyl

73

ORGANIC ELECTRODE PROCESSES

groups on the a-carbon of acetamide improved ease of reduction.
However, the best results are obtained when the aryl group is linked
directly to the carbonyl. The reactivity is decreased as the carbonyl
and aryl group are separated by CH 2 groups. It appears that the
greater the electron donating effect of the substituents on the nitrogen
and the a-carbon, the easier it is for reduction to occur.

Compounds of the morpholinone series which are essentially cyclic
amides, can be reduced electrolytically (62). The first step is a cleavage,
via a two-electron change, to form the N-substituted acetamide ; then
a four electron change to yield the secondary amine, thus :

H a

N 1 C-O

2H

H

! N CCH 3

; N CH 2 CH 3
H

The electrolytic reduction of diketopiperazines is of interest, as it
is by this method that it is possible to obtain a-aminoaldehydes.
Chemical methods for the preparation of such compounds are not
generally successful. Upon reduction, the carbon-nitrogen bond is
usually ruptured, resulting in two a-aminoaldehyde molecules (63).

O H

II I
C N

R C H

H C R 4 H + 4 e 2

O

II
CH

R C--H

N-
H

NH a

O

In many respects the cathodic reduction of imides is similar to that
of the amides. Whereas an intermediate has never been isolated from the

74

CATHODIC REDUCTION OF CARBONYL TYPE COMPOUNDS

reduction medium of an amide, a partially reduced hydroxy-phthali-
midine has been obtained by reduction of N,N-diethylaminoethyl tetra-
chlorophthalimide in an acetic-hydrochloric acid medium at a mercury
cathode, utilizing a reference potential of 1-2 volts v. a S.C.E. (64).

rl O p . HOH

Cl it Cl I /

J\ ^\ I/

cif y-cx cl r V c \

>NCH 2 CH 2 N(C 2 H 5 ) 2 2H+2e >NCH 2 CH 2 N(C a H B ) 2

ciL J-c/ > ciL J-c/

ci I ci I

C1 o C1 o

The same results were obtained with a similar phthalimide at a lead
cathode in an aqueous-acetic sulphuric acid medium, at a reference
potential of 0-68 volt v. a S.C.E. (3). The preparation of hydroxy-
phthalimidine has also been accomplished by performing the reduction
in an acid medium at a copper electrode (65).

In the literature can be found many examples wherein phthalimides
have been successfully reduced to the phthalimidine, and thence on to
the isoindoline (66), thus :

In many cases, although satisfactory yields of the desired isoin-
dolines are obtained at a high overpotential electrode in an aqueous
acid medium, the current efficiency is generally quite poor. As an
example, we might consider the results obtained in the reduction of
a suspension N-methyl phthalimide at a lead electrode in an aqueous-
sulphuric acid medium (67). At a current density of 0*0445 amp./cm. 2
a 577 per cent yield of the respective isoindoline was obtained with a
14*4 per cent current efficiency. At this current density, even in the
initial period of electrolysis, there is considerable hydrogen evolution,
which explains the reason for low current efficiency. The simple
expedient of increasing the solubility of the compound and the over-
potential of the electrode, by addition of glacial acetic acid to the
medium, increased the current efficiency of the reaction to 42 per cent

75

ORGANIC ELECTRODE PROCESSES

with an 82 per cent material yield of N-methylisoindoline (68). Thus it
appears that, as would be expected, by increasing the concentration, the
"limiting current" for the reaction was increased, with the net result
being that less of the current was used for gas production and more for
the actual reduction of the carbonyl groups. It is possible that the
current efficiency might have been improved still further by decreasing
the current density as the reduction proceeded, to the true " limiting
current " for the particular concentration of depolarizer; of course,
the increase in electrode overpotential in this medium, as compared
with that in an aqueous medium, must not be overlooked as a con-
tributing factor in the improved results. N,N-Dimethylaminoethyl
tetrachlorophthalimide, upon reduction in an aqueous-acetic- sulphuric
acid medium at a lead cathode and at a current density of 0-049 amp./
cm. 2 , results in a 91 per cent material yield of the N,N-dimethyl-
aminoethyl tetrachloroisoindoline (3).

8e

>NCH 2 CH 2 N(CH 3 ) 2 > I I >NCH 2 CH 2 N(CH 3 ) 2

CX 8H

II
O

The efficiency of the reduction, based on the material yield of
isoindoline, is 46 per cent. The same reduction, performed at a controlled
potential of iio volts v. S.C.E., results in a 77-7 per cent efficiency,
based on the material yield of the isoindoline. The current density,
when the electrolysis is performed under these conditions, diminishes
as the electrolysis progresses. This increased efficiency is to be expected,
for from the concept of " limiting current, " which is directly related to
the concentration of the depolarizer, only that current is being supplied
which is necessary for the reduction of the phthalimide. As a conse-
quence there is a negligible amount of hydrogen evolved.

In the same report (3), are the results obtained with three high
overpotential electrodes; namely cadmium, lead and mercury. These
are shown in Fig. iv:i. Despite the high overpotential
characteristics of the mercury electrode, the best yield obtainable was
approximately 80 per cent, whereas with lead, a slightly lower over-
potential electrode, under these experimental conditions, gave yields of
over 90 per cent.

76

CATHODIC REDUCTION OF CARBONYL TYPE COMPOUNDS

FIG

02 04 06 08 .10 12 .14 16 18 20 22 24 26 28 .30
CURRENT DENSITY (omp./cm2)

iv : i Dependence of isoindoline yield on current density

This would tend to indicate a possible catalytic effect of this electrode
material on the reduction. A cadmium electrode, which is the lowest
overpotential electrode of the three studied, gave relatively poor yields
of the desired isoindoline. In all cases in which high yields of the
isoindoline were not obtained, the partially reduced phthalimidine was
also isolated from the reaction mixture.

The two carbonyl groups in the tetrachlorophthalimide molecule are
not reduced simultaneously. It is possible to obtain excellent yields of
the phthalimidine by merely permitting the electrolysis to proceed for
only 50 per cent of the time required for complete reduction to the
isoindoline.

If a completely aqueous system is used the desired isoindoline may
not be obtained. This holds with phthalimides, which have halogen
substituents in the benzene ring. As an example, we might consider the
electrolytic reduction of N,N-dimethylaminoethyl tetrachlorophthal-
imide in a 5-7 per cent aqueous sulphuric acid medium, using a lead
electrode (69). The medium, current density, temperature, and ampere
hours per mole were the same as used to successfully reduce N-
methylphthalimide to N-methylisoindoline, with, however, the accom-
panying low current efficiency (67). Throughout the course of the
reaction gas evolution was excessive, with the suspended imide finally

77

ORGANIC ELECTRODE PROCESSES

going into solution. From the catholyte was isolated a compound
which, upon analysis, and from degradation studies, was found to be
N,N-dimethylaminoethyl-3,4,6-trichloroisoindoline.

Cl

H 2

Cll

C

H 2
Cl

'\NCH 2 CH 2 N(CH 3 ) 2
/

If the electrolysis is allowed to proceed for approximately one-
third the time used to obtain the trichloroisoindoline, three products
are isolated: N,N-dimethylaminoethyl tetrachlorophthalimide (I), N,N-
dimethylaminoethyl-3,4,5,6-tetrachloro-2-hydroxyphthalimidine (II),
and the N,N-dimethylaminoethyl-3,4,6-trichloroisoindoline (III). Pro-
longed electrolysis did not remove more than the one ring halogen. It
might be of interest to deviate slightly and discuss briefly the mechan-
ism of this reaction, so as to better understand how, under certain
conditions, competitive electrochemical reactions are possible, because
of a lack of energy sufficient to drive the reaction in the desired direction
in an efficient manner.

Based on the results, the mechanism shown on page 79 was proposed.

As the hydroxyphthalimidine (II) was obtained from the short
period electrolysis it may be assumed to be an intermediate prior to
dehalogenation. As a result of the low efficiency of the system the
subsequent reduction of the hydroxyphalimidine (II) to the isoindoline
(IV) is relatively slower than under more efficient conditions, i.e. an
aqueous-acetic-sulphuric acid medium (p. 75). Therefore, the inter-
mediate suggested by (III), which is possibly stabilized in this />-
quinoid form by adsorption on the cathode surface, undergoes nucleo-
philic substitution on carbon 5, with a two-electron change, and the
addition of hydrogen. As the energy required for the molecule to
assume an o-quinoid structure may be too great, the probability of any
amount of the 4,5,6-trichloro-2-(2-dimethylaminoethyl)-isoindoline
being formed is quite unlikely. The proposed mechanism also accounts
for th& fact that only one chlorine is removed from the ring, for as soon
as the carbonyl group is reduced resonance with the benzene ring is
impossible. Were this not so, the expected product would be the 4,7-
dichloroisoindoline. This was confirmed by the fact that N,N-dimethyl-

CATHODIC REDUCTION OF CARBONYL TYPE COMPOUNDS

NCH 2 CH 2 N(CH 3 )o

iion

V

b

C1

sr
o

HOH

II

H OH

2C

Cl

Cl '-'
6e -f 611

ci 'y
in

lo| e

a

H

H

Cl

NCH 2 CH 2 N(CH 3 ) 2

x f

Cl

H '

IV

aminoethyl tetrachlorophthalimidine, when reduced under identical
conditions, yielded the trichloroisoindoline.

This demonstrates an instance in which an electrolysis on a low
efficiency level permitted the energy within the molecule to dominate
the path of reaction. Using more efficient conditions, the carbonyl
reduction was fast enough to minimize the effect of resonance with
resultant substitution.

The cyclic imides also respond satisfactorily to reduction at a high
overpotential electrode in an acidic medium. Succinimide is readily
reduced to pyrrolidone at a lead electrode (70).

o

79

ORGANIC ELECTRODE PROCESSES

At a zinc-amalgam electrode the completely reduced product, pyrroli-
dine (71) is obtained. Substituted succinimides, such as the N-methyl,
N-ethyl, N-isopropyl, and N-phenyl, have been reduced to their
respective pyrrolidines (72), The glutarimides also respond to cathodic
reduction in the same manner (73).

We might consider purine compounds as belonging in the general
category of imides. Generally, the carbonyl group of the acid is reduced
preferentially to the urea carbonyl group, and thus reduction of bar-
bituric acid at a lead cathode in a 50 per cent sulphuric acid medium
gives dihydrouracil and finally trimethylene urea (74).

NH CO

NH CO

4 H

NH CH 2

CO CH 2

I I
NH CO

4 H

NH CH 2

I I

I I
NH CH 2

However, if we activate the urea carbonyl by replacement of the
methylene group by a >C(C 2 H 5 ) 2 the urea carbonyl group is reduced,
in preference to the carboxyl carbonyls (75). Thus 5,5-diethyl-
barbituric acid will give 5,5-diethyl-4,6-dioxy-i,2-dihydropyrimidine.

NH CO

I I
CO C(GJ1 5 ) 2

NH CO

4 H

NH CO

Z H 5 )

NH CO

Xanthine and its derivatives, such as theophyllin, behave in a
manner very similar to that of barbituric acid, in that the urea carbonyl
remains unaffected by cathodic reduction (76). Theophyllin will, on
reduction at a lead cathode in an aqueous sulphuric acid medium, yield
desoxy-theophyllin.

CH 3 N CO

H

CO C N

CH 3 N C N

>CH

4 H

4 e
80

CH 3 N CH 2

H

CO C N

CH 8 N CN

H

CATHODIC REDUCTION OF CARBONYL TYPE COMPOUNDS

REFERENCES

(1) Shima, Mem. Coll. Sci. Kyoto, 1928, n, i.

(2) Tsrfd and Schmitz, Z. Elektrochem., 1902, 8, 281; Escherich and Moest,
ibid., 1902, 8, 849; Willstater and Veraguth, Ber., 1905, 38, 1985; Bredt,
et al., Annalen, 1906, 348, 199; 1909, 366, i; J. pr. Chem., 1911, 84,
786; Law,.?, chem. Soc., 1906, 89, 1512; 1907, 91, 748; 1911, 99, 1113;
Tafel, Ber., 1909, 42, 3146; Tafel and Schepps, Z. Elektrochem., 1911,
17, 972; Schepps, Ber., 1913, 46, 2564; Muller, Dissertation, Dresden
(1921); Penfold and Morrison,.?. Roy. Soc. N.S. Wales, 1923, 57, 215;
Schall and Kirst, Z. Elektrochem., 1923, 29, 537; Shima, Mem. Coll. Set.
Kyoto, 1930, i3A, 315, 327; Swann, Univ. of Illinois Eng. Exp. Stat.
Bull., 1931, No. 236; Trans. Amer. electrochem. Soc., 1932, 62, 153;
1933, 6 4 3i3; 1933, 65, 239; McKee and Brockman, ibid., 1932, 62,
203; Price and Mueller, J. Amer. chem. Soc., 1944, 66, 634; Allen,

y. org. Chem., 1950, 15, 435; J. Amer. chem. Soc., 1950, 72, 3797;
I95 1 , 73t 3503; y. chem. Soc., 1951, 1598; Allen and Corwin, J. Amer.
chem. Soc., 1950, 72, 114; 117; Levine and Allen, J. chem. Soc., 1952,
254; Allen, Fearn and Levine, ibid., 1952, 2220.

(3) Allen and Ocampo, y. Amer. electrochem. Soc., 1956, 103, 452.

(4) D.R.P., 274201, 1912; 277393, 1913; 328342, 1917; B.P., 140115,
I 9 I 8; 335638, 1929; Hibbert and Read, J. Amer. chem. Soc., 1924,
46, 983; Shima, Mem. Coll. Set. Kyoto, 1925, 9A, 183; Kagan and
Kryukov, J. Chem. Ind. (Moscow), 1933, 10, 34.

(5) F. Muller, Diss., loc. cit., ref. 2.

(6) Swann, Onstott and Baastad, Trans. Amer. electrochem. Soc., 1955, 102, 113.

(7) Findlay and Williams, Trans. Faraday Soc., 1922, 17, 453; Creighton,
Trans. Amer. electrochem. Soc., 1939, 75, 289; U.S. P., 1612361, 1653004,
1712951, 1712952, 1926; Swiss P., 213617, 1929; F.P., 760507, 1934.

(8) B.P., 140115, 1918; 140527, 1919; Pascal, Swiss P., 88118, 1921.

(9) Baur, Z. Elektrochem., 1931, 37, 255; F.P., 455972, 1912; D.R.P.
292865, 1912; U.S.P., 1181555, 1190845, 1916.

(10) Elbs and Brand, Z. Elektrochem., 1902, 8, 784, 786; Tafel and Schmitz,
Z. Elektrochem., 1902, 8, 288; Tafel, ibid., 1911, 17, 975; Schepss, Ber.,
1913, 46, 2566; Muller, Z. Elektrochem., 1927, 33, 253; Kirchhof and
Stepanow, Chem-pharm. Ind. (russ.) 1932, 21 ; Kaneyama and Takahashi,
y. Soc. Chem. Ind. Japan, 1933, 36, 174; Foresti, Atti. Soc. ital. Progr.
Sci., 1939, 5 346; Wilson and Wilson, Trans. Amer. electrochem. Soc.,
1941, 80, preprint 30; D.R.P., 113719, 1899; 252759, 1911; 303303,
306304, 306523, 310023, 324919, 324920, 1917.

(n) D.R.P., II37I9, 1899.

(12) Tafel, Ber., 1911, 44, 327.

(13) Goldach, Helv. Chim. Ada, 1931, 14, 1436.

(14) Tafel, Ber., 1912, 45 448.

(15) Elbs and Brand, loc. cit., ref. 10.

(16) Godchot and Taboury, Bull. Soc. chim., 1913, 13, [47], 12.

(17) Kaufmann, Z. Elektrochem., 1895, 2, 365; 1898, 4, 461; Law, J. chem.
Soc., 1906, 89, 1512; 1907, 91, 748; 1911, 99, 1113; Tafel and Schepss,
Ber., 1911, 44, 2148; Schepss, ibid., 1913, 46, 2564; Shima, Mem.
Coll. Sci. Kyoto, 1928, uA, i, 407, 419; 1929, nA, 69, 73; Rapson
and Robinson, J. chem. Soc., 1935, 1537-

O E.P. Ol ( -

ORGANIC ELECTRODE PROCESSES

(18) Law, J. chem. Soc. t 1907, 91, 748.

(19) Nelson and Collins, J. Amer. chem. Soc., 1924, 46, 2256.

(20) Law,J. chem. Soc., 1911, 99, 1115.

(21) Shima, Mem. Coll. Sci. Kyoto, 1928, n, i.

(22) Allen, y. org. Chem., 1950, 15, 435.

(23) Allen, Fearn, and Levine, .? chem. Soc., 1952, 2220.

(24) Allen, y. Amer. chem. Soc., 1950, 72, 3797.

(25) Shima, Mem. Coll. Sci. Kyoto, 1928, n[A], 407.

(26) Schepss, Ber. t 1913, 46, 2570.

(27) Albert and Lowy, Trans. Amer. electrochem. Soc., 1939, 75, 367.

(28) Woehler, Annalen, 1844, 51, 153; Kauffman, Z. Elektrochem.,
1898, 4, 461; Elbs and Brand, Z. Elektrochem., 1902, 8, 784; Haber
and Russ, Z. physik. chem., 1904, 47, 297; Kaufler, Z. Elektrochem.,
1910, 16, 236; Law, y. chem. Soc., 1912, 101, 1550; Swann and Nelson,
Trans. Amer. electrochem. Soc., 1935, 67, 201 ; Swann, et al., ibid., 1944,
85*231.

(29) Swann, Trans. Amer. electrochem. Soc., 1933, 64, 245.

(30) Swann and Edelmann, Trans. Amer. electrochem. Soc., 1930, 58, 179.

(31) Allen and Corwin, y. Amer. chem. Soc., 1950, 72, 114, 117.

(32) Leonard, Swann, and Fuller,^. Amer. chem. Soc., 1953, 75, 5127.

(33) Allen, y. chem. Soc., 1951, 1598; y. Amer. chem. Soc., 1951, 73,

3503.

(34) Allen, Fearn, and Levine, loc. cit., ref. 23.

(35) Levine and Allen, y. chem. Soc., 1952, 254.

(36) Leonard, Swann and Dryden, y. Amer. chem. Soc., 1952, 74, 2871.

(37) Leonard, et al., y. Amer. chem. Soc., 1952, 74, 4620, 6251; 1954,
76, 3193.

(38) U.S.P., 867575, 1907.

(39) Baur, Z. Elektrochem., 1919, 25, 102.

(40) Marie, Marquis and Birkenstock, Bull. Soc. chim., 1919, 25, 512; Inoue,
y. Chem. Ind., Japan, 1921, 24, 906.

(41) Allen, unpublished results.

(42) Royer, Comp. rend., 1869, 69, 1374; Avery and Dales, Ber., 1899, 32,
2237; Tafel and Friedrichs, ibid., 1904, 37, 3187; Baur, Z. Elektrochem.,
1919, 25, 102; Mohrschulz, ibid., 1926, 32, 434; D.R.P., 163842, 1903;
190438, 1907; 210693, 1909; 239312, 243746, 1910; F.P., 456165, 1912;
U.S.P., 873038, 1907; 1013502, 1059740, 1911; 1227706, 1917.

(43) U.S.P., 1181555, 1190845, 1916.

(44) Allen and Corwin, unpublished results.

(45) Royer, C.r. Accid. Sci. Paris, 1870, 70, 731; Bach, ibid. 9 1898, 126,
479; Coehn and St. John, Ber., 1904, 37, 2836; Ehrenfeld, ibid., 1905,
38, 4138; Fisher and Priziza, ibid., 1914, 47, 256; Rabinowitsch and
Mashowetz, Z. Elektrochem., 1930, 36, 846; Ukrain. Chem.J., 1931, 6,
(Sci), 217; U.S.P., 1185028, 1916.

(46) Fester and Schivazappa, Z. anorg. Chem., 1928, 171, 163.

(47) Mettler, Ber., 1905, 38, 1745; 1906, 39, 2933; Inoue, y. Chem. Ind.
Japan, 1921, 24, 906; Decans and Dufour, Bull. Soc. chim., 1925, 37,
1*167; Somlo, Z. Elektrochem., 1929, 35, 769; Fichter and Stein, Helv.
Chim. Ada, 1929, 12, 821 ; Mayer, Schafer, Rosenbach and Schonfelder,
Arch. Pharm., 1929, 267, 571.

(48) Mettler, loc. cit., ref. 47.

(49) Somlo, loc. cit., ref. 10.

82

CATHODIC REDUCTION OF CARBONYL TYPE COMPOUNDS

(50) Weil, Ber., 1908, 41, 4147; Mettler, ibid., 4148; Tesh and Lowy,
Trans. Amer. electrochem. Soc., 1924, 45, 37; Rutovski and Korolev,
Trans. Sci. Chem. Pharm. Inst. (Moscow), 1928, 19, 177; Kawada and
Yosida, Bull. Hyg. Research Inst. (Japan), 1929, 35, 261.

(51) Allen, J. Amer. chem. Soc., 1951, 73, 353-

(52) Rodionow and Lewtschenko, Chem. Zbl. y 1938 II, 42 14.

(53) Sakurai, Bull. chem. Soc. Japan. , 1932, 7, 127; Define and Somlo,
IX Congr. int. Chim., 1934 IV, 360; Rodionow, Lewtschenko and
Zvorykina, Bull. Soc. chim. France, 1937, [5] 4, 463.

(54) Mettler, Ber., 1906, 39, 2940.

(55) Tafel and Emmert, Z. Elektrochem., 1911, 17, 569.

(56) Tafel and Jtirgens, Ber., 1909, 42, 2554.

(57) Tafel, et al., Ber., 1907, 40, 3312; 1909, 42, 2548; 1912, 45, 437; Stenzl
and Fichter, Helv. Chim. Ada, 1934, 17, 665.

(58) Tafel and Friedrichs, Ber., 1904, 37, 3197; Mettler, ibid., 3695.

(59) Kindler, Ber., 1924, 57, 773; Arch. Pharmaz., 1927, 265, 389.

(60) Kindler, Ber., 1923, 56, 2063.

(61) Gawrilow and Koperina, J. Gen. Chem. (U.S.S.R.), 1939, 9, 1394;
Koperina and Kljutscharewa, ibid., 1941, n> 51.

(62) Lees and Shedden,^. chem. Soc., 1903, 83, 750.

(63) Heimrod, Ber., 1914, 17, 338.

(64) Allen and Ocampo, unpublished results.

(65) Sakuari, Bull. chem. Soc. Japan, 1930, 5, 184; Dunet and Willemart,
Bull. Soc. chim., 1948, 887.

(66) Spath and Breusch, Monatsh., 1928, 50, 349; Sakuari, Bull. chem. Soc.
Japan, 1930, 5, 184; 1932, 7, 155; Clemo, Raper, Tenniswood, J. chem.
Soc., 1931, 429; Cook and France, J. phys. Chem., 1932, 36, 2383;
Lukes and Smetachova, Coll. Czech, chem. comrn., 1933, 5, 61 ; Spath,
Kuffner and Kittel, Ber., 1939, 72, 1109; Sakurai, Bull. chem. Soc.

Japan, 1939, 14* i?3 J Spath and Hillel, Ber., 1939, 72, 1577.

(67) Cook and France, J?. phys. Chem., 1932, 36, 2383.

(68) Allen and Ocampo, unpublished, results.

(69) Allen and Ocampo, Trans. Amer. electrochem. Soc., 1956, 103, 682.

(70) Tafel and Stern, Ber., 1900, 33, 2224; Tafel and Emmert, Z. physik.
Chem., 1906, 54* 433; Breusch, Monatsh., 1929, 50, 349.

(71) Sakurai, Bull. chem. Soc. Japan, 1935, 10, 311.

(72) Tafel and Stern, Ber., 1900, 33, 2232; Craig, J. Amer. chem. Soc., 1933,
55> 295; Sakurai, Bull. chem. Soc. Japan, 1936, n, 41; ibid., 1938, 13,

35-

(73) Lukes, Coll. trav. chim. Techec., 1932, 4, 351; Lukes and Smetackova,
ibid., 1933, 5> 61 ; Sakurai, Bull. chem. Soc. Japan, 1938, 13, 482.

(74) Tafel, et al., Ber., 1900, 33, 3378; 1907, 40, 4489.

(75) Tafel and Thompson, Ber., 1907, 40, 4490.

(76) Tafel, et al., Ber , 1900, 33, 3370; 1901, 34, 1170; 1907, 40, 3752,

CHAPTER V

MISCELLANEOUS CATHODIC
REACTIONS

(i) ADDITION OF HYDROGEN TO UNSATURATED COMPOUNDS

THE addition of hydrogen to unsaturated compounds has been accom-
plished electrolytically. However, the ease with which this occurs,
especially with double bond carbon compounds, is not as great as one
might anticipate. Generally, to obtain saturation of a double bond a
catalytic type electrode such as platinum,, covered with spongy platinum,
or nickel with a spongy nickel coating is required. At times it is
possible to achieve the desired results by use of an uncoated nickel
electrode and a mercury or lead electrode. Reactions of this type have
been performed in both acidic and basic media.

Acetylene has been reported to yield ethylene when subjected to
electrolysis at a platinized-platinum electrode in a basic medium.
Increasing the potential of the reaction results in the formation of
ethane (i). Under similar conditions, 2-methyl-3-butyn-2-ol gives a
95 per cent yield of dimethyl-vinyl-carbinol (2) :

CH-CC(OH) (CH 3 ) 2 2e -f aH > CH 2 -CH C(OH) (CH 3 ) 2

In the manner already discussed, it has been possible to hydrogenate
many simple unsaturated acids (3), thus :

CH 3 - CH =CHCOOH ae + aH > CH 3 CH 2 CH a COOH

Crotonic acid Butyric acid

CH 2 -CH CH 2 CH 2 COOH > CH 8 (CH 2 ) 3 COOH

Allyl acetic acid n- Valeric acid

CH 3 ~CH 2 CH -CH CH 2 CH =CH CH 2 CH =CH (CH a ) 7 COOH

Linoleic acid

> CH 3 (CH 2 ) 16 COOH

Stearic acid

8 4

MISCELLANEOUS CATHODIC REACTIONS

CH-COOH CH 2 COOH

CH-COOH CH 2 COOH

Maleic acid Succinic acid

Some of the more complex acids also respond to electrolytic methods.
For example, 3,a-dehydro-3-phthalideacetic acid, when reduced in an
aqueous sodium carbonate medium, will yield 3-phthalideacetic acid (4) :

CH CH 2 COOH
O

Using similar conditions enabled these investigators to prepare 2-cyano-
hydrocinnamic acid, thus:

CN CN

' J

2e + 2 H , <> ~~ CH 2 CH 2 COOH

The diene acid, j8, y-diphenylmuconic acid, has been reduced with
excellent yield to the monene, /?, y-diphenyl-dihydromuconic acid, via a
i,4-hydrogen addition (5):

C 8 H 5 C-CHCOOH C e H 6 CCH 2 COOH

I ae + 2H ||

C 6 H 5 C-CHCOOH > C 6 H 5 CCH 2 COOH

The reduction of cinnamic acid in an acid medium has been reported
to yield the dimeric acid, j8, y-diphenyl-adipic acid, thus (6) :

C 6 H 5 CHCH 2 COOH
2 C 6 H 6 CH - CHCOOH 2 e + aH |

"^ C 6 H 5 CHCH 2 COOH

Reduction, at least partly, of the aromatic character of the benzene
ring has been found possible. If benzoic acid is reduced in a basic
medium at a high overpotential electrode, the A 2 -tetrahydrobenzoic
acid is obtained (7) :

/ \COOH 4 e + 4 H /"NcOOH

\==s ~ ^ \ _ /

This same reaction performed in an acid medium results in the reduction

85

ORGANIC ELECTRODE PROCESSES

of the carboxyl group to a hydroxy methyl, without any attack on the
aromatic character of the benzene ring. Phthalic acid, however, has its
ring structure attacked in both acidic and basic media. In the former
medium, at a lead electrode, the less highly reduced structure 1,2-
dihydrophthalic acid is obtained (7), whereas in the basic medium at a
mercury cathode A 2 -tetrahydrophthalic acid is the main product (8) :

/COOK

/
// \

\_/ COOH
/COOH * ~~~~~ X ~" X

/ \_rOQH

If the monoethylester of phthalic acid is similarly treated at a mercury
cathode, and in an aqueous potassium chloride medium, the mono-
ethylester of 3,4-dihydrophthalic acid is obtained (9).

It has also been reported that phenol can be reduced at a platinum
electrode coated with platinum black in an aqueous sulphuric acid
medium, to yield cyclohexanol (10). Using a lead electrode and an
aqueous sulphuric acid medium, pyridine is readily reduced to
piperidine, and quinoline to either the dihydro- or tetrahydro-quinoline.
Dihydropyrrole and indoline are readily obtained by electrolytic
reduction of pyrrole or indole respectively (n).

Unsaturated compounds of the imino type (>C N ) are readily
reduced in an acidic medium at a high overpotential electrode such
as lead or mercury. Schiff bases, for example, upon electrolytic
hydrogenation, yield the respective secondary amine (12).

zH

R 1 CH-N R > R 1 CH 2 NHR

2e

Thus auramine gives a 90 per cent yield of the reduction product,
leucauramine (13):

[(CHs) 1 NC 6 H 4 ] l C=NH > [(CH 3 ) 2 NC 6 H 4 1 2 CHNH 2

k,

Oximes, such as acetoxime, add four hydrogens to form isopropyl-
amine (14):

(CH 3 ) 2 C N OH 4 H + 4e (CH 3 ) 2 CHNH 2 + H 2 O

86

MISCELLANEOUS CATHODIC REACTIONS

However, glyoxime, upon reduction, only adds two hydrogens to each
oxime group, with resultant formation of the j8-ethylene dihydroxyl-
amine (15), thus:

__
NOH NOH NHOH NHOH

Acetylacetone dioxime will yield the 3,5-dimethyl pyrazolidine (16):

CH 3 C CH 2 C CH 3 ---- > CH 3 CHCH 2 CH CH 3

II II II

NOH NOH HN - NH

Cathodic reduction of phenylhydrazone results in cleavage of the
>N N< bond with formation of the respective amine and aniline

(17):

R 2 C=N NHC 6 H 5 4 H + 4e __ v R 2 CHNH 2 + C 6 H 5 NH 2

The reduction of nitriles is generally accomplished at a high over-
potential electrode, such as lead or mercury in an aqueous-acid medium,
to yield the respective primary amine and some ammonia (18):

RCN + 4 e + 4 H -- > RCH 2 NH 2

In this manner nitriles such as propionitrile, benzonitrile, />-tolu-
nitrile, and succinonitrile, have yielded n-propylamine, benzylamine,
4-methyl-benzylamine, and putrescine, respectively.

The reduction of benzene diazonium salts at a mercury cathode in
an acid medium yields the corresponding phenylhydrazines. Thus
benzene diazonium chloride yields phenylhydrazine (19):

(CH 6 N - N)+C1- -- -> C 6 H 5 NHNH 2

(2) REDUCTION OF ALKALOIDS

In this complex group of compounds there are many focal points
of attack by electrolytic reduction. If oxygen is present as a carbonyl
group, its reduction is possible. Double bonds in the molecule can also
be reduced, as well as the reductive cleavage of rings containing nitrogen.
An illustration of the latter is the reduction of pegan at a lead electrode
in an aqueous sulphuric acid medium to yield N-[o-aminobenzyl]-
pyrrolidine (20) :

8?

ORGANIC ELECTRODE PROCESSES

/

p

/ CH *\

r*I4

...

CH

CH 2

Pegene-9-one-(i) which contains both a ring nitrogen and a carbonyl
group can be cathodically reduced to N-[o-aminobenzyl]pyrrolidine
under the same conditions as the pegan (21):

,CH Z

8e + 8H

CH/

CH 2N

N-

CH 2

-CH 2
CHj

H 2

NH 2

An example for the hydrogenation of the double bond can be seen
from the cathodic reduction in an acidic medium at a lead cathode of
berberine, which yields tetrahydroberberin (22) :

O CH 2 " O CH 2

CH 3

CH a

OH

Using similar conditions of electrolysis, strychnine (C 21 H 22 O 2 N 2 )
has been reduced to tetrahydrostrychnine (C 21 H 26 O 2 N 2 ), brucine
(C 23 H 26 O 4 N 2 ) to tetrahydrobrucine (C 23 H 3() O 4 N 2 ), and methoxy-
methyl-dihydrobrucidine (C 25 H 34 O 4 N 2 ) to methoxy-methyl-tetra-
hydrobrucidine (C 25 H 36 O 4 N 2 ) (23).

(3) REDUCTION OF SULPHUR, SULPHONYL CHLORIDE,
AND ARSENIC-CONTAINING COMPOUNDS

The literature does not indicate an extensive amount of research in
this category of compounds. The application of the electrochemical
technique does give relatively pure products in many of the reactions

88

MISCELLANEOUS CATHODIC REACTIONS

investigated. The reduction of the esters of a-bromo aliphatic acids in
the presence of sodium thiosulphate and an aqueous alcoholic medixim
of potassium carbonate at a low overpotential electrode, such as platinum,
leads to a dithio acid (24). The reduction of the methyl ester of a-
bromoacetic acid under such conditions yields the dimethyl ester of
dithiodiglycolic acid :

CH 2 BrCOOCH 3 + Na 2 S 2 O 3 - > H 3 COOCCH 2 S S CH 2 COOCH 3

Reduction of the free acid a, a'-dithiodiglycolic acid at a higher
overpotential electrode, namely lead, in a 2N sulphuric acid medium
gives a good yield of thioglycolic acid, thus (25) :

S CH 2 COOH

| 2e + 2H 2 HSCH 2 COOH

S CH 2 COOH *

Using a mercury cathode and a basic medium, it has been possible
to quantitatively convert homocystine to homocysteine (26) :

S CH 2 CH 2 CH(NH 2 )COOH

I 2e -f 2H 2 HSCH 2 CH 2 CH(NH 2 )COOH

S CH 2 CH 2 CH(NH 2 )COOH - "*

The reduction of the dimethochloride of bis-[4-dimethylamino-
phenylj-disulphide also responds satisfactorily to the electrolytic
method, to yield the corresponding thiol (27) :

2e

2H

Sulphonyl chlorides are reduced in a stepwise manner, first to the
respective sulphinic acid, and thence to the mercaptan. For example,
/>-toluenesulphonyl chloride at a lead electrode in an alcoholic-acid
medium is first reduced to />-toluene sulphinic acid and then to the
mercaptan, thus (28) :

~ H 8 C/ \SO 2 H

4 e + 4H_^ H 3 CSH -f 2H 2 O

ORGANIC ELECTRODE PROCESSES

Caution must be used in the isolation of these mercaptans, for they are
generally very easily " air oxidized " to the disulphide. It is possible, if
there is a nitro group present in the compound, to preferentially reduce
it to the amino group at a low potential electrode. Using a higher
potential electrode such as lead, will cause both the nitro group and the
sulphonyl chloride group to be reduced, according to the following
scheme (29) :

CH a

CH 3

CH 3

N0 a

6H+6e

-NH 2

S0 2 C1

-NH 2

SO 2 H

In an acidic medium at a high overpotential electrode such as lead,
zinc amalgam, or mercury, arsanic acids will be reduced to the respective
arsine, which, in the presence of air, is readily converted to the respective
arseno compound. An example is the^ reduction of 4-aminophenyl-
arsonic acid in a aN aqueous hydrochloric acid medium at a zinc
amalgam cathode, to yield 4-aminophenylarsine (30) :

>AsH 2

With a higher acid concentration (nN) the same compound will
yield />-arseno-anilin (31):

H 2 N/ SAs0 3 H 2 nN HC1 H 2 N/~\As

>NH 2

In a dilute acid medium this product can be reduced back to the 4-
aminophenylarsine. An interesting application of the electrolytic
process has been in the preparation of Salvarsan (32). The reduction
of 3-nitro-4-hydroxyphenylarsonic acid at a mercury cathode yields the
3,3'-diamino4,4'-dihydroxyarsenobenzene dihydrochloride (Salvarsan) :

NO 2

s0 3 H 2

90

MISCELLANEOUS CATHODIC REACTIONS

(4) DEHALOGENATION

The electrolytic removal of aliphatic halogen poses no great problem
and is readily accomplished in an acidic medium with a low over-
potential electrode such as copper, which is used to achieve partial
dehalogenation, and a high overpotential electrode such as lead or
mercury, to obtain complete or further dehalogenation (33). Carbon
tetrachloride, when reduced at a copper electrode, will yield chloroform,
which can be further reduced to dichloromethane with a lead electrode.

CC1 4 > CHC1 3 > CH 2 C1 2

In a similar manner, trichloroacetic acid can be reduced step wise to
chloroacetic acid :

C1 3 CCOOH > C1 2 CHCOOH > C1CH 2 COOH

Chloral hydrate responds similarly, to yield acetaldehyde :

CC1 3 CH(OH) 2 > CHC1 2 CHO > CH 2 C1 CHO > CH 3 CHO

In many instances, especially where the halogen is on adjacent
carbons, the unsaturated compound is obtained. Thus hexachloro-
ethane will yield tetrachloroethylene :

CC1 3 CC1 3 > CC1 2 -CC1 2

Dibromoethylene will yield ethylene at a lead electrode :
CH 2 Br C!~I 2 Br ~^~ CH 2 C.H 2

Apparently, if there is halogen present on a terminal position of an
aliphatic chain attached to a ring, the compound can, on reduction,
dimerize to form either the acetylene or ethylene type compound, thus
i,i,i-trichloro-2,2-diphenylethane can be reduced at a lead electrode,
in hot alcoholic-hydrochloric acid, to the tetraphenylbutyne :

2 (C.H 5 ),CHCC1, > (C 6 H 5 ) 2 CH C^C CH(C 6 H 5 ) 2

Using similar conditions it is possible to obtain a mixture of p y p'~
dichlorostilbene and i , i ,4,4-(tetrachlorophenyl)-2,3-dichlorobutene-2
from 1,1,1 -trichloro-2, 2-bis-(p-chlorophenyl) ethane :

(C1C 6 H 4 ) 2 CHC1 3 > C1C 6 H 4 CH=CH C 6 H 4 C1 +

rC!C 6 H 4 ) 2 CH CC1-CC1 CH(C 6 H 4 C1) 2

The removal of an aromatic halogen cannot generally be anticipated
when utilizing the electrochemical method, although the reduction of
bromobenzene to benzene has been reported (34) as well as the reduction
of N,N-dimethylaminoethyl tetrachlorophthalimide to the respective
trichloroisoindoline (35). In such a compound, containing a ring

9'

ORGANIC ELECTRODE PROCESSES

halogen as well as other reducible groups, one would generally not
expect a reduced compound lacking one or more of the originally
present halogen atoms.

(5) DESULPHONATION

There have been a number of patents issued for the desulphonation
of aromatic compounds using a high overpotential electrode in a neutral
or basic medium, which indicates the practicality of such an electro-
chemical process in the dyestuff field. The reactions are generally
performed in a basic medium at a mercury cathode (36). The following
are a few examples illustrating the reaction :

O

NH

S0 2
Saccharin

HOOC

NH 2
Sulpho-anthranilic acid

O 3 H

i -Aminonaphthalene-3 ,8-
disulphonic acid

HO 3 S OH

I S0 8 H

S0 3 H

i -Hydroxynaphthalene-
2,4,8-trisulphonic acid

CONH 2

Benzamide

HOOC

NH 2
Anthranilic acid

l S0 3 H

i -Aminonaphthalene-
3-sulphonic acid

OH

SO 3 H

SO 3 H

i -Hydrpxynaphthalene-
2,4-disulphonic acid

MISCELLANEOUS CATHODIC REACTIONS
NH 2 NH 2

X-N. / \\

V- S0 3 H

H0 3 S

S0 3 H

i -Aminonaphthalene-
2,4,6-trisulphonic acid

O S0 3 H

S0 3 H

i -Aminonaphthalene-
2,4-disulphonic acid

O

H0 3 S O

Anthraquinone-i ,5-disulphonic
acid

Anthraquinone

REFERENCES

(1) Coehn and Billitzer, Z. Elektrochem., 1901, 7, 683; Billitzer, ibid., 1901,

7 959-

(2) Favorsky and Lebedewa, Bull. Soc. chim., 1939, 6, 1347; See also Fokin,
Z. Elektrochem. , 1906, 12, 761 ; Golowtschanskaja, ^. Chim. gen., 1940,

I0 >435-

(3) D.R.P., 126446, 132223, 1899; 187788, 189332, 1906; Marie, Comp.
rend., 1903, 136, 1331; Z. Elektrochem., 1905, n, 644; Bull. Soc. chim.,
1908, 3, 202 ; Folsin, f&iYf., 1906, 12, 749 ; Meyer, et al., Monatsh, 1913,
34, 1667; Pomilio, Bull. Soc. chim., 1915, 21, 444; Lapworth and Royle,
y. c/zem. 5oc., 1920, 117, 746; Norris and Cummings, Ind. Eng. Chem.,
1925, J 7t 305; U.S.P., 1927289, 1929; I945i75,i93o; Siebert, Z. Elektro-
chem. , 1938, 44, 768; Woodruff, .7. Amer. chem. Soc., 1942, 64, 2857.

(4) Edwards, .?. chem. Soc. y 1926, 813.

(5) Fierz-David, Blangey and Uhlig, Helv. Chim. Acta, 1949, 32, 1414

(6) Wilson and Wilson, Trans. Amer. electrochem. Soc., 1943, 84, 153.

(7) Mettler, Ber., 1906, 39, 2934.

(8) Fichter and Simon, Helv . Chim. Acta, 1934, 17, 1219.

(9) Rodionow and Zvorykina, Bull. Soc. chim., 1938, 5, 840.

(10) Fichter and Stocker, Ber., 1914, 47, 2015; Bancroft and George, Trans.
Amer. electrochem. Soc., 1930, 57, 399.

(n) Ahrens, Z. Elektrochem., 1896, 2, 577; D.R.P., 90308, 1896; 104664,
1898; 310023, 1916; Tafel, Z. physikal Chem., 1900, 34, 187; Carrasco,
Gazetta, 1908, 38, ii, 301 ; Marie and Lejeune, J. Chim. physique, 1925,
22, 59; Drozdov, y. Gen. Chem. (U.S.S.R.), 1933, 3, 251; U.S.P.,
1947732, 1934 ; Kobosew and Monblanowa, Acta physicochem. (U.S.S.R.),
1935, 4 395-

93

ORGANIC ELECTRODE PROCESSES

(12) Brand, Ber., 1909, 42, 3460; Knudsen, ibid., 1909, 42, 3994; Brand and
H6ing, Z. Elektrochem., 1912, 18, 745 ; Law, .7* chem. Soc., 1912, 101, 154.

(13) Hoffman, Diss. Giessen, 1914.

(14) Tafel and Pffermann, Ber., 1902, 35, 1510; Ishibashi, Mem. Coll. Set.
Kyoto, 1923, 7 A, 39; Trans. Amer. electrochem. Soc. y 1924, 45, 335;
Anziegin and Gulewivich, Z. physiol. Chem., 1926, 158, 32 ; Kaplansky,
Ber., 1927, 60, 1842; Ruzicka, Goldberg, Hiirbin and Boekenoogen,
Helv. Chim. Acta, 1933, 16, 1329; Sugasawra, Satoda, and Yanagisawa,
J.pharm. Soc. Japan, 1938, 58, 29.

(15) Tafel and Pffermann, loc. cit., ref. 14.

(16) Tafel and Pffermann, Ber., 1903, 36, 219.

(17) Tafel and Pffermann, Ber., 1902, 35, 1512.

(18) Ahrens, Z. Elektrochem., 1896, 3, 99; Ogura, Mem. Coll. Sci. Kyoto,
1929, i2A, 339; Ohta, Bull. chem. Soc. Japan, 1942, 17, 485; J. chem.
Soc. Japan, 1942, 63, 1762.

(19) McClure and Lowry, Trans. Amer. electrochem. Soc., 56, 445 ; See also
Fichter and Willi, Helv. Chim. Acta, 1934, 17, 1416; Takayanagi, J. chem.
Soc. Japan, 1936, 57, 64.

(20) Spa*th, Kuffner and Lintner, Ber., 1936, 69, 2055.

(21) Spath and Platzer, Ber., 1936, 69, 392.

(22) Freund and Fleischer, Liebigs. Ann., 1915, 409, 188.

(23) Tafel, Liebigs. Ann., 1898, 301, 291 ; Tafel and Naumann, Ber., 1901,
34, 3291; Clemo, Perkin, Jr., and Robinson, .?. chem. Soc., 1927, 1589;
Gulland, Perkin, Jr., and Robinson, ibid', 1927, 1635 ; Leuchs and Wegener,
Ber., 1930, 63, 2219; Leuchs and Beyer, ibid., 1931, 64, 2157, 2162.

(24) Price and Twiss, J. chem. Soc., 1908, 93, 1650; 1909, 95, 1052.

(25) Larsson, Ber., 1928, 61, 1439; Sv. kem. Tidskr., 1928, 40, 149.

(26) Allen and Steinman, J. Amer. chem. Soc., 1952, 74, 3932.

(27) Schwarzenbach and Kuden, Helv. Chim. Acta, 1939, 22, 371.

(28) Fichter and Bernoulli, Z. Elektrochem., 1907, 13, 210; Fichter, Verh.
Naturf. Ges. Basel, 1908, 19, 37 ; Fichter and Tamm, Ber., 1910, 43, 3032.

(29) Fichter and Beck, Ber., 1911, 44, 3636.

(30) Fichter and Elkind, Ber., 1916, 49, 239.

(31) Matsumiya and Nakata, Mem. Coll. Sci. 'Kyoto, 1927, 10, 199; D.R.P.,
270568, 1911; U.S.P., I7I3475, 1927-

(32) Matsumiya and Nakata, loc. cit., ref. 31.

(33) Fokin, Z. Elektrochem., 1906, 12, 761; Askenasy and Vogelsohn, ibid.,
1909, 15, 7773; Brand, et al., ibid., 1910, i6 t 669; Ber., 1913, 46, 2938,
2945J 1921, 54, 2017; J. Prakt. Chem., 1930, (2), 127, 240; Ber., 1939,
72, 1029, 1668; Midgley, Jr., Hochwalt and Calingaert, J. Amer. chem.
Soc., 1923, 45, 1821 ; Maynard and Howard, ]r.,J. chem. Soc., 1923, 123,
960; Yoshitomi,^- Pharm. Soc. Japan, 1924, 3 ; Streight and Hallonquist,
Trans. Amer. electrochem. Soc., 1929, 56, 485; Hood and Imes, J. phys.
Chem., 1932, 36, 927 ; Sandonnini and Borghello, Atti Accad. naz. Lincei,
1934, 20 > 334; iQ35 2I > 30; Riccoboni, Atti. 1st. Veneto. Sci. Lett., 1935,
96!!, 183; Melnikow and Rokitzkaja, J. Chim. gin., 1937, 7, 2596; Hein,
Klein and Nebe, Ber., 1938, 71, 2381; Plump and Hammet, Trans.
Am^r. electrochem. Soc., 1938, 73, 523.

(34) Busch and Weber, Ber. y 1937, 70, 744.

(35) Allen and Ocampo, Trans. Amer. electrochem. Soc., 1956, 103, 682.

(36) D.R.P., 147228, 1902; 248527, 255724, 1911; B.P., 273043, 1926;
Matsui and Sakurada, Mem. Coll. Sci. Kyoto, 1932, 15 (A), 151, 181.

94

CHAPTER VI

ANODIC OXIDATION

(i) MECHANISM OF ANODIC OXIDATION

THE oxidation of an organic compound by electrolytic means is in
many respects quite similar to the action of a strong oxidizing agent.
The exact mechanism involved in the oxidative process is still, however,
not a certainty. One proposal was put forth by Glasstone and Hickling
(i) in which it was suggested that, in aqueous solution, hydroxyl ions
are discharged at low potentials to form hydroxyl radicals, which
combine to form hydrogen peroxide. This then may react with the
organic depolarizer with resultant oxidation, or it may decompose to
give oxygen and water in the following manner :

__
2 :OH 2e . 2 -OH

oxidation
H 2 O + i

It is also possible that the intermediate hydroxyl radical may be respon-
sible for the oxidation process (2), or that oxygen or the metallic oxides
formed at the electrode surface are the responsible agents (3). Where
hydroxylation occurs, as will be seen in the oxidation scheme for benzene
(Chap. VII), it may very well be that the reaction is due to the presence
of hydrogen peroxide or of the hydroxyl radical. However, in instances
such as the oxidation of iodate to periodate, which is more readily
accomplished at a lead dioxide electrode than at a platinum electrode,
it is quite likely that this process is not due to the presence of hydrogen
peroxide, but to active oxygen. This assumption is based on the fact
that lead dioxide is an excellent agent for the decomposition of hydrogen
peroxide. Unfortunately, no one idea explains every reaction, therefore
each requires individual study to evaluate the oxidation mechanism.
There is no doubt that far more work must be done before the
knowledge of anodic processes is on a par with that of the cathode.

95

ORGANIC ELECTRODE PROCESSES

(2) ANODIC OXIDATION OF FATTY ACIDS AND THEIR SALTS

This is one of the aspects of electrolytic oxidation that has been
investigated quite extensively. As early as 1834, Faraday observed that
upon electrolysis of an acetate solution he obtained some hydrocarbon
(4). However, with the initiation of investigation by Kolbe, detailed
results became available. He showed that upon electrolysis of an
alkali-metal acetate, a hydrocarbon subsequently known as ethane, and
two volumes of carbon dioxide were obtained (5) :

2 CH 3 COO~ > 2 CO 2 -f C 2 H 6 + 2e

This process, known as the Kolbe reaction, was considered by its
discoverer to involve the oxidation of the carbonyl group by oxygen
obtained from the electrolysis of water :

2 CH 3 COOH + O > C 2 H 6 + 2 CO 2 -f H 2 O

A further investigation of this reaction by others showed it to be
generally applicable to salts of aliphatic monobasic acids (6). The
dicarboxylic and aromatic acids were found not to undergo this type
reaction.

One of the interesting facts about the reaction is that in an aqueous
medium the formation of the hydrocarbon occurs to an appreciable
extent only at a smooth platinum or iridium electrode. If a platinized
platinum, palladium, gold, nickel, or iron anode is used, the efficiency
for hydrocarbon formation is considerably reduced. If a platinized
electrode is poisoned with mercury, the efficiency improves tremen-
dously. The presence of catalysts for hydrogen peroxide decomposition,
either those which give deposits on the anode, for example lead or
manganese salts, or those which do not, for example iron and copper
salts, brings about a decrease in the formation of ethane when an
aqueous solution of an acetate is submitted to electrolysis. The product
in these cases consists essentially of methanol. When using a platinum
electrode, the most favourable results are obtained by using tempera-
tures below 50 C. and anode current densities over 0-25 amp./cm. 2 .
A high concentration of salt also favours this reaction (7). If low
current densities are utilized, paraffins and olefins which possess only
half the carbon atoms ordinarily associated with products of the Kolbe
reaction are obtained (8).

When the conditions are such that the acetate is oxidized to ethane,
the anode potential is about + 2'2 volts v. S.C.E. However, when the

96

ANODIC OXIDATION

process is suppressed in favour of methyl alcohol formation, the
potential is about + i'8 volts v. S.C.E. The reaction in which alcohols
are formed preferentially is referred to as the Hofer-Moest reaction,
and is generally favoured by the use of an alkaline solution and by
addition to the medium of various inorganic salts, such as perchlorates,
sulphates, and bicarbonates (9).

The electrolysis of solutions of the fatty-acid salt in the correspond-
ing anhydrous fatty-acid gives results which are almost identical with
those obtained in an aqueous solution (10). Methanol has also been
found to be a most suitable solvent for the Kolbe reaction (11). With
this medium, apparently, conditions such as electrode material, current
density, temperature, and concentration of depolarizer, are not as critical
as in an aqueous system. In addition, in this particular medium the
yield of Kolbe product is somewhat better. The conductivity of the
methanol medium can be improved by the addition of a small amount
of water without detracting from its effectiveness in the Kolbe
reaction.

(a) Mechanism of the Kolbe Reaction

There have been several theories proposed to account for the products
obtained as a result of the Kolbe electrosynthesis. The first theory, the
peroxide theory, was proposed by Schall and developed by Fichter (12).
These investigators were of the opinion that the acetate ions were
discharged directly at the anode, and the resulting radicals reacted to
form the diacetyl peroxide, which then decomposed to form the alkyl
radicals and thence the hydrocarbon :

CH 3 COO: CH 3 COO CH 3 COO CH 3 \

> > | >z C0 2 + >->CH 3 CH 3

CH 3 COO: CH 3 COO* CH 3 COO CH 3 '/

As alcohols are obtained as side products of the Kolbe reaction, the
explanation for their formation was suggested by decomposition of the
peroxide to give a per-acid. This per-acid hydrolyzes, especially in an
alkaline solution, to yield the alcohol :

CH 3 COO

| + HaU CH 3 COOH + CH 3 CO 2 OH

CH 3 COO >

CH 3 C0 2 OH > CH 3 OH + CO 2

When the Kolbe reaction is carried out in an acidic medium, the

ORGANIC ELECTRODE PROCESSES

formation of olefins and esters is explained by dehydration of the alcohol,
and reaction of the alcohol with the free acid respectively. This is a
rather unlikely explanation, as the conditions used in the Kolbe electro-
synthesis are generally not conducive chemically to such reactions.

The advocates of the peroxide theory have claimed that, by using
low temperatures during their reaction, they have been able to detect
small quantities of diacetyl peroxides and per-acids during the electro-
lytic process (13). This does not necessarily mean that these are bona-
fide intermediates in the process, since we may also look upon diacetyl
peroxide and the per-acid as by-products obtained under abnormal
experimental conditions. The fact, too, that many diacyl peroxides
decompose on heating to give products similar to that obtained in the
Kolbe electrosynthesis is also not sufficient evidence for the claim
of its being intermediate in the process. The only factor in common
between the electrolysis and the decomposition of the diacetyl peroxide
by heating is that both presumably yield an alkyl free radical. The
effects of anode material, of catalysts for hydrogen peroxide decom-
position, and of added salts cannot, however, be readily explained by
this theory.

A viewpoint which consolidates some of the essential features of
the previously mentioned theory, and at the same time accounts for
many of the facts which these theories fail to explain, has been outlined
by Glasstone and Hickling, and is referred to as the " Hydrogen
Peroxide Theory " (14). As a result of detailed studies of electrolysis
of acetates in aqueous solution, these investigators suggested that
hydrogen peroxide, which is first formed at the anode from hydroxyl
radicals, or these hydroxyl radicals themselves prior to combination
in pairs, react with the acetate ions to form acetate radicals. These
then combine, possibly with the intermediate formation of diacetyl
peroxide, to yield ethane and carbon dioxide, thus:

CH 3 COO : + -OH > CH 3 COO + :OH

or
2 CH 3 COO: + H 2 2 > CH 3 COO + a:OH

2 CH 3 COO > (CH 3 COO) 2 > CH 3 CH 3 + 2 CO 2

It is possible that some acetate radicals are formed by the direct dis-
charge of the ions, as is the case in nonaqueous solutions; but an
additional mechanism must be introduced, such as the one just proposed,
to account for the influence of electrode material and of catalysts for
hydrogen peroxide decomposition. It is significant that the anodes,

ANODIC OXIDATION

at which there is no Kolbe reaction, consist of substances which are
either themselves catalysts or which become oxidized to substances
which are catalysts for hydrogen peroxide decomposition. Under these
conditions, as well as when access of acetate ions to the anode is prevented
by the presence of foreign anions, the reactions mentioned presumably
do not occur, but instead peracetic acid is probably formed, with
subsequent decomposition to methanol and carbon dioxide:

CH 3 COO: + z:OH > CH 3 CO 2 OH + :OH

or

CH 3 COO: H 2 2 CH 3 CO 2 OH + :OH

or

CH 3 COOH + O CH 3 CO 2 OH

thence

CH 3 CO 2 OH CH 3 OH + CO 2

One of the difficulties associated with any theory pertaining to the
Kolbe reaction is that of accounting for the fact that, although the
potential at which it occurs is so high, oxygen evolution does not take
place preferentially. In view of the possibility that normal discharge
of an hydroxyl ion leading to the anodic evolution of oxygen results
from the transfer of a proton from a molecule of water adsorbed on
the electrode surface, it is possible that the water molecules have been
largely replaced by acetate ions. The discharge of hydroxyl ions to
form oxygen is rendered difficult, and thus it does not occur until the
potential is high enough for the discharge of acetate ions, and their
detachment from the anode takes place. If the acetate ions are prevented
from reaching the electrode by introducing neutral salts, their adsorption
is relatively small, and the discharge of hydroxyl ions at the normal
potential occurs. The presence of catalysts for hydrogen peroxide
decomposition presumably favours the discharge of hydroxyl ions,
since they facilitate the formation of oxygen. The low potentials
observed under these latter conditions can thus be explained.

In nonaqueous solutions the Kolbe synthesis takes place with high
efficiency at platinized platinum and gold, as well as at smooth platinum
anodes. Increased temperature and the presence of catalysts for hydrogen
peroxide decomposition, both of which have harmful effects in aqueous
solution, have relatively little influence. The mechanism of the reaction
is apparently quite different in aqueous and nonaqueous solution ; in the
latter no hydroxyl ions are present, and so neither hydroxyl radicals nor
hydrogen peroxide can be formed. It is generally agreed that in the
nonaqueous medium the initial anodic process is the discharge of the

99

ORGANIC ELECTRODE PROCESSES

acetate ions. If the primary anodic reaction is different in an aqueous
and in a nonaqueous medium, it is difficult to explain the fact that the
Kolbe reaction with acetates in an anhydrous ethylene glycol medium
requires an anodic potential very similar to that for an aqueous solution
(15). The resulting radicals probably combine in pairs, as in an aqueous
solution, to form acetyl peroxide, which decomposes in a manner
already described.

Attempts to duplicate the Kolbe reaction by addition of hydrogen
peroxide to solutions of acetates have led to negligible yields of hydro-
carbon being obtained (f6). This, of course, offers most serious
objection to the hydrogen peroxide theory. However, it must be borne
in mind that conditions of a chemical experiment are entirely different
from those obtained electrochemically. This type of oxidation of
acetates has been effected by use of either sodium persulphate or
fluorine, with the result that products very similar to those obtained
electrochemically were achieved (17). Addition of carbonates to the
fluorine reaction medium yields Hofer-Moest type products (18). As
in the anodic process, when a catalyst for hydrogen peroxide decom-
position is added to the acetate-persulphate mixture, the formation of
hydrocarbon is inhibited (19).

Clusius and his associates have contributed a great deal to a better
understanding of the mechanism of some of the side reactions associated
with the Kolbe reaction. They performed electrolysis of fatty acids in
light and heavy water and found, for example, that acetates in an aqueous
medium gave methane at low current density, and ethane at high current
density. From their experimental results they concluded that the
products obtained depended only on the acetate, which was isotopically
labelled, and not on the solvent medium (20). Thus the results obtained
verify the concept that methyl radicals are formed initially. When
these radicals are present in high concentrations, they dimerize to
ethane, or, when present in low concentrations, they abstract hydrogen
from the acetate ion or acetic acid, to give methane, thus :

CH 3 - + CH 3 COOH > CH 4 + -CH 2 COOH

The acetic acid radical is subsequently oxidized to carbon dioxide and
watej*.

The electrolysis of propionates and butyrates yields an olefin as the
main product, and a small quantity of a paraffin composed of the same
number of carbon atoms as the olefin (21). Thus the olefin, propene,

100

ANODIC OXIDATION

is obtained from the free radical intermediate by elimination of a hydro-
gen from the /?-carbon atom of the acid:

CH 3 CH 2 CH 2 - > CH 3 CH - CH 2

The paraffin obtained in the same reaction was probably produced as
a result of some disproportionation of the alkyl free radical. The
formation of alkyl free radicals does not preclude the validity of either
the peroxide or the hydrogen peroxide theories. However, Clusius is
more favourably inclined to the view originally put forth by Brown and
Walker (22), that the carboxylate ion is oxidized to the carboxylate
radical, which then decarboxylates to yield the alkyl radical:

RCOOH > RCOO : > RCOO > R- + CO 2

This free radical mechanism also offers reason for the formation of 4-
phenylbenzoic acid and 4-phenylpyridine from the anodic electrolysis
of benzoic acid in a pyridine medium (23).

It has been shown (24) that aryl radicals, if produced, tend to attack
the solvent molecules with resultant polymerization. This may
explain the essential absence of the usual Kolbe dimer when aryl
carboxylic acids are subjected to electrolysis. It is quite possible that
the mechanism illustrated above holds in both aqueous and nonaqueous
media. Therefore it may be that, in either case, we are dealing with a
situation analogous to the reductions which give radicals at a high over-
potential electrode (Chap. III). At this high overpotential electrode
we have a relatively weak anode carboxylate bond, with the result that
the molecule remains for a period sufficient only for a one-electron
change, with resultant formation of the free radical :

R:C:0:

i

Anode

R- + CD 2 < - R :^:0 :

The oxygen atom in this radical tends to fill its octet, thus the R : C
bond is weakened, resulting in cleavage to give the alkyl radical and

101

ORGANIC ELECTRODE PROCESSES

carbon dioxide. The alkyl radical then either dimerizes or forms the
olefin, depending on its concentration in solution.

(>) Monocarboxylic Acids

Of the straight chain fatty acids, acetic acid is the only low carbon
acid which gives a high yield, of approximately 90 per cent of Kolbe
product (25). Comparably good yields are obtained with acids contain-
ing six or more carbon atoms (26). The intermediate aliphatic acids
give larger quantities of esters, olefins, and other by-products, which
can be suppressed in favour of the Kolbe product by use of high
concentrations of alkaline salts of the carboxylic acid. Unfortunately,
due to the occurrence of side reactions, and the evolution of oxygen, the
current efficiency is occasionally quite poor (27).

When propionates are subjected to anodic oxidation, the products
obtained consist of ethylene, ethyl propionate, carbon dioxide, oxygen,
and butane (28). The latter is anticipated from the Kolbe reaction.

2 C 2 H 5 COO: > H 5 C 2 C 2 H 5 f 2 CO 2 + 20

At low current densities the amount of butane formed is quite small.
As the current is increased the amount of oxygen evolved at the lower
current density diminishes, with formation of an increasing amount of
ethylene. Further increase in current density results in the formation
of the Kolbe product, butane. Electrolysis of potassium propionate
at a gold or smooth platinum anode, in a medium of propionic acid and
methanol, will give approximately a 70 per cent yield of butane. How-
ever, the former electrode does not give as satisfactory results when
used in an aqueous medium (29). When propionates are electrolysed
in the presence of inorganic ions, ethyl alcohol is obtained with low
efficiency (30), as contrasted with the high yield of methyl alcohol from
the electrolysis of acetates.

The products obtained from the electrolysis of salts of normal
butyric acid consist of propylene, propyl butyrate, isopropyl butyrate,
isopropyl alcohol, hexane, and carbon dioxide (31). The last two
compounds are formed by the normal Kolbe process.

2 C 3 H 7 COO: > H 7 C 3 C 3 H 7 + 2 CO 2 -f 2e

The J&opropyl alcohol and its ester obtained from the reaction might
very well be attributed to the addition of hydrogen and hydroxyl during
the electrolysis.

102

ANODIC OXIDATION
CH 3 CH: CH 2 + H + + -OH > CH 3 CH CH 3

OH

Under ordinary circumstances the mere bubbling of propylene through
water does not accomplish this result. The isobutyrate gives only a
trace of the Kolbe dimer, with the major organic products consisting
of isopropyl alcohol, propylene, and hexane (32).

Normal valerates upon electrolysis yield the Kolbe products, octane
and carbon dioxide,

2 C 4 H 9 COO:

H 9 C 4 C 4 H 9 + 2 C0 2 + 2e

in addition to quantities of alcohols, butyl valerates, and various other
oxidation products (33). Isovalerates also give the anticipated di-
isobutyl in addition to other compounds, such as trimethyl carbinol
butylenes, etc. (34).

As mentioned previously, the higher fatty acids give quite satisfac-
tory yields of the Kolbe dimer, with paraffin hydrocarbons up to C 34 H 70
having been prepared by this method (35). Some of the results com-
piled by Moser (36) are given in Table I.

TABLE I
Kolbe Electrolysis of Aliphatic Monobasic Carboxylates

Acid

Structure

Percentage Yield

Paraffin

Olefin

Ester

Acetic .

CH 3 COOH

85

2,

2

Propionic

C 2 H 5 COOH

8

66

5

n-Butyric

C 3 H 7 COOH

14*5

53

10

iso-Butyric

(CH 3 ) 2 CHCOOH

trace

62

10

n-Valeric

C 4 H 9 COOH

50

18

4

tso-Valeric

(CH 3 ) 2 CHCH 2 COOH

43

42

5

Methyl ethyl J
acetic 1

(CH 3 ) v
>CHCOOH
(C 2 H 6 )/

10

42

10

Trimethyl

acetic

(CH 3 ) 3 C COOH

13

52

o

Caproic

C 6 H U COOH

75

7

i'5

Laurie .

C n H 23 COOH

45

Myristic

C 13 H 27 COOH

33'7

Palmitic .

C 15 H 31 COOH

30

Stearic .

C 17 H 36 COOH

27-6

.

103

ORGANIC ELECTRODE PROCESSES

There is no really valid reason as yet as to why there should be any
difference in yield of the Kolbe dimer. One opinion offered has been
that the differences observed might be due to the ease with which the
acid peroxide is hydrolysed. The more readily the hydrolysis occurs,
the easier will be the formation of the per-acid, with resultant production
of alcohols, esters, and olefins (37).

(c) Mixtures of Carboxylic Acids

When two carboxylic acids are subjected to anodic electrolysis,
three different Kolbe dinners may be anticipated, thus :

RCOO : + R'COO : -> R R + R' R' -f R R' 1 2 CO 2

The amounts of each dimer formed can more or less be controlled by
the relative amounts of each acid present initially in the electrolysis
medium. If the concentration of one acid is increased, the conversion
of the other acid to the unsymmetrical dimer is improved considerably
(38). A considerable difference in the size of the acid molecules can
be tolerated; however, if each acid itself is not capable of forming a
symmetrical dimer, it will not generally, form its unsymmetrical counter-
part, although there are a few exceptions to this reported in the
literature (39).

A considerable amount of experimental work has been performed
by Matsui and Arakawa (40) in studying the formation of mixed hydro-
carbons. These workers performed their experiments in an aqueous
alcoholic medium at high current densities and at temperatures of
approximately 70 C. It was found, for example, that when a mixture
of the potassium salts of propionic acid and palmitic acid, the former
being present in a 50 per cent molar excess, was electrolysed, the major
component isolated was w-heptadecane, in an amount of 40 per cent
more than the dimer, triacontane, from palmitic acid. The butane
formed went off with the gaseous products.

C 2 H 5 COOK 4- C 15 H 31 COOK > C 17 H 36 -}- C 30 H 62 + C 4 H 10 + 2 CO 2

Mixtures of keto-acids (41), as well as the half-esters (42), also yield
the crossed dimer of the keto-acid and half-ester respectively. The
coupling of two different types of compounds has been achieved by
anodic oxidation, using mixtures of keto acids (43), cyano acids (44),
half-esters (45), and phenyl substituted acids (46) with a fatty acid.

A general method has been developed for the synthesis of unsym-
metrical fatty acids which involves coupling a half-ester with a mono-
carboxylic acid (47).

104

ANODIC OXIDATION

MeO 2 C(CH 2 ) x CO 2 H + RCO 2 H > R(CH 2 ) X CO 2 H + MeO 2 C(CH 2 ) 2X CO 2 Me

+ R R

From the mixture consisting of the hydrocarbon and the mono and
di-ester, the unsymmetrical acid R(CH 2 )xCO 2 H can generally be
separated by distillation. These same workers have prepared stearic
acid by coupling such acids as decanoic with sebacic half-ester and
myristic with adipic half-ester. This method has also been applied to
the synthesis of trideuteropropionic acid by anodic coupling of the
mono ethyl ester of malonic acid and trideuteroacetic acid (48).

The electrochemical method has been found extremely satisfactory
for the synthesis of branched-chain fatty acids (49) as found in nature,
which have demonstrated marked biological activity (50). One such
acid, () 3,13,19-trimethyltricosanoic acid, has been reported to
cause granulomatous lesions in the guinea-pig (51). This acid has been
synthesized from 3-methylheptanoic acid (52), thus:

CH 3 (CH 2 ) 3 CH(CH 3 )CH 2 C0 2 H + HO 2 C(CH 2 ) 3 CO 2 CH 3
CH 3 (CH 2 ) 3 CH(CH 3 )(CH 2 ) 4 C0 2 H + HO 2 CCH 2 CH(CH 3 )CH 2 CO 2 CH 3
CH 3 (CH 2 ) 3 CH(CH 3 )(CH 2 ) 5 CH(CH 3 )CH 2 COOH + HO 2 C(CH 2 ) 7 CO 2 CH 3

CH 3 (CH 2 ) 3 CH(CH 3 )(CH 2 ) 5 CH(CH 3 )(CH 2 ) 8 C0 2 H + HO 2 CCH 2 CH(CH 3 )

I CH 2 C0 2 CH 3

CH 3 (CH 2 ) 3 CH(CII 2 ) 5 CH(CH 2 ) 9 CHCH 2 CO 2 H
CII 3 CH 3 CH 3

(d) Branched-Chain Acids and Half-Esters

One of the most important restrictions on the type of acid which
can be used is that this acid should not be substituted in the a-position.
Although small yields of Kolbe product have reportedly been obtained
with such a-alkyl and cycloalkyl acids (53), far better yields are obtained
with acids whose alkyl substituent is well removed from the carboxyl
group (54). The following are a few of the substituted acids which
give a 50 per cent or better yield of the desired Kolbe product:

EtOOC-CMe 2 -CH 2 -COOH

MeOOC-CH 2 -CHMe-CH 2 -COOH

MeOOC-CH 2 -CMe 2 -CH 2 -COOH

MeOOC-CH 2 -CHMe-(CH 2 ) 2 -COOH

EtOOC(CH 2 ) 2 -CHMe-(CH 2 ) 3 -CHMe-(CH 2 ) 2 -COOH

EtOCH 2 -CHMe-(CH 2 ) 3 'CHMe-(CH 2 ) 2 -COOH

ORGANIC ELECTRODE PROCESSES

The salts of oc-substituted acids, when electrolysed in the presence
of potassium carbonate, undergo a reaction similar to the Hofer-Moest
reaction, in that the carboxyl group is replaced by a hydroxyl group to
give the alcohol. Potassium cyclohexanecarboxylate upon electrolysis
yields cyclohexanol (55) :

H 2 H 2 **2 **2

c c c c

H 2 C HCCOO : + : OH -- > H 2 C CH(OH) + CO 2

C

H 2 H 2 ' H 2 H 2

The salt of cyclopropanecarboxylate under similar conditions yields
allyl alcohol, with cyclopropanol, its probable intermediate, undergoing
an immediate isomeric change to give the isolated alcohol (56).

CH 2 , CH 2X

| >CHCOO : + : OH - > CO 2 + | >CHOH
CH/ CH/

I
CH 2 =CHCH 2 OH

The other main product isolated from the reaction mixture was the
cyclopropane carboxylic acid ester.

(e) Unsaturated Acids

An additional limitation of the Kolbe reaction is that anodic coupling
cannot be consummated when either an a, /? or a /?, y-unsaturated acid
is used. If sodium aery late were to undergo the normal Kolbe synthesis,
the expected product would be butadiene. Instead, however, acetylene
is obtained, possibly via the following reaction,

2 CH 2 : CHCOO: ---------- > CH=CH -f- CH a = CHCOOH + CO 2

together with quantities of carbon dioxide, carbon monoxide, oxygen,
and ethylene (57). The salts of the higher unsaturates undergo a true
Kolbe synthesis, to yield the dimer. Thus potassium oleate gives the
olefin C 34 H 66 , and potassium undecenoate gives C 2a // 38 in addition to
small quantities of acetylene hydrocarbons (58).

Maleate and fumarate salts, as well as the higher acids, undergo a
reaction similar to the Kolbe, to yield acetylene as one of the major
products of the reaction (59) :

CHCOO:

|| ---- > CHasCH + 2 CO 2

CHCOO :

106

ANODIC OXIDATION

Thus generally the Kolbe synthesis is not applicable and is of no
preparative value with the acids which are unsaturated in the a, /J or j3, y
positions. Unsaturation in other portions of the molecule results in no
interference, and the anticipated products are generally obtained.

(/) Aromatic Acids

Salts of aromatic acids also do not undergo the Kolbe reaction.
When benzoates are subjected to electrolysis traces of acetylene and
carbon dioxide are obtained, together with appreciable quantities of
oxygen and benzoic acid (60). In a methanolic solution benzole acid
yields benzene (61), and its diethylamine salt in a pyridine medium
gives 4-phenyl-pyridine, 4-phenyl-benzoic acid, and traces of
diphenyl (62).

When 4-tert.-butyl-2,6-dimethylbenzoic acid is subjected to anodic
oxidation in a methanolic medium an aromatic hydrocarbon and a
substituted phthalic ester are obtained (63) :

COOCH 3

4

- ^ { / - x _>-COOCH a +

CH 3

Upon electrolysis in methanol, or a mixture of pyridine and methanol,
both j8-phenylpropionic acid and phenylacetic acid will give the usual
Kolbe dimer (64).

(g) Substituted Acids

Substituents in the oc-position result in a suppression of the Kolbe
reaction, as would be anticipated from the previous discussion of a-alkyl
substituted acids. Negligible dimer formation has been obtained in
an aqueous medium on anodic oxidation of hydroxy, amino, halogeno,
acetylamino, alkoxy, phenyl, and cyano acids containing the named
substituents in the a-position (65). From the reaction mixtures is
generally obtained an unstable a-substituted acid which decomposes to
give an aldehyde or ketone and an ester.

107

ORGANIC ELECTRODE PROCESSES

The salts of monochloroacetic acid, for example, yield chlorine,
hydrochloric acid, carbon dioxide, carbon monoxide, oxygen, and
relatively smaller amounts of chloromethylchloroacetate which can be
construed as being obtained via a modified Hofer-Moest reaction. In
addition, small quantities of methylene chloride and traces of formalde-
hyde are also obtained (66).

The a-amino acids, upon electrolysis, yield an aldehyde or a ketone
and ammonia. This is also the case when the amino group is acetylated
or sulphonated. It has been shown (67) that aminoacetic acid yields
methylamine, ammonia, formaldehyde, formic acid, carbon monoxide,
and carbon dioxide. The initial reaction is :

NH 2 CH 2 COOH + O -> NH 3 + HCHO + C0 2

The ammonia and the formaldehyde react to form a mixture of methyl-
amines. In a similar manner a-hydroxy and a-alkoxy-acids yield, in
addition to other products, aldehydes containing one carbon atom less
than the parent compound. For example, glycolic acid yields oxalic
acid and formaldehyde (68). If the reaction is performed in an alkaline
medium at an iron, copper, or nickel anode, oxalic, glyoxylic, and
formic acids are obtained. At a platinum anode, however, the major
product is formaldehyde (69). Lactic acid or its salt upon electrolysis
yields acetic acid, acetaldehyde, and formic acid (70). If pyruvic acid
is the desired product, the copper salt of lactic acid is used, thus giving
copper pyruvate which precipitates as it forms, preventing further
oxidation (71). A sugar with one carbon less is obtained when a salt
of a sugar acid is subjected to electrolysis. For example, D-glucuronic
acid yields D-arabinose (72).

COO: CH(OH) 2 CHO

| > | _> |

(CHOH) 4 - C0 2 (CHOH) 3 - H 2 O (CHOH) 3

CH 2 OH CH 2 OH CH 2 OH

The electrolysis of the substituted fatty acids mentioned above, under
the experimental conditions used in this earlier work, cannot be con-
sidered ideal in light of more recent investigations, in which anhydrous
metl^nol or methanol-pyridine media were used. With this media,
oc-aryloxy and a-phenyl acids will undergo anodic coupling to yield the
Kolbe product (73). On occasion, it is even possible to obtain anodic
coupling with ot-acetylamino-acids using the same media (74).

1 08

ANODIC OXIDATION

Substituents, which generally do not result in Kolbe products when
in the a-position, exert only a minor effect when removed to another
position in the acid molecule. As an example, anodic couplings have
been obtained with alkoxy (75), hydroxy (76), keto (77), acylamino (78),
amino, cyano (79) acids, and also with aldehyde acid acetals (80).

(/O Dicarboxylic Acids

Dicarboxylic acids do not undergo reactions of the Kolbe type, but
considerable oxidation occurs, with the result that numerous degradation
products are formed. These products include both saturated and
unsaturated acids through loss of one carbon atom, and alcohols and
olefins containing two carbons less than the original starting material (8 1) .
Anodic oxidation of malonic acid at a platinum electrode in an alkaline
medium yields oxalic and formic acids :

COOII

CH 2 > COOH - HCOOH

I I

COOH COOH

The alkali salts of the monoester dicarboxylic acid do give a Kolbe type
reaction, with the products being higher dicarboxylic acid esters, thus:

COO : COOR

I I

2 (CH 2 )n - (CH 2 )n + 2 CO 2

I I

COOR (CH 2 )n

COOR

This reaction is generally referred to as the Brown-Walker synthesis
(82). For example, the potassium salt of ethyl malonate in aqueous
solution yields, upon anodic electrolysis, diethyl succinate :

COOC 2 H 5 COOC 2 H 5

2 CH 2 > CH 2 +2C0 8

COO : CH 2

COOC 2 H 5

The higher dicarboxylic acid mono esters also behave in a like
manner, thus making this synthesis an excellent tool for the preparation

109

ORGANIC ELECTRODE PROCESSES

of long chain carboxylic acid diesters (83). The ideal conditions for
this reaction are a high current density, high concentrations of the
depolarizer, and a low temperature, which are in essence the same as
those which favour the formation of ethane in the electrolysis of acetates.
As the process is performed in an aqueous medium, the diester formed
usually precipitates out, thus making it invulnerable to further oxidation.
Since the anolyte must be prevented from becoming alkaline, it is
desirable to use a membrane between the anode and cathode com-
partments. An alternative is to use an excess of carboxylic acid.
This reaction has much in common with the Kolbe reaction, in that it
does not occur either at a gold or a platinized platinum anode (84).
The mechanism of this reaction is probably similar to that of the Kolbe
reaction.

As a result of electrolysis of the ester salts, by-products such as
olefins and alcohols are formed. For example, ethyl succinate gives, in
addition to its major product, diethyl adipate (I), some hydroxyacrylate
(II), and ethyl acrylate (III) (85).

COO:

I
2 (CH 2 ) 2

COOC 2 H 5

COO:

|

(CH 2 ) 2 + OH

COOC 2 H 5

COO:

2 (CH 2 ) 2 ~
COOC 2 H 5

COOC 2 H 6
(CH,) t

COOC 2 H 5
(I)

CH 2
C

2 CO 2

:ooc 2 H 5

(ID

CH 2

CH

I
COOC 2 H 5

an)

4-

COOH

(CH 2 ) 2
COOC 2 H 5

C0 2

A mixture of two different esters gives an unsymmetrical diester, in
addition to the products obtained from each ester separately. Thus,
the salts of ethyl succinate and ethyl malonate yields diethylglutarate
as the unsymmetrical product in addition to quantities of diethyl
adipate and diethyl succinate (86):

no

ANODIC OXIDATION
COO : COO : COOC 2 H 5

! I ' I

(CH 2 ) 2 4- CH 2 > (CH 2 ) 3 + 2 CO 2

I I I

COOC 2 H COOC 2 H 5 COOC 2 H 5

Sodium malonic ester and similar compounds couple in a manner
similar to the Kolbe reaction, with the exception that no carbon dioxide
is eliminated (87) :

COOC 2 H 5 COOC 2 H 5 COOC 2 H 5

2 CH 2 > HC CH

I I I

COOC 2 H 5 COOC 2 H 5 COOC 2 H 5

(t) Acetoxylation, Alkoxylation, and Alkylation

Of interest in connection with the reactivity of acids at the anode
is the acetoxylation which has been reported to occur when the mono-,
di-, or triphenyl acetic acids are subjected to electrolysis in a sodium
acetate-acetic acid medium. Under these conditions the respective
phenyl substituted methyl acetate is obtained in satisfactory yield (88).
Comparable acetoxylation also takes place when a-acylamino acids are
subjected to electrolysis under similar conditions (89). Anisole gives
a mixture of 0- and p-acetoxyanisoles in a 2 : i ratio when subjected
to electrolysis in a potassium acetate-acetic acid medium at a graphite
or platinum anode (90).

Acids can also be alkoxylated in a manner similar to acetoxylation.
Thus in a methanolic medium, di- (91) and tri- (92) phenyl acetic acid
yield upon electrolysis the respective methyl ethers :

Ph 2 CHCOOH > Ph 2 CHCOCH :}

Monophenylacetic acid does not behave in a like manner, but instead
gives dibenzyl (93). Anodic electrolysis of N-acylglycines and N-acyl-
a-alanines in an alcoholic medium resulted in excellent yields of the
respective amides (94).

It might be of interest to mention that a small quantity of trinitro-
m-xylene has been isolated from an anodic electrolysis medium consist-
ing of trinitrotoluene in sodium acetate-acetic acid (95) :

in

ORGANIC ELECTRODE PROCESSES
NO 2 NO

2 N ^ 5N0 2 2 N

CH 3

The above reaction is possibly the result of a direct attack on the benzene
ring by the methyl free radicals formed during the Kolbe reaction.

REFERENCES

(1) Glasstone and Hickling, Chem. Rev., 1939, 25, 407; Trans, Amer. electro-
chem. Soc., 1939, 75, 333; Hickling and Richards, J. chem. Soc., 1940,
256; Hickling and Rodwell, ibid., 1943, 90.

(2) Klemenc, Z. physikal. Chem., 1939, 185, i; Glasstone, Trans. Amer.
electrochem. Soc., 1943, 8 4 preprint 10.

(3) Fichter,^. Soc. chem. Ind. y 1929, 48, 325T, 34iT, 347T.

(4) Faraday, Pogg. Ann., 1834, 33, 438.

(5) Kolbe, Annalen, 1849, 69, 257.

(6) Wurtz, Ann.Chim. Phys., 1855, 44, 291 ; Kolbe, Annalen, 1860, 113, 244;
Lapschin and Techanowitsch, Neue Peters. Acad. Bull., 1861, 4, 81 ;
Kekule, Annalen, 1864, 131, 79; Englebach, Kirt. Z. Chem., 1864, 653;
Brester, Z. Chem., 1866, 2, 60; Jahresb. Chem., 1866; Bourgoin, Compt.
rend., 1867, 65, 892, 998; Ann. Chim. Phys., 1868, 14, 157; Dumas,
Compt. rend., 1877, 84, 1231, 1395; Renard, Ann. Chim. Phys., 1879,
17,289; Miller, J.pr.Chem., 1879, 19,328; Bunge, J. Russ. Phys. Chem.
Soc., 1880, 12, 415; Bartoli and Papasogli, Gazzetta, 1883, 13, 22;
Jahn, Ann. Physik., 1889, 37, 408; Bourgoin, Bull. Soc. chim., 1889, 33,
417; L6b, Z. Elektrochem., 1895, 2, 296.

(7) Bunge, jf. Russ. Chem. Soc., 1890, 21, 525; Murray, J. chem. Soc., 1892,
61, 10 ; Petersen, Z. physikal Chem., 1900, 33, 108; Glasstone and
Hickling, ,?. chem. Soc., 1934, 1878.

(8) Shukla and Walker, Trans. Faraday Soc., 1931, 27, 35 ; See also Ruzicka
and Stoll, Helv. Chim. Acta, 1934, 17, 1308.

(9) Hofer and Moest, Annalen, 1902, 323, 285; Fichter and Siegrist, Helv.
Chim. Acta, 1932, 15, 698; Glasstone and Hickling, loc. cit., ref. 7.

(10) Hopfgartner, Monatsh., 1911, 32, 523.
(n) Salanze, Bull. Soc. chim., 1925, 37, 522.

(12) Schall, Z. Elektrochem., 1896, 3, 83 ; Fichter, Bull. Soc. chim., 1934, i,
1585; Trans. Amer. electrochem. Soc., 1939, 75, 309.

(13) Fichter and Zumbrunn, Helv. Chim. Acta, 1927, 10,869; Fichter and Buess,
Helv. Chim. Acta, 1935, 18, 445; Hallie, Rec. Trav. Chim., 1938, 57,

;b 152; Denina, Gazzetta, 1938, 68, 443.

(14) Glasstone and Hickling, loc. cit., ref. i, 7.

(15) Glasstone and Hickling,.?. chem. Soc., 1936, 820.

(16) Glasstone and Hickling, y. chem. Soc., loc. cit., ref. 7; Trans. Amer. Electro-
chem. Soc., 1939, 75, 333; Hickling, Quart. Reviews, 1949, 3, 121.

112

ANODIC OXIDATION

(17) Fichter and Humpert, Helv. Chim. Ada, 1926, 9, 692; Glasstone and
Hickling, J. chem. Soc., loc. cit., ref. 7; Fichter and Heer, Helv. Chim.
Ada, 1935, 18, 704; Hickling and Westwood, J. chem. Soc., 1938, 1039.

(18) Fichter and Brunner, Helv. Chim. Ada, 1929, 12, 573

(19) loc. cit., ref. 16.

(20) Holemann and Clusius, Z. physikal. Chem., 1937, 35, B, 261; Clusius
and Schanzer, ibid., 1943, 192, A, 273.

(21) Kruis and Schanzer, Z. physikal. Chem., 1942, 191, A, 301 ; Holemann
and Clusius, Ber., 1937, 70, 819; Schanzer and Clusius, Z. physikal.
Chem., 1941, 190, A, 241.

(22) Brown and Walker, Annalen, 1891, 261, 107; Trans. Roy. Soc. Edin.,
1891, 36, 291.

(23) Fichter and Stenzl, Helv. Chim. Ada, 1934, 17, 535.

(24) Hey, Ann. Reports, 1944, 41* 181 ; 1948, 45, 139.

(25) Fichter, Organische Elektrochemie, Dresden and Leipzig, 1942; Denina,
Gazzetta, 1938, 68, 443; Fichter and Meyer, Helv. Chim. Ada, 1933,
16, 1408.

(26) Petersen, Z. Elektrochem., 1906, 12, 141 ; Schmidt, Schoeller and Eberlein,
Ber., 1941, 74, 1313.

(27) Brockman, Electro-Organic Chemistry, New York, 1926.

(28) Lob, Z. Elektrochem., 1923, 29, 264; Petersen, loc. cit., ref. 26; Preuner
and Ludlam, Z. physikal. Chem., 1907, 59, 682; Fairweather and Walker,
y. chem. Soc., 1926, 3111.

(29) Fairweather and Walker, loc. cit., ref. 28.

(30) Hofer and Moest, Annalen, 1902, 323, 284; Z. Elektrochem., 1904, 10, 833.

(31) Lob, loc. cit., ref. 28; Hamonet, Compt. rend., 1896, 123, 252; Petersen,
Z. physikal. Chem., 1900, 33, 115 ; Fichter and Burgin, Helv. Chim. Ada,
1931, 14, 90.

(32) Hamonet, loc. cit., ref. 31 ; Petersen, loc. cit., ref. 31.

(33) Petersen, Z. physikal. Chem., 1900, 33, 295; Z. Elektrochem., 1906, 12,
141.

(34) Kolbe, Mem. andProc. Chem. Soc., 1848, 3, 378 ; Petersen, loc. cit., ref. 31

(35) Rohland, Z. Elektrochem., 1897, 4, 120; Petersen, ibid., 1906, 12, 142.

(36) Moser, Elektrolytischen Prozesse, 1910.

(37) Fichter, .J. Soc. chem. 2nd., 1929, 48, 351.

(38) Von Miller and Hofer, Ber., 1895, 28, 2427 ; Greaves, et al.,J. chem. Soc.,
1950, 3326.

(39) Brown and Walker, Annalen, 1893, 274, 42; von Miller, Z. Elektrochem.,
1897, 4 55-

(40) Matsui and Arakawa, Mem. Coll. Sci. Kyoto, 1932, 15 A, 189.

(41) Stoll, Helv. Chim. Ada., 1951, 34, 1817.

(42) Vanzetti and Coppadora, Gazzetta, 1904, 34, i, 154; Atti R. Accad.
Lincei, 1903, 12, 209; Carmichael, J. chem. Soc., 1922, 121, 2545.

(43) Hofer, Ber., 1900, 33, 655; Wellstatter and Pfannenstiel, Annalen, 1921,
422, i ; Fichter and Lurie, Helv. Chim. Ada, 1933, 16, 885 ; Hunsdiecker,
Ber., 1942, 75> 447-

(44) Asano, Kameda and Wada, J. pharm. Soc. Japan, 1944, 64, (8A), 25.

(45) von Miller and Hofer, Ber., 1895, 28 > 2427; von Miller, Z. Elektrochem.
1897, 4, 55; Asano, Kameda and Wada, y. pharm. Soc. Japan, 1945, 65,
(7/8A), 5; Greaves, eJ a/., J. chem. Soc., 1950, 3326.

(46) Asano, Kameda and Alada, J. pharm. Soc. Japan, 1945, 65, (4A), 15.

(47) Greaves, et al., loc. cit., ref. 45.

ORGANIC ELECTRODE PROCESSES

(48) Holemann and Clusius, Ber., 1937, 70, 819.

(49) Asano, Kameda and Wada, loc. cit., ref. 44; Stallberg-Stenhagen, Arkiv
Kemi, 1950, 2, 95; Linstead, Lunt and Weedon, J. chem. Soc. y 1950,
3331; Moroe, jf. pharm. Soc. Japan, 1950, 70, 54; 1951, 71, 123.

(50) Wietzel, Angew. Chem., 1948, 60, A, 263; Brownlee, Ann. Reports, 1948,
45, 292; Truter, Quart. Reviews, 1951, 5, 393, 394.

(51) David, Polgar and Robinson, J. chem. Soc., 1949, 1541

(52) Linstead, et al., Quart. Rev., 1952, 6, 391.

(53) Petersen, Z.physikal. Chem., 1900, 33, 294; Swann, Trans. Amer. electro-
chem. Soc., 1929, 56, 457.

(54) Karrer, et al., Helv. Chim. Ada, 1932, 15* I399J I933. x ^ 2 97J Leonard
and Mader, J. Amer. chem. Soc., 1950, 72, 5388; Linstead, Lunt and
Weedon,^. chem. SOC./IQSO, 3333 J Stallberg-Stenhagen, loc. cit., ref. 49;
Birch, et al., J. chem. Soc., 1952, 1363; Linstead, et al., J. chem. Soc.,
1952, 3261.

(55) DemjanofT, J. Russ. Phys. Chem. Soc., 1904, 36, 314; Demjanoff and
Dojarenkow, Ber., 1907, 40, 2596.

(56) Fichter and Reeb, Helv. Chim. Acta, 1923, 6, 450; Demjanoff,,? Russ.
Phys. Chem. Soc., 1929, 61, 1861.

(57) Petersen, Z. Elektrochem., 1912, 18, 710; Bouis, Bull. Soc. chim., 1923,
33, 1081; Petrov and Vyakhirev, J. Gen. Chem., U.S.S.R., 1939, 9, 513.

(58) Rohland, Z. Elektrochem., 1897, 4, 120.

(59) Kekule, Annalen, 1864, 131, 88; Bourgin, Ber., 1874, 7, 1039; Aarland
and Carstanjen, J. pr. Chem., 1871, 4, 376; Aarland, ibid., 1873, 6, 256;
Bull. Soc. chim. 1873, 19, 258; Behal, Ann. Chim. Phys., 1897, 16, 367;
Henrich and Schenk, Ber., 1919, 52, 2120; Henrich and Herzog, ibid.,
1919, 52, 2126; Fichter and Petrovich, Helv. Chim. Acta, 1941, 24,

549-

(60) L8b, Z. Elektrochem., 1895, 2, 296.

(61) Fichter and Meyer, Helv. Chim. Acta, 1934, 17, 535.

(62) Fichter and Stenzl, Helv. Chim. Acta, 1939, 22, 971.

(63) Fichter and Meyer, loc. cit., ref. 61.

(64) Fichter and Stenzl, loc. cit., ref. 62; Linstead, Shephard and Weedon,
J. chem. Soc., 1952, 3624.

(65) von Miller and Hofer, Ber., 1894, 27, 461; Petersen, Bull. Acad. roy.
Danemark, 1897, 387; Elbs and Kratz, J. pr. Chem., 1897, 55, 502;
Kaufler and Herzog, Ber., 1909, 42, 3870; Fichter and Schmid, Helv.
Chim. Acta, 1920, 3, 704; Fairweather, Proc. Roy. Soc. Edin., 1925, 45>
23; Fichter and Schnider, Helv. Chim. Acta, 1930, 13, 103.

(66) Lasser-Cohn, Annalen, 1889, 251, 341; Bunge, J. Russ. Phys. Chem.
Soc., 1892, 24, 690; Troeger and Evers, J. pr. Chem., 1898, 58, 121;
Kaufler and Herzog, Ber., 1909, 42, 3858.

(67) Fichter and Schmid, loc. cit., ref. 65.

(68) von Miller and Hofer, Ber., 1894, 27, 469; 1895, 28, 2487; Walker,
y. Chem. Soc., 1896, 69, 1278; Baur, Z. Elektrochem., 1919, 25, 102.

(69) Tommila, Suomen Kern., 1933, 6B, 70; Ann. Acad. Sci., Fenn., 1933,
38A, No. 6.

(7o).^Kolbe, Annalen., 1860, 113, 244; von Miller and Hofer, loc. cit., ref. 68;
'"Walker, loc. cit., ref. 68; Hamonet, Compt. rend., 1901, 132, 259.

(71) Smull and Subkow, Chem. and Met. Eng., 1923, 28, 357.

(72) Neuberg, Biochem. Z., 1908, 7, 527; Neuberg, Scott and Lachmann,
ibid., 1910, 124, 152.

114

ANODIC OXIDATION

(73) Fichter, Stenzl and Bealinger, 1938, 21, 375; Fichter and Stenzl, ibid.,
1939, 22, 971; Linstead, Shephard and Weedon, J. chem. Soc., 1952,
3624.

(74) Linstead, Shephard and Weedon, J. chem. Soc., 1951, 2854.

(75) Noyes, J. Amer. chem. Soc., 1897, 19, 775; Hamonet, Bull. Soc. Mm.,
1905, 33, 513; Fairweather, Proc. Roy. Soc. Edin., 1925, 45, 23.

(76) Kitaura, Bull. Inst. Phys. Chem. Res., Tokyo, 1937, 16, 765.

(77) Hofer, Ber., 1900, 33, 655; Willstatter and Pfannenstiel, Annalen, 1921,
422, i ; Fichter and Lurie, Helv. Chim. Acta, 1933, 16, 885.

(78) Offe, Z. Naturforsch, 1947, 26, 182; Linstead, Shephard and Weedon,
loc. cit., ref. 74.

(79) Offe, Z. Naturforsch, 1947, 26, 185.

(80) Wohl and Schweitzer, Ber., 1906, 39, 890; Stoll and Rouve, Helv. Chim.
Acta, 1937, 20, 525.

(81) Lob, Z. Elektrochem., 1895, 2, 296; Oettell, ibid., 1894, 2 9; Petersen,
Z. physikal. Chem., 1900, 33, 698; Alkersberg, Z. anorg. Chem., 1902,
31, 161; Salzer, Z. Elektrochem., 1902, 8, 893; Walker and Wood,
J. chem. Soc., 1906, 89, 598; Vanzetti, Atti R. Accad. Lined, 1904, 13,
ii, 112; 1906, 15, i, 574; 1907, l6 tii, 79, 139; 1908, 17, ii, 33 1 J Doumer,
Compt. rend., 1927, 184, 747; Tommila, Ann. Acad. Set., Fenn., 1932,
36A, No. i ; Sihvonen, Suomen Kem., 1933, 6B, 64.

(82) Brown and Walker, Annalen, 1891, 261, 107; Trans. Roy. Soc. Edin.,
1891, 36, 291.

(83) Brown and Walker, Annalen, 1893, 274, 41; Komppa, Ber., 1901, 34,
900; Walker and Lumsden, J. chem. Soc., 1901, 79, 1197; Bouveault,
Bull. Soc. Mm., 1903, 29, 1038; J. Walker and A. Walker, J. chem. Soc.,

1905, 88, 961; Walker and Wood, ibid., 1906, 89, 598; Crichton, ibid.,

1906, 89, 929; Stosius and Wiesler, Biochem. Z., 1920, 108, 75; Walker,
J. Soc. chem. Ind. y 1921, 40, 169!*; Fairweather, Proc. Roy. Soc. Edin.,

1925, 45t 23, 283; 1926,46,71.

(84) Robertson, J- chem. Soc., 1925, 127, 2057.

(85) Brown and Walker, Walker and Lumsden, Bouveault, Crichton, loc. cit.,
ref. 83.

(86) Vanzetti and Coppadoro, Atti R. Accad. Lincei, 1903, 12, ii, 209.

(87) Mulliken, Amer. chem. y., 1893, 15, 523; Weems, ibid., 1894, 16, 569;
Alpiani and Gasperoni, Gazetta, 1902, 32, ii, 235; Ulpiani and Rodano,
ibid., 1905, 35, ii, 365; Atti R. Accad. Lincei, 1905, 14, i, no; 1905,
14, ii, 601.

(88) Linstead, Shephard and Weedon, J. chem. Soc., 1952, 3624.

(89) Linstead, Shephard and Weedon, J. chem. Soc., 1951, 2854.

(90) Hayashi and Wilson, unpublished results.

(91) Van der Hoek and Nauta, Rec. Trav. Chim., 1942, 61, 845.

(92) loc. cit., ref. 88.

(93) Fichter and Stenzl, loc. cit., ref. 73 ; Linstead, Shephard and Weedon,
loc. cit. t ref. 88.

(94) loc. cit., ref. 89.

(95) Fieser, Clapp and Daudt, J. Amer. chem. Soc., 1942, 64, 2052.

CHAPTER VII

ANODIC OXIDATION OF ALIPHATIC
HETEROCYCLIC AND AROMATIC

COMPOUNDS

(i) OXIDATION OF ALIPHATIC COMPOUNDS

THE straightforward oxidation of aliphatic compounds does not
generally lead to as interesting results as one would like, simply because
the powerful effects of anodic oxidation have not as yet been completely
mastered. Considerable care must be exercised in applying this tech-
nique, for it is much too easy to obtain carbon dioxide and water as
end products. Considerable investigation has been done on the effects
of anodic oxidation on the alcohols, aldehydes, ketones, sugars, and
acids. A survey of this work may serve as a clue to future efforts.

(a) Alcohols

The anodic oxidation of an alcohol in an acidic medium generally
leads to the formation of an aldehyde, with subsequent oxidation to the
acid (i). To obtain the aldehyde it is necessary to utilize low current
densities, or, if high current densities are used, it is necessary to perform
the reaction at elevated temperatures, in order that the aldehyde may
distill off as it is formed, and thus be protected from further oxidation
(2). Should it be desired to oxidize the alcohol to the acid, precautions
must be taken to keep the intermediate aldehyde which is formed in
solution, in order to enable its further oxidation. Thus a dilute sulphuric
acid solution of ethanol can be oxidized to acetic acid using a platinum
electrode, a current density of 0-2 amp./cm. 2 , and maintaining the
temperature at approximately 30 C. The current efficiency in this
instance is about 78 per cent (3). Anodic oxidation of isopropyl
alcohol under comparable conditions yields acetone. These results
duplicate those obtained by chemical means (4).

116

ANODIC OXIDATION

When the alcohol is subjected to anodic oxidation in a basic medium,
no definite predominating product results, for although the aldehyde
may be formed in large quantities, it condenses rather easily to
resins and other condensation products. In addition the presence of
acidic material, saturated hydrocarbon, and hydrogen in the reaction
mixture is not unusual (5). An explanation for this reaction has been
put forth in the " hydroxidation " theory (6). In solution the alcohol
dissociates into the hydrogen ion and the ethoxide ion, in the case of
ethyl alcohol. The ethoxide ion is oxidized at the electrode surface via
a one-electron change, to yield the ethoxy radical, which splits, in an
effort to assume a stable state, to the aldehyde or hydrocarbon :

CH 3 CH 2 OH - - v CH 3 CH 2 O : + H

CH 3 CH 2 O- - CH 3 CH 2 O-

CH 3 CHoO- * CH 3 CHO H II-

* H,

------ - CH 4

CH 3 CH 2 - ------ > I ICHO -f CH, f CH

The assumption of the existence of an ethoxy radical is entirely possible,
although the presence of small concentrations of the ethoxy ions and
relatively large concentrations of hydroxyl ions, as would be anticipated
with the medium used, makes this theory questionable.

The higher alcohols are generally oxidized to the corresponding
acid in an acidic medium (7). For example, propyl alcohol, when
subjected to anodic oxidation at a lead dioxide or platinum anode in a
dilute sulphuric acid medium, results in a 90 per cent yield of propionic
acid, together with a very small quantity of the aldehyde. Isovaleric
aldehyde is obtained as a result of anodic oxidation of isoamyl alcohol
in an aqueous sulphuric acid medium at a relatively low current density,
with either a lead dioxide or nickel anode. At the higher current
densities isovaleric acid is the main product (8) :

(CH 3 ) 2 CH CH 2 CH 2 OH ------ > (CH 3 ) 2 CH CH 2 CHO ---------- ->

(CH 3 ) 2 CH CH 2 COOH

The secondary alcohols such as menthol, borneol, and tropine, when
subjected to anodic oxidation in an acidic medium, yield the respective
ketones menthone (9), camphor (10), and tropinone (n):

117

ORGANIC ELECTRODE PROCESSES
CH 3 CH 3

CH 2 CH CH 2 CH 2 CH CH 2

II II

CH 2 CH CH(OH) > CH 2 CH C=

CH 3 CH CH 3

menthol

CH 3 CH CH 3
menthone

CH 2

CH 3
-C CH(OH)

CH 3 C CH 3

-CH CH 2

. borneol

CH 2

CH 3
-C

CH 3 C CH 3

CH 2

-CH CH 2

camphor

CH 2

CH

CH 2
I I

N(CH a ) CHOH

CH

CH

tropinone

In an acid electrolyte at a lead dioxide electrode, glycol yields glycol
aldehyde, glycolic acid, formaldehyde, trioxymethylene, a polymer of
formaldehyde, formic acid, and a small quantity of a hexose, depending
on the electrolysis conditions (12).

HOCH 2 CH 2 OH > HOCH 2 CHO > HCHO -> (CH 2 O) 3

HCOOH
(6) Aldehydes

In an acid medium the aldehydes are generally oxidized to their
corresponding acids, with some evolution of carbon monoxide and
dioxide. Thus from acetaldehyde and propionaldehyde, acetic acid
and propionic acid are the respective predominating products (13). In
a basic medium acetaldehyde will give a satisfactory yield of formic
acid (14), with the higher aldehydes giving some amount of saturated
hydrgcarbon (15). Of interest are the results obtained when formalde-
hyde is subjected to anodic electrolysis in a basic medium at a platinum
anode. In this case there is no gas evolution observed at the anode.
However, if a copper anode is substituted for the platinum anode,

118

ANODIC OXIDATION

hydrogen is formed at the anode in an amount equal to that at the
cathode, thus (16):

HCHO + O > CO 2 -f H 2 -f 2e

The mechanism proposed for this reaction assumes an initial hydration
of the formaldehyde, followed by the formation of an ion which
decomposes to hydrogen and carbon dioxide :

_ /OH

H 2 C = O > H 2 C 01 + HOH > H 2 C<;

. Nl

OH
2H+

Ho -f CO 2 2e H 2 C< ~
< \OI

This result can also be obtained by utilizing anodes coated with finely
divided copper or silver. However, in these instances, the evolution of
hydrogen from the anode is spasmodic. The proposed mechanism is
open to question since it is known that some metal oxides will cata-
lytically decompose formaldehyde in a basic medium, with resultant
hydrogen formation (17).

(c) Ketones

To date very little effort has been devoted to the effects of anodic
oxidation on aliphatic ketones. Utilizing a platinum anode at low
current densities in an acid medium, acetone will yield acetic and
formic acids (18). At higher current densities, in addition to the afore-
mentioned acids, methane, ethane, unsaturated hydrocarbons, mesityl
oxide, carbon dioxide and monoxide are obtained (19). Similar results
were also obtained using an alternating current (20).

(d) Sugars

Considerable effort has been put forth on the effects of anodic
oxidation of sugars. One of the earlier investigations involved the
oxidation of glucose at a lead dioxide anode in an aqueous-sulphuric
acid medium (21). The results led the investigator to suggest that the
initial oxidation resulted in a degradation of the molecule to gluconic
acid which then oxidized to both arabinose and saccharic acid. The
arabinose was further oxidized to its acid, thence to trihydroxyglutaric
acid, formaldehyde, and finally formic acid, thus :

119

ORGANIC ELECTRODE PROCESSES

H H OH H H H OH H

I I I I I I I I
HOCH 2 C C C C CHO >HOCH 2 C C C C COOH

I I I I I I I I

OH OH H OH OH OH H OH

H H OH H H OHH

III I I I I

HOH 2 C C C C CHO HOOC C C C C COOH

III I I I I

OH OHH , OH OHH OH

I
H H OH H H OH

HOH 2 C C C C COOH -> HOOC C C C COOH ~> HCHO ->
III III HCOOH

OH OH H OH OH H

It was found that in a medium containing calcium carbonate and a small
quantity of calcium bromide, arabinose could be oxidized at a platinum
anode to the calcium salt of its acid (22) :

H H OH

I I I
2 HOH 2 C C C C CHO

I I I
OH OHH

H H OH

I I I i

HOH 2 C C C C COO Ca

III
OH OH H

The oxidation of xylose, 3 -methyl-glucose, mannose, galactose, and
maltose to their respective acids has been reported with varying results

(23).

0) Carboxylic Acids

In the previous chapter the Kolbe synthesis was discussed. How-
ever, in passing it might be of interest to briefly mention the straight-
forward oxidations that occur when a carboxylic acid is subjected to
anodic oxidation at a platinum anode in a sulphuric acid medium.
Acetic acid can be oxidized to glycolic acid, glyoxylic acid, and thence to
oxalic acid, with formaldehyde and formic acid as side products (24):

CH 2 OH CHO COOH

CH 3 COOH > I > I > I

Acetic
acid

COOH
Glycolic
1 acid

HCHO >
Formaldehyde

COOH
Glyoxylic
acid

HCOOH
Formic acid

COOH
Oxalic
acid

120

ANODIC OXIDATION

In a similar manner lactic acid has been oxidized to acetaldehyde, and
pyruvic acid (25), and malic acid to malonic acid and oxalic acid
or acetaldehyde (26).

CH 3 CH(OH)COOH > CH 3 CHO

Lactic acid Acetaldehyde

I

CHgCOCOOH
Pyruvic acid

COOH
CH(OH)

COOH

I

COOH

^ |

CH 2

> \
COOH
Malonic
acid

COOH

Oxalic
acid

CH 2

COOH
Malic acid

CH 3 CHO
Acetaldehyde

(/) Alicyclic Compounds

Compounds of this type are quite amenable to anodic oxidation,
and in many instances, under controlled conditions, will yield the desired
product. Thus cyclohexane in a sulphuric acid medium at a lead
dioxide anode will yield cyclohexanol, cyclohexanone, and ultimately
under more drastic conditions, adipic acid, thus (27) :

CH 2 CHOH

/\ /\

CH 2 CH 2 CH 2 CH 2

CH 2 CH 2 CH 2 CH 2

\/ \/

CH 2 CH 2

Cyclohexane Cyclohexanol

I

O

II
COOH C

I /\

(CH 2 ) 4 < CH 2 CH 2

COOH CH 2 CH 2

Adipic acid \ /

CH 2
Cyclohexanone

121

ORGANIC ELECTRODE PROCESSES

Under similar conditions the same investigators oxidized cyclohexene
and obtained the following products :

C O

CH
CH 2 CH
CH 2 CH 2

CH 2
Cyclohexene

H 2 C C
H 2 C CH 2

CHOH
H 2 C CHOH
H 2 C CH 2

CH a

Cyclohexene
oxide

CH 2
Cyclohexanediol

O

li

C

/\ COOH

CH 2 CH 2 |

| | > (CH,),

CH 2 CH 2 |

\y COOH

CH 2
Cyclohexanone Adipic acid

Cymene has been prepared by anodic oxidation of a-pinene, using
a lead dioxide anode and performing the reaction in a dilute aqueous
sulphuric acid medium (28). The side reaction in this anodic process
leads to terebic acid.

CH 8
C

HC - '"CH

| (CH 3 )C(CH 3 )

H

CH 8
C
CH

H 2 C

CH 2

CH

HC CH

Y

CH(CH 3 ) Z
Cymene

\

a-Pinene

(CH 3 ) 2 C O

HOOC CH C

\/
C

O

Terebic acid

122

ANODIC OXIDATION

The anodic oxidation of the hydroxy group of borneol, menthol, and
tropine, with the resultant isolation of the ketonic compounds camphor,
menthone, and tropinone respectively, has already been mentioned
(p. 117). In spite of the relative complexity of the molecule, the yield
of ketone is quite good, indicating that the electrolytic process can be
a nondestructive one.

(g) Miscellaneous Aliphatic Anodic Oxidations

Ammonium formate and formamide will, on oxidation in liquid
ammonia or in an ammonium hydroxide medium at a platinum anode,
yield urea in satisfactory quantities (29) :

O

NH,(liq.) ||

HCOONH 4 > H 2 NCNH 2

O O

II NH 4 OH ||

HCNH 2 > ]

Under comparable conditions, ammonium acetate will also give urea
(30). Ethyl thiocyanate, when oxidized in an acetic-hydrochloric acid
medium at a platinum anode, will yield ethyl sulphonic acid (31):

C 2 H 5 SCN > C 2 H 5 S0 8 H

Formamidine disulphide has been obtained as a salt from the anodic
oxidation of thiourea in an acid medium at a platinum electrode (32).
The salt obtained depends on the acid used in the electrolysis medium.

H 2 NCSNH 2 -

Sc s s c/

H 2 N/ \NH 2

A 50 per cent yield of 6/s-(j3-y-dichlorophenyl)-formamidine disulphide
dihydrochloride has been claimed from the anodic oxidation of allyl
thiourea (33) :

Pt
CH 2 -CH CH 2 NH CS NH 2

HC1

rClH 2 C CHC1 CH 2 NH C SI

II -2HC1

L HN J 2

In addition, the above reaction also involves anodic halogenation,
which will be discussed in greater detail later.

123

ORGANIC ELECTRODE PROCESSES

(2) OXIDATION OF HETEROCYCLIC COMPOUNDS

Compounds of this type can give interesting and useful products,
if the techniques of anodic oxidation are applied with caution. Under
rigorous conditions the oxidation will yield only small fragments of the
starting product. Anodic oxidation of furfuraldehyde in a dilute
sulphuric acid medium at a lead dioxide anode yields maleic acid. If
the electrolysis is performed without a diaphragm succinic acid is
obtained as a result of reduction of the double bond (34).

CH-C CHO .- HC COOH CH 2 COOH

O II ->

CH = CH HC COOH CH 2 COOH

Pyridine at a platinum anode in an aqueous sulphuric acid medium
yields 2-pyridyl-pyridinium sulphate, which then gives glutacon-di-
aldehyde (35), thus:

=1 -

S0 3 H

X CH CHO

HC<f

X CH-CHOH

If a lead dioxide anode is used, complete decomposition occurs, with the
products consisting of a mixture of formaldehyde, formic acid, methyl
amine, ammonia, nitric acid, carbon monoxide and dioxide (36).
Anodic oxidation of methyl N-methyl-pyridinium sulphate and methyl
N-methyl-quinolinium sulphate in a basic medium at an iron electrode
in the presence of a small amount of potassium ferricyanide will result in
good yields of N-methyl-a-pyridone and N-methyl quinolone respec-
tively (37):

_

s0 4 CH t \ N-CH

S0 4 CH 3

ANODIC OXIDATION

The oxidation of piperidine at a lead dioxide electrode in a sulphuric
acid medium results in the formation of 8-amino-valeric acid, with
smaller quantities of glutaric and succinic acids (38):

/CH 2 CH 2 , /CH 2 COOH

H 2 C< >NH > H 2 C< y NH 2

\CH a CH 2 / \CH a CH 8 /

CH 2 COOH /CH 2 COOH

CH,COOH CH 2 COOH

Under carefully controlled conditions nicotine can be oxidized to
nicotinic acid using a lead dioxide or platinum anode in an aqueous
sulphuric acid medium. More drastic conditions result in complete
decomposition (39).

CH CH 2 CH 2 H

\ 1 I / \

CH C CH CH 2 > CH C COOH

I II \ / I II

CH CH N CH CH

% /I \ /

N CH 3 N

(3) OXIDATION OF AROMATIC COMPOUNDS

The anodic oxidation of aromatic compounds usually leads to more
interesting results than those obtained with the aliphatic compounds.
Experiments on this series of compounds started with the initial
investigations of Renard (40), who did not go further with his work
other than to record the observation that electrolytic oxidation of
benzene in an ethanolic-sulphuric acid medium yielded unidentifiable
substances. As a result of extensive investigations (41) subsequent to
Renard 's initial studies it is possible to indicate the general reaction
process :

' \ -> / S OH -> < ; S OH -> Decomposition
=/ \P=/ \=/ products

Benzene Phenol Catechol

HO/ -OH -> O = < =-O -* HOOC CH=CH COOH

Hydroquinone Quinone Maleic acid

Although phenol has not been isolated from the reaction mixture, it
is quite probable that it is the intermediate to the formation of hydro-

"5

ORGANIC ELECTRODE PROCESSES

quinone and catechol which can be obtained using a platinum, lead
dioxide, or graphite anode in a sulphuric acid medium or a phosphoric
acid medium (42). Phenol, however, can be obtained in small yields
by electrolysis of benzene in an acetic acid-sodium acetate medium
using a platinum anode (43). Catechol, too, can only be detected in
small amounts, as it is easily decomposed at the anode. It is not too
difficult to obtain high yields of quinone by performing the electrolysis
on an aqueous sodium sulphate or sodium acetate suspension of benzene,
using a lead dioxide anode (44). The quinone is readily converted to
hydroquinone by performing the electrolysis without a diaphragm. A
method which has been proposed for the preparation of hydroquinone
involves an anodic oxidation of a suspension of benzene in an acetic-
sulphuric acid medium, which is then passed into the cathode compart-
ment containing a lead or zinc electrode together with a finely divided
lead sulphate catalyst (45). Further oxidation of quinone yields maleic
acid as well as other decomposition products (46).

Another aromatic which has been investigated quite extensively is
toluene. According to the information available as to the oxidation
products obtained (47), the following scheme can be used to represent
the oxidative process :

0=

HO

Toluene

o-Cresol

. CH 3

OH
Toluhydroquinone

=0

CH,

HO <f V- CH,

*/>-Cresol

Quinone

complete degradation

OH

Homopyrocatechol
126

ANODIC OXIDATION

The initial investigations of toluene indicated that the aromatic
ring was quite resistant to oxidative change. However, it was found
that the methyl group underwent oxidation more readily, to yield
benzaldehyde (48). There are indications that benzyl alcohol is inter-
mediate to the formation of the aldehyde, but, as the alcohol is quite
unstable, it is immediately oxidized to the more stable benzaldehyde
(49). Benzaldehyde may be obtained in approximately 20 per cent
yield by anodic oxidation at a platinum anode and at low temperature,
from an emulsion of toluene in a mixture of dilute nitric and acetic
acids (50). This yield can be improved considerably by the use of an
oxygen carrier (51). Similar results were obtained when toluene was
dissolved or " salted-in " in an aqueous solution containing sodium
xylenesulphonate (52). Other oxidative products, such as benzoic acid,
are also to be found in the reaction products, together with hydroxylated
derivatives of toluene, such as o-cresol and />-cresol, which are inter-
mediate to the formation of toluquinone and quinone (53). Tolu-
quinone can be further oxidized at a lead dioxide anode to a mixture
of acids consisting of formic, mesaconic, and maleic acid (54).

Oxidation of chlorobenzene proceeds according to the following
scheme :

COOH

I

; > HO \^y~ Cl > O^Z/^ > ( CH 2) 2

COOH
p-Chlorophenol Quinone Succinic acid

The succinic acid is only obtained in the absence of a diaphragm
(55). Treatment of />-fluortoluol in a similar manner yields p-fluor-
benzoic acid in approximately 10 per cent yield, with the remainder
being essentially decomposition products (56). The oxidation of
p-chlortoluene to p-chlorbenzoic acid in an acetic-nitric acid medium
at a platinum anode has been reported (57). In this instance the value
of electrolysis is questionable, as the nitric acid itself can oxidize
j!>-chlorotoluene to the respective benzoic acid derivative. The anodic
oxidation of />-nitrotoluene in an acetic-sulphuric acid medium using a
platinum anode will give a 40 per cent yield of />-nitrobenzyi alcohol,
in addition to smaller quantities of />-nitrobenzaldehyde and ^-nitro-
benzoic acid (58).

Ethylbenzene yields numerous products, either as a direct result of

127

ORGANIC ELECTRODE PROCESSES

oxidation as a suspension in a dilute sulphuric acid medium at a lead
dioxide anode, or as a consequence of secondary reactions (59). The
following scheme illustrates the reaction paths :

>-C 2 H 5

Ethyl benzene Methyl phenyl
carbinol

>CH(OH)CH 3 -> C VCOCH, > VCHO

\ .-X \ /

Acetophenone Benzaldehyde

\

CH 8 CH 3

2,3-6n-phenyl-2,3-butanediol
(formed at the cathode in the
absence of a diaphragm)

p-Ethylphenol

Quinone

OH
Ethylhydroquinone

2,2 / -dihydroxy-5,5 / -diethyl diphenyl
O

^ -C 2 H 6

6

Ethylquinone

The oxidation of 2,4-dinitrotoluene at a lead dioxide anode in a
dilute aqueous sulphuric acid medium at temperatures of 90 to 95 C.
will give approximately 26 per cent, 2,4-dinitrobenzoic acid (60). This
yield 45 can be considerably improved (74 per cent yield) by using a
platinum anode, a 50 per cent aqueous sulphuric acid medium, a
temperature of 75 C., and 2 per cent chromic acid as a catalyst (61).

128

ANODIC OXIDATION

The anodic oxidation of the xylols has also been extensively investigated
(62). The products obtained can be illustrated by the scheme for
p-xylol.

H 3 C-< VcH.

p-Xylol -Toluylaldehyde

OH OH O

I jj Terephthalaldehydic acid

r^T-T <v. T-J C* " xs^TU ^ T.J

-ChL 3 > H 3 C-\' /~^H 3 > H 3

OH O

/>-Xylenol p-Xylohydroquinone />-Xyloquinone

IIOOC-

Terephthalic acid

/ V^ J. J.

CH 3

4,4 / -dihydroxy-2,5,2 / ,5 / -tetramethyl-diphenyl

It is not unreasonable to express the opinion, at this point, that the
diphenyl compounds formed in these instances, and in the illustrated
example on the oxidation of ethyl benzene, came about as a result of
free-radical formation. It is also of interest to mention that if the
oxidation of />-xylene is performed in the presence of acetone the
terephthalaldehydic acid condenses with the acetone, to form/>-carboxy-
benzal acetone, thus :

H 3 CCOCH 3 -> HOOC-/^VcH=CH-COCH t

v=/

The oxidation of naphthalene will yield a-naphthol, naphthoquinone,
and finally phthalic acid, when the reaction is performed on a suspension
of the material in a dilute sulphuric acid medium (63). The efficiency
of this reaction is, as with all emulsion type reactions, extremely poor.
However, the addition of an oxygen carrier such as Ce(S0 4 ) 2 improves

0.tP. 129 <

ORGANIC ELECTRODE PROCESSES

the efficiency considerably (64). The process is illustrated in the follow-
ing reaction scheme :

OH

OH

COOH
COOH

OH

Naphthalene cc-Naphthol Naphthohydro- Naphtho-

quinone quinone

Phthalic acid

When oc-methyl naphthalene is oxidized anodically in an acidic acetone
medium at a lead dioxide electrode, the methyl group is apparently
unattacked, with the major reaction product being i,i'-dimethyl-4,4'-
dinaphthyl (65) :

H 3 C

HoC

CH 3

The electrolytic oxidation of anthracene to anthraquinone is of
considerable importance, as the latter compound is an intermediate in
the production of alizarin (66). The oxidation of a suspension of
anthracene occurred with approximately 50 per cent efficiency when
the reaction was performed at a platinum anode at 40 C., with current
densities of from o-oi to 0-03 amp./cm. 2 , in an acetone-sulphuric acid
medium. The efficiency of this process was increased to more than 80
per cent by the addition of small quantities of either cerium, chromium,
or manganese salts, which served as oxygen carriers. Similar results
have also been obtained with a lead dioxide anode and a 20 per cent
sulphuric acid medium (67). Phenanthrene, upon oxidation in the
presence of an oxygen carrier, will yield phenanthraquinone, which,
on further oxidation, will yield diphenic and benzoic acids (68) (p. 131).

As previously shown in the discussions on benzene oxidation,
hydr9xylated intermediates are formed prior to the quinone type com-
pounds. It was found that when phenol was oxidized at a lead dioxide
electrode in a sulphuric acid medium, diphenyl compounds, probably
attributable to free-radical formation, were also formed. As a result of

130

ANODIC OXIDATION
\

:OOH

COOH

Phenanthraquinone

W COOH

Diphenic
acid

Benzole
acid

the various investigations (69) the following is the reaction process
involved in the oxidation of phenol.

OH

OH

*>/

OH

Catechol

OH

OH

Monophenyl-

ether of
pyrocatechol

OH

OH

4,4 -d ihydroxy-
diphenyl

OH

OH

O

Phenol

Ilydro-
quinone

Quinone

OH

Y

THCf

V^IA VxV-

!!

CHCOOH

Maleic acid

2,4'-dihydroxy-
diphenyl

One group of investigators demonstrated that the nature of the
products obtained in the oxidation of phenol depends to a great extent
on the current density used (70). This being the case, it is not unreason-

131 2

ORGANIC ELECTRODE PROCESSES

able to assume that the reaction path is dependent on the potential at
the electrode surface.

The anodic oxidation of o-cresol at a lead dioxide electrode in a
sulphuric acid medium results in the formation of the following products

CH< CH 3

-C-

/~~-OH -> HO-~-OH -> O=<;>==O -> HOOC-CH-C-COOH

\- - / \-=/ f NTT /

o-Cresol Toluhydroquinone Toluqinone Mesaconic acid

Y

Decomposition products

esol Toluhydroqumone Toluqmone Mesaconi

I J

| Decomposition ]

H 3 C CH 3

HO -

o-Dicresol

It is claimed (72) that anodic oxidation of o-cresol in a concentrated
alkali hydroxide at a lead or copper anode will give a good yield of
salicylic acid.

CH 3 COOII

Benzyl alcohol in an aqueous sodium carbonate medium at a platinum
anode will give a high yield of benzoic acid (73). However, at a nickel
anode in a sodium benzenesulphonic acid medium, a mixture of
benzaldehyde and benzoic acid is obtained (74), which is what might
be expected of this lower potential anode. As it is known that benzalde-
hyde is quite susceptible to air-oxidation, it may be in this instance
that the acid obtained is a result of air-oxidation, and not a direct
consequence of the electrolysis.

(a) Agomatic Aldehydes and Ketones

The anodic oxidation of aromatic aldehydes and ketones offers
generally a synthetic route of little interest. Benzaldehyde can be
oxidized almost quantitatively to benzoic acid, utilizing a copper oxide

133

ANODIC OXIDATION

anode and an aqueous sodium benzenesulphonic acid medium (75).
In an acetone-sulphuric acid medium, benzoin is oxidized in a stepwise
manner according to the following scheme (76) :

C 6 H 6 CH(OH)~ CO C 6 H 6
Benzoin

C 8 H 6 CO--CO C 6 H 5
I Benzil

C 6 H 6 CHO

Benzaldehyde

C 6 H 5 COOH
Benzoic acid

This oxidation, when performed in an alcoholic-sodium hydroxide
medium, yields only a small quantity of benzoic acid. In this instance
the poor yield can possibly be attributed to the reactivity of the carbonyl
groups in a basic medium, with resultant formation of condensation
products. In an alcoholic-hydrochloric acid medium an excellent yield
of benzil is obtained (77). Of interest is the oxidation of anthraquinone
to give a mixture of alizarin and quinizarin in 90 per cent yield, when the
electrolysis is performed at a platinum anode in a concentrated sulphuric
acid medium. In addition there are obtained smaller quantities of
further oxidation products such as purpurin and alizarin-cyanin (78).

OH

O

II
C\

O

II

O

Anthraquinone

\

O OH

Quinizarin

HO O OH

Alizarin-cyanin

(ft) Aromatic Carboxylic Acids

As in other types of aromatic compounds discussed, the initial step
in the anodic oxidation of an aromatic acid is the introduction of a

133

ORGANIC ELECTRODE PROCESSES

hydroxyl group, either ortho or para to the carboxyl group. This is
followed by further hydroxylation, with eventual decomposition upon
too prolonged an exposure to electrolysis.

The oxidation of benzoic acid at a platinum anode in a dilute
sulphuric acid medium illustrates the aforementioned (79) :

:OOH

OH

Catechol

HO

/Benzoic acid Salicylic 2,5-Dihydroxybenzoic

acid acid

OH

HO-/ V-COOH ~> HO-/~\-COOH -> HO-/~V-COOII

\=/ \==S Np=r=/

HO HO

p-Hydroxybenzoic 3,4-Dihydroxybenzoic 2,4,5-Trihydroxybenzoic
acid acid acid

J

Q Decomp.

Quinone

The reports indicate that the isolation of any one particular
compound is not large enough to make this an interesting preparative
method. Of passing interest is the anodic oxidation of o-toluic acid at
a lead dioxide anode in an acetone-sulphuric acid medium to yield
approximately 5 per cent phthalic acid, plus a considerable quantity of
decomposition products (80) :

CH S COOH

COOH > /V-COOH + decomp. products

Similar conditions of electrolysis, when used on />-toluic acid, result
in approximately 14 per cent terephthalic acid (81).

^34

ANODIC OXIDATION

(c) Miscellaneous Aromatic Oxidations

The electrochemical oxidation of a methyl group attached to an
aromatic nucleus is considerably facilitated by the presence of a cyano-
group in the molecule. Thus ortho-, meta-, and para-tolunitrile when
oxidized at a lead dioxide anode in a sulphuric acid or acetone-sulphuric
acid medium will give the respective cyanobenzoic acid in yields of
6-2 per cent, 28 per cent, and 37 per cent respectively. In addition,
the para-tolunitrile also yields approximately 8 per cent of terephthalic
acid (82).

Primary aromatic amines, when subjected to anodic oxidation,
generally yield imines or azo compounds, depending on conditions of
electrolysis. Thus the electrolytic oxidation of asymmetric w-xylidene
at an iron anode in an alcoholic potassium hydroxide medium will yield
azo-w-xylene (83) :

H 3 C <^_^-NH 2 > H 3 C

CH 3 CH 3 CH.

Utilizing a lead dioxide or a platinum anode and a dilute sulphuric acid
medium the products obtained consist of approximately 50 per cent
imino-2,4-dimethylquinol (I), 20 to 30 per cent 2,4-dimethylquinol (II),
and a small quantity of 3-hydroxy-2,6-dimethylbenzoquinone-(i,4)
(III). If elevated temperatures are used, the major product obtained
is />-xyloquinone (IV), thus (84):

CH 8

TT p /=\ heat } s.

- HO>C> ^ H -^H

I

H 3

I II IV

CH 3 \

O CH 3

<

-OH

O
III

In a concentrated sulphuric acid medium the amino group is

J 35

ORGANIC ELECTRODE PROCESSES

apparently unaffected, with the major product being 4-amino-2,5-
xylenol :

H 3 C

\_
H 3 C <f V- NH 2 > HO/ V- NH 2

X CH 8 X CH 3

The secondary and tertiary amines are not as vulnerable to attacks
as are the primary amines.. The electrolytic oxidation of such com-
pounds has been of interest in the dye-stuff industry. The anodic
oxidation of the leuco-base of malachite green to malachite green has
been accomplished, using a nickel anode in a dilute sulphuric acid
medium containing a catalytic quantity of uranyl sulphate as an oxygen
carrier (85):

CHC 6 H 5 > C-Celfr,

(CH 3 ) 2 N <f >/ (CH 3 ) 2 N-<

It might be of interest to mention that one investigator (86) has
developed a method in which the leuco-base was impregnated into the
cloth fibres and then submitted to anodic oxidation. The leuco-base
of brilliant green has also been oxidized anodically by pressing the
reduced form of the dye into the pores of a carbon anode and performing
the electrolysis in an aqueous sodium sulphate medium (87). There is
no doubt that the oxidation of the leuco-bases proceeds through a
free-radical mechanism. This was demonstrated chemically by
Michaelis et al. (88), who demonstrated that organic dyestuffs undergo
a two-step oxidation. The first is the half-oxidized form or free radical,
and the second is the completely oxidized or quinoid form. These
investigators also found these changes to be pH dependent. This
phenomenon was also found to be the same at an electrode surface.
Investigations at a rotating platinum micro-electrode showed that when
the leuco-bases of either crystal violet or malachite green are subjected
to electrolysis in a strongly acidic aqueous hydrochloric acid medium,
only % one-electron change occurred. The product in the case of the
leuco-base of crystal violet gave every indication of being the free-
radical tris-(/> : p' : />''-dimethylamino)triphenylmethyl trihydrochloride,
thus (89):

136

ANODIC OXIDATION

HN(CH 3 ) 2

H
(CH 3 ) 2 N

HN(CH 3 ) 2

+

HN(CH 8 ),

HN(CH 3 ) 2

-f

It is very possible that the free radical obtained could not be further
oxidized to the quinoid form at the low pH used in the investigation.
All the nitrogens were in the ammonium ion form, and thus the lower
oxidation state was stabilized. However, as soon as the/>H was raised,
the free radical oxidized readily in the presence of air to the quinoid
form. This indicates that under less acidic conditions the unshared
pair of electrons from one of the nitrogen atoms was made available for
contribution to the formation of a quinoid form, thus permitting a higher
state of oxidation.

The toluenesulphonic acids are readily oxidized anodically at a lead
dioxide electrode, to yield a sulphobenzoic acid. Thus ^-toluene-
sulphonic acid, when subjected to electrolysis in a 20 per cent sulphuric
acid medium at 50 to 60 C., results in a 30 per cent yield of p-sulpho-
benzoic acid, plus quantities of other degradation products (90) :

SO.H

HOOC

S0 3 H

Anodic oxidation of o-toluene sulphonic acid (I) is claimed to give a
per-acid intermediate (II), which then is further oxidized to a hydroxy-
lated carboxylic-sulphonic acid, 3,4-dihydroxy-6-sulphobenzoic acid
(III), and finally to mesaconic acid (IV) (91):

CH 3

CH 3

((A-SO.H

II

COOH CH 3 C COOH

HOOC C H
III IV

137

ORGANIC ELECTRODE PROCESSES

In connection with the oxidation of toluene sulphonic acids it might
be of interest to briefly discuss the anodic oxidation of o-toluene-
sulphonamide to saccharin. In this process the sulphonamide is
oxidized to the o-carboxyl-sulphonamide, which then loses a mole of
water to form the saccharin :

A-CH, _

l^j S0 2 NH 2

The first reported process (92) utilized a 10 per cent solution of sodium
o-toluene-sulphonamide in a 2 per cent solution of sodium hydroxide
at current densities of 0-5 to 0-9 amp./cm. 2 . As other conditions were
not specified, the validity of this method was questionable. Subsequent
investigation, however, demonstrated that oxidation of o-toluene-
sulphonamide could readily be accomplished using a platinum anode
in a 2N sodium carbonate medium at a temperature of 60 C. (93).
In this process it was found unnecessary to use a membrane to separate
the anolyte from catholyte if a lead cathode were used. This procedure
was reported to result in a 40 per cent yield of saccharin. Addition of
an oxygen carrier resulted in an increase to a 75 per cent yield (94).
Thus with the attainment of such satisfactory yields the preparation of
saccharin by an electrolytic process has become a practical procedure.

REFERENCES

(1) Jaillard, Compt. rend., 1864, 58, 203; Almeida and Deherain, ibid., 1865,
59, 214; Habermann, Monatsh., 1886, 7, 259; Bacquerel, Compt. rend.,
1875, 81, 1002; Dony-Henault, Z. Elektrochem., 1900, 6, 533; Elbs and
Brunner, ibid., 1900, 6, 604; Elbs and Foerster, ibid., 1900, 7, 341;
Moser, Elektrolytischen Prozesse, 1910, p. 66; Koidzumi, Mem. Coll. Sci.
Kyoto, 1922, 5, 359; 1928, ii, 391.

(2) Moser, op. cit., p. 65

(3) D.R.P., 274032, 1911 ; Askenasy, Leiser and Griinstein, Z. Elektrochem.,
1909, 15, 846; Trans. Amer. electrochem. Soc., 1924, 45, 129.

(4) Elbs and Brunner, loc. cit., ref. i.

(5) Miller and Hochstetter, Z. Elektrochem., 1914, 20, 367 ; Miiller and
Ruis y Miro, ibid., 1921, 27, 54; Miiller, ibid., 1921, 27, 563; Tanaka,
ibid., 1929, 35, 38; Koidzumi, Mem. Coll. Sci. Kyoto, 1928, n, 391.

(6) Miiller, Z. Elektrochem., 1922, 28, 101.

(7) Elbs and Brummer, loc cit., ref. i .

'3*

ANODIC OXIDATION

(8) Elbs and Brunner, loc. cit., ref. i; Koizumi,^, chem Soc. Japan, 1921,
42, 928; Jcrschow and Pjatnitzkaja, Neuheiten Techn., 1940, 9, 29.

(9) D.R.P., 166357, 1902.

(10) U.S.P., 875062, 1907; P.P., 387539, 1907; D.R.P., 217555, 1908;

297019, 1915.
(n) D.R.P., 118607, 1900; 128855, 1901.

(12) Renard, Ann. Chimie., 1879, 17, 313; Neuberg, Biochem. Z., 1909, 17,
270 ; Lob and Pulvermacher, ibid., 1909, 17, 343 ; Muller, Z. Elektrochem.,
1921, 27, 565; Tommila, Ann. Acad. Sci. Fenn., 1934, 39, No. n.

(13) Law, y. Chem. Soc., 1905, 87, 198; Heimrod and Levene, Ber., 1908, 41,

4443-

(14) Heimrod and Levene, loc. cit., ref. 13.

(15) Law, loc. dt., ref. 13.

(16) Muller and Hochstetter, loc. cit., ref. 5; Muller, Z. Elektrochem., 1923,
29, 264; Fresno, Anal. Fis. Quim., 1924, 22, 121.

(17) Fry and Payne, J. Amer. chem. Soc. y 1931, 53 t 1975.

(18) Elbs and Brunner, Z. Elektrochem., 1900, 6, 609.

(19) Brandt and Opp, Trans. Amer. electrochem. Soc., 1931, 59, 237.

(20) Shipley and Rogers, Canad.J. Res., 1939, 17, (B), 147.

(21) Lob, Biochem. Z., 1909, 17, 132; Z. Elektrochem., 1910, 16, i.

(22) Steiger, Helv. Chim. Acta, 1936, 19, 191; Fink and Summers, Trans.
Amer. electrochem. Soc., 1938, 74, 638.

(23) Kiliani, Ber. y 1932, 65, 1269; 1933, 66, 118; Bernhauer and Irrgang,
Biochem. Z. y 1932, 249, 216 ; Isbell and Frush, J. Res. nat. Bur. Standards,
1935, *4t 359; Fink and Summers, loc. cit., ref. 22; Schmidt and Simon,
J.prakt. Chem., 1939, 152, 196.

(24) Schall and Melzer, Z. Elektrochem., 1922, 28, 474; Schall and Markgraf,
Trans. Amer. electrochem. Soc., 1924, 45, 161 ; Schall and Thieme, Z.
Elektrochem., 1927, 33, 571; Schreiner, ibid., 1930, 36, 953; Kravtzoff,
Compt. rend., 1936, 202, 1036; Baur, Helv. Chim. Acta, 1939, 22, 1120.

(25) Kolbe, Annalen, 1860, 113, 244; Miller and Hofer, Ber., 1894, 2 7> 467;
Smull and Subkow, Chem. Mettalurg. Engng., 1923, 28, 357 ; Carpeniseanu,
Compt. rend., 1934, 148, 460.

(26) Bourgoin, Bull. Soc. Mm., 1868, 9, (2), 427; Miller and Hofer, Ber.,
1894, 27, 470; Tommila, Ann. Acad. Sci. Fenn., 1932, 36, (A), i.

(27) Yokoyama, Bull. chem. Soc., Japan, 1933, 8> 7 1 -

(28) Fichter and Schetty, Helv. Chim. Acta, 1937, 20, 1304.

(29) Fichter, Z. Elektrochem., 1912, 18, 651; Fichter, Stutz and Grieshaber,
Verh. Naturf. Ges. Basel, 1912, 23, 249.

(30) Fichter and Lindenmaier, Helv. Chim. Acta, 1929, 12, 570.

(31) Fichter and Wenk, Ber., 1912, 45, 1374.

(32) Fichter and Wenk, loc. cit., ref. 31; Fichter and Braun, Ber., 1914, 47,
1526; Shipley and Rogers, Canad.J. Res., 1939, 17, (B), 147.

(33) Fichter and Braun, Ber., 1914, 47, 1529.

(34) B.P., 253877, 1926.

(35) Baumgarten and Damman, Ber., 1933, 66, 1637.

(36) Yokoyama and Yamamoto, Bull. chem. Soc., Japan, 1932, 7, 28.

(37) Fisher and Neundlinger, Ber., 1913, 46, 2544; Neundlinger and Chur,
y. prakt. Chem., 1914, 89, 466; Fisher and Chur, ibid., 1916, 93, 363.

(38) Takayama, Bull. chem. Soc., Japan, 1933, 8, 137.

(39) Yokoyama, Bull. chem. Soc., Japan, 1932, 7, 103; Fichter and Stenzl,
Helv. Chim. Acta, 1936, 19, 1174.

139

ORGANIC ELECTRODE PROCESSES

(40) Renard, Compt. rend., 1880, 91, 125.

(41) Fichter, Trans. Amer. electrochem. Soc., 1924, 45, 75 ; J. Soc. chem. Ind.,
1929, 48, 325T.

(42) Fichter and Stocker, Ber., 1914, 47, 2012; Inoue and Shikata, J. Chem.
Ind. Japan, 1921, 24, 567; Ghosh, Bhattacharyya, Muthanna and Mitra,
J. sci. industr. Res., 1952, nB 356.

(43) Archibald, Trans. Roy. Soc. Canada, 1932, 26, III, 69.

(44) D.R.P., 117251, 1899; Inoue and Shikata, loc. cit., ref. 42.

(45) D.R.P., 117251, 1899; Seyewetz and Miodon, Bull. Soc. chim., 1923, 33,
449; See also Shipley and Rogers, Canad.J. Res., 1939, 17, (B), 147.

(46) Kempf,^. prakt. Chem., 1911, 83, 329.

(47) Fichter, loc. cit., ref. 41. '

(48) Renard, Compt. rend., 1880, 91, 175; James, jf. Amer chem. Soc., 1899,
21, 890; Merzbacher and Smith, ibid., 1900, 22, 723; Puls, Chem. Ztg.,
1901, 25, 263; Law and Perkin, Trans. Faraday Soc., 1904, i, 31 ;
Chem. News, 1905, 91, 54; 1905, 92, 69; Fichter, Z. Elektrochem., 1913,
19, 781 ; Fichter and Stocker, Ber., 1914, 47, 2007; Fichter and Acker-
mann, Helv. Chim. Acta, 1919, 2, 595.

(49) Kordzumi, Mem. Coll. Sci. Kyoto, 1928, nA, 383.

(50) Mann and Paulson, Trans. Amer. electrochem. Soc., 1925, 47, 31.

(51) Mitchell, Trans. Amer. electrochem. Soc., 1929, 55, 495.

(52) McKee and Heard, Trans. Amer. electrochem. Soc., 1934, 65, 328.

(53) Fichter, et al., loc. cit., ref. 48.

(54) Yokoyama and Ishikawa, Bull. chem. Soc., Japan, 1931, 6, 275.

(55) Fichter and Adler, Helv. Chim Acta, 1926, 9, 279.

(56) Fichter and Rosenzweig, Helv. Chim. Acta, 1933, 16, 1155.

(57) Dunbrook and Lowy, Trans. Amer. electrochem. Soc., 1924, 45, 89.

(58) Elbs, Z. Elektrochem., 1896, 2, 522; Labhardt and Zschoche, ibid., 1902,
8 > 93J Coehn, ibid., 1903, 9, 643; Kaufler, ibid., 1907, 13, 636; Fichter
and Bonhote, Helv. Chim. Acta, 1920, 3, 395; Dunbrook and Lowy,
loc. cit., ref. 57.

(59) Law and Perkin, Chem. News., 1905, 92, 67; Ono, Helv. Chim. Acta,
1927, 10,45.

(60) Sachs and Kempf, Ber., 1902, 35, 2711.

(61) Brown and Brown, Trans. Amer. electrochem. Soc., 1939, 75, 393.

(62) Law and Perkin, Trans. Faraday Soc., 1904, i, 31 ; J. chem. Soc., 1907,
91, 258; Fichter and Stocker, Ber., 1914, 47, 2009; Fichter and Grisard,
Helv. Chim. Acta, 1921, 4, 935; Fichter and Meyer, ibid., 1925, 8, 74;
Fichter and Rinderspacher, ibid., 1926, 9, 1097; Shipley and Rogers,
loc. cit., ref. 45.

(63) Bottens, Z. Elektrochem., 1902, 8, 673; Ono, J. chem. Soc. Japan, 1921,
4 2 > 38, 559 J Schetty, Diss. Basel, 1925; White and Lowy, Trans. Amer.
electrochem. Soc., 1932, 62, 223.

(64) D.R.P., 152063, 1902.

(65) Fichter and Hergzbein, Helv. Chim. Acta, 1928, u, 1264.

(66) U.S.P., 1397562, 1920; Thatcher, Trans. Amer. electrochem. Soc., 1927,
5i> 175-

(67) I$e Blanc, Z. Elektrochem., 1900, 7, 290; Fontana and Perkin, ibid.,
1904, n, 99; Rasch and Lowy, Trans. Amer. electrochem. Soc., 1929,
56, 477; Drosdow and Drosdow, J. Appl. Chem. (U.S.S.R.), 1933, 6 >
897-

(68) D.R.P., 152063, 1902.

140

ANODIC OXIDATION

(69) Fichter and Stocker, Ber., 1914, 47, 2014; Fichter and Brunner, Bull.
Soc. chim., 1916, 19, 281.

(70) Shields and Coull, Trans. Amer. electrochem. Soc., 1941, 80, Preprint 7.

(71) Fichter and Stocker, Ber., 1914, 47> 2017; Fichter and Ackermann, Helv.
Chim. Acta., 1919, 2, 595; Yokoyama and Ishikawa, Bull. chem. Soc.
Japan, 1931, 6, 275.

(72) Barth, Annalen, 1870, 154, 360; B.P., 103739, 1916; F.P., 487595, 1918.

(73) Koidzumi, Mem. Coll. Set. Kyoto, 1928, n (A), 383.

(74) McKee and Heard, Jr., Trans. Amer. electrochem. Soc., 1934, 65, 316.

(75) McKee and Heard, Jr., Trans. Amer. electrochem. Soc., 1934, 65, 301.

(76) Law, y. chem. Soc., 1906, 89, 1440.

(77) James, y. Amer. chem. Soc., 1899, 21, 893.

(78) D.R.P., 74353, 1892; Perkin, Diss. Berlin, 1899; Rasch and Lowy,
Trans. Amer. electrochem. Soc., 1932, 62, 167.

(79) Lob, Z. Elektrochem., 1895/96, 2, 663; 1896/97, 3, 3; Fichter and Uhl,
Helv. Chim. Acta, 1920, 3, 22.

(80) Fichter and Grisard, Helv. Chim. Acta, 1921, 4, 937.

(81) Zschoche, Z. Elektrochem., 1902, 8, 93 ; Fichter and Grisard, Helv. Chim.
Acta, 1921, 4 t 934; Allmand and Puttick, Trans. Faraday Soc., 1927,
23, 641 ; Fichter and Simon, Helv. Chim. Acta, 1934, J 7 7 J 8-

(82) Fichter and Grisard, Helv. Chim. Acta, 1921, 4, 929.

(83) Christeller, Diss., Basel, 1900.

(84) Fichter and Muller, Helv. Chim. Acta, 1925, 8, 290.

(85) Lowy and Haux, Trans. Amer. electrochem. Soc., 1921, 40, 123.

(86) Franke, Z. ges. Textilind., 1932, 35, 193.

(87) White and Lowy, Trans. Amer. electrochem. Soc., 1932, 6i t 305.

(88) Michaelis, y. biol. Chem., 1932, 96, 703; Michaelis and Schubert, Chem.
Rev., 1938, 38, 437.

(89) Allen and Powell, Trans. Faraday Soc. y 1954, 50, 1244.

(90) Sebor, Z. Elektrochem., 1903, 9, 370.

(91) Fichter and Stocker, Helv. Chim. Acta, 1924, 7, 1076.

(92) D.R.P., 85491, 1895-

(93) Pamfilov, Bull. Inst. Polyt. Ivanovo-Vozesensk, 1921, 4, 167.

(94) B.P., 9322, 1903; Swiss P., 78277, 1918; Fichter and Lowe, Helv.
Chim. Acta, 1922, 5, 60; Swiss P., 94223, 1921 ; Halla, Z. Elektrochem.,
1930, 3 6 96; Holl. P., 41338, 1935-

141

CHAPTER VIII

ANODIC SUBSTITUTION

ANOTHER phase of electro-organic chemistry that has been studied,
but not as extensively as electrolytic oxidations and reductions,
is the process of anodic substitution. However, there are certain
fundamental factors which are understood to be intimately related to
these processes. Firstly, in anodic substitution there is no net change
in electrons in the substitution product. Secondly, the hydrogen atom
in the molecule is replaced by another atom or group, such as a halogen
atom or nitro group, which has been freed from the solution by the
electrolytic process. There are generally considered to be three processes
by which such a substitution may occur. The first may be due to
oxidation of a substance in solution to yield an active agent which attacks
the organic compound. This type of process would be essentially
electro-chemical in nature, and may be affected by the anode potential,
current density, and electrode material. In this instance, too, the
compound should behave as a typical depolarizer, by lowering the anode
potential, and the products obtained may differ from those obtained by
ordinary chemical methods. The second process would be analogous
to a chemical reaction, with the current merely functioning as a release
mechanism for the substituting agent from a solution of one of its
compounds. Thus, for example, the liberation of bromine from a
solution of a bromide salt is only dependent on the passage of current
as a generating agent, and generally its ability to act as a substituting
agent is independent of the anode potential and the electrode material.
The net result in this instance would be essentially the same as if
bromine gas were bubbled through a solution of the organic compound.
A thiid process associated with the substitution reaction may be related
predominantly to the concentration of an active specie at the anode
surface, which enables substitution to occur where, under ordinary
conditions of concentration, such a reaction would not result. To

142

ANODIC SUBSTITUTION

illustrate this point we can consider a suspension of naphthalene in a
dilute aqueous solution of nitric acid. Under ordinary circumstances
no noticeable reaction results. However, with the passage of an electric
current, nitre-naphthalene is obtained at the anode. This end result
can be attributed not to the discharge of the nitrate ions at the electrode,
but to the large concentration of these ions at the electrode surface.
In this instance the factors generally associated with an anodic oxidation
or a cathodic reduction, i.e. electrode potential and electrode material,
would be of minor consequence as compared to the current density and
the diffusion coefficient of the organic compound subjected to anodic
substitution.

(i) ANODIC HALOGENATION

It might be desirable, before discussing the reactions themselves,
to briefly outline some aspects of anodic halogenation. If a halogen
solution is subjected to electrolysis at an electrode such as platinum
containing a trace of iridium or rhodium to render the electrode inert
to the corrosive action of free halogen, the liberation of halogen will
occur generally at the decomposition potential of the halogen in solution.
Thus, the liberation of iodine, bromine, and chlorine should occur at a
potential of about 0-52, 0-94, and 1-31 volts respectively. However, if
there is present in solution a depolarizer which is capable of reacting
with the halogen being liberated, the potential for halogen liberation
will be somewhat lower. With the proper regulation of the anode
potential, there is, in most instances, little danger of obtaining a mixed
reaction. Although several values have been obtained for the decom-
position potential of water, a steady evolution of oxygen and hydrogen is
only obtained at about 1-7 volts. At this potential the hydroxyl ions are
discharged at the anode with the formation of water and the evolution
of gaseous oxygen. However, there is always the possibility that the
substance being subjected to anodic substitution may also be a satis-
factory oxygen depolarizer, with the final result a product which has
undergone both anodic oxidation and substitution. If the anodic
substitution is performed in a basic medium there is a certain degree
of polarization which leads to oxygen evolution, with the result that
the product obtained is a consequence of dual action. This is especially
so in instances in which the organic depolarizer is not present in
sufficient quantity to react immediately with the liberated halogen. In
this situation the halogen reacts with the hydroxyl ion to form a hypo-

ORGANIC ELECTRODE PROCESSES

halogen acid (HOX) which can act both as a halogenating agent and an
oxidizing agent. This is especially true in a concentrated basic medium
where the oxygen potential is quite low, and therefore the halogen
liberated readily forms hypohalous ions. An example of the different
action in acid and alkaline solution is found in the electrolysis of
solutions of acetone. In a dilute hydrochloric acid medium, mono-
chloroacetone is obtained.

CH 3 COCH 3 + 2 Cl > CH 3 COCH 2 C1 + HC1

In an alkaline medium ,the reaction does not stop at the monohalogen
substitution product, but continues on to form a trihalogen derivative,
thus:

CH 3 COCH 3 + 3 NaOCl > CH 3 COCC1 3 + 3 NaOH

CH 3 COC1 3 + NaOH > CHC1 3 + CH 3 COONa

Therefore, before attempting an anodic halogenation, considerable
thought should be given to the experimental condition prior to onset
of an investigation.

(a) Chlorination

The addition of chlorine to organic depolarizers has been the subject
of detailed investigation by a number of workers (i). As a result of their
experimental findings it is possible to elucidate three different mechan-
isms which may serve to explain the process of anodic chlorination
which results from the conversion of chlorine ions to atomic chlorine.

The first mechanism proposed is that the liberated chlorine reacts
directly with the organic compound in the following manner :

RH + 2 Cl -* RC1 + HC1

In the second instance, if a carrier such as iodine were added to the
electrolysis medium the reaction would proceed in the following way :

I 2 4- 2 Cl > 2 IC1

RH + 2 IC1 > RC1 + HC1 + I 2

Lastly, the chlorine atom might combine with another of its kind to
form gaseous chlorine, which would then react to yield the halogenated
compound :

2C1 > C1 2

^ RH + C1 2 > RC1 + HC1

It may well be that in the first two proposed mechanisms the anode
potential will be lowered. However, in the last mentioned instance, the

144

ANODIC SUBSTITUTION

addition of the organic substance should have no effect on the anode
potential. Therefore, this process will be essentially comparable to a
chemical addition of chlorine gas to a solution of the organic compound.
Unfortunately, though we may theorize in subsequent discussions, there
are to date no clear cut examples which would definitely prove an anodic
chlorination to be purely electrochemical in nature.

The first record of the introduction of a halogen in place of hydrogen
was reported as far back as 1830 by Liidersdorf (2), who electrolysed a
mixture of ethyl alcohol and concentrated hydrochloric acid, with the
resultant separation of a heavy oily layer possessing a sweet ethereal
odour. This product was probably chloral, whose commercial prepara-
tion from ethanol has been reported (3), using a carbon anode and an
aqueous hydrochloric acid or potassium chloride medium at a tempera-
ture of 1 00 C. The chloral distills off, thus preventing its further
oxidation. More recently the preparation of chloral with a 61 per cent
efficiency was reported (4), using aqueous potassium chloride as the
anolyte, a carbon anode, a copper cathode, a current density of 0-08
amp./dm. 2 , and a temperature of 70 to 80 C. A subsequent group of
investigators (5) found they were unable to obtain the efficiency claimed
by the previous investigator (4) and, therefore, conducted a more system-
matic study on the preparation of chloral using a porous carbon tube
as an anode, a copper cathode and various chloride media. They pointed
out that the reaction involves the simultaneous anodic oxidation of
alcohol and substitution of chlorine, thus : \F?S

^
CH 3 CH 2 OH + 8 Cl > CC1 3 CHO + 5 HC1

These, as well as the previously mentioned investigators, also reported
the complication of the over-all reaction process by the possible side
reactions which yielded chloral alcoholate, monochloroacetic acid, ethyl
acetate, monochloroacetaldehyde hydrate and its alcoholate, and
chloroether. The last three named compounds were considered to be
intermediates in the production of chloral. According to their findings
they concluded that a viscous solution of calcium chloride, plus a cata-
lytic quantity of cyanuric acid at reflux temperature gave best results.
An excessive amount of water in the medium caused a decrease in yield.
Additional findings showed that platinum was a more effective anode
than carbon, with the maximum current efficiency obtained being 61
per cent. They claimed from the results obtained that the anode
potential is the main controlling factor in the process. However, the

O.E.P. 145 L

ORGANIC ELECTRODE PROCESSES

experimental evidence presented does not quite substantiate this claim.
Another interesting chlorine derivative of alcohol is ethylene chloro-
hydrin, which is obtained by passing ethylene into the anode compart-
ment of an electrolysis cell in which chlorine is being liberated, as a
result of the electrolysis of either an aqueous hydrochloric acid or sodium
chloride solution. The reaction is performed at a temperature of 100
C., with improved yields being obtained by the addition of catalytic
amounts of either charcoal or platinum black (6). Recently a group of
investigators (7) conducted detailed studies on this particular reaction,
using porous carbon tubes as anodes, and a sodium chloride solution
as the electrolyte. They report that in this reaction both ethylene
chlorohydrin and ethlylene glycol are produced, the yields depending on
the experimental conditions. The over-all reaction involves essentially
a simultaneous oxidation and chlorine addition:

CH 2 =CH 2 + OH- + Cl- > CH 2 (OH)CH 2 C1 + 20

The conditions resulting in the optimum yield of ethylene chlorohydrin
(84 per cent) involved the use of a 10 per cent solution of sodium
chloride, a flow rate of 5 -55 litres of ethylene/hour/dm. 2 of anode surface,
and a current density of 2-25 amp./dm. 2 at a temperature of i C. At a
temperature of 90 C. the recorded yield dropped to i per cent. At
this temperature the quantity of glycol obtained amounted to 17 per
cent, whereas at i C. a 5 per cent yield of glycol was obtained. The
evidence presented indicates that a higher electrode potential is necessary
in order to obtain the optimum yield of ethylene chlorohydrin. Lower
electrode potentials favour the formation of ethylene glycol. In this
instance it appears that ethylene is a better depolarizer for oxygen than
for chlorine ; thus the reason for oxidation of ethylene to its respective
glycol at the lower potential.

The early methods for the preparation of chloroform which involved
the electrolysis of a dilute alkaline chloride solution resulted mainly in
the oxidation of the alcohol, with the chloroform obtained being a
minor side product (8). These results are not unexpected in view of
the considerably lower oxygen liberation potential in an alkaline medium
as compared to that in an acidic medium. Some improvement in yield
was obtained using an aqueous calcium chloride medium at a current
dtasity of 0-08 amp./cm. 2 , and a temperature of 50 to 70 C. (9). In
this manner the alkalinity of the medium was controlled as a result of the
precipitation as calcium hydroxide. However, there is disadvantage

146

ANODIC SUBSTITUTION

here, too, in the fact that the calcium hydroxide tends to occlude the
passage of current by plating on the electrode. One investigator (10)
conducted quite an extensive survey of the conditions necessary to
obtain chloroform by electrolysis. Utilizing a platinum anode, a current
density of approximately I amp. /cm. 2 , and Roush neutralization
cathode (i i), and an electrolysis medium containing a 20 per cent solution
of potassium chloride, yielded chloroform with current efficiencies from
65 to 75 per cent. This method was satisfactory for the conversion of
both ethanol and acetone. However, when ethanol was being converted
it was possible to utilize temperatures of approximately 30 C., whereas
with acetone, a temperature of 15 C. was found more desirable. Feyer
further found that a current efficiency as high as 95 per cent from
ethanol or acetone is possible, if a nickel or copper electrode, which is
first etched by nitric acid, is used as the cathode in an aqueous calcium
chloride medium. The calcium hydroxide which formed on this type
of cathode did not cause occlusion of the electrode surface, with the
result that the resistance to the passage of current remained low. We
may look upon the general reaction in the following manner. The
initial step can be considered electrochemical, thus :

Cl- + OH~ > HOC1 + 2e

The hypochlorite formed reacts chemically with the ethanol or the
acetone to yield the desired compound in the following manner:

CH 3 CH 2 OH + 5 HOC1 > CHC1 3 + 4 H 2 O + 2 HC1 + CO 2

CH 3 COCH 3 + 3 HOC1 > CHC1 3 + 2 H 2 O + CH 3 COOH

A number of other modifications in the electrolyte have been reported
(12), but without any advantage over that discussed previously. Propyl
alcohol, as well as methyl ethyl ketone, when subjected to electrolysis
in a calcium or potassium chloride medium, reportedly also yields
chloroform (13). Electrolysis of acetone in a concentrated hydrochloric
acid medium, utilizing a platinum anode and a membrane to separate
the anode and cathode compartments, results in satisfactory yields of
chloroacetone (14). Propionic acid under similar conditions will yield
predominantly j8-chloropropionic acid, with a small quantity of <x-
chloropropionic acid (15).

In studies on the anodic halogenation of aromatic compounds it was
thought that it might be much easier to distinguish the electrochemical
halogenation process from the purely chemical one. However, the

147 L 2

ORGANIC ELECTRODE PROCESSES

results are such that here, too, there is available no clear picture from
which one can form a definite opinion concerning the process involved.
When dealing with a suspension of the organic compound to be sub-
stituted, there is very little likelihood of a true electrochemical process.
The anodic chlorination of benzene was first attempted on a suspension
in concentrated hydrochloric acid, from which was obtained a small
quantity of an oily chlorinated product, with the balance of the material
being unchanged benzene ( 1 6). This investigator found that the addition
of iodine did not improve the experimental results. However, sub-
sequent research (17) showed, for example, that when benzene was
electrolysed as a suspension in aqueous hydrochloric acid, especially
in the presence of a carrier such as iodine, chlorobenzene was obtained
with efficiencies as high as 75 per cent, A group of Indian investigators
(18) found they were able to obtain current efficiencies as high as 89
per cent utilizing a porous carbon anode, a medium consisting of
benzene in a hydrochloric acid-monochloroacetic acid solution, a current
density of 4-3 amp./dm. 2 , a temperature of 38 C., and a catalytic amount
of cyanuric acid. These results cannot, however, be attributed to an
electrochemical phenomenon, as it is possible to obtain almost as
satisfactory yields by direct chemical chlorination of benzene. If one
were to obtain an appreciable amount of a further chloro-substituted
benzene, then this might be considered an electrochemical process.
Van Name and Maryott (19) made a careful study of the effects of anodic
electrolysis of benzene in an acetic acid-lithium chloride medium, and
in an ether-zinc chloride medium, using graphite as electrodes. The
results obtained were very discordant, which made impossible any
conclusion as to whether the process which yielded various chloro-
benzenes was strictly anodic. A subsequent investigation on the anodic
substitution of benzene by another group of workers (20) claimed that
the contradictory results obtained by the previous investigators could
be directly attributed to the non-homogeneity of the benzene in the
electrolyte. The results obtained from electrolysis of benzene in a
glacial acetic-hydrochloric acid medium left no doubt in these authors*
minds that the processes which yielded the different chloro-substituted
products under varying current density conditions were electro-chemical
in nature. Utilizing suitable conditions it was possible to achieve
maximum chlorination, although the products obtained were always
somewhat contaminated by quantities of oxidation products. In addition
to chloro-, />-dichloro-, $yw-tetrachloro-, and hexachlorobenzene, such

148

ANODIC SUBSTITUTION

oxidation products as pentachlorophenol and chloranil were obtained,
according to the following scheme :

OH
Cl/^Cl

OH\

\

Cl
Cl

C V C1

Cl

O

Cl
Cl

It was found that platinum was the most suitable electrode material for
the chlorination. With magnetite and carbon anodes, lower efficiencies
and considerable oxidation of the depolarizer were obtained; at the
platinum anode, the higher current densities resulted in excellent
conversion to the hexachlorobenzene. If the current density was
maintained below 0-26 amp./cm. 2 , it was not possible to demonstrate
the presence of the highly chlorinated product. In addition to the
necessity for high current densities, it was also desirable to perform
the electrolysis at elevated temperatures and a low concentration of
benzene, to obtain the maximum conversion to the hexachlorinated
product.

The anodic electrolysis of a hydrochloric acid suspension of toluene
at a platinum electrode and room temperature will yield a mixture of
approximately 70 per cent o-chlorobenzene, and 30 per cent />-chloro-
benzene (21). If this reaction is performed at reflux temperature and
at a carbon anode, the major product is />-chlorobenzene, with only a
small quantity of the o-chloro-substitution product (22). The results
obtained in these electrolyses are indicative of an electrochemical
process, for, if the substitution were performed utilizing gaseous chlorine

149

ORGANIC ELECTRODE PROCESSES

under almost identical conditions, the main product obtained, benzyl
chloride, would be a consequence of substitution on the methyl group.
The anodic substitution of toluene by chlorine was also investigated
in an acetic-hydrochloric acid medium (23), with the results obtained
being represented by the following scheme :

CH

CHClo

CH ;

OH

This process also appears to be a true electrochemical one, at least
partly, for the methyl group is not substituted with the halogen until
the 2,4,5-trichlorotoluene is formed.

The anodic substitution, by chlorine, of such compounds as cyclo-
hexanone, benzophenone, acetophenone, and propiophenone in an
acetic-hydrochloric acid medium has been investigated (24), with the
resultant formation in each instance of a monochloro-substituted
product. Cyclohexanone and benzophenone yielded a monochloro-
cyclohexanone and a monochlorobenzophenone respectively, with the
position of the chlorine not being established. The acetophenone and
propiophenone substitutions occurred on the aliphatic portion of the
molecule, the former compound yielding o>-chloroacetophenone and
the latter a mixture of a- and j8-chloropropiophenones.

The substitution of chlorine in aromatic amines is generally accom-

150

ANODIC SUBSTITUTION

panied by some oxidation. Initial studies on the electrolysis of aniline,
under conditions which were not too carefully controlled, resulted in
the isolation from the hydrochloric acid reaction medium of aniline
black or chlorinated aniline and chlorinated quinone (25). A more
recent study (26) indicated that, if conditions were carefully controlled,
it was possible to obtain a certain degree of selectivity in the substitution
process. Thus, using a carbon anode, a concentrated hydrochloric
acid medium and low temperatures, it was possible to obtain a satis-
factory yield of the 2,4,6-trichloroaniline. At a temperature of 35
to 40 C. in a 10 per cent hydrochloric acid medium, the major product
isolated was the tetrachloroquinone. A more concentrated hydro-
chloric acid medium, i.e. 20 per cent, and a temperature of 5 to 50 C.,
resulted in the formation of 2,4,6-trichloroquinone. Thus it appears
that under conditions in which the hydroxyl ion might be considered
to be at a minimum, that is in the concentrated hydrochloric acid
medium, the anodic oxidation process is comparatively a minor one. It
has also been reported (27) that the anodic substitution of indigo (I)
by chlorine, using a concentrated hydrochloric acid medium, is accom-
panied by an oxidation process which results in the formation of chloro-
isatin (II), dichloroisatin (III), and trichloroaniline (IV):

Clf

c-o

N

I
H

II

C,Q ;

\/

ci

III IV

(b) Bromination

The early attempts to prepare bromoform by anodic substitution of
ethanol in an aqueous potassium bromide-sodium carbonate medium
resulted only in the oxidation of the alcohol to acetaldehyde and acetic

ORGANIC ELECTRODE PROCESSES

acid, and finally to carbon dioxide and water. This can be attributed
to the greater ease of oxidizability of the bromide ion to bromate, which
in turn reacts with the alcohol under these experimental conditions (28).
Utilizing a sodium bromide-calcium bromide medium results in the
formation of some bromoform, but in this instance, too, the major
process is the oxidation of the alcohol (29). The results obtained with
acetone, however, are considerably better. The electrolysis can be
performed at a platinum anode, in a cell containing a diaphragm to
separate the anode and cathode compartments. The anolyte consists
of 10 ml. acetone and 25 gm. of potassium bromide in 75 ml. of water.
During the course of the electrolysis, which is performed at current
densities of 0-02 to 0-04 amp./cm. 2 , sodium carbonate is added from time
to time to compensate for the acid formed. This process results in
almost a quantitative yield of the desired product, according to the
following equation (30) :

CH 3 COCH 3 + 6 Br- + H 2 O > CHBr 3 + 3 HBr + CH 3 COOH

It has also been reported (31) that this process can be successfully
accomplished without the use of a diaphragm, provided that the alkali
formed at the cathode is neutralized by addition of carbon dioxide to
the medium, and that the reduction of the acetone at the cathode is
prevented by addition of a substance such as chromate ion, which is
more readily reduced. Such a process involved the electrolysis of an
aqueous solution containing I4oml. acetone, 385 gm. potassium bromide,
and 2*i gm. of potassium chromate per litre. The current efficiency
in this instance is approximately 90 per cent at a current density of
O'oy amp./cm. 2 .

Unfortunately little work has been done on the anodic bromination
of the aromatic nucleus. Some studies have been carried out on
the anodic substitution of benzene as a suspension in an aqueous
hydrobromic acid solution, using either a platinum or carbon anode
(32). The maximum efficiency (23 per cent) for the conversion to
bromobenzene is attained if an iron carrier is used in the process. As
this product can be obtained in the same yield by passing bromide gas
through a solution of benzene, it is highly questionable that this can be
considered a true electrochemical process. The electrolysis of toluene
in ^ hydrobromic acid medium at either a carbon or platinum electrode
results in substitution in the nucleus, to give bromotoluene. How-
ever, this result is obtained only if the electrolysis cell is kept in

ANODIC SUBSTITUTION

the dark during the course of the reaction. If the same reaction is
performed in bright light the final product is benzyl bromide (33).
Here too it is quite questionable whether or not this is a direct electro-
chemical process. A certain amount of bromination of benzene and
toluene also occurs when the molecular compounds AlBr 3 .3C 6 H 6 and
AlBr 3 .3C 7 H 8 are electrolysed (34). The products obtained as a result
of electrolysis of phenol in a hydrobromic acid medium depend to a great
extent on the concentration of acid used. In a relatively low concen-
tration medium, e.g. i N, the primary process is oxidation of the
phenol, probably due to the greater ease with which phenol acts as an
oxygen depolarizer as compared to its action on bromine. However, if a
4 N acid solution is used, a 20 per cent yield of />-bromophenol is
obtained, due to the fact that the bromine potential is lowered sufficiently
for oxidation to be somewhat inhibited (35). This investigator reports
that the phenol had a definite depolarizing action, which resulted in a
depression of the anode potential. Similar results were obtained upon
anodic chlorination of phenol in a hydrochloric acid medium. Low
concentrations of phenol in the 4 N hydrobromic acid medium results
in the formation of 2,4,6-tribromophenol as one of the reaction products

(36).

Aniline, when subjected to anodic bromination in an aqueous hydro-
bromic acid medium, yields, in addition to oxidation products, an
appreciable quantity of 2,4,6-tribromoaniline (37). The accompanying
oxidation can probably be attributed to the fact that the decomposition
potential of bromine is quite close to the oxidation potential for
aniline under the experimental conditions used. Of interest is the
reported preparation of eosin by electrolysis of fluorescein in an aqueous
potassium bromide medium (38), thus:

=O

Br

As mentioned previously, the inhibition of oxidation during the
bromination of phenol can be accomplished by utilizing a high concen-

'53

X S4

ANODIC SUBSTITUTION

tration of bromide in the medium. This factor has been utilized in the
preparation of 5,5'-dibromoindigo (II), with excellent yields, by anodic
substitution, of indigo (I), at a carbon anode in a 51 per cent hydro-
bromic acid medium at a temperature of 20 to 22 C. (39). If the
water concentration is increased the oxidative process accompanies
that of bromination, as indigo is also a good oxygen depolarizer, with the
net result being the formation of 5-bromoisatin (IV), 5,7-dibromoisatin
(V), and 2,4,6-tribromoaniline (VI). Under these conditions oxidation
of the indigo to isatin (III) probably precedes the halogenation (p. 154).
In the reaction in which indigo is subjected to anodic chlorination,
(p. 151) the potentials for oxidation and chlorination are so close that the
formation of 5,5'-dichloroindigo is almost impossible. It has also been
reported that satisfactory yield of the dibromoindigo can be obtained by
utilizing a nonaqueous medium consisting of pyridine dihydrobromide
in nitrobenzene (40). As dibromoindigo can be obtained by chemical
means, this initial bromination cannot be looked upon as a primary
electrochemical reaction. In fact the only possible electrochemical
process which may be considered a primary one in this instance is not
that of halogenation, but of oxidation.

(c) lodination

Probably the most outstanding example of the application of anodic
iodination is in the preparation of iodoform. This compound can be
prepared chemically by the action of iodine on an aqueous solution of
alcohol or acetone, in the presence of a base, according to the following
equation :

CH 3 CH 2 OH + 5 I 2 + H 2 O > CHI 3 + CO 2 -f 7 HI

From this equation it can be seen that 70 per cent of the iodine is not
used in the process. This iodine must be recovered by use of a chlorine
or hypochlorite solution which oxidizes it again to the free iodine required
in the process. Thus the practicality of this process is questionable.
In the electrolytic process the medium consists of an alkaline solution
of the alcohol containing potassium iodide. The passage of current
serves merely to liberate the iodine ions at the anode, as potential
measurements indicate that the alcohol is not a depolarizer in these
circumstances (41). The net anodic reaction can be assumed to be the
following :

CH 8 CH 2 OH + xo I + H 2 O > CHI 8 + CO 2 + 7 HI + xoc

ORGANIC ELECTRODE PROCESSES
and the cathodic reaction :

10 K+ -1- ioH 2 O + ice > ioKOH+5H 2

10 KOH + 5 H 2 O + 5 CO 2 > 5 K 2 CO 8 + 10 H 2 O

The potassium iodide, which forms as a result of neutralization of the
hydriodic acid by the potassium hydroxide, is again electrolysed to free
iodide ions, thus making the process somewhat continuous, until all the
iodide is exhausted. The electrolytic preparation of iodoform has been
the subject of some very comprehensive investigations dating back to
patents issued to the Schering Co. in 1884 (42). As a result of sub-
sequent investigations (43) the factor which has been found most critical
in the process is the pH of the medium. It has been found that the
optimum condition for the formation of the iodoform is the presence of
a quantity of hypoiodite ion, which of necessity requires slightly alkaline
conditions (44). Conditions which result in too high a hydroxyl-ion
concentration cause decomposition of the hypoiodite ion to the iodate,
with resultant decreased efficiency for iodoform production. The
optimum degree of alkalinity is usually achieved by use of a potassium
carbonate medium. In addition to this factor it has been found that
the efficiency of the process is enhanced by use of elevated temperatures,
high concentrations of alcohol and iodide, and low concentrations of
carbonate and current density (45). Utilizing the conditions in which
the electrolysis is performed at a platinum anode at a current density
of o-oi to 0-02 amp./cm. 2 , on a solution containing, per 100 ml. of water,
10 gm. of ethanol, 5 gm. of sodium carbonate, and 17 gm. of potassium
iodide at a temperature of 60 to 70 C., results in efficiencies as high as
98 per cent (46). The process can be made almost continuous by
adding further quantities of ethanol and potassium iodide, and removing
the relatively pure iodoform, which precipitates as the electrolysis
proceeds.

It might be of interest to briefly discuss the reactions which may
occur at the anode, as a result of this electrolytic process. As indicated
earlier, ten iodide ions take part in the reaction with the ethanol
to form the iodoform. As a certain amount of hypoiodite is required
for the process we may consider the initial reactions at the anode to be
th^ following :

4 I~ + 2 H 2 O > 2 IO- + 2 I- + 4 H. + 4 e

61- > 3l2 + 6e

156

ANODIC SUBSTITUTION

This electrochemical process is followed by a reaction of the ethanol
with the electrolysis products, thus :

CH 3 CH 2 OH + 2 ICT + 3 I 2 > CHI 3 + H 2 O + 5 I" + 3 H+ 4- CO 2

with the overall reaction being :

CH 8 CH 2 OH + 3 I" + H 2 > CHI 3 + 7 H+ + CO 2 + 10 e

The above equations show that the total reaction requires ten atoms of
iodine, with seven of these being used to form seven equivalents of acid.
Of the ten equivalents of alkali produced at the cathode, seven are used
to neutralize the acid and two are used to neutralize the carbon dioxide
formed. This leaves an excess of one equivalent of alkali, which will
increase the concentration of hydroxyl ions to such an extent as to cause
the hypoiodous acid to react to form increasing amounts of iodate, with
the consequential decrease in iodoform production efficiency. This free
alkali is removed by passage of a slow stream of carbon dioxide gas in
the vicinity of the cathode. The amount is regulated so that the alkali
is converted to the carbonate, as indicated by the presence of an amber
yellow colour in the electrolyte.

According to the Schering patents (42) for the preparation of iodo-
form from ethanol, acetone may replace the latter in the process. The
details are not given, but the reaction probably proceeds according to
the following scheme :

CHaCOCHa + 3 I" + H 2 O > CHI 8 + 3 H+ + CH 3 COOH + 6 e

The use of acetone in place of alcohol has the advantage in that it
requires only 60 per cent of the energy required for ethanol to produce
the iodoform. In actual practice however, this process has a number of
disadvantages. The control of alkalinity is more difficult, since of the
six equivalents liberated at the cathode only four are used up at the
anode. In addition this alkalinity factor is more critical with acetone,
as this substance tends to form condensation products in a basic
medium. Therefore it is absolutely essential that the reaction medium
be kept almost neutral during the electrolysis, either by the addition of
small quantities of iodine (47) or acetone (48). Another method which
has been suggested as being better is the use of the " Roush neutraliza-
tion cathode " (n), in which two cathodes are utilized. One of these
is enclosed in a porous diaphragm cup and, by utilization of proper
resistances, the amount of current to each cathode is controlled. As

157

ORGANIC ELECTRODE PROCESSES

a result the alkali liberated at the enclosed cathode does not mix with
the bulk of the solution, and that amount liberated at the other cathode
can be so regulated as to be sufficient to neutralize the acid formed at
the anode. Utilizing this procedure it is possible to obtain a current
efficiency of 95 per cent for the production of iodoform.

Attempts to iodinate benzene by electrolysis in a hydriodic acid
medium have yielded negative results (3 1). Phenol also will not respond
to iodination under these conditions. However, if the reaction is
performed in an alkaline medium, the hydrogen of the phenolic group is
replaced by iodine to yield the iodoxy- derivative (35). Thymol, when
electrolysed in an alkaline potassium iodide solution, yields the dithy-
moldiiodide which has been used as an antiseptic (49).

C 3 H 7 C 3 H 7 C 3 H 7

OH

4- 3

OI IO

+ 4 HI

Phenolphthalein, when iodinated under similar conditions, is not
attacked at the hydroxyl hydrogen, as is the case with thymol, but instead
is substituted on the ring to yield tetraiodophthaleine (50) :

I I

NaOi

NaOi

COONa

COONa

V/ V

This has also been found to be the case with fluorescein, whose product,
as a result of electrolysis, is the tetraiodofluorescein (51). It has also
been reported that z-amino pyridine can be electrolysed to the 2-amino-
5-iodopyridine in an aqueous methanolic potassium iodide medium
(52). Satisfactory yields are obtained if the temperature is maintained
between o and 5 C. Electrolysis of 8-hydroxyquinoline in an aqueous
soaium carbonate-potassium iodide medium at a platinum anode
results in a mixture of 7-iodo-(I) and 5,7-iodo-8-hydroxy quinoline
(II) (S3)-

158

ANODIC SUBSTITUTION

OH

OH

(d) Fluorination

The investigation and application of anodic fluorination has its
origin in the relatively recent experimental work of Simons and his
co-workers (54). In the electrochemical method for the preparation of
fluorocarbons, elementary fluorine is neither produced nor employed
in the preparation. The chemical methods involved the use of elemen-
tary fluorine, which was reacted directly with the hydrocarbon, or used
in the preparation of intermediate metal fluorides such as cobaltic
fluoride and silver fluoride (55).

The general technique for performance of the electrochemical
process consists of passing an electric current through a one-compart-
ment cell containing an electrically conducting solution of hydrogen
fluoride and the organic compound to be subjected to anodic fluorina-
tion. The cell is constructed of any material, such as iron and copper,
which is resistant to attacks by hydrogen fluoride. These same
conductor materials may be used as cathodes as well as nickel, which is
also used as an anode in this process. No corrosion of the electrodes
occurs if the applied voltage is kept below that which will cause the
liberation of elementary fluorine. The preferred voltage for the process
is generally between 5 and 6 volts, at a current density of approximately
0*02 amp./cm. 2 . Temperatures are kept low to prevent escape of the
hydrogen fluoride. Utilizing this technique, various types of organic
compounds such as carboxylic acids, alcohols, ethers, amines, hydro-
carbons, etc., have been successfully fluorinated. The carboxylic acids
yield saturated fluorocarbons corresponding to the hydrocarbon radical,
in addition to large quantities of the hydrogen-containing product.

Acetic acid, when subjected to anodic substitution in the hydrogen
fluoride medium, yields, in addition to carbon dioxide and hydrogen,
a mixture of carbon tetrafluoride and fluoroform in a ratio of about
two to one. Acetyl fluoride requires the addition of sodium fluoride to
the medium to overcome the solution resistance brought about by the
reaction with hydrogen fluoride to yield the poor conductor, acetyl

'59

ORGANIC ELECTRODE PROCESSES

fluoride. Using these modified conditions results in the formation of
equal quantities of fluoroform and carbon tetrafluoride. Acetone will
also yield equal quantities of fluoroform and carbon tetrafluoride upon
electrolysis in a hydrogen fluoride medium.

Propionic acid will yield C 2 F 6 H and butyric acid quantities of
C 3 F 6 H 2 and CF 7 H. "Some of the oxygen-containing organic substances
yield quantities of OF 2 , particularly if water is present in the medium.
Many nitrogen-containing starting materials will yield a quantity of
NF 3 . Under certain conditions carbon dioxide, and to a lesser extent
carbon monoxide, may b'e obtained as a result of the electrolytic process.
In some cases fluorinated resins are formed but, as these are generally
soluble in the medium, they do not interfere with the isolation of the
products. The product in greatest amount is usually that containing
the same number of carbon atoms as the starting material. On occasion,
products will be obtained which contain either a larger or lesser number
of carbon atoms.

Electrolysis of pyridine results in the formation of nitrogen tri-
fluoride and perfluoropentane, thus^

I |j + 15 HF > NF 8 4- C 6 F 12 + 10 H a

W

In addition to the major reaction products a small quantity of 2-fluoro-
pyridine was also isolated from the reaction medium. Under similar
conditions of electrolysis quinoline yielded a quantity of nitrogen
trifluoride and a compound C 9 F 18 .

Normal propyl alcohol and normal amyl alcohol were found to yield
mixtures of fluorinated products which could be fractionated by
distillation. The normal amyl alcohol was found to yield such products
as CF 4 , CF 3 H, C 2 F 6 , C 4 F 10 , and C 5 F 12 . The last two products,
perfluorobutane and perfluoropentane, were the major fluorinated
products.

Anodic fluorination of the olefins, when attempted in the anhydrous
hydrogen fluoride medium, results in addition of hydrogen fluoride to
the double bond with subsequent polymerization of the olefin, which in
many cases is quite undesirable. This polymerization can be minimized
t< a great extent by performing the reaction in a glacial acetic acid-
potassium hydrogen fluoride medium. The cell used is divided by a
diaphragm to separate the anode from cathode. In this manner the
organic fluoro-carbon is prevented from coming in contact with the

, 160

ANODIC SUBSTITUTION

cathode, where it is suspected that the polymerization reaction is
instigated. Using this method it was possible to obtain a 70 per cent
yield of i,i-diphenyl-i,2-difluoroethane from the respective stilbene
(56):

F F

H H

The mechanism of anodic fluorination is suspected to be very
similar to that of other types of anodic halogenation (57). However,
considerable work will have to be published before any final conclusions
can be arrived at concerning the general applicability of this anodic
process to the fluorination of organic compounds.

(2) ANODIC THIOCYANATION

It has been demonstrated that electrolysis of thiocyanate salts in
the presence of organic compounds in the anode chamber of a cell
generally results in satisfactory yields of the organic thiocyanate. In
this manner, many derivatives of phenol, and of primary, secondary,
and tertiary aromatic amines can be obtained. For example, phenol,
when subjected to anodic substitution at a graphite electrode at current
densities of 0-03 to 0-04 amp./cm. 2 in a sodium thiocyanate-hydro-
chloric acid medium at a temperature of o to 5 C., yields the
respective thiocyanate (58) :

OH OH

f J + NaSCN + HC1 >

SCN

Under similar conditions the thiocyano derivatives have been obtained
from carvacrol, o-, m-, and ^-cresols, thymol, and guaiacol (59). As
previously mentioned, aromatic amines also respond to anodic thio-
cyanation. Thus aniline yields 4-thiocyanoaniline, N-methylaniline
the 4-thiocyano-N-methylaniline, and N,N-dimethylaniline the N,N-
dimethyl-4-thiocyanoaniline (60). Electrolysis of o- and m-toluidine
results in substitution of the thiocyanate para to the amino group to
yield respectively 2-methyl-4thiocyanoaniline and 3-methyl-4-thio-
cyanoaniline (58). The electrochemical thiocyanation of the aromatic
amines is of great interest because these thiocyano derivatives are
readily converted to the respective aminobenzthiazoles, which have

O.E.P.

ORGANIC ELECTRODE PROCESSES

been found useful as intermediates in the synthesis of dyestuffs,
sensitizers, pharmaceuticals, and moth-proofing agents. To illus-
trate the process, we may consider the preparation of 2-amino-6-
methylbenzothiazole, which involves the intermediate thiocyanation of
/>-toluidine. The anolyte was a solution of 1-5 gm. of />-toluidine,
3-2 gm. of ammonium thiocyanate, and 2 ml. concentrated hydrochloric
acid in 35 ml. aqueous alcohol. After the passage of 2 amp. hr. of
electricity at a current density of 0-03 to 0-04 amp./cm. 2 , the electrolyte
was diluted with water. A heavy oily layer, which was essentially the
thiocyanotoluidine, formed. This was not isolated, but instead the
mixture was boiled under reflux for a period, to accomplish the rearrange-
ment to the respective benzothiazole. Utilizing these conditions, other
benzothiazoles, namely 2-amino-6-methoxy-benzothiazole, 2-amino-6-
ethoxybenzothiazole, and 2-amino-6-carboethoxy-benzothiazoles were
also prepared (61).

(3) ANODIC NITRATION

Earlier in this chapter it was pointed out that electrochemical
nitration is the result of an increase in concentration of nitrate ions at
the electrode surface, which results in the formation of the desired
nitration product. This type of reaction is especially advantageous in
cases in which a high concentration of nitric acid in the bulk of the
solution would result in decomposition, either of the starting material
or of the product. Thus it is possible to use a dilute solution of nitric
acid which would have no effect on the substances in solution. Relatively
high current densities would seem to be desirable in this process as they
would result in an increase in the concentration at the anode surface.
However, too high a current density may counteract this condition, due
to dispersion of this concentration state by oxygen evolution.

One of the earliest applications of this process was in the nitration
of naphthalene, in which a suspension of this compound in dilute nitric
acid yielded the respective mono- and di-nitronaphthalenes (62). A
subsequent group of investigators (63) concluded from their experiments,
which were performed at a current density of 0-3 amp./cm. 2 and a
temperature of 95 to 98 C., that the reaction was not a result of
increased concentration of nitrate ions at the electrode surface, but
of local heating at the anode from the oxidation reactions taking place
therein. This concept, however, is contradicted to some extent by the
results of another group of investigators (64), who found that a suspen-

162

ANODIC SUBSTITUTION

sion of naphthalene could be nitrated electrolytically using a current
density of 0-2 amp. /cm. 2 , while maintaining the temperature below
45 C. These investigators found that, although there was a relative
increase in nitric acid concentration in the bulk of the solution which
was, however, insufficient to accomplish the nitration, the reaction did
occur within the highly concentrated diffusion layer at the electrode
surface.

The process of anodic nitration has also been utilized for the
preparation of nitrocellulose [C 6 H 7 O 5 (NO 2 ) 3 ], nitrobenzene, nitro-
toluene, nitro-w-xylol, and nitroglycerin (65).

(4) ANODIC ALKOXYLATION

If furan is treated with a methanolic solution of chlorine or bromine,
the addition of two methoxyl groups occurs, with the net result being
the formation of 2,5-dimethoxy-2,5-dihydrofuran (66):

2 CH 3 OH + C1 2

+ 2 HC1

OCH 3

As the product obtained by the chemical method is often contaminated
with a small quantity of halogen-containing impurity, the stability of
the acid-sensitive dimethoxydihydrofuran is not too great. An electro-
lytic method has been developed which is superior to the chemical
method, in that a halogen-free reaction product is obtained (67). The
furan is mixed with a methanolic solution of ammonium bromide and
the solution subjected to electrolysis at a temperature of approximately
14 C. using a nickel cathode and a platinum anode. During
electrolysis, ammonia and hydrogen are liberated at the cathode; at
the anode the bromine which forms reacts with the methanol and the
furan to give the desired compound, thus :

+ 2 CH 3 OH + 2 Br-

CH 3

f 2 HBr

\0/

OCH 3

The hydrogen bromide reacts with the ammonia being liberated at the
cathode to regenerate the ammonium bromide required in the process.
With the electrolytic method the investigators obtained a 73 per cent
yield of the dimethoxydihydrofuran and a current efficiency of 86 per
cent. This method has been applied to the methoxylation of such

163

ORGANIC ELECTRODE PROCESSES

furans as 2-methyl furan, furfuryl alcohol, and furfuryl acetate, with
the final products being the respective 2,5-dihydro-2,5-dimethoxyfuran
in each case (68). Ethoxylation of the furans has also been accomplished
by using ethanol instead of methanol in the process. The result is the
formation of 2,5-diethoxy-2,5-dihydroxyfuran from furan (69).

A more recent investigation (70) showed that it was unnecessary to
include a bromide salt in the reaction medium. The same reaction,
that is the anodic methoxylation of furan, can be accomplished by
utilizing a medium of .sodium hydroxide and methanol. Thus all
possibilities of destruction of 2,5-dihydro-2,5-dimethoxyfuran by
halogen are eliminated.

This is not the end. It is not even the beginning of the end. But it is,
perhaps, the end of the beginning.

WINSTON CHURCHILL.

164

ANODIC SUBSTITUTION

REFERENCES

(1) Zehrlant, Z. Elektrochem., 1901, 7, 501; Van Name and Maryott, Amer.
J. Sci., 1913, 35, 153; Youtz, J. Amer. chem. Sac., 1924, 46, 545.

(2) Ludersdorf, Ann. Physik., 1830, 19, 83.

(3) E. Schering, Z. Elektrochem., 1894, *> 7-

(4) Koidzumu, Mem. Coll. Sci. Kyoto, 1925, 8, 155.

(5) Ghosh, Bhattacharyya, Muthanna and Parikh, J. sci industr Res., 1952,
nB, 371.

(6) B.P., 140831, 1917; U.S.P., 1253615,^919.

(7) Bhattacharyya, Muthanna and Patankar, J. sci. industr. Res., 1952, nB,

365.

(8) Elbs and Herz, Z. Elektrochem., 1897, 4* 113; Dony-Hnault, ibid.,

1900, 7t 57-

(9) Trechzinsky, J. Russ. Phys. Chem. Soc., 1906, 38, 734.

(10) Feyer, Z. Elektrochem., 1919, 2 5> 115-

(n) Roush, Trans. Amer. electrochcm. Soc., 1905, 8, 283.

(12) Teeple, J. Amer. chem. Soc., 1904, 26, 536; Waeser, Chem Ztg., 1910,
34, 141; Bottazzi, AttiAccad. naz. Lined, 1909, 18 II, 133; Shipley and
Rogers, Canad.J. Res., 1939, *7B, 147; Russ. P. 7181, 1923.

(13) Feyer, loc. cit., ref. 10.

(14) Riche, Annalen, 1859, 112, 321; Richard, Compt. rend., 1901, 133,
878; Szper, Bull. Soc. chim. France, 1932, 5 1 (4), 654; Shipley and
Rogers, loc. cit., ref. 12.

(15) Fichter and Ruegg, Hely. Chim. Acta y 1937, 20, 1580.

(16) Muhlhofer, Diss., Munich, 1905.

(17) Lowy and Frank, Trans. Amer. electrochem. Soc., 1923, 43, 107; Croco
and Lowy, ibid., 1926, 50, 315.

(18) Ghosh, Bhattacharyya, Rao, Muthanna and Patnaik, J. sci. industr. Res.
1952, nB, 361.

(19) Van Name and Maryott, loc. cit., ref. i.

(20) Fichter and Glanzstein, Ber., 1916, 49, 2473.

(21) Muhlhofer, loc. cit., ref. 16.

(22) Cohen, Dawson and Crosland, J. chem. Soc., 1905, 87, 1034.

(23) Fichter and Glatzstein, loc. cit., ref. 20.

(24) Szper, Bull. Soc. chim. France, 1932, 51, (4), 656.

(25) Elbs and Brunschweiler, J. prakt. Chem., 1895, 5 2 > (2), 559; Gilchrist,
y. phys. Chem., 1904, 8, 539.

(26) Erdelyi, Ber., 1930, 63, 1200.

(27) Fichter and Cueni, Helv. Chim. Acta, 1931, 14* 658.

(28) Elbs and Herz, Z. Electrochem., 1897, 4, 113.

(29) Dony-Hnault, Z. Elektrochem., 1900, 7, 65.

(30) Coughlin, Amer. chem.J., 1902, 27, 63.

(31) Mailer and Loebe, Z. Elecktrochem., 1904, 10, 409.

(32) Croco and Lowy, loc. cit., ref. 17.

(33) Bruner and Czarnecki, Bull. Acad. Sci. Cracow, 1909, 322; Isgarysev and
Polikarpov, Chem. Zbl., 1941, i, 885.

(34) Neminski and Plotnikov, J. Russ. Phys. Chem. Soc., 1908, 40* 39*.

(35) Zehrlant, Z. Elektrochem., 1901, 7, 501.

165

ORGANIC ELECTRODE PROCESSES

(36) Mtthlhofer, loc. cit., ref. 16.

(37) Gilchrist, J. phys. Chem., 1904, 8, 539.

(38) D.R.P., 108838, 1899.

(39) D.R.P., 149983, 1902; Fichter and Cueni, loc. cit., ref. 27.

(40) D.R.P., 239672, 1909; Fichter and Cueni, loc cit., ref. 27.

(41) Dony-H6nault, Z. Elektrochem., 1900, 7, 57.

(42) D.R.P., 29771, 1884; U.S.P., 372940, 1885.

(43) Elbs and Herz, Z. Elektrochem., 1897, 4 113; Foerster and Meves, ibid.,
p. 268; Dony-Hnault, loc. cit., ref. 41; Perkin, Pract. Meth., 1905,
263; Fink, Trans. Amer. electrochem. Soc., 1919, 36, 348; Lazzarini,
Ind. Mm., 1929, 4> 77i ; Vyskocil, Chem. Listy, 1929, 23, 212, 249;
Shipley and Rogers, Canad.J. Res., 1939, 17 (B), 147.

(44) Pieroni, Gazetta, 1912, 42, i, 534-

(45) Elbs and Herz, loc. cit., ref. 43.

(46) Foerster and Meves, loc. cit., ref. 43.

(47) Teeple,^. Amer. chem. Soc., 1904, 26, 170.

(48) Abbott, J. phys. Chem., 1903, 7, 84.

(49) D.R.P., 64405, 1891; Bougault, J. pharm. Chem., 1918, 17, (7), 221;
Moles and Marquina, Am. Coc. Espan. Fisica. Quim., 1919, 17, 59;
Shipley and Rogers, loc. cit., ref. 43.

(50) Classen and Lob, Ber., 1895, 28 l6 5-

(51) D.R.P., 108838, 1899.

(52) D.R.P., 526803, 1926.

(53) Brown and Berkowitz, Trans. Amer. electrochem. Soc., 1939, 75, 385.

(54) Simons, et al., Trans. Amer. electrochem. Soc., 1949, 95, 47, 53* 55, 59, 64;
see also Simons, Fluorine Chemistry, 1950, Vol. I, p. 414; 1954, Vol. II,
p. 340, (Academic Press, New York); Silver and Cady, J. Amer. chem.
Soc., 1952, 74, 5792; Clifford, El-Shamy, Emeleus and Haszeldine,
J. chem. Soc., 1953, 2372; Schmidt and Schmidt, Z. Anorg. U. Allgem.
Chem. y 1955, 279, 289; Gramstad and Haszeldine, J. chem. Soc., 1956,

(55) Ruff, Z. Angewandte Chem., 1928, 41, 738; RufT and Ascher, Z. anorg.
allgem. Chem., 1929, 183, 197; Fisher and Jaenchner, Z. Angewandte
Chem., 1929, 42, 810; Ruff and Keim, Z. anorg. allgem. Chem., 1931,
201, 245; Fredenhagen and Cadenbach, Ber., 1934, 67, 928; Miller,
Calfee and Bigelow, .?. Amer. chem. Soc., 1937, 59, 198; Young, Fukuhara
and Bigelow, ibid., 1940, 62, 1171; Fukuhara and Bigelow, ibid., 1941,
63, 2792; Cady, et al., Jnd. Eng. Chem., 1947, 39, 290; Fowler, et al.,
ibid., 1947, 39, 292; Park, et al., ibid., 1947, 39, 354; Miller, et al., 1947,

39 40 !

(56) Schmidt and Schmidt, Chem. Techn., 1953, 5, 454.

(57) Schmidt and Schmidt, J. prakt. chem., 1955, 2, 250.

(58) F.P., 702829, 1930; Mel'nikov, Sklyarenko and Cherkasova,./. Gen. Chem.
(U.S.S.R.), 1939, 9 1819.

(59) Fichter and Schonmann, Helv. Chim. Acta, 1936, 19, 1414; Mel'nikov,
Sklyarenko and Cherkasova, loc. cit., ref. 58.

(69) Fichter and Schonmann, loc. cit., ref. 59; Cherkasova, Sklyarenko and
Mel'mkov, JF. Gen. Chem., (U.S.S.R.), 1940, 10, 1373; Kaufmann,
^ Angew. Chem., 1941, 54t 195-

(61) Mernikov and Cherkasova, Zhur. Obshchei. Khim., 1944, 14* H3-

(62) D.R.P., 100417, 1897.

(63) Fichter and Pliiss, Helv. Chim. Acta, 1932, 15, 236.

166

ANODIC SUBSTITUTION

(64) Calhane and Wilson, Trans. Amer. electrochem. Soc., 1933, 63, 247.

(65) D.R.P., 203377, 1906; U.S.P., 874564, 1907; 880373, 1908; Medinski,
Chem. Zbl. t 1934 II, 2523; Kirk and Bradt, Trans. Amer. electrochem.
Soc., 1935, 67, 209; Atanasiu and Belcot, Bull. Sect. Sci. Acad. roum.,
1937, 19, 101 ; Atanasiu, Bull. chim. pura aplic., 1940, 39, 71.

66) Clauson-Kaas, KgL Danske Videnskab. Selskab, Mat-fys. Medd., 1947,
24, 6; Clauson-Kaas, Limborg and Fakstorp, Ada chem. scand., 1948,
2, 109; B.P., 59504i> 1948.

(67) Belg. P., 500356, 1951 ; Clauson-Kaas, Limborg and Glens, Acta chem.
scand., 1952, 6, 531.

(68) Clauson-Kaas, Limborg and Dietrich, Acta chem. scand., 1952, 6, 545;
See also Clauson-Kaas, ibid., 1952, 6, 556; Elming, ibid., 1952, 6, 572;
Clauson-Kaas and Tyle, ibid., 1952, 6, 667.

(69) Clauson-Kaas, Acta chem. scand., 1952, 6, 569.

(70) Hayashi and Wilson, personal communication.

167

INDEX

ACETALDEHYDE, 13, 1 6, 59, Q I, 1 1 8, 121

/>-Acetamidobenzaldehyde, 63
/>-Acetamido-w-dimethyl-amirfopropio-

phenone, 66

4,4'-&-/>-Acetamidohydrobenzoin, 63
3 ,4-fo'$-/>-Acetamidophenyl- i ,6-bis-di-

methylaminohexane-3, 4-diol, 66
Acetic acid, 12, 13, 16, 102, 103, 118, 119
Acetoacetic acid, 72
Acetone, 61, 116, 144
Acetophenone, 128, 150
Acetoxime, 86
o-Acetoxyanisole, in
/>-Acetoxyanisole, in
Acetylacetone dioxime, 87
Acetylene, 60, 84, 106
Acetyl fluoride, 159
j8-Acetyl pyridine, 65
Adipic acid, 121, 122
Agitation, effect of on efficiency of electro-
lysis, 15
Aliphatic monobasic carboxylates, Kolbe

electrolysis of, 103
Alizarin, 130, 133
Alizarin-cyanin, 133
Allyl acetic acid, 84
Allyl alcohol, 106
Allyl thiourea, 123
Alternating currents, effect of on D.C.

electrolysis, 17
Aminoacetic acid, 108
/>-Aminoacetophenone, 6, 65
a-Aminoaldehydes, 74
m y m / -6w(m-Aminobenzalamino)-hydro-

benzoin, 63

m-Aminobenzyl alcohol, 70
o-Aminobenzyl alcohol, 70
N-(o-Aminobenzyl)-pyrrolidine, 87, 88
2-Amino-6-carboethoxy-benzothiazole,

162

2-Amino-6-ethoxybenzothiazole, 1 62
2-Amino-5-iodopyridine, 158
2- Apiino-6-methoxy-benzothiazole, 1 62
2-Amino-6-methylbenzothiazole, 1 62
i-Aminonaphthalene-2, 4-disulphonic

acid, 93

i -Aminonaphthalene-3 , 8-disulphonic
acid, 92

i- Aminonaphthalene-3 -sul phonic acid,

92 .

i-Aminonaphthalene-2, 4, 6-trisulphonic

acid, 93

2-Amino-6-nitrotoluene, 54
/>-Aminophenol, n, 13, 52, 53
4-Aminophenylarsine, 90
4-Aminophenyl-arsonic acid, 90
2, 3-6i's-(p-Aminophenyl)-2, 3-butanediol

65

^-Aminopropiophenone, 65
2-Amino pyridine, 158
a- Amino- valeric acid, 125
4-Amino-2, 5-xylenol, 136
Ammonium acetate, 123
"Ammonium formate, 123
Amplidyne generator, 22
Aniline, 6, n, 49, 51, 52, 151, 153, 161
Aniline black, 13
Anisaldehyde, 62
Anisole, in
Anthracene, 12, 130
Anthranilic acid, 92
Anthraquinone, 12, 93, 130, 133
Anthraquinone- 1 , 5-disul phonic acid, 93
Arabinose, 108, 119, 120
^-Arseno-anilin, 90
Auramine, 86

Automatic controlled potential devices 2 1
Azobenzene, 13, 51, 53
Azoxybenzaldehyde, 58, 62
Azoxy benzene, 51
o-Azoxychlorobenzene, 1 1
Azo-m-xylene, 135

BARBITURIC ACID, 80

Benzaldehyde, 12, 62, 126, 127, 128, 133

/>-Benzaldehyde sulphonic acid, 17

Benzamide, 92

Benzene, 62, 85, 91, 107, 125, 148

Benzene diazonium chloride, 87

Benzidine, 52

Benzil, 133

Benzoic acid, u, 12, 85, 107, 130, 132,

133, 134
Benzoin, 133
Benzonitrile, 87

1 68

INDEX

Benzophenone, 12, 64, 150
Benzpinacol, 64
Benzpinacone, 64
Benzyl alcohol, 62, 132
Benzylamine, 87
Benzyl chloride, 150
Benzylidine-hydroxyl-amine, 50
Berberine, 88
Borneol, 117, 118, 123
a-Bromoacetic acid, methyl ester, 89
a-Bromo aliphatic acids, esters of, 89
Bromobenzene, 91
m-Bromobenzoic acid, 70
m-Bromobenzyl alcohol, 70
Bromoform, 151
5-Bromoisatin, 155
-Bromophenol, 153
Bromotoluene, 152
Brown- Walker synthesis, 109
Brucine, 88
Butane, 72, 102, 104
2,3-Butanediol, 60
Butylacetoacetic acid, 72
4-ter.-Butyl-2,6-dimethylbenzoic acid,

107

Butyric acid, 84
150-Butyric acid, 103
n- Butyric acid, 102

CAMPHOR, 117, 118, 123

Caproic acid, 103

Carbon dioxide, 102

Carbonic acid, 70

Carbon monoxide, 70

Carbon tetrachloride, 91

Carbon tetrafluoride, 159

-Carboxyacetophenone, 66, 71

/>-Carboxybenzal, acetone, 129

2-Carboxybenzaldehyde, 71

2-Carboxybenzyl alcohol, 71

o-Carboxy-sulphonamide, 138

Carvacrol, 161

Catalytic effects, of added substances, 10

of electrodes, 9
Catechol, 125, 131, 134
Chloral, 145
Chloral hydrate, 91
Chloranil, 149
/>-Chlorbenzoic acid, 127
Chloroacetic acid, 91
Chloroacetone, 147
a>-Chloroacetophenone, 150
Chloroamine, 53
o-Chloroaniline, 53
-Chloroaniline, 53
Chlorobenzaldehydes, 62
Chlorobenzene, 127, 148
o-Chlorobenzene, 149

^-Chlorobenzene, 149
m-Chlorobenzoic acid, 70
o-Chlorobenzoic acid, 70
-Chlorobenzoic acid, 70
w-Chlorobenzyl alcohol, 70
o-Chlorobenzyl alcohol, 70
/>-Chlorobenzyl alcohol, 70
Chloroform, 91, 147
5-Chloro-2-hydroxybenzoic acid, 70
5-Chloro-2-hydroxybenzyl alcohol, 70
Chloroisatin, 151
Chloromethylchloroacetate, 108
^-Chlorophenol, 127
Chloropicrin, 55
j3-Chloropropionic acid, 147
j>-Chlorotoluene, 127
Cinnamic acid, 85
Constant current density electrolysis, 8

equipment for, 20

Controlled potential electrolysis, advan-
tage of, 21

automatic devices for, 22

manual device for, 21
o-Cresol, 126, 132, 161
7M-Cresol, 161
-Cresol, 64, 126, 161
Crotonic acid, 84
Crystal violet, 136
Cuminaldehyde, 62
Current efficiency, 15
2-Cyanohydrocinnamic acid, 85
Cyclohexane, 121
Cyclohexanediol, 122
Cyclohexanol, 86, 106, 121
Cyclohexanone, 121, 122, 150
Cyclohexene, 122
Cyclohexene oxide, 122
Cyclopropanol, 106
Cymene, 122

DECOMPOSITION POTENTIAL, 3

of aqueous halogen acid solutions, 4
of aqueous solutions of acids and bases,

4

3-a-Dehydro-3-phthalideacetic acid, 85
Depolarizer, i, 8

concentration effect of, on electrolysis,

14

Desoxy-theophyllin, 80
Diacetyl peroxide, 98
3,3 / -Diamino 4,4'-dihydroxyarseno-

benzene dihydrochloride, 90
2,6-Diaminotoluene, 54
Dibenzyl, in
Dibromoethylene, 91
3,5-Dibromo-2-hydroxybenzoic acid, 70
3 ,5-Dibromo-2-hydroxybenzyl alcohol,

70

169

INDEX

5,5'-E>ibromoindigo, 155
5,7-Dibromoisatin, 155
p-Dichlorobenzene, 148
Dichloroisatin, 151
Dichloromethane, 91
fo*$-(|3-Y-Dichlorophenyl)-formamidine

disulphide dihydrochloride, 123
,^>'-Dichlorostilbene, 91
o-Dicresol, 132
Diethyl adipate, no
N ,N-Diethylaminoethyl tetrachloro-

phthalimide, 75
5,5-Diethylbarbituric acid, 80
5,5-Diethyl-4,6-dioxy-i,2-dihy r dropyri-

midine, 80
Diethylglutarate, no
Diethyl succinate, 109, no
Diffusion coefficient, 14
Diffusion layer, 14

2 ,5 -Dihy dro-2 , 5 -dimethoxyf uran , 1 64
i,2-Dihydrophthalic acid, 86
3,4-Dihydrophthalic acid, monoethyl

ester, 86

Dihydropyrrole, 86
Dihydro-quinoline, 86
Dihydrouracil, 80
2,5-Dihydroxybenzoic acid, 134
3,4-Dihydroxybenzoic acid, 134
2,2'-Dihydroxy-5,5'-diethyl diphenyl, 128
2,4'-Dihydroxydiphenyl, 131
4,4'-Dihydroxydiphenyl, 131
3,4-Dihydroxy-6-sulphobenzoic acid, 137
4,4'-Dihydroxy-2,5,2 / ,5'-tetramethyl-

diphenyl, 129
Diketopiperazines, 74
2,5-pimethoxy-2,5-dihydrofuran, 163
p-Dimethylaminoacetophenone, 12, 48, 66
/>-Dimethylaminobenzaldehyde, 63
N ,N-Dimethylaminoethyl-3 ,4,5 ,6-tetra-

chloro-2-hydroxyphthalimidine, 78
N ,N-Dimethylaminoethyl tetrachloroiso-

indoline, 78
N ,N-Dimethylaminoethyl tetrachloro-

phthalimide, 15, 76, 77, ?8, 9*
N,N-Dimethylaminoethyl tetrachloro-

rhthalimidine, 78
N,N-Dimethylaminoethyl-3,4,6-tri-

chloroisoindoline, 78
4,4 / -6w-Dimethylaminohydrobenzoin, 63
w-(4-Dimethylaminophenyl)-disulphide

dimethochloride, 89

ime thylaminophenyl- 3 -/>-methoxy-

phenylbutane-2,3-diol, 66
tris*{p : p' : p''-Dimethylamino)triphenyl-

methyl trihydrochloride, 136
N,N-Dimethylaniline, 161

-, 130
N ,N-Dimethylphenylacetamide, 11,73
3, 5 -Dimethyl pyrazolidine, 87

2 ,4-Dimethylquinol ,135

N ,N-Dimethyl-4-thiocyanoaniline, 1 6 1

Dimethyl-vinyl-carbinol, 84

2,4-Dinitrobenzoic acid, 128

Dinitronaphthalene, 162

Diphenic acid, 130

Diphenyl, 107

Diphenyl acetic acid, in

jS, y-Diphenyl-adipic acid, 85

i , i -Diphenyl- 1 ,2-difluoroethane, 1 6 1

j3, Y-Diphenyl-dihydromuconic acid, 85

j5, Y-Diphenylmuconic acid, 85

a, a'-Dithiodiglycolic acid, 89

Dithiodiglycolic acid, dimethyl ester, 89

Dithymoldiiodide, 158

ELECTRODE OVERPOTENTIAL, 5
Electrode potential, 3, 6, 22
Electrodes,

alloy, 10, 41

for oxidation, 5

for reduction, 4

preparation of, 40

Electrolysis, general technique of, 45
Electrolysis cells, various types of, 33
"Electrolytes, used in electrolysis, 13
Eosin, 153
Ethane, 84
Ethoxylation, 164
Ethyl acetate, 145
Ethyl acrylate, no
Ethyl alcohol, 16, 60, 116, 145 _
Ethylbenzene, 127
Ethylene, 84, 91, 102, 146
Ethyl ene chlorohydrin, 146
0-Ethylene dihydroxylamine, 87
Ethylene glycol, 146
Ethylhydroquinone, 128
Ethylhydroxylamine, 55
Ethyl malonate, 109, no
^-Ethylphenol, 128
Ethyl phenyl carbinol, 62
Ethyl propionate, 102
Ethylquinone, 128
Ethyl succinate, no
N-Ethyl succinimide, 80
Ethyl sulphonic acid, 123
Ethyl thiocyanate, 123

FARADAY'S LAW, 18

Pick's Law, 15

Fluorescein, 153, 158

Fluorides, effect of on overpotential, 12

Formaldehyde, 55, 59, 69, 70

Formaldehyde oxime, 55

Formamide, 123

Formamidine disulphide, 123

170

INDEX

Formic acid, 13, 69, 70, 108, 109, 1 18, 119

/>-Fluorbenzoic acid, 127

Fluoroform, 159

/>-Fluortoluol, 127

Furan, 163

Furfural dehyde, 124

Furfuryl acetate, 164

Furfuryl alcohol, 164

GALACTOSE, 120

Gluconic acid, 119

Glucose, 60, 119

D-Glucuronic acid, 108

Glutacon-dialdehyde, 124

Glutaric acid, 125

Glycol, 118

Glycol aldehyde, 118

Glycolic acid, 69, 108, 118, 120

Glyoxime, 87

Glyoxylic acid, 61, 69, 108, 120

Guaiacol, 161

tt-HEPTADECANE, 104

Hexachlorobenzene, 148
Hcxachloroethane, 91
Hexane, 102

Hofer-Moest reaction, 97
Homocysteine, 89
Homocystine, 89
Homopyrocatechol, 126
Hydrazobenzene, 51, 53
Hydroanisoin, 62
Hydrobenzoin, 62
llydrocuminoin, 62
Hydrodiphthalyl, 71
Hydrofuroin, 64
Hydrogen carriers, n
Hydroquinone, 125, 131
Hydrosalicyloin, 62
Hydrovanilloin, 62
/>-Hydroxyacetophenone, 65
Hydroxyacrylate, no
2-Hydroxy-5-amino benzenesulphonic

acid, 53

7-Hydroxyazacyclodecane, 68
/>-Hydroxybenzaldehyde, 64
w-Hydroxybenzoic acid, 70
/>-Hydroxybenzoic acid, 134
ra-Hydroxybenzyl alcohol, 70
^-Hydroxybenzyl alcohol, 64
3-Hydroxy-2,6-dimethylbenzoquinone,

135

Hydroxylamine, 6, 50, 52, 55
i -Hydro xynaphthalene-2 ,4-disulphonic

acid, 92
i,-Hydroxynaphthalene-2,4,8-trisul-

phonic acid, 92

Hydroxyphthalimidine, 75
8-Hydroxyquinoline, 158

lMINO-2,4-DIMETHYLQUINOL, 135

Indigo, 10

Indigo-white, 10

Indole, 86

5,7-Iodo-8-hydroxy quinoline, 158

7-Iodo-8-hydroxy quinoline, 158

9-(o-Iodophenyl) acridine, 7

IR drop, 44

Irreversible electrode processes, 2

Isoamyl alcohol, 117

Isobutane, 60

Isobutene, 60

Isobutylacetoacetic ester, 73

Isobutyl alcohol, 60

Isobutyraldehyde, 60

Isohydrofuroin, 64

Isophthalic acid, 72

Isopropyl alcohol, 61, 102, 116

Isopropylamine, 86

Isopropyl butyrate, 102

N-Isopropyl succinimide, 80

Isovalerates, esters of, 103

Isovaleric aldehyde, 117

7-KETO-i-AZABiCYCLO 6.4.0 dodecane, 68
Kolbe reaction, 96

LACTIC ACID, 108, 121

Laurie acid, 103

Lead di-isopropyl, 61

Lead tetra-isopropyl, 61

Leucauramine, 86

Leuco-bases,

crystal violet, 136
malachite green, 136

Limiting current density, 14

Linoleic acid, 84

MALACHITE GREEN, 136

Maleic acid, 85, 124, 125, 131

Malic acid, 121

Malonic acid, 109, 121
ethyl ester, 105

Maltose, 120

Mannitol, 60

Mannose, 60, 120

Medium,

effect of, on electrolysis, 12
types used in electrolysis, 13

Membranes, used in electrolysis, 31

171

INDEX

Menthol, 117, 123
Menthone, 117, 118, 123
Mesaconic acid, 132, 137
Methanol, 69
^-Methoxyacetophenone, 48

effect of constant c.d. on reference

potential in reduction of, 7
Methoxylation, 163
Methoxy-methyl-dihydrobrucidine, 88
Methoxy-methyJ-tetrahydrobrucidine, 88
Methylamine, 55
N-Methylaniline, 161
Methyl anthranilate, 10
4-Methyl-benzylamine, 87
2-Methyl-3-butyn-2-ol, 84
Methylene chloride, 108
Methyl ethyl acetic acid, 103
i-Methyl-2-ethyl-3-hydroxypiperidine,

67

Methyl ethyl ketone, 61, 147
Methyl ethyl pinacol, 61
i-Methyl-2-ethyl-3-piperidone, 67
2-Methyl furan, 164
3-Methyl-glucose, 120
3-Methyl heptane, 73
3-Methylheptanoic acid, 105
N-Methyl-heptylamine, 67
Methylhydroxylamine, 55
N-Methylisoindoline, 76
Methyl-N-methyl-pyridinium sulphate ,

124
Methyl-N-methyl-quinolinium sulphate,

124

a- Methyl naphthalene, 130
N-Methyl-a-phenylacetamide, 73
Methyl phenyl carbinol, 128
N-Methyl-j3-phenylethylamine, 73
N-Methyl phthalimide, 75
i-Methyl-2-propyl-2-pyrroline, 68
N-Methyl-a-pyridone, 124
N-Methyl quinolone, 124
N-Methyl succinimide, 80
2-Methyl-4-thiocyanoaniline, 161
3-Methyl-4-thiocyanoaniline, 161
Monochloroacetaldehyde hydrate, 145
Monochloroacetic acid, 108, 145
Monochloroacetone, 144
Monochlorobenzophenone, 1 50
Monochlorocyclohexanone, 150
Mononitronaphthalene ,162
Monophenylacetic acid, in
Morpholinone, 74
Myristic acid, 103

NAPHTHALENE, 12, 129, 143, 162
i,5-Naphthalene-diamine, 54
a-Naphthol, 50, 129
Naphthoquinone, 12, 129

a-Naphthylamine, 50
Nernst equation, 15
Nicotine, 125
Nicotinic acid, 125
Nitramines, 55
m-Nitroaniline, 54
m-Nitrobenzaldehyde, 58, 62
Nitrobenzene, 6, 49, 52, 53, 163
w-Nitrobenzoic acid, 70
o-Nitrobenzoic acid, 70
o-Nitrobenzoic acid, methyl ester, 10
-Nitrobenzyl alcohol, 127
Nitrocellulose, 163
Nitroethane, 55
Nitrogen trifluoride, 160
Nitroglycerin, 163
3-Nitro-4-hydroxyphenylarsonic acid,

90

Nitromethane, 55
Nitro-naphthalene, 143
w-Nitrophenol, 54
Nitrosamines, 55
Nitrosobenzene, 6, 49
/>-Nitrotoluene, 127
Nitro-m-xylol, 163

OCTANE, 72, 73, 103
Overpotential, 3

Oxalic acid, 69, 108, 109, 120, 121
Oxygen carriers, 12

PALMITIC ACID, 103, 104
Paraldehyde, 12
Pegan, 87
Pegene-9-one, 88
Pentachlorophenol, 149
Perfluoropentane, 160
Phenanthraquinone, 130
Phenanthrene, 130
-Phenetidine, 53
Phenol, 86, 125, 131, 161
Phenolphthalein, 158
Phenylacetamide, 73
Phenylacetic acid, 69, 107
4-(Phenyl-azo)- 1 -naphthol , 50
4-(Phenyl-azo)- 1 -naphthylamine, 50
4-Phenylbenzoic acid, 107
2,3-fo's-Phenyl-2,3-butanediol, 128
j3-Phenylethylamine, 73
Phenylhydrazine, 87
Phenylhydroxylamine, u, 49, 53
Phenylhydroxylamine, effects of acids on,

hienylpropionic acid, 107
4-Phenylpyridine, 107
N-Phenyl succinimide, 80

172

INDEX

Phthalic acid, 72, 86, 129, 134

monoethyl ester, 86
Phthalic anhydride, 72
Phthalide, 71, 72
3-Phthalideacetic acid, 85
Pinacol, 61
a-Pinene, 122
Piperidine, 86, 125
Polarization, i

Potassium cyclohexanecarboxylate, 106
Potassium undecenoate, 106
Propane, 61
Propionaldehyde, 118
Propionic acid, 103, 117, 118, 147
Propionitrile, 87
Propiophenone, 150
Propyl alcohol, 117, 147
Propylamine, 87
Propyl butyrate, 102
Propylene, 102
Propylhydroxylamine, 55
Purpurin, 133
Putrescine, 87
Pyridine, 86, 124

2,3-6-(j3-Pyridyl)-2,3-butanediol, 65
2-Pyridyl-pyridinium sulphate, 124
Pyrocatechol, monophenyl ether, 131
Pyrrole, 86
Pyrrolidone, 79
Pyruvic acid, 108, 121

QUINONE, 13, 125, 126, 127, 128, 131
Quinizarin, 133
Quinoline, 86, 160

REFERENCE ELECTRODES,

saturated calomel, 42

silver-silver chloride, 43
Reference potential, 3
Reversible electrode processes, i
" Ripple voltage," 20
Roush neutralization cathode, 147, 157

SACCHARIC ACID, 119

Saccharin, 92, 138

Salicylaldehyde, 62, 71

Salicylic acid, 71, 134

Salt bridges, 45

Salvarsan, 90

Schiff bases, 86

Sodium acrylate, 106

Sodium benzenesulphonic acid, 132

Sodium cymenesulphonate, use of, to

obtain desired solubility of organic

compound, 13

Sodium -toluene-sulphonate, use of, to
obtain desired solubility of organic
compound, 13

Stearic acid, 84, 103, 105

Stilbene, 62, 161

Strychnine, 88

Succinic acid, 85, 124, 125, 127

Succinimide, 79

Succinimidcs, substituted, 80

Succinonitrile, 87

Sulpho-anthranilic acid, 92

/>-Sulphobenzoic acid, 137

TEMPERATURE, EFFECT OF, ON ELECTRO-
LYSIS, 1 6

Terebic acid, 122

Terephthalaldehydic acid, 129

Terephthalic acid, 17, 129, 134, 135

$;yw-Tetrachlorobenzene, 148

i,i,4,4-(Tetrachlorophenyl)-2,3-
dichlorobutene-2, 91

A 2 -Tetrahydrobenzoic acid, 85

Tetrahydroberberin, 88

Tetrahydrobrucine, 88

A 2 -Tetrahydrophthalic acid, 86

Tetrahydro-quinoline, 86

Tetrahydrostrychnine, 88

Tetraiodofluorescein, 158

Tetraiodophthaleine, 158

Tetraphenylbutyne, 91

Theophyllin, 80

4-Thiocyanoaniline, 161

4-Thiocyano-N-methylaniline, 16 1

Thioglycolic acid, 89

Thiourea, 123

Thymol, 158, 161

Time, influence of, on electrolytic process,

17

Toluene, 12, 126
^-Toluene sulphinic acid, 89
o-Toluene-sulphonamide, 138
o-Toluene sulphonic acid, 137
-Toluenesulphonic acid, 17, 137
/>-Toluenesulphonyl chloride, 89
Toluhydroquinone, 126, 132
o-Toluic acid, 134
/>-Toluic acid, 17, 129, 134
7-Toluidine, 161
/>-Toluidine, 162
m-Tolunitrile, 135
o-Tolunitrile, 135
/>-Tolunitrile, 87, 135
Toluquinone, 126, 132
/>-Toluylaldehyde, 129
2,4,6-Tribromoaniline, 153, 155
2,4,6-Tribromophenol, 153
Trichloroacetic acid, 91
2,4,6-Trichloroaniline, 151

173

INDEX

i,i,i-Trichloro-2,2-6w-C#)-chlorophenyl)- UREA, 123

ethane, 91

i,i,i-Trichloro-2,2-diphenylethane, 91 - VALERIC ACID, 103

2,4,6-Trichloroquinone, 151 n- Valeric acid, 84, 103

2,4,5-Tnchlorotoluene, 150 Vanillin, 62

Trideuteroacetic acid, 105 Voltammetric curves, 8
Tndeuteropropionic acid, 105
2,4,5-Trihydroxybenzoic acid, 134

Trihydroxyglutaric acid, 119 WETTING AGENTS, ..,_,.,. .

Trimethyl acetic acid, 103 use of * obtain de f ired solubility of

Trimethylene urea, 80 or g anic compound, 14
( ) 3 > i 3 > 1 9-Trimethyltricosanoic acid ,

105 f XANTHINE, 80

Trinitrotoluene, in />-Xylenol, 129

Trinitro-m-xylene, in m-Xylidene, 135

Trioxymethylene, 118 />-Xylohydroquinone, 129

Triphenyl acetic acid, in jp-Xylol, 129

Tropine, 117, 118, 123 />-Xyloquinone, 129, 135

Tropinone, 117, 123 Xylose, 120

174