ORGANIC E LECTRQC H EM ISTRY M. Fleischmann, B.Sc., Ph.D., A.R.C.S., and D. Pletcher, BSc., Ph.D., A.R.I.C. . . Chemistry Department, The University, Southampton SO9 5N H The special characteristics of electrode processes Methods of investigation, 90 Reactive intermediates in electrode reactions . . .. . . . . 94 Uncharged radicals, 95 Carbanions and carbanion radicals, 97 Carbonium ions and carbonium ion radicals, 98 Other intermediates, 100 . . .. 85 Correlations between electrochemical data and other physical pro- Some further examples of electrochemical reactions of organic perties . . .. .. .. . . .. . . . . . . 103 compounds . . .. . . . . . . . . . . . . 106 The control of electrode reactions, 106 Oxidations, 109 Reductions, 1 12 References .. . . . . . . .. . . .. . . .. 114 At the end of the last and the beginning of this century there was considerable interest in the possibility of synthesizing organic compounds by electro- chemical methods. Much of this work was summarized by Fichter.1 In some respects this is a depressing summary, as the early hopes of achieving selective reactions by electrochemical methods were clearly not realized, mainly because the electrolyses were carried out under conditions where the electrode potential, pH etc. were not controlled. Since the middle 1920s many of the advances in organic electrochemistry have been made by polarography, a technique in which a dropping mercury electrode is used to provide a highly reproducible surface in the construction of currentlpotential curves.However, polarographers have generally been reluctant to prepare and isolate the products of the electrode reaction, a factor which contributed to failure in communication between electro- chemistry and organic chemistry. In recent years, the forecast of a falling cost of electricity due to the advent of nuclear power and a realization that large scale industrial electrosynthesis is a real possibility have led to a resurgence of interest in organic electro- chemistry and research has been directed to the development of new selective routes and the improvement of existing electrosynthetic methods. At the same time it has become feasible to study the electrochemical oxidation and reduction of organic compounds in greater detail, partly because of the development of electronic control systems known as potentiostats, partly Fleischmann and Pletcher 87 because of the recognition and characterization of reaction intermediates and partly because of the considerable development in the understanding of electrode kinetics which has followed the development of numerous new techniques for studying electrode reactions.In fact, it is'now possible to study complex reaction sequences in aetail. The first section of this review (p. 88) deals briefly with the special characteristics of electrode reactions and with the most common methods of investigation, while the second (p. 94) illustrates the generation, nature and manner of reaction of intermediates of electrode reactions. The third (p.103) and fourth (p. 106) sections deal with the correlation of electrochemical data with other physical properties and with some further examples of electro- chemical reduction, oxidation and substitution reactions which illustrate features of electrosynthesis additional to those described in the second section. For obvious reasons, this review is intended to be illustrative rather than comprehensive, and for fuller expositions of various aspects of these subjects, the reader is referred to reviews of electrode mechanisms72J organic polaro- graphy,4-6 synthesis,7-llJl reductions712J3 oxidations,l4-16 substitntion,l7 and the Kolbe synthesis.18-20 Text-books on relevant topics of electrode processes include references 22-27.THE SPECIAL CHARACTERISTICS OF ELECTRODE PROCESSES It is well known that the standard free energy change of a reaction such as is given by Rd + 0, - o x O x + H, - Rd AGO = -zFEO where z is the number of electrons transferred in the process. If we refer immediately to a scale of free energies or electrode potentials (Fig. 1) it is evident that spontaneous reactions using oxygen or air as the oxidant, or hydrogen as a reducing agent, are only possible (in electrochemical terms) within the potential range limited by the reduction of oxygen and the oxidation of hydrogen. This driving force therefore amounts to roughly 0.5 eV or 42 kJ mol-1. By contrast, it is possible to carry out electrochemical reactions between +3.5 V and -2.5 V, even in aqueous solutions, if suitable electro- lytes are chosen (e.g.perchlorates for oxidations and quaternary ammonium salts for reductions). The large driving force for electrochemical processes is therefore of the order of 3 eV or 250 kJ mol-1. Strictly, it is not valid to compare the available driving force to the standard free energy change of any chosen reaction since most processes are irrever- sible; we must therefore estimate the rate constants instead. At the surface potential, 4, and in transferring a charge zFmol-1 a maximum amount of between the metal and an electrolyte solution, there is a difference in electrical work (YZ~F can be done in reaching the activated complex, where a(O< (Y < 1) is the transfer coefficient.The rate of an electrochemical reaction is therefore R.I.C. Reviews 88 - + 4.0 O 3 + 3 H 2 0 + 6 e - - 6OH- ClOh + e - -ClO, - + 3.0 - + 2.0 Ag2++ e- - Ag+ Z10-+H20+2e- 0,+2H20+4e- - C I - + 2 0 H - - 4 0 H - - + I .o radical cation - -H+ radical I I - e - 2H++2e- - H, 1 1 - e - - ov substrate -- -- 2.0 radical t anion - +H + radical 1 +e+ carbanion Zn2++ 2e- - Z n Na++ e- -Na Solvated electrons Therefore, current -- 3.0 - - 4.0 Fig. I . Representative electrode potentials or log I = const. + -.$ azF JRT at constant concentrations, where the product determines the slow step, vi is the order of the reaction with respect to species i which is present at a con- centration ci (for more exact expressions of the effect of the potential difference on the rates of electrode processes see reference 26).Equation 2 shows that we can modify the energy of activation, Ex, of electrochemical reactions to such an extent that they will take place at room temperature even if Ex is large. Fleischmann and Pletcher carbon i u m I ion I - e S + 89 Pot en t ios t a t 1 Osci I I bsco pe t m I / m Y W --L- Pulse profiles Potentiometers F a - Counter electrode Reference t Osci I lator - ---I- - - Fig. 2. Circuit diagram for a potentiostat We can summarize this first point in the following way: electrochemical reactions enable one to introduce a considerable amount of energy into molecules at low temperatures.The order of magnitude of this energy is in fact comparable to that of the strength of chemical bonds and this explains why many of the chemical changes which are observed can take place. It is not surprising that many 'high energy' chemicals which are used as oxidants or reductants in synthesis are made electrochemically and it is clearly of some interest to avoid such intermediate steps in synthesis. Methods of investigation Many methods of investigating electrode reactions which have been developed during the past decade are based on the regulation of the potential of the working electrode with respect to an unpolarized reference electrode using a feedback amplifier (Fig. 2), the current being passed between the working and a subsidiary electrode.Such feedback amplifiers are known as potentio- stats. The potential to be applied between the working and reference elec- trodes is set at the second input of the amplifier. A succession of constant potentials may be chosen to construct a current/potential curve ; alternatively, a slow linear sweep of potential with time will give this curve (potentio- dynamic method). When a dropping mercury electrode is used, the current/ potential curve is known as a polarogram. Figure 2 shows that other potential/time profiles may be imposed on the working electrode. In the case of a triangular potential/time profile with a high sweep rate [0.1-1000 V s-11, the current/potential plot may be recorded oscillographically. This technique is known as cyclic voltammetry and is discussed later.A sine wave of small amplitude applied to the input gives Lissajou figures on the oscilloscope from which the electrode impedance may be calculated; alternatively the impedance may be found using a bridge. R.I.C. Reviews 90 I .5 3.0 Fig. 3. Polarogram of anthracene in acetonitrile containing 0. I M Et,NCIO, When a potential step is applied to the input, a current/time transient is obtained; Fig. 2 shows that square pulses may be also applied. Electrode reactions may also be studied by applying a constant current and measuring the potential/time curve or the potential in the steady state. While this method requires only simple apparatus, controlled potential methods have been increasingly used in the last few years.The advantages as com- pared to constant current techniques will be apparent, since the rates of electrode processes are controlled by the potential, as may be seen from equations 2 and 3. In addition, in preparative work a measure of selectivity in the reactions may be achieved (equation 1). Equation 3 shows that the current would increase indefinitely with potential; in practice, the diffusion of the electroactive species to the electrode, and possibly also of the products away from the electrode, will become rate determining and the current reaches a limiting value. A typical current/ potential curve obtained with a dropping mercury electrode, that is by polarography, is shown in Fig. 3. The portion AC illustrates the addition of one electron to an aromatic hydrocarbon and in the region BC the electrode process is diffusion controlled.The potential at which the current is half the diffusion current is known as the half-wave potential, E,. Polarographic waves are classified as being reversible or irreversible. Jn the first case, the rate of the electrode reaction is fast compared with diffusion so that the shape of the wave is controlled by mass transfer of the reagent and product and the equilibrium at the surface Rd O x + ze-- Therefore ( 5 ) E E, + RT ~ In aox ~ zF aRd Fleischmann and Pletcher 91 where aox and aRd are the activities of the oxidized and reduced forms at the electrode surface. It will be seen that in this case E, h EO. In the second case, the rate of the electrode process is slow compared with diffusion and E, is no longer equal to EO.However, with increasing potential, the current again becomes diffusion controlled. The decision as to whether the electrode process is reversible or irreversible may be made by considering the shape of' the wave. In the second case the wave is drawn out along the potential axis compared to a reversible wake (for exact criteria, see reference 25). In the example shown, and for many other hydrocarbons, the wave AC is reversible but that for the addition of a second electron CD is irreversible. In electrode reactions, it is frequently found that a chemical process precedes electron transfer and only one species is electroactive in a potential region.For example,28 CH,(OH), HCHO + 2e- + 2HS HCHO - + H,O CH,OH ( b ) ( a ) A limiting current (which is controlled by the rate of the preceding reaction) may therefore be realized when b becomes fast compared to a. Such waves are known as kinetic waves and their distinction from diffusion controlled waves is described el~ewhere.~5 Other reaction sequences can also be investi- gated (e.g. following reaction^).^^ The steady-state current/potential data can further be analysed by plotting the initial part of the region AB in Fig. 3, according to equation 4. The linear relation between log I and + is known as a Tafel plot. The variation in position of these plots with the concentration of electroactive species gives information directly about the slow step of the electrode reaction.For measurements under non-steady state conditions, such as in cyclic voltammetry, all the information which has been described above may again be obtained23925 and further data of the reaction sequence may be derived. In the first place, the rate of diffusion in the non-steady state is greater than for steady polarization and the current can therefore reach higher values, as in section AB in Fig. 4. (This increase in current allows the study of the kinetics of faster electrode reactions than is possible by steady-state methods.) Non-steady state diffusion then becomes rate controlling and the current falls (section BC). On reversal of the potential sweep, the starting material is regenerated in the case of simple reactions and a peak, DEF, is observed in the oxidation current.Secondly, for a sequence such as R + e- - R - R ' - + X f A RX' - 3 (RX) 2 some of the intermediate is removed by the chemical reaction and the reoxida- tion peak for R'-- R + e - is therefore diminished.30 In some cases peaks for the oxidation of RX' and (RX)2 may also be observed. In general, the shape of the peaks, and in particular their height and separation and their dependence on the sweep rate, is directly related to the overall kinetics of the electrode reaction. It is R. I. C. Reviews 92 B E Fig. 4. Cyclic voltammogram of anthracene in acetonitrileZ9 containing 0. I M Et,NCIO,, sweep rate 0. I5 V s-l of considerable importance that in this way the kinetics of fast reactions succeeding a slow electron transfer may be measured.Other pulse techniques give essentially similar information and the choice of method will be governed largely by convenience although the interpretation of potential step experiments will usually require fewer assumptions than the interpretation of cyclic voltammograms and may therefore be preferable. Thus, in a single-step experiment, the initial part of the current/time transient gives the rate of the electron-transfer step and the whole transient the rate of preceding steps.27 Using a double step, say first to the cathodic and then to the anodic direction, the kinetics of a following reaction can again be deter- mined. For example, the rate constant, k, for the rearrangement of hydrazo- benzene to benzidine can be determined in this way.31 Further information about the kinetics of electrode processes may be obtained using various translating electrodes which increase the rate of mass transfer.The most widely used of these is the rotating disc electr0de.3~ Information concerning chemical reactions succeeding electron transfer may be obtained using the ring-disc electrode.33 This consists of a disc electrode closely encircled by a ring electrode. The intermediate is generated at the disc and the fraction of the intermediate which is removed at the ring is related to the rate of removal of intermediate by chemical reaction during its transit from the disc to the ring. The illustrations which have been given have assumed that the inter- mediates are not adsorbed and that the coupled reactions take place in 93 FIeischmann and Pletcher solution.Frequently, this will not be the case but it is possible to investigate purely heterogeneous steps in precisely similar ways and indeed simultaneous heterogeneous and homogeneous pathways.34 It will be apparent that very detailed information on the reaction mecha- nisms (i.e. the rate of preceding steps, the rate of electron transfer, the rate of succeeding reactions and data concerning adsorption) can now be obtained by electrochemical methods. As yet, such investigations have been carried out only on a very small proportion of known electrode reactions and although, where this information is available, the intermediates are fully consistent with those postulated in the next section and are in line with those postulated in physical organic chemistry, in other cases the existence of intermediates must be deduced from non-kinetic data such as the nature of products, esr spectra etc.REACTIVE INTERMEDIATES IN ELECTRODE REACTIONS Early workers in the field of mechanistic organic electrochemistry believed that electrochemical reductions took place via the production of ‘active hydrogen’ at the cathode and that the potential at which an organic com- pound was reduced was dependent on the potential energy required by the ‘active hydrogen’ in order to react with the organic compound. Similarly, electrochemical oxidations were believed to proceed via the production of ‘active oxygen or hydroxyl radicals’ at the anode.In recent years it has become clear that the intermediates produced during an electrode reaction are frequently the same as those commonly encountered in other fields of organic chemistry-carbanions, carbanion radicals, radicals, carbonium ion radicals, carbonium ions and, less commonly, intermediates such as biradicals, solvated electrons, transition-metal ions, surface oxide layers and inorganic radicals. Three factors have been mainly responsible for the recognition of these intermediate species. Firstly, the growing use of aprotic solvents, such as acetonitrile, dimethyl formamide, dimethyl sulphoxide, and propylene carbonate, in which the lifetime of the intermediates is much longer than in aqueous solutions ; secondly, the development of improved analytical techniques which allow the identification of minor as well as major products and, lastly, the development of a number of techniques, both electrochemical and non-electrochemical, for the study of transient species.Of the electrochemical techniques, cyclic voltammetry has been the most widely exploited although a number of linear sweep and pulse techniques for the study of short-lived species have been described. Electron spin resonance is by far the most common non-electrochemical technique since it allows the detection and identification of very dilute solutions of free radicals.35 Short- lived radicals have been studied using electrochemical cells in the resonance ~avity.~6 The use of ir, visible and uv spectroscopy has been facilitated by the development of optically transparent electrodes37 which allow the electro- chemical cell to be placed directly in the light path.This technique permits the study of species close to the electrode while actual surface layers may be studied by total reflectance spectroscopy.38 Electrochemiluminescence39 is R.I.C. Reviews 94 another spectroscopic technique which has been used for the study of anion and cation radicals. The species Rf and R- are produced together, either by the use of alternating current or by placing two electrodes in close proximity, and are allowed to react to produce excited molecules, R*, which can then emit light which is characteristic of the system.In this section some of the electrochemical methods of producing inter- mediates are described and some illustrations of their reactions are given. It is important to realize that the behaviour of the intermediates will be different when they are adsorbed on the surface of the electrode from when they are free in solution; this is one of the factors which causes the electrode material to be important in controlling the products of an electrode reaction. Thus the Kolbe reaction RCOOH - RCOO' + e- + H+ - CO, + R' p R - R product R++ e--- has been shown to give widely differing products on platinumls and carbon40-42 electrodes. Moreover, although the Kolbe synthesis is believed to proceed via radical intermediates, no esr spectra have been obtained for them, presumably because the radicals are strongly adsorbed on the electrode.Uncharged radicals Radicals are intermediates in many electrode reactions, both oxidations and reductions, and a number of examples are cited as equations below. (R = alkyl, 4 = phenyl.) 43 45 46 47 RCOOH - R'+ CO, + H+ + e- A l (CH,), - CH; + Al(CH,), + e- OH 0' 0 0' OH 0 -CH (COOEt) , - CH(COOEt),+ e - + 2 C = 0 + H++ e - - q5,C'OH 49 R'+ I - RI + e----- Fleischmann and Pletcher 95 radical + RH RH + alkene I I R - C - C ' I I \ metal al kyl 50 Fig. 5. Main modes of decomposition of uncharged radicals With the exception of a few radicals such as the diphenylpicrylhydrazyl radical, these intermediates are very reactive species and can react in a number of ways.The main modes of disappearance are summarized in Fig. 5. The most common mode of disappearance is dimerization as illustrated by the normal Kolbe synthesis, the formation of pinacol from the reduction of acetone47 and the formation of diphenyl from the reduction of tetraphenyl- ammonium ions.48 However, under suitable conditions, high yields of pro- ducts other than the dimer can be obtained, for example, 4RMgl .-- RCOOH 51 4e- + 4Mg2++ 41-+ PbR, e- + H+ + CO, + R' CH2=CH-cH=CH2=_ dimer RCH,-CCH=CH-CH;- although by-products formed by competing reactions are almost always observed. At the same time, it is important to realize that the products and reaction routes will depend on the concentration of radicals present, i.e.will be dependent on the electrode potential, concentration of electroactive species and other electrolysis parameters. For example, a high concentration of radicals will favour dimerization. Other typical radical processes which have been demonstrated to occur during electrode reactions are the initiation of p~lymerization~~ and chain reactions. An example of a chain reaction is the reduction of benzyl iodide on mercury in aqueous ethanol53 where several times the coulombic yield of the product, benzylmercuric iodide, was obtained. The direct chemical reaction is very slow. Thus the mechanism postulated is R. I. C. Reviews 96 I-.- +H- solvent- +H2 So'Vent +2- Carbanions and carbanion radicals In aprotic solvents, aromatic hydrocarbons may be shown (e.g.by polaro- graphy or cyclic voltammetry) to reduce in two one-electron steps to the carbanion radical and di~arbanion.5~ In the absence of proton donors or other addition agents, the radical anions are relatively stable. On the time scale of an electrochemical experiment, which is determined by the transition of the species through the boundary layer in contact with the electrode (approx. 0.1 s), the process may be reversed by changing the potential and the parent hydrocarbon will be regenerated.29 On a longer time scale, dimeri- zation or disproportionation will take place; the dianions formed will abstract protons relatively easily even from aprotic solvents so that dihydro- aromatic or dihydrodiaromatic compounds are formed at the potential at which the first electron is added. Thus anthracene gives 9, lo-dihydroanthra- cene55 and phenanthrene gives 9,9', lO,lO'-dihydrodiphenanthrene.56 The dicarbanion produced directly at more negative potentials, naturally, also protonates readily; in this case, reversal of the polarization in cyclic volt- ammetry can lead to the radical 4H*57 so that the protonation clearly takes place in two steps.scheme t +- +e- +- +-- +-a +H - +H - +H ' + + + 2 - When a proton donor [e.g. water or benzoic acid154 is added to the aprotic media, the first reduction wave increases at the expense of the second wave until the two waves merge to form a single, two-electron wave.The reduction H'+e- - +H- w 2 #H-+H+- has been postulated, the single wave being obtained because the species 4H- has a higher electron affinity (i.e. it is more readily reduced) than the parent hydrocarbon. A similar scheme may be written for the reduction of the hydrocarbon in the presence of carbon dioxide55 Fleischmann and Pletcher 97 + + e--- +-- + .- + CO, - $COO - + 'COO- + e- - +-coo- CH CH,' + -coo-+co, - + ( c o o - ) 2 and when naphthalene is reduced in the presence of carbon dioxide, the product is 1,4-dihydro- 1,4-dicarboxynaphthalene. It is also possible to react these aromatic hydrocarbon carbanions with methyl iodide to give the dime thy1 di hydroderivative .55 The carbon dioxide addition is a general reaction of carbanions.For example, when stilbene or benzyl chloride is reduced in the presence of carbon dioxide the products are 1,2-diphenylsuccinic acid55 and phenylacetic acid5* respectively. Aliphatic carbaiiions are also common intermediates and they usually occur at more negative potentials in reactions producing radicals. For example, R-+CI- RCI + 2e--- +C-H -C-H+ +CH = CH++ 2e-- C-H, - C-HCN CH, = CHCN + 2e-- CH, = CHCN + C-H,-C-HCN + 2Hf- \C = + 2e- + H+ -- CH3\ C--OH 59 CH 3' and in most cases the final products arise by proton abstraction from the solvent. Thus, the reduction of ethyl chloride and acetone at very negative potentials gives ethane and isopropanol respectively. Compounds, such as acrylonitrile, which contain an activated alkenic group60 may be reduced to form a dianion NCCH2CH2CH2CH2CN which will react with another molecule of acrylonitrile to form the hydrodimer, adiponitrile.This reaction is the basis for a new industrial plant. 1,3-Buta- diene61 and c+unsaturated acids62 are other examples of the many alkenic compounds which reduce in the same way and it is also possible to form crossed hydrodimers from mixtures of activated alkenes.63 Carboniunz ions and carbonium ion radicals It might be expected that in an aprotic solvent, aromatic hydrocarbons would also oxidize in two one-electron steps to give a carbonium ion radical and a carbonium ion 49 55 +H - $H'++ e - +H'+ - +++ H++ e- +H~++ e - or +H.+- 98 R.I.C. Reviews However, the cation species are much more unstable than the corresponding anion species and this simple behaviour is found only for relatively few aromatic hydrocarbons such as 9, IO-diphenylanthra~ene~~.~~ and 1,8- dithionaphthalene.65 The behaviour found for most hydrocarbons consists of an ece mechanism, i.e.the first electron transfer is followed by a very fast chemical reaction, possibly dimerization or reaction between the cation radical and the solvent, to give a species which is oxidized further. The first electron transfer has been shown to be reversible by using very rapid cyclic techniques where the rate of potential change is fast compared with the following chemical reaction.66 The products which have been reported for the oxidation of hydrocarbons such as anthracene are many and varied and perhaps this is not surprising in view of the reactivity of the cation inter- mediates.However, in some cases such as hexamethylbenzene oxidized in acetonitrile good yields of a substituted amide have been reported67 C H 3 ) 3 C H 3 CH 3 + H+ + 2e- \ CH3 CH3 CH3 CH3 CH 3 CH,CN + CH3@H3 CH, CH3 CH, CH, CH3C+= N -CH2 CH3 CH, CH3 CH3 CH2 = CH- CH2R -CH2 \\ CH CO. NH.CH vlH3 - Hydrolysis Moreover, there now seems general agreement that the electrochemical substitution of aromatic species takes place via carbonium ion intermediates.68 Thus, whether the reaction studied is halogenation, cyanation, methoxylation or acetoxylation the mechanism seems to be +H - #I++ H++ 2e- x- +x and not a mechanism involving formation of radicals followed by radical substitution, since substitution only takes place at potentials where the hydrocarbon is oxidized, even when the substituting anion is oxidized to radicals at lower potential^.^^ Furthermore, the distribution of product isomers coincides more closely with a reaction between a carbonium ion and an anion rather than attack on an aromatic system by a radical.70 Aliphatic carbonium ions may be prepared by the oxidation of carboxylic acids,71 alkyl iodides,72 aliphatic hydrocarbons73 and primary amines.74 RCOOH --- R++C02+2e-+H+ CH,I - CH~++ I+ + 2e- =CH - CfHR + H++ 2e- RCH2NH2 - RCH2N'+H2 + e - - RC+H2 +NH2 Fleischmann and Pletcher 3 3 4 .3 99 ROMe ROCOCH H O R + alkene - -H+ CH CN 3 CH c += N RL CH ,CON H R RX 4 R Fig.6. Probable reactions of aliphatic carbonium ions in common solvents A summary of the probable reactions of aliphatic carbonium ions in common solvents is shown in Fig. 6. Carbonium ions formed in electrode reactions show the usual carbonium ion rearrangement. Thus, when neopentyl iodide is oxidized in acetonitrile, N-tert-pentyl acetamide can be isolated as well as N-neopentyl acetamide. 72 CH, lcHlcN CH3--C=N -CH,-CMe3 CH,-C=N CH3- C-CH, I C2H5 Other intermediates + It is also interesting to note that non-classical carbonium ions can be prepared electrochemically. For example, the anodic oxidation of exo- norbornene-2-carboxylic acid gives the non-classical carbonium ion.75 Two interesting biradicals which have been prepared electrochemically are benzyne76 and dichl~rocarbene.~~ Since both are extremely reactive, their preparation is inferred by products from reactions with trapping agents.The small yield of product may be due to low efficiency in the preparation of 100 R. I.C. Reviews the biradicals but is also likely to be due to competing trapping reactions, e.g. reaction with the solvent. The biradicals were prepared by reduction in acetonitrile, dichlorocarbene from carbon tetrachloride and benzyne from o-dibromobenzene. The trapping agents were tetramethylethylene and furan respectively. 2C I- + CC 1; CC 14 + 2e-- N O - NO; + e- Me2C=CMe2 - Me ‘c-c /Me ClO, - C l O i + e- /CH3 ‘CHj There is a number of oxidations where inorganic radicals have been postulated as intermediates.Thus, in acetonitrile, the anodic limit has been found to be strongly dependent on the inert ele~trolyte,~~ which suggests that the initial reaction is oxidation of the anion present. The following mechanism, with perchlorate as the inert electrolyte, has been proposed :78 C104+CH3CN - CH2CN+ HC104 Similarly, the methoxylation of dimethyl formamide in a nitrate base electrolyte is believed to involve the nitrate radical :79 Me/ ‘c’ ‘Me + HNO3 / CH3 NO; + HCON ‘CH3 - HCON Clearly, the intermediates in the oxidation and reduction of substituted hydrocarbons will not always be carbanions and carbonium ions. For example, the intermediates from nitrogen containing organic compounds will frequently have the charge situated on the nitrogen.However, their reactions and properties do not differ greatly from the carbon species. Examples of such charged nitrogen species are shown below. The solvated electron is a probable intermediate in the electrochemical reduction of ben~ene.8~98~ It is only possible to reduce benzene in solvents such as ammonia, hexamethylphosphoramide, ethylenediamine and amines, Fleischmann and Pletcher 8 101 80 R3N - e - + R 3 N + s R3N+H 81 CH,CONHR - e-+ CH,CON+HR SOIVenf_ CH,CON+H,R 82 + e-+ H+ NH2 NH NH - NH in which solvated electrons may be formed. Therefore, it seems likely that in the presence of Group I metal ions the reduction mechanism is Li ( s h ) + e- - Li+(sol.) + e-(soh.) - w products Metal ions in lower or higher valence states may also be used as inter- mediates in reductions and oxidations.For example, M g ( p and A1(1)8~ produced by the anodic oxidations of the metals will reduce 2-methoxy- phenyl mesityl ketone, and nitrosobenzene, azoxybenzene and azobenzene respectively. 3AI++2C6H5N0 +4H20 - 60H-+ 3A13++ c6H5N- H H l l NC6H5 such as AI - AI++ e- or CO(III) will oxidize alkenes or aromatic compounds87 in a sequence cO2+ - co3+fe- and in this case the metal ion acts as a catalyst regenerated at the surface of the electrode. Similar oxidations may be carried out with metal oxides pro- duced on the electrode surface.87 Mn2++ 2 H 2 0 - Mn02 +4HS+ 2e- Mn02 + Rd - Mn2++ Ox + 2e- An intermediate which is likely to be of considerable importance is the superoxide ion generated by the reduction of oxygen in non-aqueous soh- ti0ns.8~7~~ 0; 0 2 + e - - This species will react with organic substrates leading, for example, to the R.I.C. Reviews 102 formation of organic peroxides in a reductive process.Electrochemical auto- catalysis can also be observed.g0 The reactions which have been listed in this section have been formulated as though the intermediates are dissolved in solution. In fact, in many cases these intermediates will be adsorbed on the electrode surface and this will be particularly true for radical intermediates, for example as shown by the cis addition of methyl radicals, generated in the Kolbe reaction, to butadiene.91 Again the formation of methanol by oxidation of acetate ions in basic solu- tions is best explained by the biradical reactiong2 For the complete oxidation of hydrocarbons to carbon dioxide on platinum metal catalysed fuel-cell electrodes, the participation of OH radicals adsorbed on the surface has also been postulatedg3 and in these mechanisms one can indeed see an application of the original ideas of the mechanism of electro- oxidation.Again, it is possible to cite reduction reactions when using catalytic- ally active electrode materials such as platinum black where the original mechanisms involving hydrogen would be more appropriate (cf. hydro- genat ion). 94 AGRd h - -/- AG:;. CORRELATIONS BETWEEN ELECTROCHEMICAL DATA AND OTHER PHYSICAL PROPERTIES For any electrode reaction, a thermodynamic cycle, such as that shown in Fig.7 for the reduction of R, may be drawn up. From the cycle, it may be seen that - FA (6) where AGRd is the overall free energy change for the reduction R + e-+R-, AGEI. and AGE;. are the free energies of solvation of R and R- respectively and A is the electron affinity of R. Since the formal electrode potential, EO, (the potential on the scale of the normal hydrogen electrode) is given by EO = - AGRd/F and for a reversible reduction Eo E= EFd, it may be con- cluded that (7) Ey = A + - 1 [AGE,. - AGri.1 F Similarly for an anodic reaction we can derive the expression Fig. 7. Thermodynamic cycle for the reduction of compound R 103 Fleischmann and Pletcher where I is the ionization potential of R and AGS.is the free energy of solva- tion of R+. It is important to note that these expressions only hold for reversible electrode reactions in the absence of fast following chemical reactions removing R+ or R- and in the absence of complex formation between R+ or R- and the solvent, or ion-pair formation between R+ or R- and the base used directly to correlate Ey with ionization potentials95 and EFd with electrolyte ions. With these reservations in mind, equations 7 and 8 may be electron affinities.95 The results of such correlations are shown in Figs 8 and 9 pyrene, chrysene, phenanthrene and triphenylene. Both EFd and Ey were for a series of aromatic hydrocarbons-anthracene, 172-benzanthracene, obtained in acetonitrile and the Ey were obtained at high voltage sweep rates in order to eliminate the following reaction which R+ would undergo.To within experimental error, linear plots are obtained for both correlations. Therefore, it must be concluded that the free energy terms in equations 7 and 8 are either reasonably constant or vary linearly with potential. It would not be surprising if these free energy terms remain constant for series of similar large molecules. As the electron affinities and ionization potentials may be calculated for aromatic systems, for example by simple Hiickel theory, the correlations may also be made between E, and the orbital energy, i.e. with the level of the lowest unfilled or highest filled molecular orbital.96~97 For alternate hydro- carbons these levels are symmetrically placed with respect to the non-bonding orbital and this fact also allows correlation with the frequency for the first electronic tran~ition.~~ It has also been shown that EFd and Ey for the discharge of aromatic hydrocarbons may be correlated with the frequency of the charge-transfer transition of the molecules in the presence of donor or acceptor molecules respectively (e.g.hexamethylbenzene99 and tri- fluorenoneloo). These correlations are also based on a relation between the frequency of the transition and I or A . A number of such correlations for a variety of compounds has appeared in the literature and, despite the fact that in some cases half-wavepotentials for irreversible processes or processes including a coupled reaction have been applied, good correlations have been obtained. For irreversible cases, the term AGEi.- FA determines the energy of the final state of the reaction while AGEI. - F+ determines the energy of the initial state. There will therefore be a linear relation (in the first approxima- tion) between these energy terms and the free energy of activation, i.e. a is an example of such a relationship since + changes the free energy of activa- ‘linear free energy correlation’. Indeed, equation 4 relating log I to azF+/JRT tion in a linear manner. Correlations between Ey and A will therefore still be observed provided AGEi. and AGE,. remain constant, the reactions have similar pathways and the heats of adsorption of R and R- are also independent 104 R .I. C. Reviews I .6 I .4 + 1.2 E px (V) Fig. 8 (upper) and Fig. 9 (lower). Correlations of €yx with ionization potential and Ey with electron affinity for a series of aromatic hydrocarbons of the nature of R, since these heat terms also determine the energies of the initial and final states. Similar linear free energy relationships can be obtained between half-wave potentials for compounds RAX and the structure of the compounds as expressed by the Hammett-Taft po or p*o* terms, i.e. (AE+)X == (E+)X - (E&)H = fn,R'X for a benzenoid series ( i e . A is an aromatic series) or (AE+)x = (E+)x - (E+)H = P*~,RQ*X (9) Fleischmann and Pletcher I .8 (10) 105 for a non-benzenoid series, where (E,)x and (E& are the half-wave poten- tials of the substituted and unsubstituted molecule respectively (substituent X) and p n , ~ , the reaction constants, measure the susceptibility of R to the effect of X and are independent of the nature of X, while OX, the substituent constants, depend on the position and nature of X but are independent of R.These substituent constants are determined by solution phase reactions. Linear correlations of the type 9 and 10 therefore relate the electrode process to other known reactions and indicate uniformity of reaction mechanism; their uses have been reviewed elsewhere.lo1 102 As has already been explained (p. go), the development of electronic potentiostats allows the systematic application of this idea, for example, I 04 SOME FURTHER EXAMPLES OF ELECTROCHEMICAL REACTIONS OF ORGANIC COMPOUNDS This section is divided into two parts; the first illustrates the importance of controlling the electrode potential and solution composition to give selective reactions and the second illustrates further reaction types, many of which could be applied to synthesis.CHl-NH, 2e- t 2Hi-, @OH OH The control of electrode reactions As early as 1898 it was shown that nitrobenzene could be reduced to phenyl- hydroxylamine at low negative and to aniline at more negative potentials. CH= NH CH=NOH VH + H,O / / OH Q"" 2e-+2H+- OH I05 I 06 R.I.C. Reviews It is interesting to note that the stereochemistry of an electrode reaction can also depend on the potential.For example, in the reduction of benzil the ratio of cis to trans stilbenediol formed is potential dependent.107 Furthermore, the course of a reaction may be changed by applying a pulse profile such that several different reactions take place successively and repetitively. An example is the alternate generation of carbanions and carbo- nium ions leading to electrochemiluminescence. An application to synthesis is the variation in the reduction mechanism of nitrobenzene caused by reoxidizing the intermediate, phenylhydroxylamine, before it has time to azoxybenzene NO / 0 1 (y="o (y-yJ t H H hydrazobenzene H2N m N H 2 benzidine '08 diffuse away from the electrode surface.Thus, by choosing the potential for the reduction step it is possible to accumulate the coupling product, azoxy- benzene, or its reduction product. In effect, by applying a suitable square- wave potential/time profile, benzidine may be prepared directly from nitro- benzenelog and it is clear that many other products could similarly be made directly from nitrobenzene by choosing suitable pulse profiles. It will be apparent that the duration of the pulses is an additional control variable. A question of key importance for achieving a selective reaction is whether the desired product is itself electroactive at the potential at which the reaction is carried out. The half-wave potential for the substrate and product under the experimental conditions will be a useful guide. For example, it is clear that it will always be difficult to find selective oxidation routes for the forma- tion of phenols or acetates since these are usually more easily oxidized than the parent aromatic hydrocarbon.On the other hand, cyanation of aromatic compounds is a favourable reaction since the products have more positive half-wave potentials. Polarographic data will not necessarily be an accurate guide for the course of a reaction since large scale preparations are carried out on a longer time scale and can, therefore, be affected by disproportiona- Fleischmann and Pletcher 107 tions. Thus, dihydroaromatic compounds may be formed at the potential at which only one electron is added to the hydr~carbon.~~ The control of electrode reactions by the correct choice of the solvent has already been implied, for example in the discussion of Fig.6 and in the discussion of the protonation of radical anions in aprotic solvents. In aqueous solution, the pH is of key importance and determines whether protonated or unprotonated species will be electroactive. These effects are also apparent in aprotic media where the presence of base (pyridine) can have a marked effect on the electrode mechanism. Examples which have been observed include the simple, two-electron oxidation of anthracene in the presence of pyridine to give a stable product in absence of pyridine formation e - ' y ? ' ' C H M e M e 0 M e o a c ' y H Me dimer -Meon 3::; & M e 0 = I O M e 108 O M e CH ,CHMe M e 0 Q O M e OMe R.I.C.Reviews Oxidat ions In addition to the reactions listed earlier, other oxidation processes can be formulated as proceeding through carbonium ion intermediates, for example, ring closures, expansions and contractions CH,CH,CH,COO- -2e- -'02- CH,CH,CH; r CH2- CH, 4- other products 40 75 l R \ / Me0 CH2 and the methoxylation of furans may be formulated as 1 Me0 O : M e On the other hand, many other oxidation reactions may be formulated as proceeding through radical intermediates. For example, the Kolbe and the related Brown-Walker reactions have in recent years been used for several 109 Fleischmann and Pletcher interesting syntheses which proceed via radicals.Examples are COOH 114 2 EtOOC (CH,),COO- (EtOOC),CH - - CH, - C H (COOE t)2 formation - OEt -Ze--2Cok EtOOC(CH 2)GCOOEt Other examples include the polymerization of vinyl acetate, vinyl chloride and methyl acrylate initiated by radicals formed by the Kolbe reaction.116 Radicals may be prepared by the anodic oxidation of the sodium derivative of diethyl malonate,46 ethyl acetoacetate,46 ethyl phenylacetate46 and nitro- paraffins117 as well as less common anionic species. In the absence of reactive substrates, the radicals will dimerize, but clearly such radicals are particularly useful in synthesis, as in their reactions with alkenic bonds. A reaction scheme for the oxidation of diethyl malonate in the presence of vinyl ethyl ether118 is OEt OEt I I cH(COOEt), a dH(COOEt), CH2=CHoEt- (EtOOC),CH-CH,-tH I EtOOC I-'- I + EtO G O E t OEt (E t OOC), - CH - CH ,-CH (OEt), - So'Vent (EtOOC)2CH-CH2-CH- OEt EtO' EtO' 110 R. I .C. Reviews This scheme illustrates that radicals may dimerize, hydrogen abstract or be oxidized further to give carbonium ions which may react with the solvent, cyclize, or lose a proton, and is an example of the build up of complex mole- cules from simple starting materials. It is possible to use other alkenes such as cyclohexene, styrene or butadiene and the other carbanion species listed above .I18 Several examples of elimination reactions have been reported during electrochemical oxidations. Examples include the oxidation of p-substituted phenols.OH r- 1 - HBr MeOH 0 1 CH,=CH-CHO+4H++ H2O- CH,-CH=CH2+4Hg2++ U A number of indirect oxidations, where inorganic intermediates generated at the electrode react with substrates [cf. Mg+, Al+ in reductions; MnOz, Co3+ in oxidations], have been reported. Examples include the electro- chemical modification of the Clauson-Kaas methoxylation of furan where the bromine is generated at the anode,l21 M e 0 O k M e the oxidation of propylene where the mercury(I1) is regenerated electro- chemically, 122 the oxidation of propylene to propylene oxide by electrochemically generated hyp~chloritel~~ and the oxidative reaction of alkenes with carbon monoxide and methanol 121 I I9 Fleischmnnn and Plefcher 2 H g r 11 1 in the presence of platinum carbonyls to give methyl esters of a,,B unsaturated acids1Z4 C6HSCR =CH, +CO +-OMe - C6H,CR= CHCOOMe Another example of an anodic reaction which is extensively used industrially is the perfluorination of aliphatic hydrocarbons in anhydrous hydrogen fl~oride.1259~26 The mechanism of these reactions, which clearly must be complex, is not fully understood.Reductions Processes analogous to those described for the reduction of aromatic hydro- carbons are also observed in heterocyclic series. Thus the reduction of pyrimidine is explained by the scheme127 H H+ I29 R.Z. C. Reviews 112 In heterocyclic series, ring expansion has also been observed131 131 and several different ring closures have been reported, amongst which are H OH I H I 4e:i2+2 COO H COOH CH = C(COOH), I I Non-heterocyclic ring closures which have been observed include I32 I33 I34 I33 which presumably arises from intramolecular radical coupling.An example which may be attributed to carbanionic attack on an activated alkenic bond I35 mo ,++ CH, -CH-CH,-CH-CH, -H,O NH ___ N' H2 L 'OH CH3-C == CH -C-CH, 6e-+5H+ CH3-CH-CH2-CH-CH3 c HN'OH HONH NH2 HONH - ? = + I I CH2 CH = CHCOOEt 2H+ I is the intramolecular hydrodimerization, , CH.! II =CHCOOEt I ,CH CH2 'CH ___L I m CH2CooEt CHI ,C-H - C-HCOOEt CH2COOEt \ CH2 CH2 \ /CH =CHCOOEt CH2 Fleischmann and Pletcher 113 The reactions of activated alkenes, such as hydrodimerizations, may be used for a wide variety of organic syntheses many of which would lead to industrially useful products.What is more, such reactions can often be carried out in aqueous media using tetraalkylammonium p-toluenesulphonate (McKee’s salts) as electrolyte and solubilizing agent. As the work up and extraction of products from electrosynthesis will frequently be simple compared to conventional processes because of the absence of reagents, it is likely that selective electro-organic syntheses will find increasing application as additional and alternative reaction routes. REFERENCES 1 F. Fichter, Organische elektrochemie. Dresden: Steinkopf, 1942. See also M. J. Allen, Organic electrode processes. New York : Reinhold, 1958.2 C. L. Perrin, Prog. phys. org. Chem., 1965, 3, 165. 3 P. Zuman, Prog. phys. org. Chem., 1967, 5, 1. 4 S. Wawzonek, Analyt. Chem., Reviews section, 1958, 1960, 1962, 1964. 5 D. J. Pietrzyk, Analyt. Chem., Reviews section 1966-1968. 6 0. H. Muller, Chem. Rev., 1963, 24, 93. 7 P. Zuman, J. polarogr. SOC., 1967, 13, 53. 8 S. Wawzonek, Science, N. Y., 1967, 155, 39. 9 H. Lund, Oster. Chem. Ztg, 1967, 68, 43, 152. 10 S. Swann in Techniques of organic chemistry (A. Weissberger, ed.). New York: Inter- science, 1956. 11 S. Swann, Electrochem. Technol., 1961, 1, 308; 1966,4, 550; 1967,5, 53, 101, 393,467 547; 1968,6, 59. 12 F. D. Popp and H. P. Schultz, Chem. Rev., 1962, 62, 19. 13 P. J. Elving and B. Pullman, Adv. chem. Phys., 1961, 3, 1.14 N. Elming, Adv. org. Chem., 1960, 2, 67. 15 N. L. Weinberg and H. R. Weinberg, Chem. Rev., 1968, 68,449. 16 C. Papon-Desbuquois, Revue Znst. f r . Pe‘trole, 1964, 19, 627. 17 A. P. Tomilov, Russ. chem. Revs, 1961, 639. 18 B. C. L. Weedon, Adv. org. Chem., 1960, 1, 1. 19 B. E. Conway and A. K. Vijh, Chem. Rev., 1967, 67, 623. 20 L. Eberson, in Chemistry of the carboxyl group (ed. S. Patai). New York : Interscience, 1968. 21 A. P. Tomilov and Ya. M. Fioshin, Russ. chem. Revs., 1967, 32, 30. 22 K. J. Vetter, Electrochemical kinetics. New York: Academic Press, 1967. 23 P. Delahay, New instrumental methods in electrochemistry. New York : Interscience 1967. See also Double layer and electrode kinetics. New York: Interscience, 1965. 24 B. D.Damaskin, PrincQles of current methods for the study of electrode reactions. New York: McGraw-Hill, 1967. 25 L. Meites, Polarographic techniques. New York : Wiley, 1965. 26 J. O’M. Bockris, Modern aspects of electrochemistry, vol. 1. London: Butterworths, 1954. 27 P. Delahay in Advances in electrochemistry and electrochemical engineering, vol. 1,233. New York: Interscience, 1961. 28 R. BrdiEka, Colln Czech. chem. Commun., 1955, 20, 387. 29 T. A. Gough and M. E. Peover, Polarography-1964, 1017. London: Macmillan, 1966. 30 R. S. Nicholson and I. Shain, Analyt. Chem., 1964, 36, 706. 31 W. M. Schwarz and I. Shain, J. phys. Chem., 1965, 69, 30. 32 A. C. Riddiford in Advances in electrochemistry and electrochemical engineering, vol. 4, 47. New York: Interscience, 1968.33 For references see W. J. Albery, M. L. Hitchman and J. Ulstrup, Trans. Faraday Soc., 1968,64, 2831. 34 R. Koopmann, Ber. Bunsenges. Phys. Chem., 1968, 72, 32. 35 R. N. Adams, J. electroanal. Chem., 1964, 8, 151. 36 A. H. Maki and D. H. Geske, J. Am. chem. SOC., 1960,82,2671. 37 T. Kuwana and J. W. Strojek, Discuss Faraday SOC., 1968, 45, 134. 38 D. F. A. Koch and D. E. Scaife, J. electrochem. SOC., 1966, 113, 302. 39 T. Kuwana, Electroanalytical chemistry, vol. 1, 197. London : Arnold, 1966. 40 W. J. Koehl, J. Am. chem. SOC., 1964,86,4687. R.I.C. Reviews 114 41 V. D. Parker, Chem. Cornniun., 1968, 1164. 42 N. Sato, T. Sekine and T. Sugino, J . eiectrochem. Soc., 1968, 115, 242. 43 H. Lehmkuhl, R. Schaefer and K.Ziegler, Chemie-lngr-Tech., 1964, 36, 612. 44 F. J. Vermillion and 1. A. Pearl, J. electrochem. SOC., 1964, 111, 1392. 45 R. Berkey, S. Wawzonek, E. W. Blaha and M. E. Runner, J . electrochem. Soc., 1956, 103,456. 46 T. Okubo and S . Tsutsumi, Bull. chem. SOC. Japan, 1964,37, 1794. 47 P. Zuman, D. Barnes and A. Ryvolova-Kejharova, Discuss. Faraday SOC., 1968,45,202. 48 M. Finkelstein, R. C. Peterson and S . D. Ross, Electrochim. Acta, 1965, 10, 465. 49 F. Lantelme and M. Chemla, Electrochim. Actu, 1965, 10, 657. 50 E. M. Marlett, Ann. N.Y. Acad. Sci., 1965, 125, 12. 51 R. V. Lindsey and M. L. Peterson, J. Am. chem. SOC., 1959,81, 2073. 52 J. W. Breirenbach and Ch. Srna, Macromolecular chemistry, Proceedings, Montreal, 1961. London: Butterworths, 1962.53 N. S. Hush and K. B. Oldham, J . electroanal. Chem., 1963, 6, 34. 54 For refs see M. E. Peover, Electroanalytical Chemistry, vol. 2 , l . London: Arnold, 1967. 55 S. Wawzonek, R. Berkey, E. W. Blaha and M. E. Runner, J. electrochem. SOC., 1955, 102, 235. 56 S. Wawzonek and D. Wearring, J. Am. chem. SOC., 1959, 81, 2067. 57 R. Dietz and M. E. Peover, Trans. Faraduy SOC., 1966,62, 3535. 58 S . Wawzonek, R. C. Duty and J. H. Wagenknecht, J. electrochem. SOC., 1964, 111, 74. 59 T. Sekine, A. Yamura, K. Sugino, J. electrochem. SOC., 1965, 112, 439. 60 M. M. Baizer, J . electrochem. SOC., 1964, 111, 215. 61 M. M. Baizer and J. D. Anderson, J. electrochem. SOC., 1964, 111, 226. 62 M. M. Baizer, J . electrochem. SOC., 1964, 111, 223.63 M. M. Baizer and J. D. Anderson, J. org. Chem., 1965, 30, 3138. 64 R. E. Visco and E. A. Chandross, J. Am. chem. SOC., 1964, 86, 5350. 65 A. Zweig and A. K. Hoffmann, J. org. Chem., 1965, 30, 3997. 66 M. E. Peover and B. S . White, J. electroanal. Chem., 1967, 13, 93. 67 L. Eberson and K. Nyberg, Tetrahedron Lett., 1966, 2389. 68 L. Eberson and S. Nilsson, Discuss. Faraday SOC., 1968, 45, 242. 69 V. D. Parker and B. E. Burgert, Tetrahedron Lett., 1965, 4065. 70 L. Eberson, Chemistry of the carboxyl group. London: Interscience, 1968. 71 L. Eberson and K. Nyberg, Acta. chem. scand., 1964, 18, 1567. 72 L. L. Miller and A. K. Hoffmann, J. Am. chem. SOC., 1967,89, 593. 73 M. Fleischmann and D. Pletcher, Tetrahedron Lett., 1968, 6255. 74 K. K. Barnes and C.K. Mann, J . org. Chem., 1967,32, 1474. 75 E. J. Corey, N. L. Bauld, R. T. La Londe, J. Casanova and E. T. Kaiser, J. Am. chem. SOC., 1960, 82, 2645. 76 S. Wawzonek and J. H. Wagenknecht, J. electrochem. SOC., 1963, 110, 420. 77 S. Wawzonek and R. C. Duty, J. electrochem. SOC., 1961, 108, 1135. 78 H. Schmidt and J. Noack, 2. anorg. allg. Chem., 1958,296, 262. 79 S . D. Ross, M. Finkelstein and R. C. Peterson, J. Am. chem. SOC., 1967, 89, 4088. 80 C. D. Russel, Analyt. Chem., 1963, 35, 1293. 81 J. F. O’Donnell and C. K. Mann, J. electroanal. Chem., 1967, 13, 157. 82 S. Wawzonek and T. W. McIntyre, J . electrochem. SOC., 1967, 114, 1025. 83 H. W. Sternberg, R. E. Markby and I. Wender, J. electrochem. SOC., 1966, 113, 1060. 84 R.A. Benkeser and E. M. Kaiser, J. Am. chem. SOC., 1963,85, 2858. 85 M. D. Rausch, F. D. Popp, W. E. McEwen and J. Kleinberg, J . org. Chem., 1956, 21, 212. 86 T. T. Tsai, W. E. McEwan and J. Kleinberg, J. org. Chem., 1960, 25, 1186. 87 M. Fleischmann et al., unpublished work. 88 M. E. Peover and B. S . White, Chem. Commun., 1965, 183. 89 D. L. Maricle and W. G. Hodgson, Analyt. Chem., 1965; 37, 1562. 90 M. E. Peover et af., unpublished work. 91 W. G. Smith and H. G. Gelde, J. Am. chem. SOC., 1961, 83, 1355. 92 G. Atherton, M. Fleischmann and F. Goodridge, Trans. Faraday SOC., 1967, 63, 1468. 93 E. Gileadi and B. Piersma, Modern aspects of electrochemistry, vol. 4, 47. London: Butterworths, 1966. 94 S. H. Langer and H. P. Landi, J. Am. chem.SOC., 1963, 85, 3043. 95 B. Case, N. S. Hush, R. Parsons and M. E. Peover, J. electroanal. Chem., 1965,10,360. 96 A. Streitwieser and J. Schwager, J. phys. Chem., 1962, 66, 2316. 97 G. J. Gleicher and M. K. Gleicher, J . phys. Chem., 1967, 71, 3693. 98 A. T. Watson and F. A. Matsen, J. chem. Phys., 1950,18, 1305. 99 M. E. Peover and J. D. Davies, Polarography--1964, 1003. London: Macmillan, 1966. 115 Fleischmann and Pletcher 100 E. S. Pysh and N. C. Yang, J. Am. chem. SOC., 1963,85, 2124. 101 P. Zuman, Substituent efects in organic polarography. New York: Plenum, 1967. 102 F. Haber, Z. Elektrochem., 1898, 4, 506. 103 L. Holleck and H. Schmidt, 2. Elektrochem., 1955, 59, 56, 1039. 104 H. Lund, Acta chem. scand., 1964, 18, 563. 105 E. R. Atkinson, H. H. Warren, P. I. Abell and R. E. Wing, J . Am. chem. SOC., 1950, 72, 915. 106 P. Ruetschi and G. Trumpler, Helv. chim. Acta, 1953, 36, 1649. 107 A. Vincenz-Chodkowska and Z . R. Grabowski, Electrochim. Acta, 1964, 9, 789. 108 M. Fleischmann, I. N. Petrov and W. F. K. Wynne-Jones, Electrochemistry-Pro- ceedings of 1st Australian Conference, 500. London : Pergamon, 1964. 109 H. Lund, Acta chem. scand., 1957, 11, 1323. 110 J. J. O’Connor and I. A. Pearl, J . electrochem. SOC., 1964, 111, 335. 111 A. J. Baggaley and R. Brettle, J. chemc. So. (C), 1968, 2055. 112 N. Clauson-Kaas, F. Limborg and P. Dietrich, Acta chem. scand., 1952, 6, 545. 113 A. F. Vellturo and G. W. Griffin, J. Am. chem. Soc., 1965, 87, 3021. 114 M. Ya. Fioshin, A. J. Kamneva, Sh. M. Itenburg, L. I. Kazakova and Yr. A. Ershov, 115 M. Ya. Fioshin, L. A. Mirkin, L. A. Salmin and A. G. Kornienko, Zh. vses. khim. Khim. Prom., 1963, 263. Obshch., 1965, 10, 238. 116 W. B. Smith and H. Gilde, J . Am. chem. SOC., 1960,82, 659. 117 C. T. Bahner, Ind. Engng Chem., 1952,44, 317. 1 18 H. Schafer, Meeting of Gesellschaft Deutscher Chemiker. Fachgruppe, ‘Angewandte Elektrochemie’, Lindau, October 1968. 119 M. D. Hawley and R. N. Adams, J. electroanal. Chem., 1964, 8, 163. 120 W. K. Snead and A. E. Remick, J. Am. chem. SOC., 1957,79, 6021. 121 N. Clauson-Kaas, F. Limborg and K. Glens, Acta chem. scand., 1952, 6, 531. 122 F. Goodridge, R. E. Plimley, I. D. Robb and J. M. Coulson, British Patent application 16364167. 123 J. A. M. LeDuc, US patent 3 288 692, 1966; Chem. Abstr., 1967, 66,4323813. 124 T. Inoue, S. Tsutsurni, J. Am. chem. SOC., 1965, 87, 3525. 125 J. Burdon and J. C . Tatlow, Adv. fluorine Chem., 1960, 1, 129. 126 S. Nagase, Fluorine chem. Revs., 1967, 1, 77. 127 D. L. Smith and P. J. Elving, J. Am. chem. SOC., 1962,84,2741. 128 H. Lund, Acta chem. scand., 1963, 17, 2325. 129 H. Lund, Proc. 4th Int. Congr. on Polarography, Prague, 1966. 130 H. Lund, Acta chem. scand., 1959, 13,249. 131 H. Lund, Tetrahedron Lett., 1965, 3973. 132 T. Wohlfarth, J. prakt. Chem., 1902, 65, 295. 133 H. Lund, Conference on electroorganic chemistry, Durham, N. Carolina, USA, 1968. 134 J. Tafel and E. Pfeffermann, Ber. dt. chem. Ges., 1903, 36, 219. 135 J. D. Anderson, M. M. Baizer and J. P. Petrovich, J. org. Chem., 1966, 31, 3890. R.I.C. Reviews 116