7 Electro-organicChemistry ByJ. H. P. UTLEY Departmentof Chemistry Queen Mary College Mile End Road London E 1 4NS 1 Introduction In this Report the emphasis is on new and preparatively significant reactions and related observations.* Two authoritative and comprehensive books have appeared during 1975.Iv2 2 Anodic Processes Oxidationof Neutral Organic Compounds.-Full details have been reported3 of the construction and use at 20-50 A of an undivided 800 cm2concentric capillary-gap cell which uses a graphite anode. A 3000 cm2 version with a production volume of cu. 1kg h-’ has also been built and tested. These cells may be purchased and can be used in combination with standard large-scale glassware and commercially available power supplies. Progress continues to be made on the functionalization viu anodic oxidation of normally unreactive aliphatic molecules.In fluorosulphuric acid solution containing potassium fluorosulphate alkanoic acids are converted4 at a platinum anode and at 273 K into lactones and unsaturated cyclic ketones. Proton loss is at positions remote from the carboxylic acid group. Immediate work-up allows isolation of lactones but delay leads to further chemical reaction (Scheme 1).In MeCN-LiC10 solutions U Me(CH,),CO,H work-up 1 ‘Techniques of Chemistry Vol. 5 Parts 1 and 2 Techniques of Electroorganic Synthesis’ ed. N. L. Weinberg Wiley-Interscience New York 1974. 2 S. D. Ross M. Finkelstein and E. Rudd ‘Anodic Oxidation’ Academic Press New York 1975. 3 L. Cedheim L. Eberson B.Helgee K. Nyberg R. Servin and H. Sternerup Actu Chem. Scand. (B), 1975,29,617. D.Pletcher and C. Z. Smith J.C.S. Perkin I 1975,948. * Help in locating the several hundreds of references which were reviewed came from a current-awareness service provided by Anne Jarvis and Stephen Shaw (Queen Mary College Library). 151 152 J.H.P. Utley containing a little water aliphatic ketones are oxidized5 at relatively low anodic potentials [ca. 2.2 V (us. AglAg+)]. As with carboxylic acids proton cleavage from straight-chain compounds is remote from the carbonyl group and the expected acetamides are formed.5a However from the radical-cations of cy -branched ketones comparatively stable carbenium ions may be produced by carbon-carbon bond cleavage.5b Similar considerations apply to the oxidation of secondary and tertiary alkylphenylcarbinols.For instance,5c acetophenone (83%)is obtained from the anodic oxidation in acetonitrile of 1-phenylethanol (via H' loss) whereas benzaldehyde (61%) and N-t-butylacetamide (47%) are obtained from l-phenyl- 2,2-dimethylpropanol (via Me$' loss). Examples of these reactions are given in Scheme 2. MeCO(CH ,),Me 2.2 V (us. AglAg') + MeCO(CH,),CH(Me)NHCOMe MeCN-LiCIO (40%) 2.5 V (us. AglAg+) Bu'NHCOMe + MeC0,H Bu'CO Me ---+ MeCN-LiCIO (80% (25 %) Scheme 2 A clean and simple alternative to cleavage of glycols using lead tetra-acetate and periodic acid has been suggested by Shono and co-workers.6 Anodic cleavage at readily accessible potentials at a carbon rod anode gives respectable yields of the corresponding acetals (e.g.Scheme 3). 2.02 V (us s.c.e.) CH(OMe) MeOH-Et,NOTs graphite anode (51 %) (14 %) Scheme 3 Since P. S. Skell's early work on the anodic cleavage of alkyl halides [cf. Ann. Reports (B),1969,66,2197] much work has been directed towards elucidating the detailed mechanism of the reaction. Skell's work suggested that a 'hot' lightly solvated carbenium ion was formed following ejection of halogen atom from the alkyl halide radical-cation. Laurent's more recent and comprehensive studies involving comparisons of anodic and solvolysis reactions [cf.Ann. Reports (B),1974 71,2231 suggest that for iodides in acetonitrile the halogen loss is solvent-assisted. This theme is continued in an investigation7 of the products of anodic oxidation in MeCN-LiClO of the series of 2-adamantyl bromides (1)-(3).Compound (3)is not Me Me Me ..iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii &NHCOMe gBr &Br DBr Me MeCONH (1) (2) (3) (4) (5) (a)J. Y. Becker L. R. Byrd and L. L. Miller J. Amer. Chem. Soc. 1975,97,853;(b)J. Y. Becker L. L. Miller and T. M. Siegel ibid. p. 849; (c)E. Mayeda ibid. p. 4012. T. Shono and Y. Matsumura J. Amer. Chem. SOC.,1975 97 2546. F. Vincent R. Tardivel and P. Mison Tetrahedron Letters 1975 603. Electro -organic Chemistry oxidized at S3.1 V (us.AglAg’). Compounds (1)and (2) are oxidized at 2.5 V but whereas (2) predominantly undergoes one-electron cleavage to give after work-up the acetamide (4),the unsubstituted bromide (1)undergoes two-electron cleavage to give the acetamide (5);from which the bromine is not lost.It has been pointed out that (2) is solvolysed thirty times faster than (1);this result has been interpreted in terms of Laurent’s hypothesis. However for the iodides an alternative mechanism has been shown to be worth very serious consideration (Scheme 4).Iodine is oxidized in acetonitrile consuming 2 F mol-’ and giving an ill-defined ‘iodinium species’ that can be loosely represented as ‘I”. Addition of 2-iodo-octane to a freshly electro- lysed solution of iodine in acetonitrile resulted* in the regeneration of iodine and the formation after work-up of 2-acetamido-octane (6)(54%) and 3-acetamido-octane (7) (22%). The plausibility of the mechanism in Scheme 4 is thereby established although both direct and indirect oxidation must presumably operate at the begin- ning of an electrolysis.1.8v Me(CH,),CH(MeU ‘2 MeCN ’ ’I + Me(CH,)&HMe 2F mol~’ JMeCN-H,O Me(CH,),CH(Et)NHCOMe + Me(CH,),CH(Me)NHCOMe (7) (6) Scheme 4 The oxidation of aliphatic amines has also been much studied and at conventional electrochemical concentrations amine radical-cations can undergo cleavage’ to carbenium ions (Scheme 5). At very high concentrations of t-butylamine a useful preparation of 2,2‘-azobutane (8) results’’ from the enhanced effectiveness of nucleophilic trapping of the radical-cation uis h vzs cleavage (Scheme 5). What is H2N + Me,e -+ Me,COH + Me,C=CH Me,&H / Me3&d l:bMe,C N =NC Me (8) Scheme 5 probably the first example of the capture of the hitherto unknown aza-ally1 cation (10)comes from a study” of the anodic oxidation of the aziridine (9).At 3 F mol-’ propiophenone is the major product of constant-current oxidation at platinum in methanol-NaC10,-Na2CO at 273 K. (It is readily oxidized further to the corre- sponding acetal). Treatment of a freshly electrolysed solution in the manner described in Scheme 6 gave (12) which suggested the initial formation of (11)via the L. L. Miller and B. F. Watkins Tetrahedron Letters 1974 4495. K. K. Barnes and C. K. Mann J. Org. Chem. 1967,32 1474. lo A. U. Blackham S. Kwak and J. L. Palmer J. Elecfmchern. Soc. 1975,122 1081. l1 P. G. Gassman and I. Nishiguchi J. Amer.Gem. SOC.,1975 97 1600. 154 J. H.P.Utley H N 3Frnol-I LiAIH,, R anode (11) Et20 PhCH(Et)NHMe + PhC(OMe),Et qpr 0 "C MeOH-NaCIO, (12) Na,CO . + 7 Ph(Et)C-N=CH2 ----*Ph(Et)C=NCH,OMe MeOH (10) (11) Scheme 6 aza-ally1 cation (10). Many attempts to purify an unstable product of electrolysis to 3 F mol-' gave material of 60-70% purity and the spectral data were consistent with structure (11).One of the limitations of the use of N-methyl-amides as solvents for anodic electrolyses is the ready oxidation to the isomerically stabilized cations of the type R'CONR2CH2+.This reaction has been put to good preparative use in the formation12of novel phosphonium salts and cyclic oxammonium salts (Scheme 7). + HCON(Me)CH,PPh ClO Scheme 7 Shono's work on the oxidative cleavage of enol acetates has been extended [cf.Ann. Reports (B) 1974 71 2231 and it is particularly noteworthy that the competition between two mechanistic pathways is greatly infl~enced'~ by the choice of supporting electrolyte (Scheme 8). Unsaturated ethers may also be cleaved K" Et,NOTs; nil KOAc; 60% HOAc KOAc or Et,NO% -H+,X Et,NOTs ;90 % KOAc; 25% Scheme 8 l2 M. Genies Bull. Soc. chim. France 1975 389. l3 T. Shono M. Okawa and I. Nishiguchi J. Amer. Chem. Soc. 1975,97,6144. Electro-organic Chemistry anodically. For instance cis-or trans-dimethoxystilbenes are oxidized' at platinum [1.8 V (us. AglAgI)] in MeCN-LiClO or DMF-LiC10 to (13)and (14). In this case (13) (14) the proportions of products are markedly dependent on concentration.From electrolysis of the cis-isomer in DMF the ratio of (13):(14) is ca. 1 at a starting concentration of 2.5 X lo-' mol I-l whereas the ratio is 6.5 at 0.5 X lo-' mol I-'. A variant of the long-established anodic substitution of furans has been introduced and presented as a new route" to substituted maleic anhydrides (Scheme 9). The method seems to be simple and to offer soinewhat better yields than previously used chemical methods. Several examples are given other than that illustrated in Scheme 9. Et Et Et Et Pt anode Jones Oxidation' NaHCOJaq.) 6 (84%) Scheme 9 Several interesting applications have been reported of the anodic addition of alkoxide to aromatic compounds.The synthesis of 1,4,9,12-tetraoxadispiro[4,2,4,2]tetradeca-6,13-diene (15) has been repeated,'6a and it appears that the reaction does not go to completion by electrolysis as originally reported.'6b The intermediate formed anodically must be converted by acid into the desired product (Scheme 10). Virtually identical conditions of electrolysis were II MeO. ,OCH,CH,OH nn k.2 MeOH-KOH(l%) \ ).' Et,O W OCH,CH,OH 0x0 Scheme 10 chosen for the key cyclization and addition step in an elegant and short synthesis" of 4a,Sa-homonaphthodiquinone(16) (Scheme 11). In similar vein amperostatic anodic oxidation in an undivided cell of 4-methoxyphenol in MeOH-LiClO led to l4 M. A. Michel P. Martigny and J. Simonet Tetrahedron Letters 1975 3143. J.Froborg G. Magnusson and S. Thoren J. Org. Chem. 1975,40 122. l6 (a)P. Marguretha and P. Tissot Helu. Chim. Actu 1975,58 933; (b) C. G. Fink and D. B. Summers Trans. Electrochem. Soc. 1938,74,325. f7 W. Bornatsch and E. Vogel Angew. Chem. Internat. Edn. 1975,14,420. 156 J. H. P.Utley f Me0 OMe Pt anode Mt:O&Me H,O+ & + MeOH-KOH(l%) Me0 OMe c 55 % Scheme 11 the preparation" of 4,4-dimethoxycyclohexa-2,5-dienone(17) in 97% isolated yield and with high purity. It is claimed that this procedure is much better than the alternative oxidation with Ce'". 0 Me0 OMe 0 (17) Many bright ideas for electrochemical reactions are foiled because during the key electrolysis fouling of the electrode prevents the passage of current.In this context a remarkable claim has been made. For the oxidation at platinum of benzhydrol to ben~ophenone,~~ in strongly alkaline solution the addition of cationic surfactants increases current efficiency from 1% to an unbelievable 326% (presumably based on 2 F mol-l although this is not clear). Excluding the sort of explanation which also allows Saints to cross the Irish Sea on floating altar stones it may well be that the cationic surfactant is in alkaline solution merely promoting a hydride transfer (Scheme 12). H OH-/J Ph,CHOH Ph,C \>-S"' -+ Ph,CO + SH('-')+ SH("-I)+ anode __* S"+ + H'; S"' = surfactant. Scheme 12 Anodic cleavage of the carbon-oxygen bond is well established and a useful example is the electrolysis of 4-methoxybenzyl ethers (Scheme 13).Such ethers of a range of alcohols are cleaved" under relatively mild conditions and in high yields MeCN-H,O, ROCHz00.. + 1.65 V (us. s.c.e.) ROH + OHCQOMe LiCIO Scheme 13 A. Nilsson A. Ronlan and V. D. Parker Tetrahedron Letters 1975 1107. I9 T. Franklin and L. Sidarons J.C.S. Chem. Comm. 1975,741. 2o S. M. Weinreb and G. A. Epling J. Org. Chem. 1975,40 1356. Electro -organic Chemistry (75-98%). Work-up is simple; the unwanted aldehyde is removed by extraction with saturated aqueous sodium bisulphite. The mechanism is assumed to be that proposed by earlier workers [Ann. Reports (B) 1972 69 3101 which involves proton loss from the initially formed radical-cation followed by a second electron transfer to give stabilized carbenium ions such as (18).Whilst plausible there is little + ArCHOR evidence that this mechanism holds generally; indeed for electrolysis of dibenzyl ether in dry acetonitrile detailed coulometric and '80-labelling experiments have shown that it is the ether radical-cation that undergoes cleavage.2' An interesting advance in anodic substitution reactions is the realization of trifluoroacetoxylation of usually unreactive substituted benzenes.'* Some key results are given in the Table and it seems that electron transfer from the aromatic Table Anodic trifluoroacetoxylation 22 of substituted benzenes PhX YieldsOf XC6H4OH (UkZ XC6H40COCF3) X o (%) rn (Yo) p (%) Current yieZd (YO) H - - - 27 C02Me 51 34 15 65 NO2 22 59 19 60 CF3 35 47 18 31 COMe 54 32 14 67 COPh 70 18 12 21 CN 45 30 25 10 compound is the prior step.The initialiy formed trifluoroacetates are hydrolysed during work-up to produce phenols and these reactions provide an effective means of anodic hydroxylation. In related oxidation of aromatic carbonyl com- pounds in CH,CI,-CF,CO,H solvent gives nuclear hydroxylation and nuclear acetamidation occurs in wet acetonitrile. Comparison of the results of these two studies reveals an interesting medium effect. For acetophenone and methyl ben- zoate oxidation in CF3C02H-CF3C02Na (1mol I-') gives2' substantial amounts of the rnetu-substituted products whereas oxidation in CH,C1,-CF3C0,H-Et4NBF4 (0.1mol I-') gives23 no detectable metu-substitution.Full details have now appeared of the preparatively useful inn~vation'~ of anodic cyanation of aromatic compounds in aqueous emulsions of methylene chloride containing phase-transfer agents. The possibility of an indirect route for the conversion of benzene into phenol has led to a detailed study of the optimization of yields for formation of diphenyliodonium salts from anodic oxidation of i~dobenzene~~ in the presence of 21 R. Lines Ph.D. Thesis London University 1975. 22 Z. Blum L. Cedheim K. Nyberg and L. Eberson Acru Chem. Scund. (B),1975,29 715. 23 Y.-H. So J. Y. Becker and L. L. Miller J.C.S. Chem. Comm. 1975,262. 24 L. Eberson and B. Helgee Actu Chem. Scand. (B),1975,29,451. 25 H. Hoffelner H. W. Lorch and H. Wendt J. Electroanulyt.Chem. Interfacial Electrochem. 1975,66 183. 158 J. H. P.Utley benzene. Such salts may be hydrolysed to phenol and iodobenzene can be re- covered. At low temperatures (303K) low potentials and high ratios of concentra- tions of iodobenzene and benzene chemical yields of 95% are achieved for the diphenyliodonium cation. Examples of useful anodic intramolecular coupling in the synthesis of phenolic alkaloid intermediates are becoming commonplace.26 In this context the elec- trochemical preparations of (f)-kresiginone26" and (f)-cryptopleurine (19)26bare noteworthy (Scheme 14). MeCN-HBF Pt anode 1.07 V (us. s.c.e.) OMe OMe 31% Meow OMe Scheme 14 Oxidation of Organic Anions.-An important correction has been made to the literature concerning the Kolbe reaction.It has always seemed unlikely that the of the detection of the anodically generated triphenylacetoxyl radical could be correct and assessment of the work is hampered because the key e.s.r. spectrum was not published. In a reinve~tigation~~~ it has been shown that in situ,controlled-potential oxidation at platinum of triphenylacetate in acetonitrile solution does not give rise to an e.s.r. signal. However if an anodic pulse of sufficient duration is applied and the circuit is opened the e.s.r. spectrum of the triphenylmethyl radical is obtained. Alternatively if a cathodic pulse to GO.35V is applied following the anodic pulse the same spectrum results. The rationalization of this behaviour is given in Scheme 15. 26 (a)J.M. Bobbit I. Noguchi R. S. Ware K. N. Chiong and S. J. Huang J. Org. Chem. 1975,40,2924; (b)E. Kotani M. Kitazawa and S. Tobinaga Tetrahedron 1974,30 3027. 27 (a)N. B. Kondrikov V. V.Orlov V. 1. Ermakov and M. Ya. Fioshin Elekirokhimiya 1972,8,920;(b) R. D. Goodin J. C. Gilbert and A. J. Bard J. Electroanalyt. Chem. Interfacial Electrochem. 1975,59 163. Electro -organic Chemistry anode cathode Ph ,CCO ,-s,"d,+ Ph3C+ +e-Ph,C' Scheme 15 The retention of stereochemistry at sites other than the a-carbon atom is an important feature of the Kolbe reaction and one of the reasons for its usefulness in the synthesis of naturally occurring compounds. Species other than Homo supiens now appreciate the implications of this because the reaction has recently proved useful for the syntheses of the housefly sex attractant muscalure [Ann.Reports (B) 1973,70,302] and of brevicomin (20) the sex attractant of the western pine beetle. This latter28 very elegant synthesis is outlined in Scheme 16. A two-electron EtCHkHCH,CO; + MeCO(CH,),CO; Pt anode MeCO(CH,),CHkHEt (20)42 % Scheme 16 oxidative decarb~xylation~~ has proved to be an efficient method for the preparation of the bicyclic ketone (2 l) which has the correct stereochemistry for ring opening via Baeyer-Villiger oxidation and subsequent conversion into methyl (f)-jasmonate (22)-3 Cathodic Processes Radical-anions as Bases Nucleophiles and Electron-transfer Agents.-There is growing interest in the possibility of using electrogenerated bases [cf.Ann.Reports (B),1969,66 2391. Dianions may be generated cathodically according to Scheme 17 and either the radical-anion or preferaMy the dianion may be strong bases II M + M2-Scheme 17 ** J. Knolle and H. J. Schafer Angew. Chem. Znternat. Edn. 1975 14 758. 29 S. Torii H. Tanaka and T. Mandai J. Org. Chem. 1975,40,2221. 160 J. H. P. Utley useful for initiating further chemical reaction. For cyclic voltammetry on the phenyl-substituted ethylenes (23) and (24) the transfer of the first electron is Ph,C=CHPh Ph MeC=CHPh (23) (24) reversible and that of the second irreversible. The ratio of peak currents [ip(l)/ip(2)] is in HMPA independent of sweep but for (24)in DMF the ratio can vary from 45 at 0.1 V s-l to 1.1at 30 V s-l i.e.at slow sweep speeds and in the solvent that is less able to stabilize the counterion disproportionation to the dianion is significant. Thus conditions may be set up for the generation of the strongly basic dianion at the potential of the first electron-transfer reaction. A similar study31has compared the effectiveness of water and methyl iodide in reacting with azobenzene radical-anion. Methyl iodide is very much better than water at quenching reversibil-ity of the first electron-transfer reaction. N-Methylation and NN-dimethylation are observed following electrolysis at the first peak potential and the proportions are dependent on the electrolyte and solvent. The extremes are N-methylation (90%)in moist DMF-LiCl and NN-dimethylation (loo”/,)in dry HMPA-LiC1.Cyclic volt-ammetry has also been to detect chemical reaction that is initiated by the cathodic generation of azobenzene dianion. Addition of azobenzene to solutions of acetophenone or benzophenone in acetonitrile causes diminution of the peak currents for reduction of the ketones. Azobenzene is reduced at less cathodic potentials than the ketones in question and it is supposed that azobenzene dianion abstracts a proton from the solvent and that the nucleophile thus formed (NCCH,) adds to the ketones thereby decreasing their concentration at potentials prior to Ph Ph (25) their reduction potentials. Benzil (25) and benzil dianil (26) are methylated by cathodic reduction at the potential of first electron-transfer in DMF containing methyl chloride33 [exemplified for (26) in Scheme 181.The products are highly -._-phHph Ph >=( Ph + Ph )=( Me DMF MeCl + NPh + PhNH(Ph)C=C(Ph)NHPh Hgcathode PhN NPh 2Fmol-’ PhN NPh PhN Ph cis and trans Me Me Me (24) Bu,NCl 12”/ 88 ”/ none LiCl 20 ”/ trace >60% Scheme 18 dependent on the cation; with LiC1 hydrogenation is preferred and this may be an example of the lithium cation carrying (by hydration) a proton donor to the vicinityof 30 T. Troll and M. M. Baizer Electrochim.Acru 1974,.19 951. 31 T. Troll and M. M. Baizer Electrochim.Acta 1975,20 33. 32 A. J. Bellamy J.C.S. Chem. Comm. 1975 944. 33 J. Simonet and H. Lund Bull. SOC.chim. France 1975 2547. Electro-organic Chemistry the cathode.Acetic anhydride is also a useful electrochemically inert electrophile which may react with electrogenerated nucleophiles. Thus the of benzophenone in MeCN containing acetic anhydride gives the acetate (27) in ca. 70% isolated yield. Ph2C-0Ac I COMe (27) Several paper^^'-^' have appeared which describe further the consequences of electron transfer by the homogeneous electron-exchange processes outlined last year [Ann. Reports (B) 1974 71,2261. For reactions which give a polarographic catalytic wave electron transfer to BX may be effected at the (lower) potential that corresponds to electron transfer to A provided that further reaction of BX' is rapid (Scheme 19). Thus the electrochemically inert ether (28) is cleaved3' at an accessible potential by the intermediacy of the radical-anion of 1-methylnaphthalene (Scheme 19).A' + PhCH,OPr' A + [PhCH,OPr']= (BX) (28) (BXL) BXI A Pr'O-+ PhCHi e- H+ b PhMe 93% w-1 (B') (BH) Scheme 19 Cathodic Cleavage.-Hydroxy-groups may be cleaved38 from unsaturated alcohols in DMF solution at a mercury cathode and at very negative potentials [ca. -2.6 to -2.9V (us. s.c.~.)]. Phenol is used as the proton donor and certainly for the reactions given in Scheme 20 loss of hydroxyl ion is faster than protonation of the -2.9 V (us. s.c.e.) Ph,COH DMF-BudNI b Ph,CH 2Frnol-' 95 % Ph,C=CHCH,OH Ph,C=CHMe 95 % -Ph2CHEt Ph,C(OH)C-CH -2.6 V 95 % Scheme 20 34 T. J. Curphey L. D. Trivedi and T. Layloff J. Org. Chem. 1974,39 3831. 35 H.Lund and J. Simonet J. Electroanalyt. Chem. Interfacial Electrochem. 1975,65,205. 36 J. Simonet M.-A. Michel and H. Lund Acta Chem. Scand. (B) 1975,29,489. 37 H. Lund M.-A. Michel and J. Simonet Acta Chem. Scand. (B),1975,29 217. 38 H. Lund H. Doupeux M.-A. Michel and G. Mousset Electrochim. Acta 1974,19,629. 162 J.H.P. Utley multiple bonds. Pinacols may also be cleaved3' cathodically provided there is at least one electro-active group that is a to the hydroxy-group. In protic media both C-0-and C-C- bond cleavage are observed e.g. from (29). In aprotic media it (29) (30) (31) is likely that C-C- bond cleavage is initiated by strong bases that are generated at the cathode with subsequent loss of hydroxyl radical from the ketone radical anion [e.g.(30) is formed in 90% yield from (31) in dry DMF at -2.3 V 0s. (AglAgI)]. Highly branched a-acetoxy-ketones are produced4' in good yield by cathodic removal of bromine from CYCY -dibromo-ketones. Straight-chain act -dibromo-ketones are reduced to the corresponding ketones but for highly branched molecules the loss by cathodic cleavage of the first bromine gives rise to an enolic intermediate (32) which is rapidly solvolysed (in acetic acid) to an a! -acetoxy-ketone (Scheme 21). This scheme is consistent with the known solvolytic behaviour of a -branched allylic bromides. BrABr -1.8V(AglAg+) &J3r Hg pool Ah HOAc-NaO Ac (32) 0 OH 78 % Scheme 21 An interesting mechanistic distinction may be made for the reductive cleavage of carbon-halogen bonds from benzenoid compounds.For electrolysis in DMF-D,O YO) the incorporation of deuterium depends on whether or not th? radical anion (MXL) diffuses from the cathode. Cleavage in bulk solution (to M ) results in H abstraction (no D incorporation) whereas cleavage at the electrode allows further electron transfer (to M-) and hence deuterium incorporation. Thus c1eavage4l from (33) gives benzophenone with no deuterium incorporation whereas (34) is reduced with loss of bromine to a product with ca. 60%deuterium incorporation. Consistent with this hypothesis is the observation of reversible one-electron transfer in the cyclic voltammetry of (33) i.e.the radical-anion has a lifetime which permits diffusion from the cathode. 39 M.-A. Michel G.Mousset and J. Simonet Electrochim. Acta 1975 20 143. 40 A. J. Fry and J. J. O'Dea J. Org. Chem. 1975,40,3625. 41 J. Grimshaw and J. Trocha-Grimshaw,J.C.S. Perkin 11 1975,215. Electro -organic Chemistry Cathodic Hydrogenation and Cydization.-Alkynes are less easily reduced than alkenes and consequently cathodic hydrogenation of alkynes usually results in complete four-electron reduction to alkane. A recent systematic of the cathodic reduction in DMF of 1-phenylhex-1-yne shows that although at low (polarographic) concentrations there is four-electron reduction to 1-phenylhexane preparative-scale electrolysis (0.002-0.01 M in alkyne) leads predominantly to PhCH=C=CHPr (35) the allene (35). The rearrangement is presumably bimolecular involving electro- generated base and therefore favoured by higher concentrations.A similar concen- tration dependence attends the cathodic reduction43 of 6-chloro- 1-phenylhex- 1-yne (36). At low concentrations (ca.2.5 X 10-4mol l-') controlled-potential electrolysis leads to the cyclic compound (37) i.e. from intramolecular displacement of chloride PhCrC(CH,),Cl x (36) U (37) in the alkyne radical anion. At higher concentrations intermolecular isomerization to the allene (see above) occurs at a potential which initiates further reduction to a complex mixture of products. For the reactions depicted in Scheme 22 the competition between hydrogenation and cyclic hydrodimerization on structure and in particular on the bulk of R; dimerization is inhibited where R is t-butyl or phenyl.H+,e- H+ Ph(R)C=C(CN) 5 Ph(R)k-C(CN) Ph(R)CHCH(CN) (R = Ph or Bu') 1 Ph cis and trans (R = H or Me) Scheme 22 42 W. M. Moere and D. G. Peters J. Amer. Chem. SOC.,1975,97 139. 43 W. M. Moore and A. Salajegheh J. Amer. Chem. SOC.,1975,97,4954. 44 A. L.'Avaca and J. H. P. Utley J.C.S. Perkin 11 1975 161. 45 A. L. Avaca and J. H. P. Utley J.C.S. Perkin 1 1975,971. 164 J. H.P. Utley StereoselectiveCathodicReactions.-Significant progress has been made recently in understanding and accomplishing stereoselective reduction. A particularly signifi- cant advance is the preparation of a chiral electr~de.~~ Graphite rods were baked in air (160°C for 36 hours) and the surface carboxylic groups so produced were converted into acid chloride groups (using SOCl in dry benzene) and then into chiral amide [viu S( -)-or R( +)-phenylalanine methyl ester].The chiral electrodes cannot be distinguished from the original carbon electrodes either by eye or by their cyclic voltammetric behaviour. Reduction of (38) at such pre-treated electrodes at -1.05 V (vs. s.c.e.) in ethanolic acetate buffer gives the optically active alcohol with an enantiomeric excess of the (-)-isomer if pre-treatment is with S(-)-phenylalanine ester and an excess of the (+)-isomer if R(+ )-phenylalanine ester is PhCOC0,Et (39) used. The optical yield is not given because there is no reliable value for the specific rotation of the optically pure alcohol. However similar reduction of (39) gives the corresponding optically active alcohol in 9.7% optical yield.The electrodes are re-usable. Stereoselectivity may also be a consequence of electron transfer under the influence of a chiral double layer and a possible example4' of this is the reduction of cobalt(rI1) trisacetylacetonate (Scheme 23). In this case the key observation is that if Hg pool -1.0 V (US.AglAgCI) ( ')-Co(acac)3 M&N-tri-N-methyl-'tco(acac),l (-)-menthylammonium = 14 perchlorate Scbeme 23 the reaction is halted before completion the recovered starting material has become optically active and furthermore the optical activity is larger the longer the reaction is allowed to continue. The supposition is that the rate of electron transfer in the chiral double layer is different for the two enantiomers.However reaction in bulk solution containing chiral salts may also result in asymmetric induction and an interesting comparison of stereoselective photochemical and electrochemical pinacolization has been made.48 Acetophenone is pinacolized stereoselectively at 25 "Cin a methanolic solution of (+)-1,4-bis(dimethylamino)-2,3-dimethoxybutane containing lithium bromide by irradiation (optical yield 3.3% ;meso-:(&)-isomer = 0.5)or cathodically (optical yield 5.6%; meso-:(*)-isomer =0.35). The similarity of these results suggests that interactions at the electrode environment are not essential for asymmetric induction. Hydrogen transfer from chiral donors is another possible mechanism for asymmetric reduction and the cathodic hydrogenation of a malononitrile adduct (Scheme 22; R =But) was from this viewpoint.This 46 B. F. Watkins J. R. Behling L. L. Miller and E. Kariv J. Amer. Chem. SOC.,1975 97 3549. 47 S. Mazur and K. Ohkubo J. Amer. Chem. SOC.,1975 97,2911. 48 D. Seebach and H. A. Oei Angew. Chem. Internat Edn. 1975,14 634. Electro -organic Chemistry compound is particularly suitable for probing the possibility because its radical- anion is produced at povtentials well clear of those required for discharge of chiral proton donors such as ephedrine hydrochloride or quinidine sulphate. This pre- cludes hydrogen-atom transfer [cf. Ann. Reports (B) 1970 67 2341 and ensures proton transfer. Within experimental error no induction of asymmetry was found for this system.In a pragmatic of the asymmetric reduction of phenylglyox-ylic acid (40)to mandelic acid (41),many experimental parameters have been varied PhCOC0,H PhCH(OH)CO,H (40) (41) systematically. In this case the hypothesis advanced is that optimum stereoselectivity (ca. 20% optical yield) is found for conditions (e.g. temperature pH nature and concentration of alkaloid inducer current density) which most favour the formation between carbanionic intermediate and alkaloid of diastereoisomeric adsorbed complexes which protonate with retention of configuration and presumably at different rates. A similar conclusion comes from the work of Horner's group and a recent and very detailed account of their studies has a~peared.~' For racemic ethylenic ketones such as (42) highly stereoselective reductive coupling has been achieved51 by careful control of pH resulting in the predominant (42) (43) formation of only the cis-threo-cis-isomers e.g.(43).The overall yields of glycols are highly dependent on pH but fall within the range 60-90%. 4y M. Jubault E. Raoult and D. Peltier Electrochim. Actu 1974 19 865. L. Horner and D. Degner Electrochim. Acru 1974,19,611. 51 E. Touboul and G. Dana Tetruhedron 1975,31 1925.