7 Electro-organic Chemistry* By J. H. P. UTLEY Department of Chemistry Queen Mary College Mile End Road London El 4NS 1 Introduction An important and lengthy review of organic electrochemistry has appeared written from a physical-organic standpoint.' On the electrochemical side an introductory account2 of the application of electroanalytical techniques is available. Volume 5 of the C.S. Specialist Report on Electrochemistry has a~peared.~ The current Report highlights cathodic and anodic reactions followed by a section of stereoselective processes. A recent development in the field is the study of 'paired reactions' i.e. electrolysis in an undivided cell is contrived so that useful products are obtained at both electrodes. A detailed study of one such reaction has been pre~ented.~ The cell is made from a bored graphite block optionally lined with lead and charged with acetonitrile tetra-n-butylammonium iodide (or KI) ethyl acrylate and diethylmalo- nate.The cathodic and anodic products are both obtained in ca. 95% yield their formation being rationalized according to Scheme 1a. An electrochemical synthesis of nitroalkanes in ca. 75% yield from organoboranes also involves' the products of both electrodes (Scheme 1b). 2 Anodic Processes The Anodic Oxidation of Carboxy1ates.-Considerable amounts of aromatic aldehyde are obtained from the Kolbe electrolysis of arylacetates [cf.Ann. Reports (B) 1971 68 3133. It was suggested6 that the aldehyde resulted via peroxides from reaction between benzyl radicals and adventitious oxygen.This suggestion has been verified7 in a careful comparison of the products of anodic oxidation (in the absence and presence of oxygen) of phenylacetate hexanoate and 2-ethylhexanoate ions. For the alkanoates (RCO,-) peroxides of the type ROzR were isolated when oxygen was passed through the electrolyte and for phenylacetate oxidation benzal- dehyde is absent from the products of electrolyses flushed with nitrogen but present if oxygen is bubbled through the solution. 1 L. Eberson and K. Nyberg Adv. Phys. Org. Chem. 1976 12,1. D. Pletcher Chem. SOC. Rev. 1975,4471. 0.R.Brown in 'Electrochemistry' ed. H. R. Thirsk (SpeciulistPen'odicaIReports),The Chemical Society London 1975 Vol. 5 p. 220. M. M. Baizer and R. C. Hallcher J.Electrochem. SOC. 1976 123,809. Y.Takahashi M. Tokuda M. itoh and A. Suzuki Synthesis 1976 616. 6 J. P. Coleman R.Lines J. H. P. Utley and B. C. L. Weedon J.C.S. Perkin U,1974 1064. J. E. Barry M. Finkelstein E. A. Mayeda and S.D. Ross J. Amer. Chem. SOC. 1976 98 8098. * This Report was compiled with the help of a current awareness service run by Ruth Hayden of Queen Mary College library. 137 138 J. H.P. Utley Cathode (a) 2CH2CHC02Et -%Et02CCH(CH2)2CHC02Et Anode 12 ,-2e 21-1 EtOZC(CH2)dCOzEt+ 2(EtO*C)2CH (Et02C)2CHCH(C02Et)2 -Pt electrodes RCH2N02 (b) R3B MeN02/R:NX ' (X = Br Cl) Cathode Anode O~NCH~ RB -,-A ~H~+o~NCH~-RX & ~x,Z-x-1 02NCH2R+ X-Scheme 1 As part of an attempt to functionalize pyridine in the 3-position the reactivity of anodically generated methyl and trifluoromethyl radicals have been compared.' The methyl radical is nucleophilic and for attack on the pyridinium ion the a:y ratio is 3 :1which is also the ratio found for attack by alkyl radicals generated by persulphate oxidation of carboxylates.Methyl attack on pyridine is more efficient (CQ. 20%)but less selective. The trifluoromethyl radical attacks all three positions of the pyridine and the pyridinium ion; the @-position is preferred (ca. 40-60%) but the overall yield of substitution products is low (ca. 5%). There can now be little doubt that alkyl radicals formed in the Kolbe reaction are essentially 'free'. Complete racemization (299.5% ) of several anodically generated chiral radicals has been demonstrated' [cf.Ann. Reports (B) 1973 70 3021. However this would also result from radicals involved in a dynamic adsorption- desorption equilibrium. That for saturated radicals the maximum energy of adsorp- tion is negligible is shown" by the completely random coupling of alkyl-substituted cyclohexyl radicals; if adsorption were important diastereoisomeric radicals would be differently stabilized which would lead to different energies of activation for coupling and thereby stereoselection. Aryl-substituted cyclohexyl radicals behave differently; further oxidation to carbenium ions is encouraged even though substitu- tion is remote and this appears to be a surface effect consistent with the known propensity for aryl groups to adsorb at the anode.Anodic acetamidation which involves the trapping of anodically generated car- benium ions has been used to demonstrate the anodic production" of vinyl cations (Scheme 2). The product mixture is complex the overall yield is 10%and of that less * J. H. P. Utley and R. J. Holman Electrochim. Acta 1976 21 987. L. Eberson K. Nyberg R. Servin and I. Wennerbeck Acta Chem. Scand. 1976,830,186; L. Eberson K. Nyberg and R. Servin ibid. p. 906. lo G. E. Hawkes J. H. P. Utley and G. B. Yates J.C.S. Perkin ZI,1976 1709. l1 E. Laurent and M. Thomalla Compt. rend. 1976 C282,441. Electro -organic Chemistry 139 COCH3 -2e -C02 MeCN/H20 / + CH2:C(Me)C02-[CH2:CMe] -Me2CHCON \C(Me):CHZ + COCHS / CH2:C(Me)CON \ C(Me):CH2 Scheme 2 than half consists of the amides shown.It is suggested that cathodic hydrogenation accounts for the unexpected saturation of one of the isopropenyl groups. Oxidative decarboxylation continues to be of use but a claim" that it is 'tedious inconvenient and limited to small quantities' has led to the development of a chemical alternative which involves heating the dicarboxylic acid at 185 "C for 36 h in a brew containing cuprous oxide 2,2'-dipyridyl quinoline and powdered glass. Yields for three methods of bis-decarboxylation are given in Scheme 3. Anodic (2e) 62% CuzO 40% Pt anode CO,H Scheme 3 removal of a single carboxylate group is a key in a synthesis of cubebol (Scheme 4a). In this case formation of a carbenium ion is involved with subsequent quenching by acetate ion.However for similar electrolysis of a series of easily oxidized substituted tetrahydroisoquinoline-1-carboxylicacids electron-transfer OCCU~S'~at -0.28 V (us s.c.e.); cf. ca. 2.0 V for Kolbe oxidation which suggests oxidation of the aromatic ring with subsequent loss of C02 in a pseudo-Kolbe reaction [cf. Ann. Reports (B) 1971 68 3131. The products are certainly radical- derived (Scheme 4b). The Anodic Oxidation of Neutral Organic Compounds.-The synthetic uses of anodic substitution reactions is the subject of an important wide-ranging and practically oriented review.I5 The interest in indirect methods of oxidation appear to have revived. Anodic acetamidation of relatively inert compounds e.g.alkanes and saturated long chain carboxylates probably proceeds via the formation of a radical by electrolyte oxidation with subsequent abstraction from carbon-hydrogen bonds. Further oxida- tion can then take place (Scheme 5). Good evidence for all this comes from a studyI6 '2 R.A. Snow C. R. Degenhardt and L. A. Paquette Tetrahedron Letters 1976 4447. '3 S. Torii T. Okamoto G. Tanida H. Hino and Y. Kitsuya,J. Org. Chern. 1976,41 166. l4 J. M. Bobbitt and T. Y. Cheng J. Org. Chem. 1976 1,443. 15 L. Eberson and K. Nyberg (Tetrahedron Report No. 21) Tetrahedron 1976,32,2185. 16 L. L. Miller and M. Katz J. Electroanalyt. Chem. Interfacial Electrochem. 1976,72 329. 140 J. H.P.Utley (72%) ( -) cubebol Reagents i Pt electrodes AcOH-BU'OH-E~~N,undivided cell b) ;:zQNR4 -e "'"aNR4-H?; ""WNR4 R20 \ R20 \ R3 C02H R3 CO,H R3 R3 H Scheme 4 electrolyteanion 3 X.RH R. 3 Rf Scheme 5 of the anodic oxidation of methyl hexanoate in acetonitrile in the presence of electrolytes containing C1- C102 C104- and BF4-. In each case the 2-and 3-acetamido products are obtained and in similar proportions. The oxidation of carboxylic acids may be achieved" directly and indirectly in fluorosulphuric acid [cfi Ann.Reports (B) 1975 72 1511. Peroxydisulphuryl difluoride [(FSO,),] may be prepared by the constant current anodic oxidation of K03SF(1M) in anhydrous FS03H. The anolyte may be used directly for the oxidation and lactonization of carboxylic acids (Scheme 6). Similar results are obtained if the acid is present R'R2CH(CH2),C02H + FS03H FS03-+ R'R2CH(CH2),C02H2+ (FS03)2 1 R 'R3C?>O e R'R2C(CH2),C02H2++FS03H I (CH,) + 1 OSOZF Scheme 6 throughout the oxidation although in this case work-up immediately after elec- trolysis is desirable because of subsequent rearrangement reactions.l7 C. J. Myall D. Pletcher and C. Z. Smith,J.C.S. Perkin I 1976 2035. Electro -organic Chemistry Electrochemically generated iodine (I) probably formed in acetonitrile as CH3C& is a powerful electrophile.'8 Typically a solution of the iodine (I)species is formed by electrolysis of iodine at +1.9 V (us. Ag/Ag+) in acetonitrile solution containing lithium perchlorate. This reaction may be carried out on a 10g scale and the aromatic compound to be iodinated is added to the preformed pale yellow solution of the electrophile.Some representative results are given in Table 1. Table 1 Iodination of substituted benzenes (PhX) with iodine (I)18 X OMe CH3 c1 CO2Et COCH3 NO2 Yield ("/o) 90 93 80 65 15 0 Large-scale anodic acetoxylation and methoxylation using the cells described earlier [Ann. Reports (B) 1975 72 1511 has provided"" a route to cyclic N-acylenamines reactive intermediates which have not previously been used in synthesis because of the lack of preparative methods.lgb The sequence of reactions is exemplified in Scheme 7. (92%) Reagents i C anode 8.4 F mol-* MeOH; ii NH4Br 140"C Scheme 7 Anodic aromatic substitution continues to produce surprises. A significant pro- duct2' of the anodic oxidation under acetoxylating conditions of the methoxyl- substituted indene (1)is the dimer (2).This is formally a 27r + 27r cycloaddition but electricity is required for the reaction. The result has therefore been rationalized in terms of Scheme 8. Another unexpected result follows the attempted" nuclear (1) -2-p (1)+ -%-+ (2)t -%(1)t+(2) Scheme 8 L. L. Miller and B. F. Watkins J. Amer. Chem. SOC. 1976,98,1515. l9 (a)K.Nyberg and R. Servin,Acra Chem. Scand. 1976 B30.640; (b)K.Nyberg Synthesis 1976,545. 2o L.Cedheim and L. Eberson Acra Chem. Scand. 1976,B30,527. K. Nyberg and L. G. Wistrand J.C.S. Chem. Comm. 1976 898. 142 J. H.P.Utley acetoxylation of 2-and 4-fluoroanisole. The corresponding chloro- and bromo- compounds yield straightforward products of anodic nuclear acetoxylation.For the fluoro-compounds fluorine is displaced and acetoxylation takes place almost entirely at the position previously occupied by the fluorine substituent. This displacement occurs with high current efficiency (70-80°/~)at low conversions; at high conver- sions the product is oxidized further. The mechanism is not clear but a plausible explanation is given in Scheme 9a. In contrast the introduction of fluorine by anodic substitution provideszz a method of preparation of fluoro-organosilanes (Scheme 9b). The reaction consumes 1F mol-' and the detection of products derived from alkyl or benzyl radicals is good evidence for the mechanism given in the Scheme. (a) ArF -5[FArOAc]. -F-+[ArOAc]+ OAc L -1 Scheme 9 Anodic methoxylation provides an entryz3 into a series of useful easily metallated vinyl halides (Scheme 10).The halide intermediates can be produced electrochemi- cally on a 3040 g scale and metallation is accomplished using the butyl-lithium in THF at -70 "C.The reaction also works well for relatively complex polyfunctional molecules e.g. (3) is produced in 80% isolated yield. Me0 OMe Br Me0 Me0 I 1 Reagents i MeOH-KOH(l%) Pt anode Scheme 10 From the many electrochemical studies of phenols and phenolic ethers it has often seemed that phenols are more easily oxidized than the corresponding ethers. This is contrary to the order expected from the relative electron releasing effects of OH and OR. A careful investigationz4 reveals that indeed the expectation from inductive effects is correct and the apparent anomaly is due to special structural features of the phenols previously studied.The results summarized in Table 2 show clearly that in acidic solution phenol oxidation becomes reversible and a valid comparison of the 22 I. Y. Alyev I. N. Rozhkov and I. L. Knunyants Tetrahedron Letters 1976 2469. 23 M. J. Manning P. W. Raynolds and J. S. Swenton J. Amer. Chem. SOC.,1976,98 5008. 24 0.Hammerich V. D. Parker and A. Ronlan Acfa Chem. Scand. 1976 B30 89. Electro -organic Chemistry Table 2 Peak potentials for phenol and phenolic ether E,/V(vs. s.c.e.) -50°C Substrate CH2C12 CH2C12-FS03H (10% VOl/VOl) 4-Me0.C6H4-OH 1.20 (irr) 1.49 (rev) 1,4-(Me0)2C6H4 1.38 (rev) 1.39 (rev) 4-Ph.C6H4OH 1.48 (irr) 1.59 (rev) 5-Ph.C6H40Me 1.55 (rev) 1.56 (rev) oxidation peak potentials with those for reversible oxidation of the ethers can be made.The same group has reported fully25 on the anodic intramolecular coupling of a series of phenolic diarylkanes (Scheme 11). The dienones are produced in high (for n = 3) Reagents i Pt anode CH&N Scheme 11 yield (75-85%) for n = 3; no cyclization is found for n = 2 or 4. It is almost certain that the favoured pathway is formation of a phenoxonium ion which attacks the adjacent aromatic ring in an electrophilic substitution step. A remarkably slow anodic coupling reaction has been studied.26 The metacyclophane (4)gives on cyclic voltammetry a quasi-reversible oxidation-reduction couple at 0.65 V (us.Ag/Ag') in acetonitrile. Controlled potential electrolysis at the potential at which the dication is formed (1.05V) gives (5) in 90%yield. The slowness of coupling relative to the 25 U. Palmquist A. Nilsson V. D. Parker and A. Ronlan J. Amer. Chem. Soc. 1976 98 2571. 26 J. Y. Becker L. L. Miller V. Boekelheide and T. Morgan Tetrahedron Letters 1976 2939. 144 J. H. P. Utley singly bridged compound is probably due to steric strain in the relevant intermediate (6). Similar anodic cyclizationz7 allows the efficient conversion of diarylamides such as (7) into dibenzazocines e.g. (8). Me0 OMe Me6 Me0 Me0 (6) (7) (8) The electrochemical production of ortho-quinones in the presence of 1,3-dicarbonyl compounds providesz8 a convenient and high-yield method for the preparation of certain oxygen heterocycles (e.g.Scheme 12).?H -4e -4Hf aoH OH +ao 0 Reagents i 1.1V (us. s.c.e.) C anode Pt cathode undivided cell H20-NaOAc (0.15M) Scheme 12 Another example of anodic cleavage with loss of acylium ion has been reportedz9 [cf. Ann. Reports (B) 1973 70 681. The electro-oxidation of N-acetyldiazines results in the production of the corresponding diazines with the consumption of 2F mol-'. The implied mechanism is given for one of the examples studied (Scheme 13). Ac Ac I I Ac Ac Ac Scheme 13 Attempts to trap the acylium ion failed presumably because the heterocyclic product is more nucleophilic than the aromatic trapping species.An important approach to specific functionalization is foreshadowed by what is probably a template-directed electrochemical chlorination of a ster~id.~' The ester 27 M. Sainsbury and J. Wyatt J.C.S. Perkin I 1976 661. 28 2.Grujic I. Tabakovic and M. Trkovnik Tetrahedron Letters 1976,4823. 29 P. Martigny H. Lund and J. Simonet Electrochim. Acta 1976 21 345. 30 R. Breslow and R. Goodin Tetrahedron Letters 1976 2675. Electro -organic Chemistry 145 (9) shows on cyclic voltammetry an oxidation peak (E,) at +2.6 V (us. Ag wire). Oxidation in acetonitrile solution in the presence of chloride ion (E ca. 1.0 V) was performed at two potentials. At 1.8V in the dark chlorine was evolved but the steroid molecule was not chlorinated; at 2.7 V conversion into (10) was achieved.After saponification 35-7 1YO isolated yields of the corresponding cholestan-3a -01 were obtained. These results are interpreted as shown in Scheme 14 i.e. in terms of (9) X =H (10) x=c1 Scheme 14 an electrochemically initiated chain reaction sustained by the anodic production of chlorine. Consistent with this is the suppression of the reaction by the addition of 02 NO or Br2. It would be interesting to see if the reaction could be sustained by short pulses at +2.7 V followed by longer pulses at +1.8 V. 3 Cathodic Processes The Cathodic Reduction of Organic Cations.-Some years ago quaternary ammonium amalgams with strongly reducing properties were prepared by the cathodic discharge of quaternary ammonium ions at mercury [Ann.Reports (B) 1968,65 2501. It now appears that the discharge of quaternary ammonium ions at graphite also produces an association which has reducing propertie~.~~ The cathode in question was constructed by glueing a small crystal of graphite to a vitreous carbon electrode. When used for cyclic voltammetry it was found that in DMF solution a number of tetra-alkylammonium salts gave reuersible reduction at ca. -1.5 V (us. Ag/AgI). Approximately 95*/o of the charge is recoverable e.g. subsequent addition 31 J. Simonet and H. Lund J. Electroanalyt. Chem. Interfacial Electrochem. 1977,75 719. 146 J. H.P.Utley of fluorenone results in formation of fluorenone radical anion. The efficiency of charging depends on the size of the quaternary ammonium cation; for the larger cations the cathode disintegrates.For the time being a suitable representation of the species involved is [graphite-&N]"-. The best direct route to the aporphine structure [e.g. (12)in Scheme 151 a high-yield cathodic cyclization (Scheme 15). Voltammetry suggests that two discrete le steps are involved hence the mechanism- given in the Scheme. It is not clear however why the intermediate (1 1) is formed with the stereochemistry required for cyclization. e R=HorMeO e. -1-1 Me ~~lPt0~ R Rjy RIy R Scheme 15 The Cathodic Reduction of Neutral Organic Compounds.-A significant advantage of the electrochemical method has been found for the hydr~genation~~ of cyclo-octatetraene tricarbonyl iron (13).Chemical methods which involve alkali-metal reduction and protonation on work-up operate on a time scale which allows disproportionation of intermediate radical-anions and thereby imposes a maximum of 50% for the yield. Cathodic reduction in the presence of trimethylamine hydrobromide as proton donor results in quantitative hydrogenation to (14). Other 32 R. Gottlieb and J. L. Neumeyer J. Amer. Chem. SOC.,1976,98 7108. 33 N. El Muir M. Riveccie and E. Laviron Tetrahedron Letters 1976,3339. Electro -organic Chemistry FdCO) FdCO1 (13) (14) cathodic hydrogenations are dealt with in the final section on stereoselective reactions. For cyclic ketones however a marked change of behaviour is associated with hindrance to the carbonyl For the series (15) to (17) 2e cathodic reduction to alcohol is found for carvomenthone (15),4e reduction to hydrocarbon is found for menthone (16)’ and camphor (17)can only be reduced (2e)under ‘solvated (15) (16) (17) electron’ conditions.The best electrolytes for the 4e reduction are ethanol contain- ing Li’ Mg2+ or Zn2’ i.e. conditions which encourage the formation of ion-pairs. Charge dispersal in an ion-pair is also helpful to further reduction. These observa- tions are accommodated in the proposed mechanism (Scheme 16). The ‘solvated Scheme 16 electron’ reduction of camphor to a mixture containing (*)-borne01 (84% ) and (It)-isoborneol (16%) is the most convenient and effective way of carrying out this transformation; chemical reductions are either unreliable or tend to give isoborneol as the major product.Cathodic substitution is becoming a useful preparative method. Two examples are discussed here in which products are accessible which are difficult to obtain by other methods. It has proved difficult to prepare 3-t-butylpyrene (18) by Friedel-Crafts alkylation or by a route involving metallation. The radical-anion of pyrene is however formed at a readily accessible potential [-1.5 V (vs. Ag/AgI)] and its in 34 R.J. Holman and J. H.P.Utley J.C.S. Perkin ZZ 1976 884. 148 J. H.P.Utley situ reaction in DMF with t-butyl gives (18) in 52% isolated yield. The carotenoid dione canthaxanthin (19) is also easily reduced36 and at preparative scale concentrations (2 x moll-’) the reactive intermediate is probably the highly stabilized retro-dianion (20).In the presence of acid anhydrides 0-acylation occurs and the corresponding retro-diacylates are formed. Electrolyte conditions may be chosen in which the diacylates hydrolyse readily [e.g. LiC104(0.2m) in 1 1 CH3CN-CH2C12] and S,S-dihydrocanthaxanthin (2 1) is formed almost quantitatively. This is the product of kinetic control treatment with base converts it entirely into an isomer 7,7’-dihydrocanthaxanthin(22). 0 0 -0 0 0 An unexpected coupling reaction resulted from the attempted cathodic cyclic hydr~dimerization~’ of dimethyl benzene- 1,2-diacrylate (23) (Scheme 17). A par-ticularly interesting implication of this work is that should steric factors dictate a-coupling may occur.The same observation has been made3* and put to prepara- tive use in the cathodic pinacolization of the a-and @-ionones [to give (24) and (25)] and of vitamin A aldehyde [to give (26)]. An early report that at the cathode crotonaldehyde gave mainly a -coupling to the glycol has been corrected.” The major product (56%) is the furan (27). 35 R. E. Hansen A. Berg and H. Lund. Actu Chem. Scund. 1976 B30,267. 36 E. A. H. Hall G. P. Moss J. H. P. Utley and B. C. L. Weedon J.C.S. Chem. Comm. 1976 586. 37 J. Anderson and L. Eberson J.C.S. Chem. Comm. 1976 565. 38 R. E. Sioda B. Terem J. H. P. Utley and B. C. L. Weedon J.C.S. Perkin I 1976 561. 39 J. C. Johnston J. D. Faulkner L. Mandell and R. A. Day J. Org. Chem. 1976,41 2611. Electro -organic Chemistry C0,Me M cc02Ee i,g C 0 2 e C0,Me ’’C02Me ” I (23) \ C0,Me Reagents i -1.25 V (vs.s.c.e.),Hg cathode 1.1F mol-’ DMF-HzO (4%) Scheme 17 (26) (27) Interest continues in the products of cathodic reduction of sulphur and selenium compounds. The reduction of ethylene trithiocarbonate (28) gives rise4’ to an interesting transformation (Scheme 18). The loss of ethylene is confirmed by addition of bromine to form lY2-dibromoethane. The involvement of the dianion (29) is suggested because when lY2-dibromo- or 1,2-di-iodo-ethane is used in the alkylation step the bicyclic compound (30) is obtained. The cathodic reduction of _ii @-+ cs e scs,-S-[p- ss (30) (29) Reagents i -1.4 V (us. s.c.e.) Pt cathode 0.98F mol-’ DMF-Bu4NBr; ii XCHZCH~X X=Br I Scheme 18 carbon diselenide with subsequent methylation has been used to prepare41 an intermediate (31) which can be coupled to prepare a tetraselenafulvalene (32) 40 F.J. Goodman and J. Q. Chambers J. Org. Chem. 1976,41 626. 41 E. M. Engler D. C. Green and J. Q. Chambers J.C.S. Chem. Comm. 1976 148. 150 J. H.P.Utley (Scheme 19). Although in similar conditions carbon disulphide gives a methylated reduction product [Ann. Reports (B) 14:4 71 2281 coupling to the fulvalene analogue does not take place. 97 MeseE"i-s. ii+MeSe CSez i IsgseIseMe MeSe Se MeSe Se Se SeMe (31) (32) Reagents i -1.35 v (DS. s.c.e.) Pt cathode DMF; ii (Me0)3P C6H6 80 "c Scheme 19 Deoxygenation by cathodic cleavage can be a mild and selective method.In this way the highly sensitive hydrocarbon axerophtene (33a) has been prepared42 from vitamin A acetate (33b). For vicinal acetates in which one of the acetate groups may (33a) X = H; (33b) X= OAc be cleaved cathodically a rapid elimination is induced (Scheme 20). Thus another sensitive compound hitherto obtained with difficulty 3,4,3',4'-tetrahydro-P- carotene (35) is obtained from the reduction of crustaxanthin tetra-acetate (34). OAc OAc AcO OAc Scheme 20 42 J. G. Gourcy M. Hodler B. Terem and J. H. P. Utley J.C.S. Chem. Comm. 1976 779. Electro-organic Chemistry 151 According to the mechanism implied in Scheme 20 alkynes should be obtainable from alkene diacetates; the reported formation43 of diphenylacetylene by cathodic reduction of a,a’-diacetoxystilbene (36) is consistent with this.A slightly different reductive elimination results from work designed to involve oxyphosphirane species Ph \OAC AcO Ph (36) such as (38). This type of intermediate has been implicated in other reactions. The products obtained from cathodic reduction44 of a,a’-dihalogenophosphinates(37) together with the known propensity of 1,3-dihalogeno-systems to cyclize strongly confirm the intermediacy of (38) (Scheme 21). Cathodic cleavage of halides in 0 0 0 II -II II PhCH(Br)P(OMe)CH(Br)Ph 2PhCPCH(Br)Ph 5(PhCH&POMe I (37) OMe 24’!’ Ph Ph (38) Reagents i 0 V (us.Ag/AgBr) Hg cathode 2F mol-’ DMSO-Et4NBr; ii Hf,2e H+ Scheme 21 aprotic solvents is also the basis of an efficient ~reparation~~ of hexamethyldisilane (Me3Si-SiMe3) from chloro-trimethylsilane.Presumably a silicon anion formed by 2e cleavage can displace halogen from a second molecule of chlorotrimethylsilane. The cathodic removal of both chlorine atoms from NN-dichloroto1uene-p-sulphonamide (39)gives in principle a nitrene (40). When cleavage was carried in the presence of dioxan the product expected from the well-known insertion reaction of nitrenes (41)is found in up to 32% yield. The other major product is the amine (42) and these results suggest strongly the pathway given in Scheme 22. An increasing number of studies are being reported of the electrochemistry of biologically important molecules and redox systems. In many of these investigations the systems are so complex or so poorly understood that meaningful progress is unlikely.An important exception is the rigorous and comprehensive treatment given4’ to the oxidoreduction mechanism of the vitamin B12r-B12ssystem. 43 P. Martigny. M. A. Michel and J. Simonet J. Electroanalyt. Chem. Interfacial Electrochem. 1976 73 373. 44 A. J. Fry and L. L. Chung Tetrahedron Letters 1976 645. 45 E. Hengge and G. Litscher Angew. Chem. 1976,88,414. 46 T. Fuchigami T. Nonaka and K. Iwata J.C.S. Chem. Comm. 1976,95 1. 47 D. Lexa and J. M. Saveant J. Amer. Chem. Soc. 1976,98,2652. 152 J. H.P. Utley TsNC12 -& [TsN:] 4TsNH i) (39) y+ (40) (41) ke. 2H+ TsNH~ (42) Reagents i Pt Cathode 1.8A dm-2 2F mol-' MeCN-dioxan-LEI04 Scheme 22 Electrochemical techniques have been applied particularly with a view to ascertain- ing the role of cobalt co-ordination by the 5,6-dimethylbenzimidazolemolecule at the end of the nucleotide side chain.This is clearly dependent upon pH and in principle three species are in equilibrium for each oxidation state (Scheme 23). N N 'NH Base on Base off Protonated base off Scheme 23 Saveant's main conclusion is that vitamin BIzris largely reduced by two pathways (i) reduction of the 'base-off' species with de-co-ordination of the cobalt being rate- limiting at pH >2.9; (ii) reduction at more negative potentials of the 'base-on' form with rate-limiting electron-transfer. 4 Stereoselectivity Asymmetric reduction of 2- and 4-acetylpyridine to the corresponding alcohol may be achieved in the presence of chiral electrolytes or at chiral electrodes [cf.Ann. Reports (B) 1975 72 1641. It is proving to be a good system for the detailed investigation of stereoselective cathodic reduction. For reduction in the presence of strychnine conditions have been optimized.48 Optical yields of 47% with current yields of 70-80% may be obtained using relatively small amounts of strychnine in an acetate buffered catholyte. It is important to minimize contamination of the mercury pool and therefore short electrolysis times at a large electrode favour stereoselectivity. Using similar conditions but in the presence of (+)-quinidine the optically active alcohol (30% optical yield) is obtained from 2-a~etylpyridine.~' Reduction of the 4-isomer is stereoselective in the presence of strychnine but it yields only racemic alcohol in the presence of (+)-quinidine.The 3-isomer has not been asymmetrically reduced under any conditions. The factor distinguishing the 48 J. Hermolin J. Kopilov and E. Gileadi J. Electroanalyt. Chem. Interfacial Electrochem. 1976,71,245. 49 J. Kopilov S. Shatzmiller and E. Kariv Electrochim. Actu 1976 21,535. Electro -organic Chemistry 153 behaviour of the three isomers is believed to be the relatively slow protonation of the stabilized intermediate [(43) in Scheme 241. It is envisaged that the proton donor effective at the cathode is the conjugate acid of the added alkaloid; slow protonation can occur therefore via diastereoisomeric transition states leading to stereoselection.Homogeneous asymmetric reduction of 2-acetylpyridine has also been eff ected5* Scheme 24 using the chiral reagent (44). In this case the presence of magnesium perchlorate is required. By analogy with the electrochemical work it is tempting to speculate that proton transfer is slowed by co-ordination between 2-acetylpyridine (or an inter- mediate) and magnesium ion. &CON H CH(Me)Ph (44) For graphite electrodes modified with phenylalanine methyl ester [Ann.Reports (B) 1975 72 1641 it is now clear that it is the edge surface which is made ~hiral.~' Using highly ordered pyrolytic graphite one modified electrode was deployed with its edge masked by silicon rubber and one was deployed with the basal surface masked.Asymmetric reduction (of 4-acetylpyridine) was observed only when using the electrode with an exposed edge. Two examples of anodic asymmetric oxidation have been provided. An enan- tiomeric excess (2.5%) of the sulphoxide (45) was obtained by oxidation5' of the (45) corresponding sulphide at a graphite anode modified with phenylalanine methyl ester. Less easily understood is the highly stereoselective anodic depicted in Scheme 25. It is claimed that the absence of the RS-dimer following coupling of the racemic starting material is due to a surface effect but this suggestion needs testing by comparing the results with those from homogeneous coiipling. The involvement of electrolyte cations in directing stereoselective electro- reduction through ion-pairing has been discussed previously [Ann.Reports (B) so Y. Ohnishi and T. Namakunai Tetrahedron Letters 1976,2699. 51 B. E. Firth L. L. Miller M. Mitani T. Rogers J. Lennox and R. W. Murray J. Amer. Chem. SOC.,1976 98 827 1. 52 J. M. Bobbitt I. Noguchi H. Yagi and K. A. Weisgraber J. Org. Chem. 1976 41 845. 154 J. H.P.Utley Me HO Me Me Me Me0 OH / Me Reagents i C felt anode +O. 16 V (0s. s.c.e.) CH3CN-H20-Et4NC104-NaOMe Scheme 25 1974 71,227 2291. A full report is now available on experiments showing the importance of ion-pairing in the stereoselective reduction of cyclic The cathodic cleavage of (*)-PhCH(Br)CH(Br)Ph which yields a mixture of cis -and trans-stilbene has also proveds4 to be a useful system for the systematic study of electrolyte effects on stereoselectivity.The cis:trans ratio can for comparable conditions be changed from 0.18 (0.012M-LiBr) to 1.27 [0.1M-(CBH17)4NBr]. It is suggested that the stereochemistry of reduction is described by Scheme 26 and that small cations neutralize charge and cause slow rotation in the carbanion inter- mediate. H cgt Ph H cit Ph (*)-PhCH(Br)CH(Br)Ph -b ph FH + H.vPh Br Br 1 I Ph Ph Ph w -Ph Reagents i -0.95 V (us. s.c.e.) Hg cathode DMF Scheme 26 53 J. P. Coleman R. J. Holman and J. H. P. Utley J.C.S.Perkin ZZ,1976 879. 54 H. Lund and E.Hobolth. Acta Gem. Scand. 1976 B30,895.