6 Electro-organic Chemistry By R. LINES Ra ychem Limited Farada y Road Dorcan Swindon SN3 5HH 1 Introduction This review is organized as in previous years according to the nature of the species which is believed to participate in the initial electrode reaction. Of general interest is a report by a Swedish group' on the advantages and use of a capillary-gap cell for large laboratory-scale organic electrosyntheses. 2 Anodic Processes The Anodic Oxidation of Carboxy1ates.-A procedure for the conversion of a-(pheny1thio)carboxylic acids into the corresponding aldehydes or acetals (Scheme 1) under mild conditions has been described by a Japanese group.2 CH3(CH&CHO +(PhS)2 CH3(CH&CH(SPh)C02H mc91O/O CH3(CH2)7CH(OCH3)2+PhS(O)OCH3 Reagents i aq.NaOH; ii MeOH LiClO 95% Scheme 1 The Kolbe reaction has also been used to advantage for the preparation of cyclic and semi-cyclic N-acyl NO-a~etals.~ The definition of semi-cyclic and cyclic N-acyl NO-acetals has been given by Nyberg quoted in ref. 3. Such compounds are precursors of N-acyl-immonium ions which in turn are key intermediates in the synthesis of some amino-acids and alkaloids. The starting materials are N-acyl- prolines or N-acyl-pipecolic acids and these are smoothly converted into the acetals in high yield by anodic oxidation in alcoholic solution (Scheme 2). COR ' COR' Scheme 2 L. Eberson J. Hlavaty L. Jonsson K.Nyberg R. Servin H. Sternrup and L.-G. Wistrand Acta Chem. Scand. (B),1979,33 133. J. Nokami M. Kawada and R. Okawara Tetrahedron Letters 1979 1045.'T. Iwasaki H. Horikawa K. Matsumoto and M. Miyoshi J. Org. Chem. 1979,44 1552. 102 R. Lines The finding that malonic acid derivatives can function as ketone synthons has been applied by a Japanese group4 to the synthesis of jasmones in 43% overall yield starting from diethyl malonate (Scheme 3). R' ' "ZH Pt electrodes ____ R~'CO,H 10-12Fmol~' R2 Scheme 3 The synthetic utility following the electrolytic decarboxylation of 1-alkyl-2-cycloalkene-1-carboxylic acids' is well demonstrated by the preparation of dl-muscone (1; R'= Me n = 12) outlined in Scheme 4. The reaction involves a regiospecific acetoxylation at the a-position of the unsaturated acids and this is used to advantage in the synthesis of dl-muscone by serving as a means of introducing a methyl group at the p -position of the cyclopentadecanone precursor.OAC Reagents i HOAc Bu'OH Scheme 4 The Anodic Oxidation of Neutral Organic Compounds.-The synthetically useful organic intermediates benzoquinone bis-acetals have been found to be readily prepared by the anodic oxidation of 2-substituted 1,4-dimethoxybenzenes in methanolic KOH solution.6 Many of these compounds were also found to undergo selective monohydrolysis at the acetal remote from the 2-substituent to give a good yield of the equally useful mono-acetals. The anodic oxidation of methoxy-naph- thalenes has also been studied in methanolic KOH solution7 and found to be a convenient route to methoxylated naphthalenes and 1,2-naphthoquinone mono- ace tals.The electro-acetoxylation of 4-methylphenyl acetate to 4-acetoxybenzyl acetate has formed the basis of a synthesis of vanillin described by Torii and co-workers.8 A detailed study of the electro-acetoxylation conditions revealed that the products in which the side-chain was oxidized could be obtained in high yield (90%) and with high selectivity (88%)by the addition of metal acetates especially copper(I1) acetate to the acetic acid-t-butyl alcohol electrolyte. The metal ions are thought' to promote the smooth oxidation of the benzyl radicals to benzyl cation intermediates. J. Nokami T. Yamamoto M. Kawada M. Izumi N. Ochi and R. Okawara Tetrahedron Letters 1979 1047. S. Torii T. Inokuchi K. Mizuguchi and M. Yamazaki J.Org. Chem. 1979,442303. D. R. Henton B. L. Chenard and J. S. Swenton J.C.S. Chem. Comm. 1979,326. 'M. G. Dolson D. K. Jackson and J. S. Swenton J.C.S. Chem. Comm. 1979,327. S. Torii H. Tanaka T. Siroi and M. Akada J. Org. Chem. 1979,443305. Elec tro -organic Chemistry Methanolic NaCN appears to be an excellent medium for the anodic cyanation of pyrroles' and methyl-naphthalenes. lo For a series of 1-alkyl-pyrroles,' substitution was found to take place pfeferentially at a vacant 2- or 5-position while the 1,2,5-triaIkyl-derivativesunderwent cyanation in the side-chain to form dialkyl- pyrrole-2-acetonitriles. An interesting feature of this reaction is that in contrast to the anodic cyanation of cyclohexene in the same reaction medium [see Ann.Reports (B) 1977 74 1551 no isocyanides or methoxylated products were observed.' Oxidation of some methyl-naphthalenes in methanolic NaCN resulted in a variety of naphthonitriles." The orientation of attack was found to be directed by the substituent effect of the methyl groups rather than by any steric factors. In many cases the yields of products from the cyanation of pyrroles and methyl-naphthalenes are sufficient to make the methods preparatively attractive; e.g. Scheme 5. CN Reagents i MeOH NaCN Scheme 5 A novel anodic dimerization has been discovered" in which the enamine (2) undergoes coupling between the @-carbon atom and the 4-position of the phenyl ring (see Scheme 6). Until now all of the reported oxidative dimerizations of enamines have occurred predominantly at the ethylenic carbon p to the nitrogen atom.Scheme 6 Another novel observation this time concerning a cleavage reaction relates to the stereoselective oxidation of (dl)-4,4'-dihydroxyhydrobenzoin to 4-hydroxy-benzaldehyde at a mild steel anode.'* Under the electrolysis conditions (aqueous NaOH pH 13.5 2-3 V us. SCE) the meso-form was found to be completely electro-inactive. An explanation advanced by the author'* is that a complex between the oxide coating on the steel anode and the hydroxy-groups of the glycol is a necessary prerequisite for charge transfer. For the meso-compound the phenyl rings prevent a favourable alignment of the hydroxy-groups thus inhibiting oxida- tion. Some preliminary findings have been published on the oxidation of a series of strained polycyclic hydrocarbons.l3 An excellent correlation was found between the E; potentials and the HOMO's of the compounds studied. For these compounds the HOMO's are associated with a carbon-carbon rather than a carbon-hydrogen bond K. Yoshida J. Amer. Chem. SOC.,1979,101 2116. '' K. Yoshida and S.Nagase J. Amer. Chem. SOC.,1979 101 4268. M. Masui T. Michida C. Ueda and H. Ohrnori Chem. and Ind. 1978,922. l2 J. H. Wagenknecht J. Electrochem. SOC.,1979,126,983. l3 P. G. Gassrnan and R. Yamaguchi J. Amer. Chem. SOC.,1979,101 1308. 104 R. Lines and indeed on oxidation in pyridine-methanol tricyclo[4.1.1 .0.02*7]heptane for example gave predominantly the acetal (3). A study has been made of the oxidation of a series of mono- and di-thioethers and of a number of mesocyclic dithioether~.'~ The latter compounds underwent facile (3) (4) reversible oxidation in contrast to the irreversible oxidation of the non-cyclic derivatives.The difference in behaviour was attrib~ted'~ to transannular inter- actions between the sulphur atoms of the products of the oxidation of the mesocyclic compounds. From their observations the author~'~ concluded that transannular interaction depended on the fulfilment of two conditions viz. the formation of two five- or six-membered rings coupled with geometric constraint. Thus for example 1,5-dithiacyclo-octane gave the dication (4) whereas dibenzo-dithiacyclododecane underwent irreversible oxidation. 3 Cathodic Processes The Cathodic Reduction of Organic Cations.-Following last years report of the formation of a pyrrolodiazepine from the electrolysis of a dihydrodiazepine salt [see Ann.Reports (B) 1978 75 1391 the reduction of the NN'-disubstituted dihy- drodiazepine salt (5) has been studied." The presence of the substituents on the nitrogen atoms precludes the reaction pathway reported previously and instead three products were isolated viz. (6) and the meso and racemic forms of (7). It has yet to be ascertained whether the formation of (6)precedes or follows the dimeriza- tion of (5) to (7). The Cathodic Reduction of Neutral Organic Compounds.-The cleavage of diary1 sulphones in aprotic solvents to the corresponding aryl sulphinate ion and aromatic hydrocarbon has until now been thought to proceed entirely uia charged inter- mediates i.e.anion radicals and dianions. Evidence has now been presented to suggest that radical intermediates are involved.16 Thus reduction of (8) in DMF gave (9) by cleavage of a carbon-sulphur bond in the anion radical of (8),followed by an intramolecular radical substitution on the adjacent phenyl ring. l4 G. S. Wilson D. D. Swanson J. T. Klug R. S. Glass M. D. Ryan and W. K. Musker J. Amer. Chem.SOC. 1979,101,1040. '' D. Lloyd C. A. Vincent and D. J. Walton Bull. SOC. chim. belges 1979,88 113. l6 J. Grimshaw and J. Trocha-Grimshaw J.C.S. Perkin I 1979,799. Elec tro-orga n ic Chemistry The electroreduction of a number of ao-9,9'-dianthryl-ethanes[m(CH2),,7r]has been st~died,'~ with a view to determining how the interactions between the aromatic centres affected the reversible electrode potentials.The results showed that strong interactions operated for the compounds with n = 0 and n = 2 but decreased as the length of the methylene chain increased. At n = 6 the two successive charge transfers (anion radical to dianion) merged into a single wave with a peak width close to that of a reversible one-electron transfer confirming the independence of the electro-active centres. In a separate study of bianthryl (i.e. n = 0),l8fragmentation into anthracene and 9,lO-dihydroanthracene was observed upon electrolysis at the potential of the second wave and above. This represents the first example of the direct electrochemical cleavage of a biaryl into the corresponding monomer.A very useful example of the selective protection of the primary or secondary hydroxy-group in a diol has been presented by van der Stouwe and Schafer." The authors found that tritylone ethers can be cathodically cleaved in high yield and under mild conditions to the corresponding alcohols. By taking advantage of the different reduction potentials of the tritylone and p -cyanobenzyl ether protecting groups the selective protection of the hydroxy-groups of pentane- 1,4-diol was demonstrated. The electroreduction of some cyclic ethers namely epoxides of structure (lo) has also shown interesting behaviour.20 The direct reduction of the epoxides in DMF at mercury afforded products (alchohols and hydrocarbons) by a two-electron pathway.Ph R2 \ / c-c R \0/ \R3 The same products were also obtained by indirect reduction using electrogenerated reducing agents and autocatalysis was observed in some cases. Thus for (10; R2= Ph) ring opening followed by elimination of OH-generated trans-stilbene which was then able to form a stable anion radical and thereby once a suitable concentration of it had been attained undergo homogeneous electron exchange with the substrate. " K. Itaya and A. J. Bard 2.Phys. Chem. (Frankfurt) 1978,112 1. l8 0.Hammerich and J.-M. Saveant J.C.S. Chem. Comm. 1979,938. l9 C. van der Stouwe and H. J. Schafer Tetrahedron Letters 1979 2643. *' K. Boujlel and J. Simonet Electrochim. Acta 1979,24,481. 106 R. Lines Two have appeared that describe an improved synthesis of retinal pinacol(l1) by the reductive dimerization of retinal (12).One of the methods22 uses CrC13.6H20 as a modifier whilst _the other21 uses carbon acids such as diethyl malonate as the proton donor. Both procedures seem to owe their success to the abilities of the additives to form complexes with retinal or its anion radical and thereby to direct the coupling reaction towards pinacol formation. Retinal pinacol (1 1) provides an entry into the C40 carotenoids and as such an efficient method of electrochemically cleaving allylic hydroxy-groups would be of great synthetic value. The leaving ability of such a group is however rather low. It seems appropriate then that the methyl ester of abscisic acid (13) has been chosen as a model compound for the study of the conditions required for hydroxyl cleavage.23 The hydroxy-group attached to C-6 is activated both by the enone and by the unsaturated side-chain ester.Contrary to expectations it was found that upon reduction the bicyclic ester (14) was obtained in good yield. It appears that the hydroxy-group by preventing conjugation throughout the molecule allows selective transfer of electrons to the side-chain. The radical anion that is so formed isomerizes and undergoes rapid cyclization. 0m 0 2 M e The formation of alkenes following the reductive elimination of acetate from vicinal diacetates can be rationalized either by a stepwise or a concerted process. The observation of hydrogenolysis of one of the acetate groups of diacetoxystilbene upon reduction in aqueous DMF24 has provided good evidence for the stepwise mechanism.In contrast with dry DMF where diphenylacetylene is the major product the aqueous medium is sufficiently acidic that the intermediate carbanion is protonated. Reduction of 9,lO-diacyloxy-phenanthrenesled to an unexpected acyl-oxygen fission.24 Phenanthraquinone and the appropriate carboxylic acid were isolated. Following a previous report of the formation of NO-dialkyl-hydroxylamines from the electrochemical reduction of nitro-compounds in the presence of alkyl halides ” L. A. Powell and R. M. Wightman J. Amer. Chem. SOC.,1979,101,4412. ’’ D. W. Sopher and J. H. P. Utley J.C.S. Chem. Comm. 1979 1087. 23 B. Terem and J.H. P. Utley Electrochim. Actu 1979,24 1081. 24 C. Adams N. M. Kamkar and J. H. P. Utley J.C.S. Perkin ZI 1979 1767. Electro-organic Chemistry 107 [see Ann. Reports (B) 1977 74 1583 a description of the efficient synthesis of NO-diacetyl-hydroxylamines has been published.25 The method'consists of directly reducing aliphatic or aromatic nitro-(and nitroso-)compounds in the presence of acetic anhydride. Isolated yields range from 40 to 80%. Another efficient transformation concerns the selective reduction of aliphatic alcohols to alkanes via the methanesulphonate esters.26 The method appears effective even when the alcohol contains other functional groups such as ester alkene nitrile or epoxide. A variety of 1,l-dichloro-alkenes have been ~ynthesized,~~ in good yields by the reduction of diethyl trichloromethylphosphonate and the appropriate carbonyl compounds (Scheme 7).The method has the advantage of avoiding the strongly basic and expensive alkyl-lithium reagents. 0 0 II Hg cathode R'R~CO C12CP(OEt)2-R'R2C=CC12 C13CP(0Et)2 -1.2 V 0s. SCE ' -11 Scheme 7 The electroreduction of some N-ethyl-maleimides has led to the discovery of a hydrodimerization reaction in aqueous Normally aprotic conditions are required for the formation of hydro-dimers from diactivated alkenes where pro- tonation of the intermediate anion radicals competes with their dimerization. In the present example a radical-radical coupling mechanism is thought to operate. Two papers have appeared on the synthesis of substituted barbaralane~.~~*~~ The first paper29 reports on the reduction of bridged 1,5-benzodiazepines (15) in acetonitrile to give substituted barbaralanes (16) in good yield.In the second paper,3o substituted barbaralanes e.g. (17) were obtained by the reduction of bicyclo[3.3. llnonadienediones (18) in the presence of acetic anhydride at the potential of the second wave. Reduction of the substrates at the first wave gave dimeric products only. Aco&Ac (17) X Y =H C1 or Me (18) 25 L. Christensen and P. E. Iversen Acta Chem. Scand. (B),1979,33,352. '.s T. Shono Y. Matsumura K. Tsubata and Y. Sugihara Tetrahedron Letters 1979 2157. '' F. Karrenbrock H. J. Schafer and I. Langer Tetrahedron Letters 1979 2915. 28. H. Zoutendam and P.T. Kissinger J. Org. Chem. 1979 44 758. 29 J. M. Mellor B. S. Pons and J. H. A. Stibbard J.C.S. Chem. Comm. 1979,761. 30 J. M. Mellor B. S. Pons and J. H. A. Stibbard J.C.S. Chem. Comm. 1979 759. 108 R. Lines Evidence has been presented for the reversible formation of the tetranegative ion of the aromatic hydrocarbon de~acyclene.~~ Cyclic voltammetry of the hydrocarbon in dimethoxyethane at 0 "C gave four reversible one-electron waves the fourth being located at -3.35 V (u.s.Agl Ag'). This ion represents the most reduced form of an aromatic hydrocarbon yet reported. Both electrogenerated anions32 and radical diani~ns~~ have been cited as electron- transfer reagents in solution. 1-Ethyl-4-methoxycarbonylpyridinium iodide32 furnished the anion upon reduction whilst the radical dianion was generated33 from the conjugated phenol (19)in the presence of a strong base.In each case a solution electron transfer was demonstrated by cyclic voltammetry for the reduction of a benzylic chloride. HoecH=CH-coR (19) Fritz and Kornr~mpf,~~ by the electrolysis of CF2Br2 at platinum electrodes in dichloromethane containing Bu4NBr are the first investigators to obtain the carbene CF electrochemically. Current efficiencies of up to 60% were recorded for the trapping of the carbene by various alkenes. Contrary to some statements that the reduction of allyl bromides and iodides proceeds in two distinct one-electron steps compelling evidence has now been presented3' to show that in fact the reduction follows a single two-electron pathway.Detailed electro-analytical studies of various allyl halides in acetonitrile revealed that peaks that had previously been attributed to the separate reduction of the allyl radical to the anion were the result of specific surface interactions of the allyl halides with the electrode materials. The reductions can thus be formalized as an elce2 process where I?; <Ey. The reduction of organic halides this time of benzene and pyridine is also the subject of a comprehensive paper by French researcher^.^^ Initial electron transfer to these compounds is slow thus placing the electrochemical reaction under kinetic control and this limits the amount of information that can be obtained from the system. The describe how by the analysis of the homogeneous redox catalytic processes occurring during the electrochemical reductions a more complete characterization of the system can be achieved.Following on from last years' report of electrochemically induced nucleophilic substitution [see Ann. Reports (B) 1978 75 1411 two paper^^'.^^ describe the reaction when conducted in liquid ammonia By this means H-atom abstraction by the intermediate aromatic radicals was suppressed so that attention could be 31 T. Saji and S. Aoyagui J. Electroanalyt. Chem. Interfacial Electrochem. 1979,102 139. 32 H. Lund and L. H. Kristensen Acta Chem. Scand. (B),1979,33,495. 33 A. M. Martre and J. Simonet J. Electroanalyt. Chem. Interfacial Electrochem. 1979,97,287. 34 H. P. Fritz and W. Kornrumpf J.Electroanalyt. Chem. Interfacial Electrochem. 1979 100 217. 3s A. J. Bard and A. Merz J. Amer. Chem. Soc. 1979,101,2959. 36 C. P. Andrieux C. Blooman J. M. Dumas-Bouchiat and J.-M. Saveant I.Amer. Chem. Soc. 1979,101 343 1. 37 C. Amatore J. Chaussard J. Pinson J.-M. Saveant and A. Thiebault J. Amer. Chem. Soc. 1979,101 6012. C. Amatore and J.-M. Saveant J. Electroanalyt. Chem. Interfacial Electrochem. 1979,103 303. Electro-organic Chemistry 109 focussed on the other major competing process to substitution viz. electron transfer to the radicals. A detailed analysis of the kinetics of the competitive electron- transfer process using cyclic voltammetry gave good evidence for the correctness of the SRNlmechanism (Scheme 8). ArX + e-$ ArF ArX7 + Ar' + X- (E; = electrolysis potential) Ar' + Y-+ ArYS ArY7 -e- $ ArY (E;,if E; > E;) Scheme 8 4 Miscellaneous Interest continues to be shown in modifying electrode surfaces by applying polymeric films.Two communications have appeared on the electrochemical polymerization of pyrrole~.~~*~~ in acetonitrile containing Et4NBF4 gave a tough Oxidation of pyrr01e~~ film of conductivity 100 R-' cm-'. Elemental analysis suggested a structure containing polymerized pyrrole units and BF ions in the ratio 4:1. The polymer could be modified further by e.g. nitration. Polymer films of N-methylpyrr~le~~ exhibited much lower conductivities n-'cm-') and thus by co-electrolysis of solutions of pyrrole and N-methylpyrrole films of intermediate conductivities could be obtained.The potentialities of using the charge-transfer complex TTF/TCNQ (tetrathiafulvalene/tetracyanoquinodimethane)as an electrode material in aqueous solution have been explored by Jaeger and Bard.41 The complex was found to be stable over a potential span of 0.9V and the oxidation and reduction of redox couples e.g. Curl/Cul could be carried out. Beyond the potential limits insoluble compounds of TTF or TCNQ with the ions of the supporting electrolyte were formed. A further application of the poly(4-nitrostyrene)-coatedelectrode referred to in last years' report [see Ann. Reports (B) 1978 75 1451 has been found for the reduction of 1,2-dibromo-1,2-diphenylethaneto tr~ns-stilbene.~~ Like the reduc- tion of oxygen on this electrode electron transfer to the dibromide via the charged polymer was found to take place 0.4V less cathodically than at a clean platinum surface.The electrode was also found to be sufficiently stable for preparative-scale electrolyses. Catalyst turnover was of the order of lo4. Chemically modified electrodes as well as being able to catalyse charge transfers can also alter isomer ratios a case in point being the selective chlorination of anisole at a cyclodextrin-modified electr~de.~~ At a graphite anode the chlorination of anisole in aqueous NaCl gave a para :ortho ratio of about 5 :1. Chemically bound a-cyclodextrin on the graphite changed the ratio to 18:1 while adsorbed cyclo- dextrin on the electrode surface increased the ratio to >25 :1.In these cases it is that the ortho-positions of anisole are blocked by the cyclodextrin struc- ture leading to preferential para -chlorination. 39 A. F. Diaz and K. K. Kanazawa J.C.S. Chem. Comm. 1979,635. 40 K. K. Kanazawa A. F. Dim R. H. Geiss. W. D. Gill J. F. Kwak J. A. Logan J. F. Rabolt and G. B. Street J.C.S. Chem. Comm. 1979 854. 4 C. D. Jaeger and A. J. Bard J. Amer. Chem. SOC. 1979,101 1690. 42 J. B. Kerr and L. K. Miller J. Electroanalyt. Chem. Interfacial Electrochem. 1979,101 263. 43 T. Matsue M. Fujihira and T. Osa J. Electrochem. Soc. 1979,126 500. 110 I?. Lines Turning to more conventional electrolytic procedures the almost quantitative conversion of 3-substituted sydnones into their 4-chloro- or 4-bromo-derivatives has been dis~overed.~~ The success of the method appears to lie with the mild conditions of the electrolysis (MeOH NaBr or Me4NCl) which minimizes any tendency for ring opening.Attempts at fluorination and iodination of 3-substituted sydnones by this method were unsuccessful. A preparatively significant synthesis of esters and ketones from primary and secondary alcohols has been accomplished by the use of iodide ions as catalytic current carriers.45 Typically electrolysis of an aqueous solution of the alcohol and KI (4 :1ratio) yields the ketone or ester in up to 84% yield. Although the authors4' suggest the iodonium ion as the key intermediate (Scheme 9) it seems unlikely that this species would be generated in an aqueous environment and at the low anode potential (0.6-0.8 V us.SCE). 0 Scheme 9 Liquid SOz as an electrochemical solvent has been re-examined by two group^.^^*^' Liquid SOz has many advantages from an electrochemical viewpoint e.g. high dielectric constant (ca.20) low nucleophilicity and high solvating ability for covalent compounds. Contamination by water can be a problem however and the solubility of most inorganic salts in the solvent is low. Anhydrous liquid SO2can be obtained by vacuum-line techniques46 or by using AlCI as a supporting ele~trolyte.~~ In the latter case the solutions are oxidizing and the cation radicals of a number of aromatic hydrocarbons can be obtained directly. Further electrochemical oxidation to the dications is also possible. In contrast anhydrous liquid SO containing Bu4N'C10i was found to have a very wide anodic range46 (ca.3.4 V us. SCE) and the dications of a variety of aromatic compounds could easily be generated. Halide ions particularly bromide (as sodium bromide) have been found to be excellent promoters of the electrolytic cross-coupling of disulphides with imide~~~ and with dialkyl (or diary]) ph~sphites~~ (Scheme 10). The mechanism is thought to involve electrogenerated bromine which reacts with either the imide or the phos- phite to give an intermediate (N-bromo-imide or phosphorobromide) that is capable of reacting with the disulphides and regenerating bromide ions. 2R1(C0),NH b 2R'(C0)2NSR + H2 2Fio,-l 2(R1O),P(O)H b 2(R10),P(0)SR + H2 2 F mol-I Reagents i RSSR,NaBr Scheme 10 44 Tien T.Nonaka and T. Sekine Chem. Letters 1979 283. H.-J. 45 T. Shono Y.Matsumura J. Hayashi and M. Mizoguchi Tetrahedron Letters 1979 165. 46 L. A. Tinker and A. J. Bard J. Amer. Chem. Soc. 1979,101,2316. 47 P.-C. Lacaze J.-E. Dubois and M. Delamar J.Electroanalyt. Chem. Interfacial Electrochem. 1979,102 135. 48 S. Torii H. Tanaka and M. Ukida J. Org. Chem. 1979,44 1554. 49 S. Torii H. Tanaka and N. Sayo J. Org. Chem. 1979,44,2938 Electro-organic Chemistry Finally an unusual mixed-valence ferrocene compound (20) has been synthesized by the reaction of ferrocenyl-lithium with boron trifl~oride.~' Structure deter- minations revealed that the complex contains one sterically and one electronically distinct ferrocene group; these are not coincident.The electronic absorption spectrum showed bands characteristic df inter-valence charge-transfer transitions and these were attributed" to a through-space rather than to a through-bond mechanism. Cyclic voltammetry of the complex in dichloromethane gave one reversible reduction wave and three reversible one-electron oxidation waves (Scheme 11). Fc = (.rr-CloH&Fe Scheme 11 50 D. 0.Cowan P. Shu F. L. Hedberg M. Rossi,andT. J. Kistenmacher J. Arner. Chern. SOC.,1979,101 1304.