6 Electro-organic Chemistry By J. H. P. UTLEY* Department of Chemistry Queen Mary College Mile End Road London El 4NS 1 General and Mechanistic It is the custom in this section to concentrate upon preparatively useful electro- organic reactions and purely mechanistic studies have previously received scant attention. Recent advances in technique and interpretation have been such however that it is timely to summarize some of the more important findings. Expert reviews have appeared concerning two burgeoning fields of study i.e. electro-catalytic reactions' and chemically modified electrodes.' The latter review is helpful in that it concentrates upon the more rigorous and reproducible work in this area. The details of electron transfer at an electrode or between species in homogeneous solution are much discussed and an important contribution has been made3 concern- ing the distance over which electrons may rapidly be transferred.The macrocycle (1)is reduced cathodically in a two-electron step i.e. its radical anion disproportion- ates readily with Kd= 0.14* 0.1. However partial reduction (with potassium) allows e.s.r. spectroscopic observation of the radical anion and allowance can be made for contamination with the dianion. The spectra show that the benzene rings of the radical anion of (1)are non-equivalent but that electron transfer between the benzene rings is detectable on the e.s.r. time-scale when the macrocycle is com- plexed with sodium or potassium ions. It was consequently determined in the range 309 to 195 K that the energy of activation for electron transfer over the inter- nuclear distance of 7 A was 5.85 kJ mol-' with a pre-exponential factor of 4.6* 0.1 x lo7s-'; at 291 K tb rate constant for electron transfer is 3.9 x lo6s-'.Theoretical treatments of the rates and energetics of electro-organic reactions are often hampered by lack of fundamental data such as reversible electrode potentials. (1) * The author gratefully acknowledges a current-awareness service provided by Helen Thomas (of Queen Mary College Library). ' J. M. Saveant Acc. Chem. Res. 1980,18,323. R. W.Murray Acc. Chem. Res. 1980,18,135. S. Mazur V.M. Dixit and F. Gerson J. Am. Chem. SOC.,1980 102,5343. 79 J. H. P.Utley In recent years voltammetric studies at fast scan rates have provided much informa- tion but the technique of choice for such measurements is second-harmonic a.c.voltammetry. A detailed account has been published4 of the scope and limitations of the technique. Parker and his group are also engaged in the useful process of simplifying (for more general use) the sophisticated voltammetric techniques and the interpretation of the relevant data. In this context it has been shown that potentials at half peak height (EPl2) may be measuredS to f1mV from a pen-and-ink trace. A large amount of work has concerned the use of plots of dE/d[log (scan rate)] and of dE/d[log (concentration)] in diagnosing the sequence of electron-transfer and chemical-reaction steps; the general conclusions of this approach have been summarized in a qualitative set of relatively easily applied rules.6 It is claimed7 that double potential-step chronoamperometry is particularly useful for discriminating between the ECE and the DISP 1mechanisms; the two mechan- isms are exemplified for hydrogenation of aromatic hydrocarbons in Scheme 1.The point at issue is whether the second electron-transfer process occurs at the electrode or in solution by homogeneous electron transfer.For hydrogenation of anthracene and naphthalene the clear conclusion is that the DISP 1mechanism is followed. The A & A7 ArX & ArX' [ArX = 2-chloroquinoline] k AT + PhOH + AH' + PhO-ArX' -b Ar' + X-AH' + A; + AH-+ A (DISP 1) Ar' + Nu-+ ArNu' [Nu-= PhS-] [or AH' + e-+ AH-(ECE)] ArNu' + ArX + ArNu + ArX' (DISP 1) AH-+ PhOH + AH;?+ PhO-[or ArNu' -e-+ ArNu (ECE)] Scheme 1 Scheme 2 general comment is made' that the ECE mechanism has not yet been observed under the conditions of voltammetric or amperometric techniques i.e.at low concentration and with non-steady-state measurements. In an example of an electrochemically induced chemical reaction (SRN1), namely that involving 2-chloroquinoline and thiophenolate anion in liquid ammonia solution (Scheme 2) a thorough analysis* of the kinetic characteristics of the reaction suggests that the DISP 1route is followed for experiments involving high rates of potential scan whereas at low scan rates the results are best accommodated by the ECE route. For this particular reaction the rate constant kl for cleavage of the initially formed radical anion is 1.7 x lo4s-l.Similar reactions take place in solutions in dimethyl sulphoxide (DMSO) and for these cases a question has been raised' concerning the likely importance of hydrogen abstraction from the solvent by intermediate aryl radicals; it has previously been reported" that a-naphthyl radicals react exclusively with thiophenolate anion with no concurrent hydrogen abstraction in DMSO solution. By generating a-naphthyl E. Ahlberg and V. D. Parker Acta Chem. Scand. Ser. B 1980,34,91. B. Aalstad and V. D. Parker J. Electroanal. Chem. Interfacial Electrochem. 1980,112 163. V. D.Parker Acta Chem. Scand. Ser. B 1980,34,359. C. Amatore and J. M. Saveant J. Electroanal. Chem.Interfacial Electrochem. 1980,107 353. C. Amatore J. M. Saveant and A. Thiebault J. Electroanal. Chem. Interfacial Electrochem. 1979,103 303. B.Helgee and V. D. Parker Acta Chem. Scand. Ser. B 1980,34,129. lo J. Pinson and J. M. Saveant J. Am. Chem. SOC. 1978,100 1506. Electro-organic Chemistry radicals in homogeneous solution the absolute rates of reaction with various solvents and with thiophenolate anion have been measured.' For the reaction between a-naphthyl radical and DMSO DMF acetonitrile and thiophenolate in DMSO the rate constants are respectively 3 x lo5 8 x lo6 2.5 x lo5 and 1.7 x lo81mol-' s-l. Rates of abstraction of hydrogen are therefore significant. The anodic pyridination of 9,lO-diphenylanthracene (DPA) has for some years been an example of at least one electrochemical reaction with a well-understood ECE mechanism.Linear-sweep voltammetry reveals,ll however that at low con- centrations the rate of disappearance of 9,lO-diphenylanthracene radical cation has a second-order dependence on its concentration. The voltammetric data are best explained by Scheme 3. \ rn Scheme 3 There is now considerable evidence to show that at least under conditions amenable to study by electroanalytical techniques alkyl-aromatics are oxidized anodically according to the DISP 1rather than to the ECE mechanism (Scheme 4). The combined results12 of electrochemical and spectroelectrochemical methods are particularly persuasive. For the oxidation of pentamethylbenzene in acetonitrile the decay with time of the corresponding benzyl cation (ArCH,') initially formed anodically was followed by reflectance spectroscopy.The observed curve nicely fits that predicted for the DISP 1mechanism and is clearly distinct from the very rapid decay predicted for the ECE route. The loss of proton from the radical cation is judged to be the slow step the subsequent oxidation in homogeneous solution being diff usion-controlled. Similar conclusions have been reached following the applica- tion13 of cyclic voltammetry and double-step chronoamperometry to the anodic RH RH? RHt-* R'+ H' (slow?) RH' + R' S Rf + RH (DISP 1) [or R' -e RC (ECE)] + R+ + MeCN + RN=CMe Scheme 4 E. Ahlberg and V. D. Parker Acta Chem. Scand. Ser. B,1980 34 97.l2 A. Bewick J. M. Mellor and B. S. Pons Electrochim. Acta 1980,25 931. l3 (a) J. Barek E. Ahlberg and V. D. Parker Acta Chem. Scand. Ser. B 1980 34 85; (b) R.S. Baumberger and V. D. Parker ibid. p. 537. J. H. P. Utley oxidation of hexamethylbenzene in methylene chloride-trifluoroacetic acid (TFA) solution or in acetonitrile with and without added TFA. In the latter study however it is suggested that the rate of deprotonation of radical cation will depend on the nature and concentration of an unspecified base and that in acetonitrile solution the rate of reaction between benzyl cation and solvent is sufficiently slow possibly to qualify as a rate-limiting step. The SRNl reaction (see above) is now well established and it is not surprising therefore that formal consideration has been given14 to electrophilic substitutions induced at electrodes.It has been suggested that the reactions outlined in Scheme 5 should be observable; in their de~ignation'~" as SoEl,SRE2 and SON29 0and R (a) SoEl (for H-D exchange) (6) SRE2 (forH-D exchange) RH+ + B + R' + BH+ RH; + BD -+ [RHD]' + B- R' + BD+ -+ RDt + B [RHD]' + B- -+ RD7 + BH RDf + RH -P RD + RHf RD' + RH -+ RD + RHT (c) SON2 (nucleophilic substitution) RXt + Nu-B [RXNu]' [RXNu]' -+ RNut + X-RNut + RX + RNu + RX? Scheme 5 stand for oxidative and reductive respectively and the other symbols have the usual mechanistic meaning. (In SRNl the R stands for radical!) Attention has been drawn14b to several possible cases of the SON2 mechanism and one of the most convincing is the previously puzzling acetoxylation of 4-fluoroanisole which involves dis- placement of fluorine [cf.Annu.Rep. Prog. Chem. Sect. B 1976,73 1421. Further work on this reaction has now shown that at low conversions and at low substrate concentration the current efficiency approaches 300% i.e. it is catalytic and these observations may be explained by the SON2 mechanism (Scheme 6). -e ArF F=? ArFt (ArF = p-MeOC6H4F) ArFt + AcO-+ [Ar/F -+ ArOAc' + F-]* 'OAc ArOAc+ + ArF -+ ArOAc + ArFt Scheme 6 l4 (a) R. W. Alder J. Chem. Soc. Chem. Commun. 1980,1184; (b)L. Eberson and L. Jonsson ibid. p. 1187. Electru-organic Chemistry A thorough re-examination has been made15 of the voltammetry of 9-diazofluorene (FIN2).The cyclic voltammetry is complex and several redox couples are observed; profound changes with temperature are seen with various couples dominating at various temperatures. The observations are well rationalized by a scheme in which the key features are the formation of a dimeric dianion which is relatively long-lived at 223 K but which at higher temperatures may lose nitrogen to give a cis-azine dianion which can then undergo isomerization and oxidation to the fluorenone azine product (Scheme 7).The loss of nitrogen from the initially formed radical anion [cf. Annu. Rep. Prug. Chern. Sect. B 1977 74 1621 is also considered unlikely as in linear-sweep voltammetry the gradients of plots of dE/d[log (scan rate)] and of dE/d[log (concentration)] would be respectively 29.6 and OmV per decade.The values now determinedI5' are 20.7 and 19.2mV per decade i.e. near to those predicted for the mechanism in Scheme 7. Fl-F1-F1-\ \/ N=N N-N -N FIN;?' I II -A dr Fl-/ N-N F1-\N=N (F1 = 9-fluorenyl) \F1-Scheme 7 That radical anions may be photo-excited to even more highly reducing species has been proved in an ingenious fashion.I6 The basic idea is that if photoexcitation induces rapid transfer of electrons from the radical anion the initial substrate is regenerated and it may accept another electron from the cathode with a concom- itant increase in current i.e.photochemical reduction is catalysed. The experimental difficulty is that cyclic voltammetric peak currents are diff usion-controlled and any incursion of convection caused by the local heating of photoexcitation will also increase peak currents.The temperature rise close to the electrode can easily be of the order of 10-15 K. Several pieces of evidence16 now support the contention that the major factor causing increase of current in these systems is electron transfer from the photoexcited state. For pyrene in DMF with chlorobenzene present as an irreversible electron acceptor concurrent U.V. irradiation and measurement of the peak current shows that around 500nm the peak-current heights vary with the absorbance. Even more compelling is the observation that for concurrent irradiation and cyclic voltammetry of dimethyl terephthalate (DTP)in the presence of chlorobenzene considerable enhancement of the second wave is found (formation of DTP2-) whereas in the same experiment only slight increases in current are seen for the first wave (formation of DTP').The explanation is that only the photoexcited dianion is capable of reducing chlorobenzene (Scheme 8). Discussion of the mechanism of the anodic cleavage of dibenzyl ether continues" [see Annu. Rep. Prug. Chem. Sect. B 1972,69,310; 1975,72,157]. As a result of l5 (a)D. Bethell P. J. Galsworthy K. L. Handoo and V. D. Parker J. Chem. Soc. Chem. Commun. 1980 534;(6)V.D.Parker and D. Bethell Acru Chem. Scand. Ser. B 1980 34,617. l6 H.S.Carlsson and H. Lund Acra Chem. Scand. Ser. B 1980,34,409. l7 (a)J. W. Boyd P. W. Schmalzl and L. L. Miller J. Am. Chem. Soc. 1980,102,3856; (b)E.A.Mayeda L.L. Miller and J. F. Wolf ibid. 1972,94,6812;(c)R.Lines and J. H. P. Utley J. Chem. Soc. Perkin Trans. 2 1977,803. J. H. P. Utley DTP & DTPT % [DTP']*(unreactive) 11.-PhCl DTP2--%[DTP2-]* -DTP' + Ph' + C1-T e Scheme 8 the most recent st~dy'~" there is better agreement on experimental facts; it is now acknowledged that in acetonitrile benzoic acid is produced during work-up and not directly as was originally rep~rted,"~ and also that benzylacetamide is a major product which was initially. The main point is whether in neutral or acidic solution the reactive intermediates are the cation (2) which is produced by deprotonation of the radical cation of the ether with further oxidation or PhCH20' and PhCH2+ which are formed by fission of the radical cation of the ether.17c From the relative amounts of labelled products following oxidation of PhCH20CD2Ph values of kH/kD depending on electrolyte conditions of 1.9k0.2 and 1.56 (error limits not given!) are and claimed as primary isotope effects connected with loss of a proton from the radical cation of the ether i.e.in support of intermediate (2). The possibility remains however that these are secondary isotope effects or the small primary effects that are characteristic of radical abstraction. PhcHOCH2Ph (2) These ambiguities combined with the imprecision of the experiments and the fact that 180-labelling show that oxygen in each of the products derives from the ether leave the question open. Hydrogen-deuterium kinetic isotope effects in the range kH/kD= 1.4 to 1.6f0.2 have also been used to support the suggestion of intramolecular abstraction of hydrogen in the anodic oxidation" of hexan-2-one (Scheme 9); similar values of kH/kD are obtained for rearrangement following photochemical excitation and electron-impact ionization.li Reagents i -e- MeCN H20 Scheme 9 l8 M. M. Green G. J. Mayotte L. Meites and D. Forsyth J. Am. Chem. SOC.,1980,102,1464. Elec tro -orga nic Chemistry 85 2 Anodic Processes The Kolbe reaction remains the most versatile method for the unambiguous syn- thesis of hydrocarbons and fatty acids. Schafer's studies in this direction continue" with the synthesis by successive Kolbe couplings of a pheromone (3)of the German cockroach.ClgH37CH(Me)[CH2I7CH(Me)COMe (3) The desilylation of cations is rapid compared with competing deprotonation and this has been put to good use in a synthesis2' of terminal alkenes. The method is highly selective seems to have considerable scope and there is good entry to the required carboxylate (Scheme 10).Similarly the oxidation21 of hydroquinone silyl ethers to quinones is very efficient. RCH=CH2 i ii (Et02C)2CH2 +Me3SiCH2CH(R)C02H-% [Me3SiCH26HR] -B + Me3Si+ Reagents i Base Me,SiCH2Cl; ii base RX; iii carbon anode MeCN-MeOH Scheme 10 A thorough study has been reported22 of the anodic alkoxylation of cyclic lactams; lowest yields (ca 40%) of the a-alkoxylated products (4) are found for n = 3 R=n-C4H9 whereas highest yields (ca 90Y0)are when n =4 R=Me.A related reaction has been used23 for the synthesis (Scheme 11)of several lactones one of which i.e. (9,is used as an intermediate in a route to (*)-eburnamonine (6). Lactonization does not take place in the absence of an angular substituent. @co2H ilJ$J0 ' 0 H [58%] R=Et (5) .. (6) H % [42'/0] R = CH~OAC [21%] R = C02Me Reagents i Pt anode MeCN-H,O constant current 4 F mol-'; ii several steps Scheme 11 l9 W. Seidel and H. J. Schiifer Chem. Ber. 1980 113,451. *' T. Shono H.Ohmizu and N. Kise Chem. Left.,1980,1517. 21 R. F. Stewart and L. L. Miller J. Am. Chem. SOC.,1980 102,4999. 22 M. Mitzlaff K. Warning and H. Rehling Synthesis 1980 315. 23 K. Irie M. Okita T. Wakamatu and Y.Ban Nouo.J. Chim.1980. 4 275. J. H. P. Utley The anodic of disulphides and diphenyl diselenide in acetonitrile solutions gives an electrophilic species that is capable of trans addition to alkenes. For disulphides at it has been supposed that the intermediate is a nitrilium + ion such as RSN=CMe. In the presence of acetate sulphides undergo" anodic a-acetoxylation which provides an efficient route to unsaturated sulphides. Examples of the application of each of these reactions are given in Scheme 12. Me(CH2)&H=CH2 Me(CH2)5CH(NHCOMe)CH2SPh[84%] " cyclohexene & mNHCOMe C0,Me AcO C0,Me C0,Me A S P h --*iii d s p h d S p h [7So/o] [94%] Conditions i Divided cell at +1.4 V (us AglAg') 2F mol-' MeCN (PhS),; ii as i except that at +1.3 V (PhSe), iii HOAc NaOAc constant current 2-3 F mol-'; iv 163-6 "C at 0.45 Torr Scheme 12 Zinc octaethylporphyrin (7) may be cyanated at the meso-positions anodically,26 via the cation radical; by variation of the oxidation potential and the extent of electrolysis it is possible to obtain optimum yields of mono- di- tri- and tetra- cyano-octaethylporphyrins.The monocyano-product may be obtained quantita- tively. Both dicyano-isomers are obtained in a 1 :1 mixture with an overall yield of 78%. At the other limit the tetracyanoporphyrin is obtained in 40% yield. (7) The anodic decarb~xylation~~ of tetrahydro-@-carbolinecarboxylic acids (Scheme 13) is achieved at very low potentials [0.13-0.77 V (us S.C.E.)]. This is almost certainly an example of the pseudo-Kolbe reaction [see Annu.Rep. Prog. Chem. Sect. B 1971,68 313; 1976,73 1391 whereby rapid decarboxylation of a radical cation is followed by further oxidation. 24 (a)A. Bewick D. E. Coe J. M. Mellor and D. J. Walton J. Chem. SOC.,Chem. Commun. 1980,51; (b)A. Bewick D. E. Coe G. B. Fuller and J. M. Mellor TetrahedronLett. 1980,21 3827. 25 J. Nokami M. Hatate S. Wakabayashi and R. Okawara TetrahedronLett. 1980,21,2557. 26 H.J. Callot A. Louati and M. Gross TetrahedronLett. 1980 21 3281. 27 J. M. Bobbitt and J. P. Willis J. Org. Chem. 1980,45,1978. Electro-organic Chemistry N Me [6O%] Conditions i Carbon felt anode MeOH-H,O divided cell phosphate buffer Scheme 13 3 Cathodic Processes 1,4-Diazepinium cations provide an e~ample~**~~ where one-electron reduction to a radical intermediate is clearly defined; formation of an anion by transfer of a second electron is possible but it occurs at ca 0.8 V cathodic of the first reduction wave.28 Controlled-potential reduction of the 5,7-diphenyl derivative in DMF at the first reduction potential gives a 1Fmol-' reaction and the most likely fate of the radical is disproportionation.In contrast the reduction of the 6-phenyl deriva- tive gives one-electron reduction with rearrangement to a single product which is isolated in 92% yield. The contrasting results are summarized in Scheme 14. [47%] (hydrolyses on work-up) H Conditions i DMF at -1.23 V (us AglAgCl) 1Fmol-' [92%] Scheme 14 28 D. Lloyd C. A. Vincent and D. J. Walton J.Chem. SOC., Perkin Trans. 2 1980,668. 29 D. Lloyd C. Nymo,C. A. Vincent and D. J. Walton J. Chem. Soc. Perkin Trans. 2 1980 1441. CH,Br / \ + Conditions i DMF A1 cathode undivided cell 0.7 A cm-* 2.5F mol-' Scheme 16 30 T. Shono Y. Usui T. Mizutani and H. Hamaguchi Tetrahedron Lett. 1980 21,3073. 31 J. H.P. Utley and A. Webber J. Chem. SOC.,Perkin Trans. 1 1980 1154. Elec tro -orga nic Chemistry A surprisingly simple and potentially large-scale method for the cathodic Birch reduction of benzene has been described.32 In an undivided cell with an electrolyte of tetra-n-butylammonium hydroxide in aqueous ethylene glycol benzene is reduced to cyclohexa-l,4-diene in 81'/o yield accompanied by about 10% of cyclohexene.Large amounts of the quaternary hydroxide are required and it must be that reduction by solvated electrons is possible at an essentially hydrophobic cathode environment with sufficient benzene dissolved in the electrolyte. Shono and his group have cleverly exploited33 cathodic elimination reactions in a method for extending the carbon chain of carbonyl compounds. Several examples are given and the yields of the enol ethers or sulphides are high (60-96%). The reaction is run at constant current and presumably cathode potentials are relatively high which will limit the method to compounds without other easily reducible functions. Some examples are contained in Scheme 17. PhCHZCH2CHO PhCH2CH,CH(OH)CH(SPh)2 & PhCH2CH2CH=CHSPh PhCH2CH2CH2CHO [92O/o ] Reagents i (PhS),CHLi; ii Pb cathode divided cell DMF-Et,NOTs; iii PhS(0Me)CHLi Scheme 17 The ketones (8) and (9) are known to react with respectively lithium in liquid ammonia and a dialkylcopper lithium reagent to give the tricyclic compounds (10) and (11).For both ketones reduction potentials are less cathodic with the sub- stituents at the 5-position than without and this suggests that cathodic cyclization might be effected.In the case of (9) cathodic to (11)is smooth provided that a good hydrogen-atom donor (e.g. isopropyl alcohol) is present to trap the intermediate radical. In the absence of such a donor the dimer (12) is obtained in high yield. The cathodic of (8) was not tried in the presence of a hydrogen-atom donor and in aqueous methanol the major product (71%) was the dimer (13).The results and their rationalization are given in Scheme 18.32 J. P. Coleman and J. H. Wagenknecht U.S. P. 4 187 156 (1980). 33 T. Shono Y. Matsumura and S. Kashimura Tetrahedron Lett. 1980 21 1545. R. A. J. Smith and D. J. Hannah Tetrahedron Lett. 1980 21 1081. 35 L. Mandell H. Hamilton and R. A. Day J. Org. Chern. 1980,45 1710. J. H. P.Utley OMS 0f"r, (9) 0&<*oyJ%% (10) (8) HO (12) X=H2 (13) X=O X Conditions i Hg cathode at -2.2 V (US S.C.E.) DMF-Pr'OH 1.05F mol-'; ii Hg cathode at -1.7 V (us S.C.E.) 50% aq. MeOH 1 Fmol-' Scheme 18 One-step reductive cyclization is achieved when suitable radical anions are allowed to react with 'dielectrophiles'. A full by Degrand and her co- workers contains several impressive examples and some of these are given in Scheme 19.The case involving reduction of an imine [at -1.8 V (us S.C.E.)]is reminiscent of the annelation described in Scheme 15; in Degrand's examples the electro-active species is more clearly defined. Ph N=N N-N + 6 c7 PhN=CHPh -b [59%] [78%] Conditions i Hg cathode DMF Br(CH,),Br cu 3 F mol-' Scheme 19 Selective cleavage at cathodes competes well with alternative methods and for potentially large-scale reactions it may be preferable to reduction with metals for environmental reasons. In this context the conversion of thiophen into 3- bromothiophen has been st~died;~' the key step is selective cleavage of 2,3,5- tribromothiophen. The results of this pragmatic investigation are that in aqueous dioxan high yields are obtained (80-93%) at Hg Pb Zn or graphite cathodes.A route to cyclopropanones has been explored which is based on earlier methods for ring formation in the reduction of 1,3-dihalides. In the latest however protection of the carbonyl function is built in from the start. In the absence of such protection but in nucleophilic solvents which permit hemiacetal formation cycliz- ation is achieved only for highly alkylated compounds (Scheme 20). 36 C. Degrand P. L. Campagnon G. Balot and D. Jacquin J. Org. Chew. 1980.45 1189. " D. Pletcher and M. Razaq J. Appl. Electrochem.,1980 10.575. '' W. J. M. van Tilborg R. Plomp R. dc Ruiter and C. J. Smit Recl. Trav. Chim. Pays-Bas 1980,99,206. Electro-organic Chemistry [75Yo ] [85'/o ] Conditions i at ca -2.5 V (us S.C.E.) MeCN 0 to -15 OC; ii MeOH-MeCN reduction as before.Scheme 20 The cathodic reactions of several classes of organosulphur compounds are proving to be of considerable Strictly the case of reductive rearrangement found for SS-diary1 benzene-1,2-dicarbothioates(14) and shown in Scheme 21 is an SRNlreaction without an external nucleophile. The product of rearrangement is isolated in 85% yield with the consumption of only 0.1F mol-'. Aromatic dithio- and thiol-esters undergo cathodic co~pling,~' with subsequent elimination to form a diarylacetylene; an example is shown in Scheme 22. The formation4* of tetrathioethylenes from dithiocarbonates in nearly quantitative yields (outlined in Scheme 23)is more straightforward involving ring-opening of the radical anion decarbonylation further reduction and finally alkylation by added alkyl halide.0-SAr 0 7orosAr ~cosAr -* Go -+ GO + ArS ' COSAr CSAr I SAr SAr (14) 0-ArS SAr I 0- [85'/o ] Scheme 21 s- s- i I 1 -2RS 2PhCSSR 2PhCSSR7 + PhC-CPh +PhCSCSPh I1 RS SR Y PhMSCSPh PhCECPh [90O/o 3 PhCSS Ph Conditions i Hg cathode; ii 2e- 2 PhCSSR; iii 2e- -2 PhCSS- Scheme 22 39 K. Praefcke C. Weichsel M. Falsig and H. Lund Acta Chem. Scand. Ser. B 1980,34,403. 40 M. Falsig and H. Lund Acta Chem. Scand. Ser. B 1980,34,585. M. Falsig and H. Lund Acra Chem. Scand. Ser. B 1980,34,591. 92 J. H. P. Utley i ii Conditions i Hg cathode at ca -1.5 V (us Ag IAgI) 2 F mol-' DMF; ii RZX Scheme 23 4 Indirect Processes An increasing number of preparatively significant electro-organic reactions may be classed as catalytic and/or indirect reactions.Particularly in those cases where electrode filming is a problem there may be advantage in using the electrode to generate and regenerate an organic or inorganic redox reagent which performs the desired reduction or oxidation in homogeneous solution. For several reactions including the important examples of anodic substitution into the side-chains of alkyl-aromatics and a-acetoxylation and alkoxylation of amides controversy has simmered for years over whether these are direct or indirect oxidations. In these cases a key step would be rapid abstraction of hydrogen by an anodically generated radical (e.g.MeO' NO3',or AcO') with subsequent anodic oxidation of ArCH2 or R'CHNHCOR'. Apart from the nitrate case there has generally been scepticism concerning the methoxyl radical (should it not be 'CH,OH?) and the acetoxyl radical (does it not have a half-life of a mere 60 ns?). In this context Eberson has recently used. bond energy-bond order (BEBO) and equibonding methods to estimate42 activation energies for relevant hydrogen-abstraction reactions and found them to be relatively small .e.g. 12.5-16.5 kJ mol-'. Where comparison with experiment is possible agreement between observed and calculated values is good; the calcula- tions are however wildly out for abstraction by cyano radicals. The indications are that the abstraction processes are likely to be so rapid that they may need more serious consideration.Several interesting examples of organic redox reactions that are catalysed by inorganic species have recently been An especially versatile system is that using43 the [(trpy)(bpy)R~(OH)~]'+-[(trpy)(bpy)RuO]'+ couple. These com- plexes were chosen because of their relative stability and indeed catalyst recovery is ca 75% after 100 catalytic cycles. Using this system dehydrogenation reactions may be effected e.g. of alcohols to carbonyl compounds as well as oxygenation reactions such as ArCH3 to ArC02H and cyclohexene successively to cyclo- hexenone and quinone. Ethylene is not oxidized in this system. The example given in Scheme 24 is relatively trivial but it does exemplify a likely mechanism.(bpy)= 2,2'-bipyridyl (trpy) =2,2',2"-terpyridyl Conditions i Pt or vitreous carbon anode at 0.6-0.8V (us S.C.E.) Scheme 24 42 L. Eberson Acfa Chem. Scand. Ser. B 1980,34,481. 43 B. A. Moyer M. S. Thompson,and T. J. Meyer J. Am. Chem. Soc. 1980,102,2310. Elec tro -organic Chemistry 93 The cathodic of alkyl bromides may be catalysed by nickel complexes. The products are dimers alkenes and alkanes which is a mixture that is consistent with the intermediacy of alkyl radicals. For dimers of the type (XCH2CH2)2 formed from reduction of XCH2CH2Br in the presence of NN'-ethylenebis(salicy1-ideneiminato)nickel(II) the yields are 55-77% with X = Ph C02Et or II-C~H~~. The reaction takes place at the reduction potential of the complex i.e.-1.75 V (vs S.C.E.) which is several hundreds of millivolts less cathodic than that of the bromide. A plausible reaction scheme is given in Scheme 25. [Ni"L2] h[Ni'LzI-(Ni'LJ + RBr + Ni"'L2 + R' + Br-+ CNi"L21 r 1- LBr J L = NN'-ethylenebis(salicy1ideneiminato) Scheme 25 The selective debromination of P-hydroxy-bromides by Barton's reagent [chromium(II) acetate DMSO and n-C4H9SH] is often accompanied by substantial elimination of both bromide and hydroxyl groups. This may be explained by electron transfer to the initially formed radical R'CHCH(OH)R* being faster than abstraction of hydrogen from the thiol. The P-hydroxy-carbanion would rapidly lose hydroxide ion. It has now been shown4' that cathodic regeneration of chromium(I1) allows the choice of conditions in which elimination is effectively suppressed.The hydroxyl groups are protected as tetrahydropyranyl ethers and the electrolyte consists of DMF containing chromium(I1) perchlorate a ligand (ethylenediamine) and butane- 1-thiol. The molar ratio of substrate :chromium(I1) species is about 4.5 and reaction is usually complete after 2 F mol-' based on hydroxy-bromide. As an example of the efficiency of this method 1-hydroxyindane is formed in 80% yield from 1-hydroxy-2-bromoindane; in this case the driving force for elimination to indene must be considerable. The anodic oxidation of iodine in acetonitrile is known to produce an electrophilic species that is capable of substituting into reactive aromatic systems [see Annu.Rep. Prog. Chern. Sect. B 1976 73 1411. The scope of this reaction has now been extended46 by the use of 1,2-dichloroethane containing 10% trifluoroacetic acid as the solvent. Under these conditions iodine undergoes two-electron oxidation at 2.0 V (vs Ag/Agf) to a highly reactive but unidentified species. One candidate iodinium trifluoroacetate is known to be less reactive than the species in question; another possibility is that iodine cation radical (I2?) is the electrophile with the second electron-transfer being a follow-up step of an ECE process. Whatever the exact nature of the electrophile it is capable of iodinating such compounds as benzaldehyde nitrobenzene iodobenzene benzene and benzotrifluoride in high yields (74-97%).Benzonitrile and 4-chloronitrobenzene are iodinated less efficiently (40 and 56% yields respectively) and puzzlingly acetophenone does not react. 44 C. Gosden and D. Pletcher J. Organomet. Chem. 1980,186,401. 45 J. Wellmann and E. Steckhan Angew. Chem. Int. Ed. Engl. 1980,19,46. 46 R.Lines and V. D. Parker Acta Chem. Scand. Ser. B 1980.34,47. J. H.P. Utley Several recent examples4749 point to the convenience and advantage for organic reactions of generating bromine in situ by anodic oxidation of bromide ion. These examples refer to the indirect generation of PhSeBr by anodic generation and regeneration of bromine in the presence of diphenyl diselenide. The reactions are typically performed at constant current in hydroxylic solvents and with a tetra- alkylammonium bromide as electrolyte.In this fashion the a-phenyl~elenylation~' of ketones and the oxyselenation of alkene~~**~~ have been achieved and examples are given in Scheme 26. Me0,C SePh mo [90%] (PhSe) -3 !PhSeBr] \& Br2 Br - i"i OH -* OMe \ [~OYO] [94%] Reagents i Et,NBr MgBr,.6Hz0 MeOH Pt anode constant current; ii OH ; iv BF * Et,O Scheme 26 In these days of looming shortages of petrochemicals the importance of the conversion of carbon monoxide and methanol into useful chemicals is self -evident. An interesting example of this which is essentially an indirect electrochemical reaction is a synthesis" of N-alkyl-formamides. The reactants are carbon monoxide (at pressures up to 100atm) methanol and (n-alky1)amine.A specially constructed pressure vessel-cum-undivided cell is used which includes a carbon anode and a stainless-steel cathode. The choice of electrolyte is important; using tetra-n-butyl- ammonium tetrafluoroborate (or bromide) conversions of 75-90% may be achieved. The mechanism is not clear but possibilities are given in Scheme 27. 2MeOH 52Me0-+ H2 (at the cathode) either M$O-+CO + MeOCO-5MeO-+HC02Me HCOzMe+ R2NH + R2NCH0+ MeOH or MeO-+ R2NH + MeOH + R2N-R,N-+CO + R2NCO-%R2NCHO+MeO-Scheme 27 S. Torii K. Uneyama and K. Handa Tetrahedron Lett. 1980 21 1863. 48 S. Torii K. Uneyama and M. Ono Tetrahedron Lett. 1980,21,2741. 49 S. Torii K. Uneyama and M.Ono Tetrahedron Lett. 1980 21,2653. D. Cipris J.Electrochem. SOC. 1980,127 1045.