J. Chem. Soc., Faraday Trans. I , 1986, 82, 495-508 Electrochemical Reduction of 2-Cyclohexen- 1 -ones in a Hydroethanolic Medium Enrique Brillas* and Adolfo Ortiz Departarnent de Quirnica Fisica, Facultat de Quirnica, Universitat de Barcelona, Avda. Diagonal 647, 08028-Barcelona, Spain Electrohydrodimerization of 2-cyclohexen-l-one, 4,4-dimethyl-2-cyclo- hexen- 1 -one and 4,4-diphenyl-2-cyclohexen- 1 -one in buffered hydroethan- olic solutions containing 50% (v/v) ethanol over the pH range 1.0-12.1 has been studied by polarography, cyclic voltammetry and controlled-potential coulometry. All substrates give two one-electron processes corresponding to the reduction of their respective protonated and unprotonated forms. At pH values < ca. 8.5 the protonated form of each substrate is reduced in a one-electron step to generate a neutral radical which couples to give the final hydrodimers.The unprotonated form, at pH > 5.5, leads to the formation of the same final hydrodimers by an initial generation of the anion radical via a one-electron reduction step, followed by protonation and dimerization of the neutral radical. In the pH range ca. 5.5-8.5 the two processes of each substrate compete and their relative contribution is governed by the rate of establishment of the protonation equilibrium between their unprotonated and protonated forms. Voltammetric results show that the rate-determining step of each process depends on solution pH. The electrochemical reduction of 2-cyclohexen- 1 -one and its alkyl deri~ativesl-~ leads to the formation of a mixture of hydrodimers as reaction products.They can be the corresponding diketones, hydroxyketones and pinacols, in different ratios depending on the initial substrate and on the reaction medium. Thus 2-cyclohexen-1-one only gives a mixture of (1,l '-bicyclohexyl)-3,3'-dione and 3 4 1 '-hydroxy-2'-cyclohexen- 1 '-y1)cyclo- hexanone, the first hydrodimer being the main product in all reaction media. In the electroreduction of its alkyl derivatives, however, a mixture of the three hydrodimers is obtained. For these last compounds, Tissot et aL5 have shown that in acetonitrile containing water as proton donor the yield of pinacols increases as long as the water content of the solvent rises, the relative amounts of diketones and hydroxyketones being simultaneously reduced.To explain this variation of the product distribution of hydrodimers, the authors propose two different electrohydrodimerization (e.h.d.) mechanisms depending on the reaction medium; these are initiated by generation of the radical anion via a one-electron reduction of the substrate. These authors then suggest that in acetonitrile the hydrodimers are formed via coupling of the radical anion with the substrate, whereas in the presence of water they consider that the same hydrodimers can be 6btained by dimerization of the neutral radical previously generated by protonation of the radical anion. In previous work6 we have studied the e.h.d. of 2-cyclohexen-1-one in buffered hydroethanolic solutions containing 20% (v/v) ethanol over the pH range 1 Lkl2.1.Voltammetric data indicate the presence of two reducible species depending on solution pH. They are assumed to be the protonated (at pH < 8) and the unprotonated (at pH > 5 ) forms of 2-cyclohexen-1-one. It is then proposed that both species undergo a one-electron reduction to give the same neutral radical, which subsequently couples to form the final hydrodimers. 495496 Electrochemical Reduction of 2-Cyclohexen- 1 -ones To gain a better understanding of the e.h.d. mechanisms of 2-cyclohexen-1-one and its derivatives we now report results on a comparative study of the e.h.d. of 2- cyclohexen- 1 -one, 4,4-dimethyl-2-cyclohexen- 1 -one and4,4-diphenyl-2-cyclohexen- 1 -one in buffered hydroethanolic solutions containing 50 % (v/v) ethanol at different pH values by polarography, cyclic voltammetry and controlled-potential coulometry .This reaction medium has been chosen owing to the low solubility of 4,4-diphenyl-2-cyclohexen- 1 -one in hydroethanolic solutions with a higher water content. Experimental 2-Cyclohexen-1-one (Merck, A.R. grade) was bidistilled at 61-62°C at a pressure of 10 mmHg* before use. 4,4-Dimethyl-2-cyclohexen- 1 -one and 4,4-diphenyl-2-cyclohexen- 1-one were synthesized by standard methods reported in the literat~re.~, Buffered hydroethanolic solutions containing 50 % (v/v) ethanol with an ionic strength of 0.50 were always employed. Clark and Lubs buffers were used for the pH ranges 1 .W.2 and 6.1-12.1 and sodium acetate-acetic acid buffers for the pH interval 4.2-6.1. Ionic strength was kept constant by adding NaC1.All chemicals employed in the preparation of buffered solutions were Merck, A.R. grade. The water used was obtained from a Millipore Milli-Q system. The polarographic measurements were performed with an Amel 47 1 multipolarograph. The cyclic-voltammetric measurements were carried out with conventional instrument- ation consisting of an Amel 551 potentiostat, a P.A.R. 175 universal programmer, a Nicolet 3091 digital storage oscilloscope and a Philips 8043 X-Y recorder. The ohmic-drop compensation of all voltammetric experiments was achieved with a positive-feedback network of the same instruments. The coulometric measurements were performed with an Amel 555A potentiostat equipped with an Amel 721 current integrator. All voltammetric experiments were carried out in a three-electrode cell.Solutions were deaerated by flushing with nitrogen presaturated with the solution to be investigated, and during the measurements this gas was passed over the solution. The temperature was kept at 25.0 "C. A saturated calomel electrode (SCE) was used as reference electrode and a Pt wire as counter-electrode. The flow rate, m, of the dropping mercury electrode employed in polarography was 0.841 mg s-l in 0.1 mol dm-3 KC1 solution on open circuit at a mercury height of 60 cm. In cyclic voltammetry the working electrode was a platinum sphere, sealed in glass, electrolytically covered by a thin layer of silver and then amalgamated by dipping in mercury. The area, A , of this electrode was 0.087 cm2 and its surface was renewed before each experiment by dipping it again in mercury.This procedure allowed us to carry out the cyclic-voltammetric measurements because no adsorption processes were observed. A substrate concentration of 1.0 mmol dmd3, a mercury height of 60 cm and drop times, t, of 1.0, 1.5, 2.0, 2.5, 3.0 and 3.5 s were used in the polarographic experiments. Drop times were regulated with a Tacussel GCMS hammer. Scan rates, v , between 0.1 and 50 V s-l and substrate concentrations of 0.2, 0.5, 1.0 and 2.0 mmol dm-3 were employed in cyclic voltammetry. The pH of the solutions was measured with a Radiometer 51 pH meter before and after each experiment, and in no case was a significant alteration noted. Controlled-potential coulometries were carried out in an H-cell under nitrogen and the temperature was kept at 25.0 "C.A mercury pool, magnetically stirred, of 15.2 cm2 was used as cathode, an SCE as reference electrode and a Pt wire as anode. Volumes of 150 and 50 cm3 of the buffered solution were introduced in the cathodic and anodic compartments, respectively. A solution of mol of substrate in 1 cm3 of 99% ethanol was added to the catholyte and exhaustive electrolysis was then carried out at * 1 mmHG = 13.5951 x 980.665 x lo-* Pa.E. Brillas and A . Ortiz 497 g 'PA I I I I -1.000 - 1.600 EIV vs. SCE Fig. 1. Polarograms corresponding to the reduction of 1 .O mmol dm-3 2-cyclohexen-1-one (-), 4,4-dimethyl-2-cyclohexen- 1 -one (. . .) and 4,4-diphenyl-2-cyclohexen- 1 -one (---) in hydroethanolic medium containing 50% (v/v) ethanol at pH values of: (1) 1.35, (2) 3.73, (3) 6.71, (4) 9.28 and (5) 12.04.Drop time 2.0 s and temperature 25.0 "C. a potential corresponding to the plateau of the polarographic wave. Electrolysis times were usually ca. 1 h. Coulometric n values were obtained by considering the background corrections, which were determined by electrolysing the buffered solutions under the same experimental conditions. Results and Discussion Polarographic Behaviour Polarograms corresponding to the reduction of 2-cyclohexen- 1 -one, 4,4-dimethyl-2- cyclohexen- 1 -one and 4,4-diphenyl-2-cyclohexen- 1 -one in buffered hydroethanolic solu- tions at several pH values and at a drop time of 2.0 s are shown in fig. 1. Either one or two waves (I and 11) were obtained for all substrates depending on the solution pH.In acidic solutions of 4,4-dimethyl-2-cyclohexen-l -one at pH < 5.0,2-cyclohexen- 1 -one at pH < 5.3 and 4,4-diphenyl-2-cyclohexen-l-one at pH < 5.7, only the corresponding first wave (I) was found. At higher pH values, waves I decreased in height with increasing pH, in the form of dissociation curves with pK' values of 6.11, 6.58 and 7.08 for 4,4-dimethyl-2-cyclohexen- 1 -one, 2-cyclohexen- 1 -one and 4,4-diphenyl-2-cyclohexen- 1 - one, respectively, disappearing at pH ca. 8.5. Simultaneously the second wave (11) of each substrate was observed at more negative potentials than wave I. Waves I1 increased in498 Electrochemical Reduction of 2-Cyclohexen- 1 -ones 1 . 5 0 1 . o o 9 -. +- 0.50 0.00 I Q I! 0.00 4 .OO 8.00 12.00 PH Fig.2. Variation of the limiting current with pH for the two polarographic waves of 1.0 mmol dm-3 2-cyclohexen-1 -one (O), 4,4-dimethyl-2-cyclohexen- 1 -one (A) and 4,4-diphenyl-2-cyclo- hexen-1-one (0) in hydroethanolic medium containing 50% (v/v) ethanol. Drop time 2.0 s and temperature 25.0 "C. height to pH ca. 8.5 in the shape of dissociation curves and then remained constant to pH ca. 10, whereupon a gradual reduction in height occurred. This behaviour of the studied 2-cyclohexen-1 -ones can be observed in fig. 2, where the variation of the limiting current, Zl, with pH for both polarographic waves at a drop time of 2.0 s is presented. Note that when waves I and I1 of a given substrate appear simultaneously, the sum of their limiting currents remains practically constant at each value of t.A similar splitting into two waves over the same pH range has also been described for the electroreduction of 2-cyclohexen-1-one in buffered aqueous solutions,l as well as in buffered hydro- ethanolic solutions containing 20% (v/v) ethanol.'j When only waves I and I1 appeared they were always diffusion-controlled because linear correlations between log Il and log t , with slopes close to 0.19, were obtained in all solutions. Values of the diffusion-current constant, Id, in such conditions are given in table 1. A progressive decrease in Id on increasing the size of the substituents in the C(4) position can be observed at each pH value, according to the gradual decrease of the diffusion coefficient, D, expected for these substrates.Moreover, the Id values reported in table 1 for the two waves of 2-cyclohexen-1-one are slightly lower than those obtained for the analogous waves in a hydroethanolic medium containing 20% (v/v) ethanol,6 in accordance with the increase in viscosity of the reaction medium with increasing ethanol content. When waves I and I1 are simultaneously observed, the kinetic character of wave I increases with rising pH. For a given substrate and when pH > pK'+0.70, the height of wave I is independent oft. Under these conditions this wave is a kinetic current, being limited by the rate of a chemical reaction preceding the electrode process p r ~ p e r . ~ The half-wave potential, E,,,, of each wave generally varies with both t and the pH. Thus linear plots of against log t were found at pH values where only one wave is499 -1.400 w -1.200 d 4 > 1 N ..- 1 .ooo E. Brillas and A . Ortiz 1 I 1 -/ 1 I I I 1.00 5.00 9-00 PH Fig. 3. Plot of half-wave potential against pH for the first polarographic wave of 1.0 mmol dm-3 2-cyclohexen- 1 -one (O), 4,4-dimethyl-2-cyclohexen- 1 -one (a) and 4,4-diphenyl-2-cyclohexen- 1-one (0) in hydroethanolic medium containing 50% (v/v) ethanol. Drop time 2.0 s and temperature 25.0 "C. observed. The slopes obtained for waves I and I1 are always close to 20 and 30 mV per decade, respectively, as can be seen in table 1. Plots of against pH for waves I and I1 of each substrate are shown in fig. 3 and 4, respectively. In fig. 3 the half-wave potential of wave I shifts with increasing pH to more negative potentials by some 60 mV per pH unit, at least up to a pH value near to the pK' of the corresponding substrate.On the other hand, two different types of behaviour of wave I1 can be observed in fig. 4, depending on the solution pH. The half-wave potential of wave I1 of 2-cyclohexen-1-one and4,4-dimethyl-2-cyclohexen- 1 -one at pH < 8.5, as well as of4,4-diphenyl-2-cyclohexen- 1-one at pH < 7.8, is pH-independent, whereas at higher pH values it is shifted with increasing pH to more negative potentials by some 30 mV per pH unit. Note also that waves I and I1 of 4,4-diphenyl-2-cyclohexen-l-one appear at more positive potentials than those of the other two substrates (see fig. 1, 3 and 4). In fact the analogous waves of 2-cyclohexen- 1 -one and 4,4-dimethyl-2-cyclohexen- 1 -one are observed in similar potential regions.Moreover, the value of each wave of 2-cyclohexen-l-one, at given pH and t values, is practically the same as in a hydroethano- lic medium containing 20% (v/v) ethanol,6 indicating that the ethanol content of the medium has little effect upon the half-wave potential. Cyclic-voltammetric Behaviour Cyclic voltammograms corresponding to the reduction of 1 .O mmol dmW3 2-cyclohexen- 1 -one, 4,4-dimethyl-2-cyclohexen- 1 -one and 4,4-diphenyl-2-cyclohexen- 1 -one in bufferedcn 0 Table 1. Voltammetric and coulometric results for the two electroreduction processes of 2-cyclohexen- 1-one, 4,4-dimethyl-2-cyclohexen- 1 -one and 4,4-diphenyl-2-cyclohexen- 1 -one in buffered hydroethanolic solutions at pH values where only one process occurs, at 25.0 “C compound PH 6 (E”p,2 - EC,)/mV /mV per decade 4,4-dimethyl- 2-cyclohexen- 1 -one 4,4-diphenyl- 2-cyclohexen- 1 -one 2-cyclohexen- 1.35 - 1 -one 1.66 2.21 3.07 3.73 4.13 4.82 5.29 1.35 1.66 2.2 1 3.07 3.73 4.13 4.82 1.35 1.66 2.2 1 3.07 3.73 4.13 4.82 5.29 5.69 1.52 1.48 1.45 1.43 1.40 1.39 1.35 1.31 1.37 1.31 1.24 1.22 1.21 1.20 1.19 1.34 1.24 1.17 1.10 1.06 1.05 1.04 1.02 1 .oo e e - - e - 0.52 0.53 0.53 0.53 0.52 e e e - - - 0.53 0.53 0.52 0.53 e - e - e - 0.53 0.52 0.53 0.53 0.52 0.53 (0.527)f process I - e 20.1 - e 19.8 e - 20.0 40 19.9 41 20.2 40 20.0 39 19.9 41 20.1 19.8 39 20.2 41 20.1 40 20.3 40 19.9 - e 20.0 - e 20.1 e - - e 20.0 - e 20.2 - e 20.1 40 19.8 41 20.0 40 19.7 39 20.1 40 19.9 41 20.3 (38.8)f (19.7)f e e - - e - 20.1 19.9 20.0 20.2 19.9 e e e - - - 20.1 19.8 20.0 20.0 e - e - e - 20.0 19.9 20.1 19.9 20.0 19.8 (19.7)f e e e - - - 20.0 20.1 19.9 20.2 20.1 e e e - - - 20.0 20.2 19.9 20.0 e e e - - - 19.8 20.0 19.9 20.2 20.1 20.0 (19.7)f 0.98 0.97 1.01 0.96 0.98 0.97 1 .oo 0.99 0.96 0.99 1.01 % Y 0.97 0 0.97 % 2 8 ? s- 0.99 0.98 0.99 1.01 c-r & 2: ;s 0.98 0.97 0.99 0.98 1 .oo 0.97 0.96 -2-c yclohexen- 1 -one 4,4-dimethyl- 2-cyclohexen- 1 -one 4,4-diphenyl- 2-c yclohexen- 1 -one 9.28 9.8 1 10.18 10.78 11.14 11.49 12.04 8.55 9.28 9.8 1 10.18 10.78 11.14 1.44 1.41 1.32 1.18 1.13 1.07 1.06 1.20 1.18 1.14 1.10 1.07 1.06 0.47 0.49 0.48 0.49 0.48 0.49 0.49 0.46 0.47 0.49 0.48 0.48 0.49 process I1 52 29.7 49 28.9 51 29.7 53 30.1 51 30.2 50 29.5 52 28.8 52 29.6 53 30.0 52 30.1 49 29.9 51 30.3 53 29.9 0 0 0 0 0 0 0 0 0 0 0 0 0 30.0 29.2 28.5 29.0 30.2 30.1 28.9 30.2 28.8 28.5 28.9 28.6 30.3 0.98 0.99 1.01 0.97 1 .oo 0.96 0.97 1.01 0.97 0.98 1 .oo 0.98 0.99 1.49 1.05 0.49 51 29.7 0 28.7 0.99 2.04 1.05 0.48 50 29.8 0 30.0 1.01 9.28 1.09 0.47 49 30.2 0 29.3 0.96 9.8 1 1.09 0.49 51 29.9 0 30.2 0.98 0.18 1.06 0.48 50 30.1 0 30.4 0.97 0.78 1.01 0.49 52 30.0 0 30.0 1 .oo 1.14 1 .oo 0.48 53 29.7 0 28.8 0.97 11.49 0.99 0.49 51 29.9 0 29.6 0.99 12.04 0.99 0.48 52 29.8 0 28.9 0.99 - (29.6)g - - (0.496)g (47.99 (29.6)g (0l9 a Id = 11/cm2/3 t1/6.Coulometrically obtained. Peak overlapped with the reduction peak of the hydrogen discharge. f Theoretical values corresponding to the mechanism of scheme 1 .lo* l1 4: = G/I;Ac (D u F/RT)lI2.Values obtained for process I up to 5 V s-1 and for process I1 up to 50 V s-l. Theoretical values corresponding to the mechanism of scheme 2.12502 I Electrochemical Reduction of 2-Cyclohexen- 1 -ones I 1.650 w % 1.550 > 1 CI .. ii 1.45 0 / B 6 .OO 8 .OO 10.00 12 .C" PH Fig. 4. Plot of half-wave potential against pH for the second polarographic wave of 1 .O mmol dmP3 2-cyclohexen- 1 -one (O), 4,4-dimethyl-2-cyclohexen-l -one (a) and 4,4-diphenyl-2-cyclohexen- 1-one (0) in hydroethanolic medium containing 50% (v/v) ethanol. Drop time 2.0 s and temperature 25.0 "C. hydroethanolic solutions at several pH values and at a scan rate of 1 V s-l are shown in fig. 5. On the cathodic portion of the voltage scan all 2-cyclohexen-1-ones exhibited either one or two peaks (R, R,,) depending on solution pH; these correspond to their first and second polarographic waves, respectively. Under all the experimental conditions investigated, R,, of each substrate is due to an irreversible overall process.However, the corresponding overall process of R, is only irreversible up to 5Vs-l, whereas at higher scan rates one anodic peak, O,, is also recorded. Fig. 6 shows the OI/RI couple found for a 1 .O mmol dm-3 solution of each substrate at a pH 4.82 and at a scan rate of 20 V s-l. At pH values where only one cathodic peak is observed, and under irreversible conditions, the corresponding peak current, I& is proportional to the initial reactant concentration, c, and to the square root of the scan rate, indicating that the two reduction processes of each substrate are diffusion-controlled. Values of the peak-current function, q5;, of the difference between the cathodic half-peak and peak potentials, 5,: - EP, and of the slopes of the linear plots of Eg against log c and EP against -log u obtained under such conditions are given in table 1.The diffusion coefficient used to calculate @, was determined in each hydroorganic medium by substitution of the corresponding I, value in the IlkoviC equation, assuming the number of transferred electrons per reacting molecule (n) to be 1 . This value of n is expected for e.h.d. reactions, and its assumption has been confirmed for each substrate in all media by the coulometric results to be presented below. Note that in acidic solutions of pH < ca.2.5 the peak R, overlaps withE. Brillas and A . Ortiz 503 I I I .72 5 V - 1 . 3 Fig. 5. Cyclic voltammograms corresponding to the reduction of 1 .O mmol dm-3 2-cyclohexen-1 -one (-), 4,4-dimethyl-2-cyclohexen- 1 -one (---) and 4,4-diphenyl-2-cyclohexen- 1 -one ( . . . ) in hydroethanolic medium containing 50% (v/v) ethanol at pH values of: (1) 4.13, (2) 6.71 and (3) 10.18. Scan rate 1 V s-l and temperature 25.0 "C. the reduction peak of the hydrogen discharge, and for this reason these cyclic voltammetric results are not listed in table 1. Cyclic-voltammetric parameters reported in table 1 for each reduction process are pH-independent. Moreover, cathodic peak potentials of R, and R,, for each substrate, at pH values where only one peak is observed under irreversible conditions, shift with increasing pH, at constant c and u, to more negative potentials by some 60 and 30 mV per pH unit, respectively. The cathodic peak potential of R,, for 2-cyclohexen-1 -one and 4,4-dimethyl-2-cyclohexen-l-one at pH < 8.5, as well as that for 4,4-diphenyl-2- cyclohexen-1-one at pH < 7.8, is pH-independent for a given c and v .In acidic solutions of pH < 5.5 and when v > 5 V s-l, different kinetic behaviour was found for RI of each substrate owing to the presence of the corresponding peak 0,. At a given pH, on increasing u from 5 to 50 V s-l the peak-current function gradually decreases, whereas the slope of the plot of 4 against - log v shows a progressive increase. The difference between anodic and cathodic peak potentials of the OI/R, couple is always ca.75-85 mV. Fig. 6 shows that the anodic peak current of this couple has a small value in all cases, although it increases with rising u. When the two cathodic peaks appear simultaneously, the corresponding peak R, is generally flat (see fig. 5, curve 2) and by increasing v its current becomes independent of the scan rate. Under such conditions the current is fully controlled by the rate of a chemical reaction antecedent to the electrode process p r ~ p e r . ~ On raising pH this kinetic504 Electrochemical Reduction of 2-Cyclohexen- 1 -ones I 100pA I - 1.000 -1.400 EIV us. SCE Fig. 6. Cyclic voltammograms corresponding to the reduction of 1 .O mmol dm-3 2-cyclohexen- 1 -one (-), 4,4-dimethyl-2-cyclohexen- 1 -one (---) and 4,4-diphenyl-2-cyclohexen- 1 -one ( .. . ) in hydroethanolic medium containing 50% (v/v) ethanol at a pH value of 4.82. Scan rate 20 V s-l and temperature 25.0 "C. control is observed at lower u values, so that for each substrate at pH > pK'+0.70, its peak RI is a kinetic current over the whole u range studied. Controlled-potential Coulometry Coulometric n values obtained for the controlled-potential reduction of 2-cyclohexen- 1 -one, 4,4-dimethyl-2-cyclohexen- 1 -one and 4,4-diphenyl-2-cyclohexen- 1 -one in buffered hydroethanolic solutions at pH values where only one process occurs are listed in the last column of table 1. An n value close to 1 can always be observed for each reduction process. Electrolyses at potentials on the plateau either of the first or second polarographic waves were also carried out in solutions of each substrate at pH values where both waves appear simultaneously.Polarograms recorded during these electrolyses show that the ratio of the wave heights remains constant, and after exhaustive electrolyses an n value near to 1 is always obtained. All these results indicate a one-electron reduction in each observed process. E.H.D. Mechanisms In buffered hydroethanolic solutionscontaining 50 % (v/v) ethanol all 2-cyclohexen- 1 -ones studied undergo two different one-electron reduction processes (I and 11) depending on the solution pH, leading to the formation of a mixture of hydr~dimers.~-~ To elucidate the e. h.d. mechanism of each process, its voltammetric results under irreversibleE.Brillas and A . Ortiz 505 conditions have been discussed on the basis of theoretically established diagnostic criteria for e.h.d. reactions under the same conditions.1°-12 When only process I of each substrate is observed, its voltammetric results under irreversible conditions are in accordance with the values calculated for a radical-radical coupling mechanism corresponding to scheme 1 ,lo, l1 as can be seen in table 1. In addition, the experimental slopes of the linear correlations of El,, against pH and EP against pH obtained for process I of all the 2-cyclohexen-1-ones under such conditions agree with the theoretical values of - 59.2 mV per pH unit expected for this mechanism. Scheme 1 shows that the final hydrodimer DH, is obtained by dimerization of the neutral radical AH', which is formed via a one-electron reduction of the protonated form AH+ previously generated by protonation of the initial compound A.In this mechanism the irreversible coupling reaction is the rate-determining step. A + H + ~ A H + AH' + e-,= AH ' 2AH *--+ DH2 Scheme 1. The kinetic control found for process I of each substrate, at pH values where both processes simultaneously appear, can be explained9$ l1 from the mechanism described above by taking into account that under such conditions the protonation reaction A + H+ -+ AH+ is the rate-determining step. This voltametric behaviour of process I allows one to exclude the formation of the neutral radical AH' by protonation of the radical anion A'-, initially generated via a one-electron reduction of A.In fact, the voltammetric diagnostic criteria established for this mechanism under irreversible conditions are the same as those reported above for the mechanism of scheme 1 ,lo, l1 but they cannot explain the presence of kinetic currents under other conditions. l1 Therefore the following e.h.d. mechanism studied (R = H, CH,, C,H,): lo1 is proposed for process I for all the 2-cyclohexen- 1 -ones H I + 01 R R R R H H I $ R R + e- I 6 R R H R R506 Electrochemical Reduction of 2-Cyclohexen- 1 -ones where the protonated form of each substrate is reduced in a reversible one-electron step to give its neutral radical, which subsequently couples, either via C(3),C(3') - , C(3),C( 1') - or C( 1),C( 1') -bonding, to form thecorresponding diketone, hydroxyketone or pinacol.2v The above-mentioned e.h.d. mechanism can also explain the behaviour observed in cyclic voltammetry when the O,/& couple of each substrate appears in acidic solutions at pH < 5.5 and at u > 5 V s-l. It can be assumedll? l3 that under such conditions the dimerization of the neutral radical competes with its reoxidation, peak OI being attributable to the last reaction. By increasing u, reoxidation of a larger proportion of the neutral radical can be expected, and therefore a higher value of the anodic peak current and a gradual decrease of the peak-current function of R, should be 0bser~ed.l~ Both variations are in accordance with the experimentally obtained results for all 2-cyclohexen- 1 -ones. The difference between anodic and cathodic peak potentials, as well as the dependence of FP on -log u, indicate that step (2) behaves as a quasi-reversible process.13 As can be seen in table 1, the voltammetric results of process I1 for all the 2-cyclohexen-1-ones studied, at pH values where only this process occurs, agree with the values calculated for an e.c.mechanism (scheme 2), in which the irreversible protonation of the electrogenerated radical anion A'- is the rate-determining step.12 Moreover, the experimental slopes of the linear plots of against pH and Eg against pH obtained for process I1 for each substrate under these conditions are in accordance with the theoretical values of -29.6 mV per pH unit expected for this mechanism. Scheme 2. At pH values where the half-wave potential and cathodic peak potential of process I1 for each substrate are pH-independent, the initial one-electron reduction A + e- -+ A'- must be the rate-determining step.Note that voltammetric results obtained for process I1 for all the 2-cyclohexen- 1 -ones exclude the possible dimerization of two radical anions A - followed by protonation of the dianion formed, because for this mechanism values of 19.7 mV per decade are calculated for the slopes of the linear correlations of El12 and 4 as functions of the different parameters considered.12 The final hydrodimers could then be formed from the neutral radical AH' through three different pathwaysllt l2 involving (i) dimerization of AH', (ii) coupling of AH' with A'- or (iii) coupling of AH' with A, which cannot be distinguished from the experimental results of process 11.Nevertheless, taking into account that, for each substrate, the same neutral radical is also formed in step (2) and subsequently couples in step (3), it seems plausible that in this case the final hydrodimers (the corresponding diketone, hydroxyketone and pinacol) are also obtained by dimerization of the neutral radical. Therefore, the following e.h.d. mechanism is proposed for process I1 (R = H,CH,,C,H,): 6 + e- R R lo I- c L 7 R R (4)E. Brillas and A . Ortiz 507 H H - hydrodimers (6) where the unprotonated form of each substrate is the electroactive species, which is reduced to its radical anion via the reversible one-electron step (4). The fact that the reversibility of this step cannot be observed by cyclic voltammetry, even at high scan rates, can be explained by assuming that the following chemical reactions are always sufficiently fast for the concentration of the radical anion near the electrode surface to remain insignificant. At pH > ca.10 the gradual decrease in the limiting current and cathodic peak current observed for process I1 with increasing pH can be attributed to the progressive increase in the conversion rates of the unprotonated forms into electroinactive species by hydration of the carbonyl group or the double bond.l49l5 Since process I1 is always controlled by diffusion, it can be assumed that both species are in equilibrium; then at a given pH the limiting current and the cathodic peak current are proportional to the equilibrium concentration of the unprotonated form.In the pH range where the two processes for each substrate are observed simultaneously, process I is gradually replaced by process I1 with rising pH. Under these conditions, reduction pathways (1 )-(3) and (4)-(6) compete, their relative contributions being controlled by the rate of establishment of equilibrium (1). The equilibrium constants, K for the dissociation reactions of the protonated forms of the 2-cyclohexen- 1 -ones in a hydroethanolic medium containing 50 % (v/v) ethanol are not available. In fact, only pKvalues of -6.8 and -3.60 have been reportedl9 l6 for the dissociation reaction of protonated 2-cyclohexen- 1 -one in an aqueous medium. This suggests that the apparent pK' = 6.58 determined polarographically for this substrate in a hydroethanolic medium is at least ten units higher than its thermodynamic pK value, which must be similar to the corresponding value in an aqueous medium.Analogous differences between the pK' and pK values for the protonated forms of 4,4-dimethyl- 2-cyclohexen- 1 -one and 4,4-diphenyl-2-cyclohexen- 1 -one in a hydroethanolic medium can also be expected. These last considerations indicate that reaction (1) takes place for all the 2-cyclohexen-1-ones studied in the vicinity of the electrode, but not in the bulk solution.'+ However, a quantitative treatmentg to compute the rate constant corresponding to the protonation reaction (1) of these substrates is at present not possible, because the values of their equilibrium constants are not accessible.Finally, note that voltammetric results allow the same e.h.d. mechanism to be estab- lished for analogous processes for each of the 2-cyclohexen- 1 -ones considered, indepen- dent of the product distribution of the resulting hydrodimers. The fact that the two processes for 4,4-diphenyl-2-cyclohexen- 1 -one appear at more positive potentials than508 Electrochemical Reduction of 2-Cyclohexen- 1 -ones those for 2-cyclohexen- 1 -one can then be attributed to the electron-withdrawing prop- erties of the two phenyl groups on the a, P-unsaturated carbonylic systems of the protonated and unprotonated forms. However, methyl substituents in C(4) positions have little effect upon the reduction potentials of these systems. References 1 E. J. Denney and B. Mooney, J. Chem. SOC. B, 1968, 1410. 2 E. Touboul, F. Weisbuch and J. Wiemann, C.R. Acad. Sci., Ser. C, 1969,268, 1170. 3 P. Tissot and P. Margaretha, Helv. Chim. Acta, 1977, 60, 1472; Electrochim. Acta, 1978, 23, 1049. 4 P. Margaretha and P. Tissot, Nouv. J. Chim., 1979, 3, 13. 5 P. Tissot, J. P. Surbeck, F. 0. Giilaqar and P. Margaretha, Helv. Chim. Acta, 1981, 64, 1570. 6 E. Brillas and A. Ortiz, Electrochim. Acta, 1985, 30, 1185. 7 C. Paris, S. Geribaldi, G. Torri and M. Azzaro, Bull. SOC. Chim. Fr., 1973, 997. 8 H. E. Zimmerman, K. G. Hancock and G. C. Licke, J. Am. Chem. SOC., 1968,90,4892. 9 Z. Galus, Fundamentals of Electrochemical Analysis (Horwood, Chichester, 1976), chap. 8. 10 C. P. Andrieux, L. Nadjo and J. M. Savkant, J. Electroanal. Chem., 1970, 26, 147. 11 L. Nadjo, and J. M. SavCant, J. Electroanal. Chem., 1971, 33, 419. 12 L. Nadjo and J. M. SavCant, J. Electroanal. Chem., 1973,44, 327. 13 L. Nadjo and J. M. Savkant, J. Electroanal. Chem., 1973,48, 113. 14 Organic Electrochemistry: An Introduction and a Guide, ed. M. M. Baizer (Marcel Dekker, New York, 15 L. Spritzer and P. Zuman, J. Electroanal. Chem., 1981, 126, 21. 16 R. I. Zalewski and G. E. Dunn, Can. J. Chem., 1968, 46, 2469; 1969,47, 2263. 1973), chap. VIII and IX. Paper 51668; Received 22nd April, 1985