Mendeleev Communications Electronic Version, Issue 2, 2000 (pp. 43–82) Effect of proton donors on the mechanism of electroreduction of -radicals of linear and cyclic ethers Alexander G. Krivenko, Alexander S. Kotkin and Vladimir A. Kurmaz* Institute for Problems of Chemical Physics, Russian Academy of Sciences, 142432 Chernogolovka, Moscow Region, Russian Federation. Fax: +7 096 515 3588; e-mail: kurmaz@icp.ac.ru DOI: 10.1070/MC2000v010n02ABEH001254 Using laser photoemission, we found two routes of the electroreduction of a-radicals (Rads) adsorbed on a mercury electrode for linear and cyclic ethers in the presence of strong proton donors (BH+), i.e., by direct one-electron transfer and with the antecedent formation of the metastable complex [Rads·BH+].According to laser photoemission data, the irreversible electroreduction1 –4 and electrooxidation4,5 of radicals adsorbed on an electrode surface (Rads) proceeds via two parallel pathways, namely, by the direct one-electron transfer and by the electron transfer on the antecedently formed metastable complex6 of Rads with a proton donor/acceptor ([Rads·BH+]/ [Rads·BOH–]) The second pathway is predominant for radicals containing active functional groups, e.g., carbonyl,1,2 carboxyl7(a) and hydroxyl5 groups, or bifunctional radicals4,6 (k0 = 107–1010 dm3 mol–1 s–1; the second group of radicals6).Such complexes were not found for hydrocarbon radicals or their halogen derivatives1 (k0 < < 103 dm3 mol–1 s–1; the first group). To understand the nature of such differences, it seems important to investigate radicals which possess less active functional groups.For example, the reactivity of C–O bonds in ethers is much lower than that of C–O and O–H bonds in aliphatic alcohols. 8 Since the electronic structures and spectrophotochemical characteristics of these two classes of a-radicals are similar,9 it is likely that their electrochemical behaviour is also similar.Electrode reactions of a-radicals1,5,10,11 and b-radicals6,7(a),(b) of alcohols were studied in detail by photoemission and other techniques based on non-electrochemical generation of intermediates. There is few data on the electrochemistry of ether a-radicals. Although we found earlier using laser photoemission3 that the 1,4-dioxane radical belongs to the second group (k0 = = 107 dm3 mol–1 s–1), this sole example is probably not typical because 1,4-dioxane is a diether and therefore the radical may be more active than in other ethers, as was observed, e.g., in diols4 with relation to corresponding aliphatic alcohols.1,5 Therefore, we decided on more typical radicals of linear and cyclic ethers, namely, radicals of diethyl ether 1, 1,4-dioxane 2, tetrahydrofuran (THF) 3, 2,5-dimethyltetrahydrofuran 4 and tetrahydropyran (THP) 5 (Table 1).Only radicals 1–3 were studied previously by the polarography of pulse-radiolysis products11 and by photomodulated voltammetry.12 Radicals were generated according to the following reactions: where e– aq is hydrated electron; RH is an aliphatic ether; ka and kOH (kH) are the rate constants of reactions (3) and (4), respectively. The generation of radicals 1–5 only by reaction (4) is provided by considerable differences in the rate constants ka for the e– aq capture by N2O, H3O+ and ether molecules (6×109, 2.3×1010 and < 107 dm3 mol–1 s–1, respectively13) (Table 1).The N2O molecules are main scavengers of e– aq at pH � 3, and the H3O+ cations are main scavengers of e– aq in more acidic solutions.The measurements at pH no higher than ~2 are difficult to perform because of the appearance of the anodic–cathodic wave of H· near –0.6 V (SCE)1 and the dark discharge of H3O+ on mercury, leading to a deterioration in the signal-to-noise ratio. The radicals formed in reactions (3) and (4) diffuse to an electrode, become adsorbed on it and participate in electrode a R+ e– – Rads + e– R– Ve' (1) Rads + BH+ (or BOH–) [Rads·BH+] (or [Rads·BOH–]) + e– products k1 = k0[BH+] (or [BOH–]) k2 Ve (2) 2.0 1.5 1.0 0.5 0.0 1.0 1.5 1.4 1.6 1.8 1.0 1.5 j/j0 (arbitrary units) –E/V (SCE) Figure 1 Typical TRWs of radicals at UV illumination of the electrode at various modulation periods: (a) radical 5, tm = 30 ms; (b) radical 4, tm = = 300 ms; (c) radical 3, tm = 30 ms.Stationary mercury electrode. Supporting electrolytes: aqueous buffer solutions with the addition of 0.5 M KCl saturated with N2O; pH � 5. (a) (b) (c) N2 O (H+) + e– aq N2 + OH–+ OH· (H·) ka OH· (H·) + RH R·+ H2O (H2), kOH (kH ) (3) (4) 1 2 3 4 5 106 104 102 100 0.4 0.8 1.2 1.6 2.0 W2 , W3 /s–1 –E/V (SCE) Figure 2 Tafel plots for the rate constants of reduction and oxidation (1)– (5) of radicals 1–5, respectively. The data for the reduction of radical 2 at pH�7.0 were taken from ref. 3. The experimental conditions are the same as in Figure 1.Mendeleev Communications Electronic Version, Issue 2, 2000 (pp. 43–82) reactions to generate photocurrent j. The values of j were obtained by the measurements and numerical Fourier transformation of signals from a photoelectrochemical cell illuminated with modulated light at the period tm = 1.0–0.001 s.This results in the voltammetric time-resolved wave (TRW) of a radical. The position of the half-wave E1/2 on the E axis depends on tm and the type of R and is controlled by the competition between the characteristic time of irreversible reduction/oxidation of Rads with the rate constants W3/W2 and the electrode illumination time.At E = E1/2, W = ktm, where k = 5.31 or 10.88 for W = W3 or W2, respectively. Thus, the measurements of W3 values can be performed in the range 5–5.8×103 s–1. In the range 3×103– 6×106 s–1, the values of W3 were determined from measurements of the kinetics of emitted charge Q = f(t).† Typical TRWs for the oxidation and reduction of the radicals are presented in Figure 1.The experimental details were described elsewhere.1–7 It is well known that H· and OH· radicals are reactive and hence non-selective agents. Therefore, in addition to a-radicals, the formation of b-radicals or, moreover, g-radicals in the case of THP, by reaction (4) cannot be excluded a priori (such possibilities were repeatedly discussed in the literature).However, the TRWs of radicals 1–5 demonstrated the electroreduction or electrooxidation of only one type of intermediates. Consequently, under the experimental conditions, the fraction of b- and g-radicals does not exceed 5%, similarly to the EtOH–N2O system.1,5 Figure 2 shows the Tafel functions lg W2,3(E) for radicals 1–5 at pH varied from weakly acidic to strongly basic values.It can be seen that all of the functions are similar and have similar slopes (Table 1); W is independent of pH in the specified range. This fact is indicative of the similarity of mechanisms of electrode processes. Direct electron transfer via reaction (1) is predominant, i.e., the reduction of Rads to carbanions (analogously to the radicals of alcohols, hydrocarbons and their halogen derivatives1,5,10) and the oxidation to carbocations (as in the cases of methyl, ethyl1 and carboxyalkyl radicals7(a)).The lg W3(E) functions plotted for radicals 1–5 are shifted to more positive E values on the addition of NH4 + ions and at pH £ pH*, where pH* is a threshold value which is characteristic of each of the radicals, e.g., pH* ~ 3.3 for radical 3 (Figure 3).The (dE/dpH)W3 and {dE/d(–lg[NH4 +])}W3 values obtained by cross-section of the Tafel plots at W3 = const (Table 1) are close to the corresponding values of (2.3RT/F)/a. This fact characterises the electroreduction as a first-order reaction with respect to [BH+]. Thus, with increasing [BH+], a transition to the quasi-reversible reduction of the complex [Rads·BH+] occurred in all of the radicals.Simultaneously the obsrate constant was a linear or exponential function of [BH+] or E, respectively. In contrast to radical 2, the slopes of the Tafel plots for radicals 1 and 3–5 remained unchanged even at minimum [BH+]. Thus, the transition to the discharge, which is limited by the rate of complex formation, occurs at observed constants higher than Wm, i.e., at k1 � Wm ª 107 s–1 (Figure 3).Consequently, the value of k0 lies between 107–1010 and 109– 1010 dm3 mol–1 s–1 for the complexes with NH4 + and H3O+ ions, respectively. In terms of the model6 it is possible to determine the difference of overvoltages Dh for radicals 1–5 and their complexes‡ with BH+. For BH+ = H3O+, this difference decreased † With the use of laser photoemission, radicals were generated at distances of ~(10–100) Å from the electrode; this allowed us to measure W up to the maximum values Wm ª 107 s–1.as the Tafel plot was shifted towards the cathodic region from ~1.0 (radical 2) to ~0.45 V (radical 1); for BH+ = NH4 +, Dh is ~0.2–0.3 V, and the above correlation does not occur. Previously, 14–17 a much lower efficiency of NH4 + ions as proton donors, as compared with H3O+, was found in the bulk protonation of NO3 2– and CO2 ·– and in the electroreduction of stable anions.An analysis of the experimental results (Table 1) and a comparison with published data demonstrate that the position of Tafel plots for radicals 1–5 on the E axis correlates with the redox potentials of pairs§ E0 (R/R+), e.g., E0 2 >> E0 1, E0 3 (ref. 12), as well as with the rates of anodic methoxylation of ethers in the presence of solid polymer electrolytes19 and, generally, with the structure and reactivity of the corresponding molecules. For example, the insertion of an additional CH2 group into a cyclic ether molecule facilitates the electrooxidation, and the replacement of this group by an electronegative O atom leads to considerable inhibition of the process.18 The Tafel plot of the electroreduction is analogously shifted by ca. 0.40 V on the replacement of an O atom in radical 2 with a CH2 group (radical 5). As in the case of the parent cyclic ether molecules,18 the reduction rate of the radicals substantially increased (Figure 2) after the insertion of additional electronegative substituents (O atoms), and it was somewhat inhibited due to appearance of additional electropositive (CH2 or Me) groups (Table 1).This effect is most pronounced for structurally similar radicals, e.g., the shifts of ‡ The maximum values of k0 and k2 are diffusion-controlled constants of ~1010 dm3 mol–1 s–1 and the reciprocal of the lifetime ~1012 s–1 for a collisional complex.The following averaged k values were chosen for the calculation: k0 = 7×109 dm3 mol–1 s–1 and k2 = 1011 s–1. § The correspondence between the differences in the redox potentials of intermediates and the Tafel plots for their electroreduction was postulated previously.1 aData from ref. 3. bThe value was obtained by extrapolation of the Tafel plot to W2 = 103 s–1.cThe reduction wave of radical 5 is overlapped by the dark discharge of H3O+ at pH < 3.5. Table 1 Electrochemical properties of linear and cyclic ether radicals and the constants of capture13 of H atoms and OH radicals by the ether molecules. a-Radical kOH (kH)/dm3 mol–1 s–1 a (b), ±0.05 (dE/dpH)W3 (dE/d[NH4 + ])W3 –E±0.05/V at W3 = 103 s–1 –E±0.05/V at W2 = 103 Ethoxyethyl 1 (2.9–4.2)×109 (4.3×107) 0.59 (0.61) 0.12 (0.132) 1.74 0.79 1,4-Dioxanyl 2 (2.5–3.1)×109 (107) 0.47a (0.53) 0.13a 1.45 0.38b Tetrahydrofuran-2-yl 3 4×109 [(3.3–7.2)×107] 0.55 (0.58) 0.13 (0.114) 1.68 0.87 2,5-Dimethyltetrahydrofuran-2-yl 4 0.46 (0.63) 0.11 (0.13) 1.83 1.52 Tetrahydropyran-2-yl 5 0.54 (0.66) —c (0.142) 1.76 0.66 108 106 104 102 100 1.2 1.4 1.6 1.8 2.0 2.2 W3/s–1 –E/V (SCE) pH 1.7 pH 2.6 pH 3.1 pH 3.3–13.5 Figure 3 Typical relationship between W3 and H3O+ concentration.Radical 4; the experimental conditions are the same as in Figure 1. Dashed lines show the calculations of W3(E, pH) in terms of the model.6Mendeleev Communications Electronic Version, Issue 2, 2000 (pp. 43–82) the Tafel plots are ~0.25±0.03 V for radicals 2 and 5 and ~0.15±0.03 V for radicals 3 and 4.Thus, the a-radicals of linear and cyclic ethers can form complexes with proton donors at the rate constants of 107–1010 s–1, which depend on the nature of BH+ only slightly. In contrast, the reduction overvoltages for the [Rads·H3O+] and [Rads·NH4 +] complexes are substantially different. The electrochemical properties of adsorbed a-radicals of ethers and alcohols, i.e., the mechanisms and rates of electrode processes, were found to be similar. In both cases, the electrode reactions are described by the model,6 i.e., electron transfer proceeds by the two parallel pathways: directly to Rads or to the antecedently formed metastable complex of Rads with a proton donor or acceptor. The rate of electron transfer to the adsorbed radicals correlates with the structure and reactivity of the parent ether molecules.This work was supported by the Russian Foundation for Basic Research (grant no. 97-03-32265). References 1 V. A. Benderskii and A. G. Krivenko, Usp. Khim., 1990, 59, 3 (Russ. Chem. Rev., 1990, 59, 1). 2 V. A. Benderskii, A. G. Krivenko and V. A. Kurmaz, Elektrokhimiya, 1987, 23, 625 [Sov.Electrochem. (Engl. Transl.), 1987, 23, 577]. 3 V. A. Benderskii, A. G. Krivenko, A. S. Kotkin and V. A. Kurmaz, Elektrokhimiya, 1993, 29, 246 (Russ. J. Electrochem., 1993, 29, 221). 4 A. G. Krivenko, V. A. Benderskii, A. S. Kotkin and V. A. Kurmaz, Elektrokhimiya, 1993, 29, 869 (Russ. J. Electrochem., 1993, 29, 741). 5 V. A. Benderskii, A. G. Krivenko and V. A. Kurmaz, Elektrokhimiya, 1986, 22, 644 [Sov.Electrochem. (Engl. Transl.), 1986, 22, 607]. 6 A. G. Krivenko, A. S. Kotkin and V. A. Kurmaz, Mendeleev Commun., 1998, 56. 7 (a) A. G. Krivenko, A. P. Tomilov, Yu. D. Smirnov, A. S. Kotkin and V. A. Kurmaz, Zh. Obshch. Khim., 1998, 68, 292 (Russ. J. Gen. Chem., 1998, 68, 266); (b) V. A. Kurmaz, A. G. Krivenko, A. P. Tomilov, V. V. Turigin, A. V. Khudenko, N.N. Shalashova and A. S. Kotkin, Elektrokhimiya, 2000, 36, 344 (in Russian). 8 A. Hynes, in Comprehensive Organic Chemistry, ed. J. F. Stoddart, Pergamon Press, Oxford, 1979. 9 M. Ya. Melnikov and V. A. Smirnov, Handbook of Photochemistry of Organic Radicals, Begell House, New York, 1996. 10 Z. A. Rotenberg and N. M. Rufman, J. Electroanal. Chem., 1984, 175, 153. 11 A. Henglein, in Electroanal. Chem., ed. A. J. Bard, Marcel Dekker, New York, 1976, vol. 9, p. 163. 12 D. D. M. Wayner, D. J. McPhee and D. Griller, J. Am. Chem. Soc., 1988, 110, 132. 13 G. V. Buxton, C. L. Greenstock, W. Ph. Helman and A. R. Ross, J. Phys. Chem. Ref. Data, 1988, 17, 513. 14 N. V. Fedorovich, Zh. Anal. Khim., 1993, 48, 1006 [J. Anal. Chem. (Engl. Transl.), 1993, 48, 700]. 15 N. V. Fedorovich, Ross. Khim. Zh. (Zh. Ross. Khim. Ob-va im. D. I. Mendeleeva), 1996, 40, 86 (in Russian). 16 V. A. Benderskii, A. G. Krivenko, E. A. Ponomarev and N. V. Fedorovich, Elektrokhimiya, 1987, 23, 1435 [Sov. Electrochem. (Engl. Transl.), 1987, 23, 1343]. 17 D. Schiffrin, Disc. Faraday Soc., 1974, 56, 75. 18 J. L. Goldman, R. M. Mank, J. H. Young and V. R. Koch, J. Electrochem. Soc., 1980, 127, 1461. 19 L. Kröner, D. Hoormann, E. Steckhan, J. Jorissen and H. Putter, 21st Sandbjerg Meeting on Organic Electrochemistry, Sandbjerg, 1999, p. 47. Received: 23th December 1999; Com. 99/15