J. CHEM. SOC. FARADAY TRANS., 1994, 90(5), 703-709 Characterization of Transients Formed in Aqueous Solutions of Substituted Alkyl Sulfides: A Pulse Radiolysis Studyt Dilip K. Maity, Hari Mohan and Jai P. Mittal Chemistry Division, Bhabha Atomic Research Centre, Trombay, Bombay 400 085,India ~ ~~~ The transient optical absorption band (A,,, = 340 nm) formed on reaction of OH radicals with substituted alkyl sulfides, S(CH,COpR),, R = CH,, C,H,, C,H,, C,H, in neutral aqueous solutions has been assigned to the OH adduct. The yield, lifetime and molar absorption coefficient of this band is found to decrease with increasing chain length of R. The OH adduct of S(CH2C02CH3), is able to undergo an electron transfer reaction with Br-, I-and SCN-. In acidic solutions, dimer radical cations are formed at high solute concentration.The observed rate constant for the decay of dimer radical cations is deconvoluted to different processes and the deprotonation is found to be the rate-determining step. The yield and lifetime are found to decrease with increasing chain length of R. The high concentration of acid required for their formation is discussed. Cli-and SO;-are able to undergo one-electron transfer reactions with S(CH2C0,CH3), with rate constants of 1.9 x lo9 and 1.4 x 10’’ dm3 mol-’ s-’, respectively. The dimer radical cation is found to be a strong oxidizing agent and is able to oxidize Br-with a rate constant of 2.2 x lo9 dm3 mol-’ s-’. Hydroxyl radical induced reactions with alkyl sulfides (R2S) in aqueous solutions have shown the formation of sulfur- centred radical cations.’ Simple sulfur-centred radical cations (R2S’+) are observed only in cases where the unpaired p- electron of oxidized sulfur is stabilised by the adjacent system or by steric influence^.^.^ These cations absorb at ca.300 nm. In other cases, simple sulfur-centred radical cations are converted to dimer radical cations (R,S);+, as oxidized sulfur has a high tendency to stabilise by co-ordination with a lone pair from the second sulfur at~m.~-~ The dimer radical cations absorb in the region 450-550 nm. The oxi- dized sulfur also has a high tendency to stabilise by coordi- nation with a lone pair of other hetero atoms such as N, 0,P and Such interactions are represented by a two-centre three-electron bond containing two bonding Q electrons and one antibonding o* electron and can take place by both inter- and intra-molecular association.In addition to electron-transfer reactions, OH radicals also react by H-atom abstraction and form a-thioalkyl radicals which absorb in the region 280-300 nm.4-7 Analysis of the transient signal obtained during and immediately after the pulse, on pulse radiolysis of an aqueous solution of ethyl sulfide showed a short-lived transient band (i,,,= 340 nm, t,,, < 1 p),which was assigned to an OH add~ct.~ This band could not be resolved as it is formed within the pulse. Because of the electrophilic nature of OH radicals and the high electron density at sulfur, the OH adduct may be imme- diately converted either to dimer radical cations or the a-thioalkyl radical via a complex sequence of reaction^.^ It is possible that lowering of the electron density at sulfur by the presence of groups of high electron-withdrawing power may increase the lifetime of the OH adduct.Pulse radiolysis of neutral aqueous solutions of dimethyl 3,3’-thiodipropionate gave a well resolved transient band (A,,, = 345 nm, tl,, = 4 p),which was assigned to the OH adduct.” On the other hand, OH radical induced reactions with thiodiglycolic acid (TDGA) led to the formation of a transient band (A,,, = 285 nm), which was assigned to a-thioalkyl radicals.12 This may, perhaps, be due to reduced electron density at sulfur as the C0,H group has very high electron-withdrawing power (Q* = +2.94).13 Re~ently,’~ it has been shown that OH radical induced reactions in aqueous solutions of methyl- ? Preliminary results were presented at the 3rd International Con- ference on Chemical Kinetics, July 12-16, 1993, Gaithersburg, USA.thiomethyl acetate lead to the formation of a transient band which has been assigned to the OH adduct formed on stabili- zation with an internal hydrogen bond between the hydroxyl hydrogen and an oxygen located either in an adjacent car- bony1 or methoxy group. Therefore, it appears that substit- uents can affect the nature of OH radical induced reactions with organic sulfur compounds. With this objective, pulse radiolysis studies have been carried out on substituted alkyl sulfides in which the electron density at sulfur is expected to be reduced by the presence of substituents of higher electron- withdrawing power.Such studies have gained importance as sulfur-centred transients are considered as possible interme- diates in redox reactions of biom~lecules~~*~ and are helpful in understanding the physico-chemical processes and redox properties in sulfur drugs, amino acids and other biological systems containing sulfur. A quantum-chemical calculation has been performed at the semiempirical (AM1) level to find the structures of different alkyl sulfides and their respective radical cations.16 The com- puted data for the net atomic charge on sulfur and ionization potential values for different sulfur compounds are used to analyse the present experimental results.Experimental Esters (dimethyl, diethyl, dipropyl, dibutyl) of 2,2’-thiodietha- noic acid were prepared by esterification of 2,2’-thiodiethanoic acid (TDGA) with the appropriate alcohols by a standard method.17 The purity of these esters was checked by thin- layer chromatography and IR analysis and they were found to be free from 2,2’-thiodiethanoic acid. Thiodipropionic acid dilauryl ester (C,0H5,04S, Registry No. 123-28-4) and 3,3’- thiodipropionic acid dioctadecyl ester (C4, H ,O,S, Regis-try No. 693-36-7) were obtained from Sigma Chemicals and used without further purification. All solutions were prepared in deionized ‘nanopure’ water. Fresh solutions were used in each experiment.The pH of the solution was adjusted with NaOH and HC104. Indian Oxygen, ‘iolar’ grade N,O, O,, N, gases were used for purgmg the solutions. The reaction of OH radicals in neutral aqueous solutions was carried out in N,O-saturated solutions where ea; are quantitatively con-verted to OH radicals (N,O + e,; -+ ‘OH + OH-+ N,) with G(0H) = 5.6. In acidic solutions, the reaction of OH radicals was carried out in 0,-saturated solutions where e- and H atoms are converted to HO, radicals (e,; + H+ + H“? , H' + 0, -+ HO;) and G(0H) = 2.9 (pH = 0-3.0). The reac- tion of 0'-was carried out in N,O-saturated solution at pH 12.5 where OH radicals are converted to 0'-('OH+ OH--+ 0'-+ H,O). Pulse radiolysis experiments were carried out with high- energy electron pulses (7 MeV, 50 ns), from a linear acceler- ator whose details are described elsewhere.'* The dose delivered per pulse (ca.1.0 x 1017 eV cm-3) was determined by use of an aerated aqueous solution of KSCN (10 mmol dm-3).19 The ac conductivity changes produced on pulse radiolysis were monitored in the dual-function cell arrange- ment and electronic detection system obtained from the Hahn-Meitener-Institut, Berlin.20 All the experiments were carried out at 25°C. The rate constant values were the average of at least three experiments and the variation was within *lo%. The atomic charge over the sulfur atom and the ionization potential of the different sulfides are computed in their fully optimized geometry.' The geometry optimization was by semiempirical molecular-orbital calculations with AM 1 para-metrization,2' as this appears to be the most successful method.Results and Discussion Reaction of OH Radicals with Dimethyl 2,2'-thiodiethaooate at pH 6.0 Fig. l(a) shows the transient optical absorption spectrum obtained on pulse radiolysis of N,O-saturated aqueous solu- tion of dimethyl 2,2'-thiodiethanoate (DMTE) (1.3 x lop3 mol dm-3, pH 6.0). It exhibits an absorption band with A,,, = 340 nm. In N,O-saturated aqueous solutions at pH 6.0, the primary reactive species produced on pulse radiolysis would be OH radicals and H atoms, The AA value at 340 nm decreased from 0.029 to 0.0028 in the presence of 0.2 mol dmV3 tert-butyl alcohol, a strong OH-radical and weak H-atom scavenger.Therefore, this band [Fig. l(a)] is attrib- uted to the reaction of OH radicals with DMTE. The small absorption observed in the presence of tert-butyl alcohol [Fig. l(b)] is attributed to the reaction of H atoms with DMTE as independent studies on the pulse radiolysis of N,-saturated aqueous solutions of DMTE (1.0 x mol dm-3, pH = 1.5, [tert-butyl alcohol] = 0.2 mol dm-3) also showed a similar absorption band. The rate constant for the reaction of OH radicals with DMTE is determined to be 2.3 x 10" dm3 mol-' SKI.The molar absorption coefficient of this band [Fig. l(a)], determined by competition kinetics using an N,O-saturated solution of KSCN (2.0 x mol dmP3, E~~~ = 7200 dm3 mol-' cm-l) is 3150+ 200 dm3 mol-' cm-'.The band is observed to decay by first-order kinetics with tl,, = 8.6 ps. J. CHEM. SOC. FARADAY TRANS., 1994,VOL. 90 300 4 00 500 600 wavelengthlnm Fig. 1 Transient optical absorption spectrum obtained immediately after pulse radiolysis of an aqueous solution of DMTE: (a) N,O-saturated, [DMTE] = 1.3 x mol dm-3, pH 6.0; (b) N,O-saturated, [DMTE] = 1.3 x lo-' mol dm-3, pH 6.0, [tert-butyl alcohol] = 0.2 mol dm-3; (c) N,O-saturated, [DMTE] = 1.3 x lo-' mol drnp3, pH = 6.0; (d) N,O-saturated, [DMTE] = 1.3 x rnol dmP3, pH = 12.5, (e) 0,-saturated, [DMTE] = 6.6 x rnol dmP3, [HClO,] = 4.5mol dm-3 The conductivity studies showed this band to be due to a neutral species as there was no change in the relative conduc- tivity of the solution before and after the pulse.Considering the neutral nature, high rate constant and previous experi- mental observations:.' ',I4 the band is assigned to the OH adduct formed according to reaction (1) (Scheme 1). It is also possible that the OH adduct may form a six-membered ring configuration with internal hydrogen bonding.I4 The OH adduct (A = 340 nm) may decay to a-thioalkyl radicals [reaction (2)] or to solute radical cations [reaction (3)]. The ambiguity between these two modes of decay was resolved when the decay of the 340 nm band was studied in the pres- ence of a proton acceptor. The decay of this band remained unaffected in the presence of 5.0 x mol dm-3 1 :1 mixture of HPOZ--H,PO,, suggesting that the decay may be reaction (2).The absence of conductivity changes would also support this process as the most probable mode of decay. The intensity of this band remained independent of solute concentration in the region (1.0-6.5) x lop3mol dm-3. At higher concentrations, the intensity of the 340 nm band reduced slightly and decay became faster. Simultaneously, another band of small intensity was observed to grow in the J. CHEM. SOC. FARADAY TRANS., 1994, VOL. 90 region of 485 nm [Fig. l(c)]. The decrease in the intensity and the faster decay of the 340 nm band at higher solute concen- tration could be due to the reaction of solute molecules with the OH adduct [reaction (7)] to form a solute dimer radical cation. The small decrease in the absorbance at 340 nm and corre- sponding increase at 485 nm [Fig.l(a) and (c)]is attributed to reaction (7). From these results, the molar absorption coef- ficient and the G value of the transient species at 485 nm are determined to be 1.6 x lo4 dm3 mol-' cm-' and 0.28, respectively. The absence of an absorption band in the region of 400 nm shows that intra-molecular orbital overlap between oxidized sulfur and oxygen is not taking place. This may be due to the unstable nature of the four-membered ring configuration. The 485 nm band is attributed to a dimeric species as it was observed only at high solute concentration. Reaction of OH Radicals with Different Esters In order to investigate the effect of the chain length of the ester group, similar studies have been carried out on different esters of 2,2'-thiodiethanoic acid (Table 1).These studies have been carried out at the same solute concentration (2.0 x 10-mol dmW3) and same dose (1.0 x lo" eV per pulse). In each case, a transient band was observed at the same wave- length (340 nm). The G value, E, and tl,, of the OH adduct decreases with increasing chain length showing that bulkier groups lower the stability of the OH adduct owing to steric influence. The G value (Table l),of the OH adduct is much less than that of G(0H) = 5.6 in N,O-saturated solutions. Therefore, the remaining fraction of OH radicals may be reacting to form the a-thioalkyl radicals, absorbing in the region 280- 300 nm. This band may lie within the broad absorption of the 340 nm band.The remaining fraction of OH radicals may not -0.02 nU300 400 500 600 wavelengt h/n m Fig. 2 Transient optical absorption spectrum obtained immediately after pulse radiolysis of an N,O-saturated solution of (a) dilauryl ester; (b) dioctadecyl ester; (c) N,-saturated DCE and (d) N,-be reacting to form the solute radical cations as the conduc- tivity studies have shown the formation of a neutral species. When the chain length between sulfur and the ester group is increased (as in the case of 3,3'-thiodipropionic acid), OH radical induced reactions showed somewhat different results. Fig. 2(a) shows the transient optical absorption spectrum obtained on pulse radiolysis of an N,O-saturated aqueous solution of thiodipropionic acid dilauryl ester.It exhibits an absorption band at 410 nm with a small shoulder at 340 nm. The 340 nm band may be due to the OH adduct and the 410 nm band may be due to radical cation 1 formed on p-orbital overlap of oxidized sulfur with oxygen making a two-centre three-electron bonding. The five-membered ring configu- ration obtained in this case is expected to increase the stabil- ity of this intra-molecular radical cation because of less angular strain. O=C-CH;!II 1 Owing to limited solubility, variation in the absorbance of transient bands with solute concentration could not be carried out. When the ester side chain is further increased from dilauryl to dioctadecyl, absorption of the transient band formed on pulse radiolysis is further reduced [Fig.2(6)]. The decreased intensity may be due to lower stability introduced by the bulky side chain. Reaction of OH Radicals with DMTE at Different pH In the pH range 0-10, there was no change in the intensity of the 340 and 485 nm bands formed on reaction of OH radicals with DMTE at low and high solute concentrations, respec- tively. Even the decay remained unaffected. At pH > 12.0, the OH radicals would be converted to 0.-.Fig. l(6) shows the transient optical absorption spectrum obtained on pulse radiolysis of an N,O-saturated aqueous solution of DMTE (1.3 x mol dm-3) at pH 12.5. This exhibits an absorption band with A,,, = 285 nm that may be due to a-thioalkyl radicals formed on H-atom abstraction by 0'-. This was observed to decay by second-order kinetics 2k/ d = 9.0 x lo5 s-'. The formation rate constant was deter- mined to be 3.0 x lo9 dm3 mol- 's-'. When the HClO, concentration was >1.0 mol dm-3, the intensity of the 340 nm band decreased and another absorp- tion band appeared in the region of 485 nm.This shows that the OH adduct is reacting with H+ only when the HClO, concentration is ~1.0mol dm-3. Fig. l(e) shows the tran- sient optical absorption spectrum obtained on pulse radiolysis of 0,-saturated solution of DMTE (6.6 x mol dm-3) in 4.5 mol dm-3 HClO,. This band was not observed in the presence of the OH radical scavenger (tert-saturated, [DMTE] = 2.5 x Table 1 ~max solute /nm DMTE" 340 DETE~ 340 DPTE' 340 DBTEd 340 rnol dm-3 in DCE Physical parameters of the OH-adduct formed from various esters of 2,2'-thoidiethanoic acid formation rate constant /1O'O dm3 mol-ls-' G 8.6 3.65 f0.2 6.5 2.85 0.2 1.7 3.27 0.3 1.1 1.95 0.4 E /dm3 mol-' cm-' 3150 f200 3130 f250 2620 f 300 2210 * 500 ~~ Dose = 1.0 x lo" eV cm-3 per pulse.Solute concentration = 2.0 x mol dm-3. Dimethyl 2,2'-thiodiethanoate. * Diethyl 2,2'-thiodietha- noate. Dipropyl 2,2'-thiodiethanoate. Dibutyl 2,2'-thiodiethanoate. 706 [HClO,]/mol dm-3 0 4.0 I I 1 3.0 0 0 2.0 4.0 6.0 8.0 10.0 [DMTE]/1O3 rnol dm-3 Fig. 3 Variation in the absorbance of 485 nm band as a function of (a) HClO, concentration, [DMTE] = 6.6 x mol dm-3; (b) DMTE concentration, [HClO,] = 9.8 mol dmP3 butyl alcohol) or in the absence of DMTE showing that the band is due to reaction of OH radicals with the solute and not to any transient species formed from radiolysis of HClO,.Fig. 3(a) shows the variation in the absorbance of the 485 nm band as a function of HClO, concentration. This may not represent the true variation of absorbance with [HClO,] as the dose absorbed by water would not remain the same at higher [HCIO,]. The rate constant for the reac-tion of OH radicals with DMTE ([HCIO,] = 9.2 mol dmP3, A = 485 nm) was determined to be 1.4 x lo9 dm3 mol-I s-'. The intensity of this band decreased with decrease in the solute concentration without any change in the position of A,,,.The absence of the transient band in the region of 400 nm at low solute concentrations again suggests that the intra-molecular radical cation is not formed, possibly because of the unstable nature of the four-membered ring configuration. The reactive species produced on radiolysis of aerated aqueous solutions containing a high concentration of HCIO, are HO,, OH and ClO, radicals. The HO, radicals could not be the source of the transient band absorbing at 485 nm as (i) the redox potential value of the HO, radical is quite low (+1.0).,, Stronger one-electron oxidants such as Bri-were unable to produce this band (see text). (ii) The intensity of this band remained unaffected in N,-saturated solutions where HO, radicals would not be formed.C10, radicals also could not be responsible for the tran-sient band observed at 485 nm as (i) they have no absorption at i> 400 nm23 and ClO, radicals generated on pulse radiolysis of a neutral aqueous solution of NaClO, (7.5 mol drnp3)containing 5 x lop3mol dmP3 DMTE do not show transient absorption similar to that observed in the presence of HClO,. The OH radicals are shown to undergo acid-J. CHEM. SOC. FARADAY TRANS., 1994, VOL. 90 dation of the OH adduct [reaction (5)J If the formation of the 485 nm band is due to acid-catalysed oxidation of the OH adduct, then a similar band should also be observed in the presence of another acid. In fact the pulse radiolysis studies in aerated aqueous solutions of DMTE (5.0 x mol drnp3)containing 5.0 mol dm-3 H,SO, showed the for-mation of a transient band at 485 nm.The absorbance at 485 nm was slightly lower in the case of H,SO, than that observed in 5.0 mol dm-3 HClO,. This should be due to the lower Hammett acidity function (H,) value for H,SO, .,' In H,SO, solution, the 485 nm band could not be due to SO;-as (i) it absorbs at 450 nm30 and (ii) the rate constant for the reaction of OH radicals with HSO, is very low (8 x lo5 dm3 mol-1 s-1 30).The increase in the intensity and lifetime of the transient band with solute concentration suggest the existence of an equilibrium [reaction (6)]. Therefore, the transient band observed in the presence of high acid and solute concentra-tion is assigned to the dimer radical cation of DMTE formed via reactions (l),(5) and (6).The absorbance of the transient band at 485 nm [Fig. 3(b)] has reached saturation at a solute concentration of 6 x mol dm-3 showing that all the solute radical cations are con-verted to dimer radical cations. Taking the molar absorption coefficient value of 1.6 x lo4 dm3 mol-' cm-', and the absorbance at 485 nm of 0.039, the G value of the transient species absorbing at 485 nm is determined to be 1.45, much less than G(0H) = 2.9 in neat water at pH = 1.0. This must be due to the much lower value of G(0H) in the presence of a high acid concentration as all the radiation energy would not be absorbed by water alone. The product of the G and E (2.32 x lo4 dm3 mol-I cm-') obtained for DMTE in this way, is close to that determined directly by independent experiment (2.26 x lo4 dm3 mol-' cm-').These studies have also been performed on the other esters (Table 2). In order to compare the stability and yield of dimer radical cation of these esters, pulse radiolysis studies have been carried out at identical conditions of dose and concen-tration. It can be seen that the yield (GE) and lifetime decreases as the chain length of the ester group is increased. It is suggested that the extra methylene group may hinder a good orbital overlap of the oxidized sulfur atom with the unoxidized molecule to form a strong two-centre three-electron bond. A red shift in the transient absorption band (Table 2) is observed with increasing chain length in the ester group.The decreased separation between o-o* orbitals due to increased electron-releasing power of the substituted alkyl group,13 would cause a red shift in A,,,. On the other hand, a large separation between these orbitals would correspond to a stronger interaction. The increased electron-withdrawing power would in turn reduce the electron density over the sulfur, and so the net atomic charge over the sulfur atom will increase (become more positive). Our quantum-chemically catalysed oxidation of and bromo-alkanes,26 and computed results (Table 3) provide quantitative data for the a number of other organic corn pound^.^^^^^ Therefore, it is net atomic charge over the sulfur atom for different com-possible that this band (485 nm) is due to H+-catalysed oxi-pounds and the data support our conclusions.Table 2 Physical parameters of the dimer radical cation formed from various esters of 2,2'-thiodiethanoic acid formation stability LX rate constant tlj2 constant solute /nm /lo9 dm3 mol-' s-' /P /dm3 mol-' DMTE 485 1.4 DETE 500 1.2 DPTE 510 1 .o DBTE 520 - [solute] = 4.0 x lop3mol dm-3. [HClO,] = 150 105 63 450 22 433 9.2 mol dm-3. Dose = 1.0 x 1017 eV ~rn-~per pulse. GE k -. 6a /lo3 dm3 mol-' /lo7 s-1 22.6 1.3 17.1 0.26 13.4 0.23 7.6 __ See Scheme 1. J. CHEM. SOC. FARADAY TRANS., 1994, VOL. 90 Table 3 Net atomic charge at sulfur and ionization potential for various substituted alkyl sulfides Taft ionization net atomic charge parameter, potential acid conc.for formation of solute at sulfur o* lev dimer cations/mol dm -3 -0.068 0.0 9.54 -0.099 -0.100 9.36 FH2C02CH3 s\ +0.066 +2.00 10.21 > 1.0 CH2C02CH3 /CH2C02C2H 5 s, +0.065 10.09 >1.0 CH,C02C2H5 FH2C02H +0.13 +2.94 10.48 >4.0s\CH2C02H Decay Kinetics The 485 nm band was observed to decay by first-order kinetics with a lifetime dependent on the solute and HClO, concentration. Asmus and co-worker~,~' have shown that the rate-determining step for the decay of the dimer radical cation of alkyl sulfides can be either the back reaction of the equilibrium (k -6) or the deprotonation of solute radical cation (k4). Using different concentrations of proton acceptor to accelerate the deprotonation of solute radical cations, they have shown from the plots of kobs us. kobs [solute] that the deprotonation of radical cation would be the rate-determining step if plots have same slope (k6)and different intercepts (k, + deprotonation rate constant x [proton acceptor]).The decay by back reaction of the equilibrium would yield the same intercept (k-6) and different slopes (k,/(k, + deprotonation rate constant x [proton acceptor])}. This method of estimation of the various rate constants associated with the decay of dimer radical cations is not pos- sible in the present case as dimer radical cations are observed at high acid concentration and a proton acceptor could not be used. However, plots of kobs us.kobs [solute] for various acid concentrations gave straight lines (Fig. 4) with the same slope and different intercept. Therefore, deprotonation reac- 0' I I I 0.1 0.2 0.3 0.4 0.5 k,,,[solute]/mol dm-3 s-' Fig. 4 kobaDS. k,, [DMTE] for various HClO, concentrations; (a) 9.2, (b) 8.3 and (c) 7.4 mol dm-3 tion should be the rate-determining step and the stability constant of the dimer radical cation would be equal to the slope of straight lines (Fig. 4) and the values are shown in the Table 2. From the formation rate constant value of 1.4 x lo9 dm3 mol-' s-', for DMTE, k-, is calculated to be 1.3 x lo7 s-l. The stability constants for the dimer radical cation gener- ated from other esters has also been evaluated by a similar procedure and the values are shown in Table 2.The stability constants for the dimer radical cation of dimethyl sulfide and dipropyl sulfide has been reported to be 2 x lo5 and 540 dm3 mol-', re~pectively.~'The lower value of the stability con- stant for DMTE explains the instability. The stability con- stant for the dimer radical cation of thiodiglycolic acid (68 dm3 mol-') was still lower than that of DMTE (105 dm3 mol-') and explains the higher concentration of acid (>4.0 mol dm-3) required for the formation of dimer radical cations of TDGA12* as compared with DMTE (> 1.0 mol dm-3). The intercept is seen to depend on H+ concentration (Fig. 4). This may be due to the fact that at a lower H+ concentra-tion (<1.0 mol dm-3), the OH adduct would decay to a- thioalkyl radicals and at a higher concentration of H+, the OH adduct can react with H+ to form radical cations.It is also possible that a-thioalkyl radicals may react with H+ at higher acid concentration. Therefore, the intercept would be represented by a complex function and it would be difficult to determine the deprotonation rate constant. The ratio of the intercept for two values of HClO, increases at a much higher rate than the increase in the ratio of the HClO, concentra-tion. This may be due to an increased contribution of reac- tion (2) at lower H concentration.+ Variation of Atomic Charge at Sulfur with Substituents It has been shown that the nature of the OH radical reaction depends on the nature of the substituents present in the mol- ecule.The net atomic charge at sulfur has been calculated (Table 3) and the results suggest that the amount of acid required for acid-catalysed oxidation of substituted sulfides by OH radicals can be related to the net atomic charge on the sulfur. The ionization potential of substituted alkyl sul- 708 I -2.5 22 E 9.0 I I I 12.0 -0.1 0.0 0.1 0.2 net atomic charge at S Fig. 5 Variation of the ionization potential of alkyl sulfides and the energy corresponding to A,,, of the corresponding dimer radical cation with net atomic charge at sulfur for various alkyl sulfides fides is also observed to increase with electron-withdrawing power or net atomic charge on the sulfur (Fig.5). This is expected as reduced electron density at the sulfur would make it difficult to ionize the compound. With simple alkyl sulfides, such as dimethyl sulfide, the dimer radical cations are observed at neutral pH. The electron-releasing power of the C,H, group in diethyl sulfide, increases the electron density (lowers the net atomic charge) at sulfur and thereby lowers the ionization potential (Table 3). As the alkyl group is replaced by substituents with high electron-withdrawing groups, the electron density at sulfur is lowered and the net atomic charge increases. The sulfide radical cation formation required higher acid concentration. The net atomic charge at sulfur in TDGA is very high, showing very low electron density at sulfur and an acid concentration >4.0 mol dm- ' was required for the electron-transfer reaction. The ionization potential is also very high.When C0,H groups are replaced by C0,CH3 groups as in S(CH,C0,CH3)2 ,electron density at sulfur is relatively higher owing to the electron-releasing power of the CH, group and OH radicals are able to form an OH adduct and dimer radical cations are formed at relatively lower H+ concentration (> 1.0 mol dm-3). The ionization potential calculated for this compound also supports the atomic charge at sulfur. For other substituted alkyl sulfides (Table 3), with net atomic charge on sulfur lower than S(CH2C0,CH3),, the amount of acid required for dimer radical cation formation and the ionization potentials are lower.Therefore, it can be concluded that the nature of the OH radical reaction is greatly influenced by the electron- releasing/-withdrawing power of the substituents. The ioniza- tion potential and amount of acid required for acid-catalysed oxidation by OH radicals increases with net atomic charge on sulfur. The position of the energy corresponding to Amax (eV) for the dimer radical cation of the substituted alkyl sul- fides is also observed to depend on the net atomic charge at sulfur (Fig. 5), which suggests that the strength of the sulfur- sulfur three-electron bond decreases with the presence of electron-withdrawing substituents in substituted alkyl sul- fides. Stability of Esters in High Acid Concentration The optical absorption spectra of an acidic aqueous solution of DMTE remained unchanged with time, suggesting that the solution is stable and does not undergo hydrolysis.The OH radicals failed to form dimer radical cations at HClO, < 4.0 mol dm3 in acidic aqueous solutions of TDGA. It is also reported3' that in presence of H+, hydrolysis of esters would be difficult and an adduct with H+ at the sulfur may be formed. Hydrolysis is expected to be faster in highly basic solutions.32 J. CHEM. SOC. FARADAY TRANS., 1994, VOL. 90 Reactions with Specific Oneelectron Oxidants The decay of Cli-, formed on pulse radiolysis of an 0,-saturated solution of C1- (2.0 x lo-, mol dm-3, pH 1.5) was observed (A,,, = 345 nm) to become faster and of first order in the presence of small amounts of DMTE [(l-7) x mol dm-3], showing that Cl;-is able to undergo an electron-transfer reaction with DMTE.The second-order rate constant for the reaction was determined from the slope of the linear plot of the pseudo-first-order rate Cl;-+ DMTE +2C1-+ DMTE+ (8) vs. DMTE concentration and the value is 1.9 x lo9 dm3 mol-' s-'. Time-resolved studies showed the formation of a new band with A,,, = 380 nm. This band may be due to a neutral three-electron-bonded species formed between DMTE+ and C1 -, 2. Three-electron-bonded species between halogen and alkyl sulfides are known to absorb in this region.33 The increased absorption in the region of 500 nm was not observed even at high solute concentration. C H2C02CH3 CI:.S / \ CHZCOZC H3 2 SO;-is also a strong oxidizing agent.The decay of SO;-, formed on pulse radiolysis of the N,-saturated solution of S20i-(2.0 x lo-, mol dm-3, tert-butyl alcohol, 0.5 mol dm-3, A = 460 nm) was found to become faster in the pres- ence of a small concentration of DMTE. The rate constant for the oxidation of DMTE by SO;-was determined to be 1.4 x 10" dm3 mol-I s-'. Time-resolved studies did not show the formation of a new band in the 350-600 nm region. This shows that although the electron transfer is taking place, the dimer radical cation is not formed even at high solute concentration. This suggests that the radical cation is not stable at this pH. At neutral pH, deprotonation of the radical cation to a-thioalkyl radicals is very fast.Br;- and 1;-were not able to undergo an electron-transfer reaction with DMTE, suggesting that the redox potential for the formation of the radical cation of DMTE is more than that of the Br;-/2Br-couple (+ 1.6 V). The transient (A = 485 nm) formed on reaction of OH radicals with DMTE was observed to oxidize Br- with a rate constant of 2.2 x lo9 dm3 mol-I s-and time-resolved studies showed the formation of a new band with A,= = 380 nm. This could be due to a Br adduct as Bri-, if formed, would have been evidenced by a transient band at 360 nm. These studies suggest that the redox poten- tial for the DMTE/DMTE+ couple is roughly between 1.6 and 2.0 V. The OH adduct was also observed to react with I-, Br-and SCN-. The time-resolved studies showed the formation of a new band (8 ps after the pulse) with A,,, = 380 nm.Detailed kinetic studies could not be carried out as the absorption bands of the OH adduct and the oxidized species are close to each other. The new band formed at 8 ps after the pulse could be due to the adduct of an oxidized atom with DMTE. Formation of Radical Cations in l,%Dichloroethane lY2-dichloroethane (DCE) has been used frequently as a solvent for the study of solute radical cations of organic com- pound~~~because of its high ionization potential. Therefore, complementary studies have been carried out in DCE to investigate the formation of radical cations of DMTE, which could be formed by the following mechanism : CH,CICH,CI --+ CH,CICH,CI+ + e-(9) CH,ClCH,Cl + e--+ CH,ClCH, + C1-(10) J.CHEM. SOC. FARADAY TRANS., 1994, VOL. 90 CH2ClCH2C1++ DMTE 4DMTE" + CH,ClCH,CI Fig. 2(4 shows the transient optical absorption spectrum obtained on pulse radiolysis of an N,-saturated solution of DMTE (2.5 x mol dm-3) in DCE.-It-exhibits absorp- tion bands at 370 nm. Pulse radiolysis of DCE does not produce this band [Fig. 2(c)J.The broad absorption in the region of 500 nm was observed at high solute concentration. Therefore, it should be due to dimer radical cations. The band at 370 nm may be due to a species of the type Cl:. S(CH,CO,CH,), formed between C1- and DMTE". The band at 370 nm was observed to decay by first-order kinetics with tl,2 = 36 p. The complex between C1- and radical cations of alkyl sulfides is reported to absorb in the region of 380 nm.33 Conclusions The OH radicals react with esters of 2,2'-thiodiethanoic acid to form an OH adduct in neutral solution and dimer radical cations in acidic solution.The deprotonation is the rate- determining step for the decay of dimer radical cations. A high concentration of acid is required for the formation of dimer radical cations owing to the reduced electron density at the sulfur. References K-D. Asmus, in Sulfur-centered Reactive Intermediates in Chem- istry and Biology, ed. C. Chatgilialoglu and K-D. Asmus, NATO AS1 Series A: Life Sciences, vol. 197, Plenum Press, New York, 1990, p. 155. K. Kim, S. R. Mani and H. J. Shine, J. Org.Chem., 1975, 40, 3857. K-D. Asmus, D. Bahnemann, Ch. H. Fischer and V. Veltwisch, J. Am. Chem. SOC., 1979,101,5322. M. Bonifacic, H. Mockel, D. Bahnemann and K-D. Asmus, J. Chem. SOC.,Perkin Tmns. 2, 1975, 675. M. Gobl and K-D. Asmus, J. Chem. SOC., Perkin Trans. 2, 1984, 691. K. Bobrowski and J. Holcman, J. Phys. Chem., 1989,93,6381. M. Gobl, M. Bonifacic and K-D. Asmus, J. Am. Chem. SOC., 1984,106,5984. R. S. Glass, M. Hojatie, G. S. Wilson, S. Mahling, M. Gobl and K-D. Asmus, J. Am. Chem. SOC., 1984,106,5382. H. Hungerbuhler, S. N. Guha and K-D. Asmus, J. Chem. SOC., Chem. Commun., 199 1,999. 10 E. Anklam, H. Mohan and K-D. Asmus, J. Chem. SOC., Perkin Trans. 2, 1988, 1297. 11 H. Mohan and J. P. Mittal, J. Chem. SOC., Perkin Trans. 2, 1992, 207.12 (a) H. Mohan and P. N. Moorthy, J. Chem. SOC., Perkin Trans. 2, 1990, 413; (b) D. K. Maity and H. Mohan, J. Chem. SOC., Perkin Trans. 2, 1993, 2229. 13 R. W. Taft, J. Chem. Phys., 1957,26,93;J. Am. Chem. SOC., 1953, 75, 423 1. 14 (a) K. Bobrowski and C. Schoneich, J. Chem. SOC., Chem. Commun., 1993, 795; (b) C. Schonich and K. Bobrowski, J. Am. Chem. SOC., 1993,115,6538. 15 C. von Sonntag, Chemical Basis of Radiation Biology, Taylor and Francis, New York, 1978, p. 353. 16 D. K. Maity, Hari Mohan and J. P. Mittal, unpublished results. 17 A. I. Vogel, Text Book of Practical Organic Chemistry, Longman, London, 1987, p. 501. 18 (a) S. N. Guha, P. N. Moorthy, K. Kishore, D. B. Naik and K. N. Rao, Proc. Id. Acad.Sci., (Chem. Sci) 1987,99,261; (b)K. I. Priyadarsini, D. B. Naik, P. N. Moorthy and J. P. Mittal, Proc. 7th Tihany Symp. on Radiation Chemistry, Hungarian Chemical Society, Budapest, 1991, p.105. 19 E. M. Fielden, The Study of Fast Processes and Transient Species by Electron Pulse Radiolysis, ed. J. H. Baxendale and F. Busi, Reidel, Boston, 1984, p. 59. 20 K-D. Asmus and E. Janata, The Study of Fast Process and Tran- sient Species by Electron Pulse Radiolysis, ed. J. H. Baxendale and F. Busi, Reidel, Boston, 1984, p.91. 21 M. J. S. Dewar, E. G. Zoebisch, E. F. Healy and J. J. P. Stewart, J. Am. Chem. SOC.,1985,107,3902. 22 B. H. J. Bielski, R. L. Arudi and M. W. Sutherland, J. Bid. Chem., 1983,258,1748. 23 S. A. Choudhri, M. Gobl, T. Freyholdt and K-D. Asmus, J. Am. Chem. SOC., 1984,106,5988. 24 H. Mohan and K-D. Asmus, J. Chem. SOC., Perkin Trans. 2, 1987,1795. 25 H. Mohan and J. P. Mittal, J. Chem. SOC., Perkin Trans. 2, 1992, 1731. 26 H. Mohan, D. K. Maity and J. P. Mittal, J. Chem. SOC.,Faraday Trans., 1993,89, 477. 27 K-D. Asmus, P. S. Williams, B. C. Gilbert, J. N. Winter, J. Chem. SOC., Chem. Commun., 1987,208. 28 M. Bonifacic and K-D. Asmus, J. Phys. Chem., 1976,80,2426. 29 M. A. Paul and F. A. Long, Chem. Rev., 1957,57,1. 30 E. Heckel, A. Henglein and G. Beck, Ber. Bunsenges. Phys. Chem., 1966,70,149. 31 J. Monig, R. Goslich and K-D. Asmus, Ber. Bunsenges. Phys. Chem., 1986,90, 115. 32 J. March, Advanced Organic Chemistry, Wiley Eastern Ltd., 1986, p. 334. 33 M. Bonifacic and K-D. Asmus, J. Chem. SOC., Perkin Trans. 2, 1980,758. 34 T. Sumiyoshi, N. Sugita, K. Watanal and M. Katayama, Bull. Chem. SOC.Jpn., 1988,61,3055. Paper 3/04990F; Received 17th August, 1993