(Alkylthio)alkyl-Type Radical Formation in - American Chemical Society

Dec 14, 2009 - techniques, were used to investigate the detailed mechanism of the •OH-induced oxidation of R-(methylthio)- acetamide (R-MTA) in aque...
0 downloads 0 Views 638KB Size
Download PDF
J. Phys. Chem. A 2010, 114, 105–116

105

Efficient r-(Alkylthio)alkyl-Type Radical Formation in •OH-Induced Oxidation of r-(Methylthio)acetamide Pawel B. Wisniowski,† Gordon L. Hug,‡,§ Dariusz Pogocki,†,| and Krzysztof Bobrowski*,† Centre of Radiation Research and Technology, Institute of Nuclear Chemistry and Technology, 03-195 Warsaw, Poland, Radiation Laboratory, UniVersity of Notre Dame, Notre Dame, Indiana, 46556, Faculty of Chemistry, Adam Mickiewicz UniVersity, 60-780 Poznan, Poland, and Faculty of Chemistry, Rzeszow UniVersity of Technology, 35-959 Rzeszow, Poland ReceiVed: July 25, 2009; ReVised Manuscript ReceiVed: NoVember 12, 2009

Pulse radiolysis with UV-vis/ESR detection and steady-state γ-radiolysis, combined with chromatographic techniques, were used to investigate the detailed mechanism of the •OH-induced oxidation of R-(methylthio)acetamide (R-MTA) in aqueous solution. The main pathway involves the formation of hydroxysulfuranyl radicals R-MTA-(>S•sOH) and R-(alkylthio)alkyl radicals H3CsSs•CHsC(dO)sNH2 (λmax e 260 and 340 nm). The latter radicals are highly stabilized through the combined effect of both substituents in terms of the captodative effect. At low pH, R-MTA-(>S•sOH) radicals undergo efficient conversion to intermolecularly three-electronbonded dimeric radical cations of R-MTA-(>S∴SS•sOH) radicals decompose via the elimination of water, formed through intramolecular hydrogen (attached to the nitrogen atom) transfer to the hydroxysulfuranyl moiety within a six-membered structure. This process leads to the formation of the imine radical H3CsSsCH2sC(dO)•NH, which subsequently decays in three independent channels. The first decay channel begins with a β-scission followed by hydrolysis and a subsequent Hofmann rearrangement. One of the end products of this first decay channel is CO2, which was detected. The second decay channel involves an intramolecular hydrogen transfer from the δC carbon atom to the radical imine site producing the R-(alkylthio)alkyl radical H2C•sSsCH2sC(dO)sNH2. In the third decay channel there is a 1,3-hydrogen shift in the imine radical which forms the radical H3CsSs•CHsC(dO)sNH2. The presence of the amide group induces more complex radical chemistry that leads unexpectedly to the degradation of the CH3SCH2CONH2 molecule into gaseous products, CO2 and NH3. These features of the mechanism of the •OH-induced oxidation of R-MTA are quite different from those seen in other organic sulfides in neutral solutions. Introduction A large effort has been put into characterizing the reactive intermediates that are involved in sulfide oxidation and into understanding their corresponding reaction mechanisms.1-4 The oxidation mechanism of simple alkyl sulfides involving •OH radicals is complex; however, it has been very well elaborated.5-9 The oxidation mechanism of sulfides containing neighboring groups has been found to be even more complex.3,9-27 The active participation of neighboring groups in the oxidation of organic sulfides has been observed for a variety of functionalities such as sulfide, amino, hydroxyl, ester, and carboxylate groups.10,11,17-19,28-30 This is due to the fact that these neighboring groups generally act by providing electron lone pairs that can stabilize sulfide radical cations (R2S+•) through the overlap of the heteroatoms’ doubly occupied p orbital of the sulfur.8,31-33 These neighboring-group participations can play important roles in several biological processes such as oxidative stress accompanying the etiologies of major neurodegenerative disorders or normal aging.34-39 * Corresponding author. Tel: (48)-(22)-504-1336. Fax: (48)-(22)-8111532. E-mail: [email protected], [email protected]. † Institute of Nuclear Chemistry and Technology. ‡ University of Notre Dame. § Adam Mickiewicz University. | Rzeszow University of Technology.

In our recent publications40,41 it was demonstrated that the β-positioned acetyl group to the sulfur atom affects the ultimate course of the •OH-induced sulfide oxidation in neutral or lowacidic aqueous solutions of S-ethylthioacetone (H3CsCH2s SsCH2sC(dO)sCH3). This was manifested by the very low yield of the respective intermolecularly three-electron bonded dimeric radical cations (>S∴S25 mM) of S-ethylthioacetone. Moreover, the R-(alkylthioalkyl) radicals of the type H3CsCH2sSs•CHsC(dO)sCH3 were found to be particularly stabilized through the combined effect of both substituents in terms of a captodative effect.42,43 Their formation was kinetically favored because of the respective transition states formed with the lowest activation energies.41 The low activation energies of the transition states found on the reaction pathways account for the efficient direct conversion of the hydroxysulfuranyl radicals of S-ethylthioacetone into C-centered radicals. At low proton concentrations, hydroxysulfuranyl radicals, even at high concentrations of S-ethylthioacetone, decompose via elimination of water, formed through intramolecular Htransfer within the six-membered structure that leads to the formation of C-centered radicals which further undergo a 1,3-H shift and/or intramolecular H-abstraction within the sixmembered structure leading to the R-(alkylthio)alkyl radicals (Scheme 1). These reaction pathways, important in neutral solutions, are responsible for the absence or low yield of dimeric

10.1021/jp9071026  2010 American Chemical Society Published on Web 12/14/2009

106

J. Phys. Chem. A, Vol. 114, No. 1, 2010

Wisniowski et al. M-1 s-1)48 resulting in a pH-dependent lowered yield of •OH radicals, and a correspondingly increased yield of •H atoms.

SCHEME 1: Mechanism of the Decay of the Hydroxysulfuranyl Radical of S-Ethylthioacetone

eaq- + H+ f •H

radical cations (>S∴SS∴S20 mM) of R-(methylthio)acetamide and by the efficient formation of the R-(alkylthioalkyl) radicals of the type H3CsSs•CHsC(dO)sNH2. The enhanced hydrogen abstraction from the methylene group results again from its specific location between the sulfide and amide functionalities. Moreover, the presence of the amide group induces more complex radical chemistry that leads unexpectedly to the degradation of the CH3SCH2CONH2 molecule into gaseous products, CO2 and NH3. This apparently simple organic sulfide, with such a neighboring group, was recently found to be of interest in another type of free-radical reaction of the type SH2.44,45 Experimental Section Materials. All the chemical compounds used for the experiments were of the purest commercially available grade and were used as received. The sulfide compound R-(methylthio)acetamide (R-MTA) was obtained from Parish Chemical Co. Perchloric acid (HClO4) was purchased from Aldrich, and sodium hydroxide (NaOH) was purchased from Fisher. Anhydrous sodium sulfite (Na2SO3) was purchased from Mallinckrodt. Chemicals were used as received. All solutions for pulse radiolysis with UV-vis detection were prepared freshly with Millipore water. For pulse radiolysis with TRESR detection, the deionized water was purified in a reverse osmosisdeionization system with an UV-irradiation unit from a ServA-Pure Co. The pH of the solutions was adjusted by adding HClO4 or NaOH. The pH 7 buffers for the TRESR experiments were made with sodium phosphate monobasic (NaH2PO4) and dibasic (Na2HPO4), both from Fisher. They were subsequently deoxygenated by purging with high purity N2 or N2O. Radiolysis of Water. Pulse irradiation of water leads to the formation of the primary reactive species shown in reaction a.46 In N2O-saturated solutions ([N2O]sat ≈ 2 × 10-2 M),47 the hydrated electrons, eaq-, are converted into •OH radicals according to eq b (kb ) 9.1 × 109 M-1 s-1).47 Reaction b nearly doubles the amount of •OH radicals available for reactions with substrates.

H2O f •OH, eaq-, •H -

•-

eaq + N2O f N2 + O

(a) •

(+H2O) f N2 + OH + OHpKa(•OH) ) 11.9 (b)

At pH < 4 the diffusion-controlled reaction of eaq- with protons becomes important (reaction c, kc ) 2.0 × 1010

(c)

The effective radiation chemical yields, G, of the primary species available for the reaction with a substrate depend on the concentration of the added substrate. The G denotes the number of micromoles of species formed or converted per 1 J of absorbed energy in aqueous solutions. For N2O-saturated solutions, the effective radiation chemical yield of •OH radicals converting a given substrate into substrate-derived radicals can be calculated on the basis of the formula given by Schuler et al.49 This formula relates the G-value of substrate radicals to the product ks[S] of the rate constant for the reaction of •OH radicals with the substrate times the substrate concentration itself. For N2-saturated solutions, the radiation chemical yield of •OH radicals was calculated on the basis of a simple relationship that relates their yields with the scavenging capacity of the system.50,51 Pulse Radiolysis with UV-Vis Detection. Experiments were performed with 2-5 ns pulses of high-energy electrons from the Lodz Technical University 6 MeV ELU-6 linear accelerator. A description of the pulse radiolysis setup and data collection system is reported elsewhere.52,53 Some experiments were performed using a new pulse radiolysis setup based on the 10 MeV electron linear accelerator (LAE 10) at the Institute of Nuclear Chemistry and Technology.54 The experiments were carried out with continuously flowing solutions. Absorbed doses were on the order of 3-5 Gy (1 Gy ) 1 J kg-1). N2O-saturated 10-2 M solutions of potassium thiocyanate were used as the dosimeter, taking a radiation chemical yield (G) of 0.635 µmol J-1 and a molar absorption coefficient (ε) of 7580 M-1 cm-1 at 472 nm. Spectral Resolutions. A multiregression analysis was done on the optical transient spectra resulting from the pulsed irradiation. In any time window, following the irradiation pulse, the absorbance of the signal is related to the radiation-chemical yield and the molar absorption coefficients by the formula11

Gε(λj) ) ∆A(λj)ε472G((SCN)2•-)/∆A472

(1)

Spectra are displayed in the figures using this parameter. Equation 1 is a convenient way to normalize the absorbance ∆A(λj) to the radiation dose, the inverse of which is proportional to ε472G((SCN)2•-)/∆A472 from thiocyanate dosimetry. The Gε(λj) are in turn related to the underlying transients via Beer’s Law since ∆A ) log(I0/I) is the absorbance of the sample. n

Gε(λj) )

∑ Giεi(λj)

(2)

i)1

In the regression analysis of eq 2, the Gi’s are the regression coefficients to be fit.55 The sets of εi(λj) are the reference spectra of the underlying transients enumerated by the ith subscript. The uncertainties that are quoted in the text for the Gi’s are computed as the square roots of the diagonal matrix elements of the covariance matrix from each linear regression. In addition, a check of the chemistry was done wherein it was required that the sum of the yields of the transient species could not exceed 15% of the sum of initial radiation chemical yields of the primary radicals for the radiolysis of water that were precursors of the secondary radicals that were observed. The standard reference spectra that were needed for input in the right-hand side of eq 2 were those for dimeric sulfur radical cations, hydroxysulfuranyl radicals, and two different types of

Radical Formation in Oxidation of Thioether R-(alkylthio)alkyl radicals. Since it was not possible to get the pure dimeric sulfur radical cation of R-MTA even at high concentrations and low pH (see below), we took its reference spectrum to be the dimeric sulfur radical cation of N-acetylmethionine amide with ε480 ) 8200 M-1 cm-1.56 This standard was chosen because dimeric sulfur radical cations have molar absorption coefficients that are somewhat dependent on substituents. N-Acetylmethionine amide was chosen to be as similar as we could find to R-MTA. The molar absorption coefficient of the standard spectrum for the hydroxysulfuranyl radical was ε340 ) 3400 M-1 cm-1,22 and that for ordinary R-(alkylthio)alkyl radicals was taken to be ε290 ) 3000 M-1 cm-1.57 The absorption spectrum for the new captodative-stabilized R-(alkylthio)alkyl radical is determined below. Pulse Radiolysis with TRESR Detection. Experiments with electron spin resonance (ESR) detection were performed with 0.5 µs pulses of high-energy electrons from the 2.8 MeV Van de Graaff accelerator at the Radiation Laboratory, University of Notre Dame. The radicals were detected with time-resolved electron spin resonance (TRESR).58-60 Solutions were flowed through the irradiated quartz cell at a rate of about 10 mL/min. The quartz cell was located at the center of the resonant, transverse-electric (TE)102 cavity for X-band microwaves. The cell was irradiated edge-on, incident to a 0.4 mm face of a rectangular quartz cell. A bubble trap61 was used in the flow system before the solution enters the quartz cell. The solution was also cooled, so that the irradiations took place at approximately 15 °C. The relative irradiation doses were determined by measuring the current of the charges collected on the outer surface of the sample cell. The concentration of the radicals was determined from the first half-life of the second-order decay of the SO3•- radicals that were formed in separate experiments in N2O-saturated solutions ([N2O] ≈ 25 mM). In these solutions, the SO3•- radicals are formed from the •OH oxidation of 5 mM sulfite solutions. The •OH yield was roughly doubled by the reaction of hydrated electrons with N2O (see reaction b). The total concentrations of primary radicals from water were approximately 18 µM under the operating conditions at the time of these experiments. The TRESR method has been described previously.44,62,63 The system was operated in both the kineticprofile and the boxcar modes. Steady-State γ-Radiolysis Combined with Gas Chromatography. The γ-radiolysis experiments were carried out in the field of a 60Co γ-source (Issledovatel, USRR) at the Institute of Nuclear Chemistry and Technology. The radiolytic conversion of R-MTA was less than 10% to avoid subsequent reactions of primary radicals with reaction products. The CO2 analysis of stable products was performed using the gas chromatographic headspace-technique on a Shimadzu 14B gas chromatograph with a thermal conductivity detector and a Porapak Q column using helium as a carrier gas. Computational Methods. The quantum chemical calculations were performed utilizing the Gaussian98W64,65 suite of programs. The input file structures of the radicals were prepared on a PC-computer using the ISIS Draw 2.2.466 graphic program. The visualization of the computation results and the molecular fitting67 were done using the gOpenMol 2.0 program.68 The file format conversions between modeling programs were done using the Babel 1.6 freeware program.69 The standard enthalpies of formation (∆Hf0) for the radicals in the gas phase were calculated using the Gaussian-3/DFT theory.70-72 This is a variation of Gaussian-3 (G3) theory in which the geometries and zero-point energies are obtained from B3LYP density functional theory73 [B3LYP/6-31G(d)]74 instead

J. Phys. Chem. A, Vol. 114, No. 1, 2010 107 of geometries from second-order perturbation theory [MP2(FU)/ 6-31G(d)] and zero-point energies from Hartree-Fock theory [HF/6-31G(d)]. In the procedure, the B3LYP/6-31G(d) optimization, frequency,75 and zero-point energy64 calculations were followed by a series of single point calculations on the QCISD(T,FC)/6-31G(d)76-78 and MP2(FC)/6-311+G(3df,2p).70,76-78 The zero-point energy was scaled by 0.96. Additionally, the higher-level correction was performed by taking into account the remaining deficiencies in the energy calculation.71 This variation, referred to as G3(MP2)//B3LYP, was assessed on 299 energies (enthalpies of formation, ionization potentials, electron and proton affinities) for the G2/97 test set.79 The G3(MP2)// B3LYP average absolute deviation from experiment for the 299 energies is 0.99 kcal/mol compared to 1.01 kcal/mol for G3 theory.71 The ESR hyperfine coupling constants for the B3LYP(IEFPCM)/6-31G(d) geometries of radicals were calculated with the B3LYP hybrid functional utilizing the EPR-III basis sets.80 For our purpose the EPR-III basis set of Barone, which was originally optimized for the computation of hyperfine coupling constants by DFT methods (particularly B3LYP), was augmented by the 6-311+G(2d)81 basis set for sulfur. This combination of basis sets from here on is referred to as EPR-III$. The theoretical estimates of the location of the dominant absorption bands (λmax), an approximate description of the intensities (electronic oscillator strength f), and the electronic composition of the vertical transitions were obtained from timedependent density functional response theory (TD-DFRT).82 The TD-DFRT-B3LYP/6-311+G(d,p)77,78,83 calculations were done at the gas-phase geometries obtained on the B3LYP/6-31G(d) level. However, this novel method has not been widely tested. It appears that it gives results convergent with the more accurate CASSCF/CASPT2 calculations, which are far more computerresource expensive.78,84 To account for the effect of the solvent on the geometry of the radicals, the gas-phase structures were reoptimized in the integral equation formalism model (IEFPCM).85-88 The free energy of hydration (∆Gaq) was calculated from the respective thermodynamic cycle. As the TD-DFRT method within Gaussian’98 does not collaborate with any implemented solvation models, we, therefore, estimated the expected solvent shift (∆λaq) from the shift of the respective root from the TD-DFRT calculation by applying the CIS procedure.89 For that purpose, we performed the CIS calculation for both gas-phase and IEFPCM optimized structures of radical ground states [i.e., CIS/6-31G(d)//B3LYP/6-31G(d) and CIS(IEFPCM)/6-31G(d)// B3LYP(IEFPCM)/6-31G(d)]. Results and Discussion Pulse Radiolysis with UV-vis Detection: Concentration Effects. The reaction of •OH radicals with R-(methylthio)acetamide (R-MTA) was investigated in acidic N2O-saturated aqueous solutions of R-MTA over the concentration range of 2 × 10-4 to 10-2 M. Depending on the concentration of the solute, pulse irradiation leads to different transient optical spectra. The spectrum at pH 1 (Figure 1) recorded 22.5 µs after pulse irradiation of an N2O-saturated solution containing 2 × 10-4 M R-MTA exhibits a spectrum with two distinct absorption features (ε ) molar absorption coefficient). These features are characterized by the red-edge of a band at or beyond 260 nm having Gε260 ) 1.31 cm2 J-1 and by a maximum at 340 nm (Gε340 ) 1.26 cm2 J-1). The spectrum at 1.2 µs was almost identical, but it had an additional weak shoulder around 480-490 nm (Gε480-490) ∼

108

J. Phys. Chem. A, Vol. 114, No. 1, 2010

Wisniowski et al.

Figure 1. Transient absorption spectrum (A) observed 22.5 µs after pulse irradiation of a 2 × 10-4 M solution of R-MTA saturated with N2O at pH ) 1. (B) Corrected for the contribution from acetamide radicals (see text). Inset: Time profiles of the transient at (O) λ ) 265 nm, (b) λmax ) 340 nm, and (2) λmax ) 480 nm. Assuming G(•OH) ) G(3), ε340 ) 4480 M-1 cm-1 for 3.

SCHEME 2: •OH-Induced Oxidation of r-MTA in Acid Solution

Figure 2. (a) Time profiles of the transient at absorption features located at (O) λ ) 270 nm, (b) λmax ) 340 nm, and (2) λmax ) 480 nm after pulse irradiation of a 2 × 10-3 M solution of R-MTA saturated with N2O at pH ) 0. Spectral resolution of the transient absorption spectra observed (b) 1.3 µs and (c) 22.5 µs after pulse irradiation of a 2 × 10-3 M solution of R-MTA saturated with N2O at pH ) 0.

0.2 cm2 J-1). The absorption traces at 260 and 340 nm were stable over the time domain of tens of microseconds (inset in Figure 1). Both absorption features were assigned as belonging to the same species (an R-(alkylthio)alkyl-type radical, CH3sSs•CHsC(dO)sNH2 (3), see Scheme 2) since they had the same kinetic behavior. For additional support of this assignment, see Table 1 below for the calculation of λmax of this radical, and see the section “Reaction Mechanism in Acidic Solution” for a discussion of the stability of this radical, further justifying this spectral assignment. For later use in spectral resolutions, a reference spectrum for transient 3 can be extracted from the spectrum in Figure 1. The shoulder at 480 nm has mostly disappeared at 22.5 µs. It was found that the acetamide radical (•CH2(CdO)NH2, Ac•) is formed by SH2 reactions with hydrogen atoms with a G ) 0.26 µmol J-1 at [R-MTA] ) 2 mM.44 For [R-MTA] ) 0.2 mM would have G ) 0.24 µmol J-1 using Schuler’s formula.49 Taking this to be G(Ac•) for the 22.5 µs spectrum in Figure 1 and the acetamide radical’s spectral parameters from earlier work,90 the spectrum for transient 3 is estimated as shown in Figure 1. The optical absorption spectrum obtained 1.3 µs after pulse irradiation of N2O-saturated aqueous solutions, pH 0, containing

2 × 10-3 M R-MTA shows three distinct absorption features with λ ) 265 nm (Gε265 ) 1.25 cm2 J-1), λmax ) 340 nm (Gε340 ) 1.04 cm2 J-1), and λmax ) 480 nm (Gε480 ) 0.93 cm2 J-1) (Figure 2b, data (/)). Since the intensity of the 480 nm band sharply increases upon increasing the initial R-MTA concentrations, one could expect that the spectrum might consist of the contribution from the intermolecularly S∴S-bonded dimeric radical cation (6) (see Scheme 2). This spectrum can be resolved into contributions from the following components: the two R-(alkylthio)alkyl radicals 3 and 4 (see Schemes 2 below) with the respective G(3) ) 0.14 ( 0.01 µmol J-1 and G(4) ) 0.03 ( 0.01 µmol J-1, the intermolecular sulfur-sulfur three-electron bonded dimeric radical cations (6) with G(6) ) 0.10 ( 0.01 µmol J-1, and acetamide radicals with G(Ac•) ) 0.36 ( 0.02 µmol J-1. The sum of G(3), G(4), and G(6) is 0.27 µmol J-1, which is in very good agreement with the yield of the precursor G(•OH) ) 0.29 µmol J-1. The yield of Ac• is in good agreement with the G value of H atoms (0.35 µmol J-1). Curve A of Figure 3 displays the optical absorption spectrum 270 ns after pulse irradiation of an N2O-saturated solution containing 10-2 M R-MTA. The spectrum recorded was dominated by the 480 nm absorption band (Gε480 ) 1.5 cm2 J-1) and consisted additionally of a broad, flat absorption band in the range 260-400 with Gε in this spectral region of 0.5 cm2 J-1.

Radical Formation in Oxidation of Thioether The G values of R-(alkylthio)alkyl radical 3 with G(3) ) 0.031 ( 0.003 µmol J-1 and the intermolecular sulfur-sulfur three-electron bonded dimeric radical cations with G(6) ) 0.172 ( 0.001 µmol J-1 were obtained from a spectral resolution of the 270 ns absorption spectrum (curve A of Figure 3, analysis not shown). This observation is consistent with the hypothesis that the intermolecularly S∴S-bonded dimeric radical cations were formed in competition with the long-lived species 3 (see Scheme 2). Pulse Radiolysis with UV-vis Detection: Time Evolution of the Spectra. Pulse radiolysis of solutions containing 2 × 10-3 and 10-2 M R-MTA resulted in the appearance of the absorption spectra depending on the time delays (Figures 2b,c, and 3). The transient absorption spectra obtained 22.5 µs after pulse irradiation of N2O-saturated aqueous solution, pH 0, containing 2 × 10-3 and 10-2 M R-MTA are shown as data (/) in Figure 2c and as curve B in Figure 3, respectively. For the 2 × 10-3 M concentration of R-MTA at 22.5 µs (Figure 2c), the intermolecularly S∴S-bonded dimeric radical cation, absorbing at 480 nm, was almost absent, and the residual absorption was similar to that observed at low concentrations of R-MTA (compare to Figure 1). The experimental optical spectrum recorded 22.5 µs after the pulse was best resolved into contributions from the two R-(alkylthio)alkyl radicals 3 and 4 with the respective G(3) ) 0.14 ( 0.01 µmol J-1 and G(4) ) 0.10 ( 0.01 µmol J-1. Other contributions to the spectral resolution were from the intermolecular sulfur-sulfur threeelectron bonded dimeric radical cations with G(6) ) 0.028 ( 0.002 µmol J-1, and acetamide radicals with G(Ac•) ) 0.23 ( 0.02 µmol J-1. A similar trend in spectral evolution was observed in solutions of 10-2 M R-MTA. The spectrum obtained 22.5 µs after the pulse exhibited three absorption maxima at λmax ) 270, 340, and 480 nm; see Figure 3 curve B. In this time domain, the absorption band with a maximum at 480 nm decays while the kinetic traces at 270 and 340 nm remain constant; see inset of Figure 3. Therefore, the 270 and 340 nm bands, which were masked by the strong absorption band of the intermolecularly S∴S-bonded dimeric radical cation (compare to Figure 3 curve A), were seen more clearly at 22.5 µs (Figure 3 curve B). The G values of R-(alkylthio)alkyl radical 3 with G(3) ) 0.062 ( 0.003 µmol J-1 and the intermolecular sulfur-sulfur three-electron bonded dimeric radical cations 6 with G(6) )

Figure 3. Transient absorption spectra observed (A) 270 ns and (B) 22.5 µs after pulse irradiation of a 10-2 M solution of R-MTA saturated with N2O at pH ) 0. Inset: Time profiles of the transient at absorption maxima located at (O) λmax ) 270 nm, (b) λmax ) 340 nm, and (2) λmax ) 480 nm.

J. Phys. Chem. A, Vol. 114, No. 1, 2010 109 0.065 ( 0.001 µmol J-1 were obtained from a spectral resolution of the 22.5 µs absorption spectrum (curve B of Figure 3, analysis not shown). Comparison of these G values with the corresponding G values measured at 2 × 10-3 M concentration again confirms that the intermolecularly SS bonded radical cations 6 were formed in competition with species 3. Reaction Mechanism in Acidic Solution. There is one striking difference that distinguishes the •OH-induced oxidation of R-MTA in acidic solutions from the corresponding reactions of other thioethers that have previously been studied. The key feature is the appearance of a long-lived 340 nm absorbing transient species. The existence of a 340 nm species has been seen after radiolysis of other thioethers, and these transient species have been assigned to the hydroxysulfuranyl radical. However, the hydroxysulfuranyl radical is quite short-lived in these other thioethers, especially at low pH. The questions then arise, (i) what is this new 340 nm absorbing species, following the radiolysis of R-MTA, which is stable for tens of microseconds, and (ii) why should it be appearing, specifically in R-MTA? Hydroxyl radicals, •OH, generally react in two ways with thioethers, and these steps were to be expected for R-MTA also. The main initial step in the reaction of hydroxyl radicals, with R-MTA (1, Scheme 2) in acidic solutions is expected to be an addition to the sulfur leading to the hydroxysulfuranyl-type radical (2) (Scheme 2, reaction 2.1). This reaction is well-known from other organic sulfides.1-4 Based on previous findings, this reaction generally accounts for 80-100% of the •OH radicals available. The remaining fraction (ca. 0-20%) of the •OH radicals is expected to react with R-MTA via the creation of various R-(alkylthio)alkyl radicals (3 and 4) through hydrogen abstractions from the carbon atoms adjacent to the primary sulfur moiety (Scheme 2, reactions 2.2 and 2.3). Under the experimental conditions with pH e 1, decomposition of the hydroxysulfuranyl radical occurs via external protoncatalyzed elimination (Scheme 2, reaction 2.4) of the hydroxide ion giving a water molecule and the monomeric radical cation (5). The rate constant for that reaction is kS ) 1.9 × 1010 M-1 s-1 (vide infra), which makes the half-life of the hydroxysulfuranyl radical at pH ) 1 to be t1/2 ≈ 0.5 ns. The monomeric radical cation (5) can associate with a second nonoxidized molecule of R-MTA to give the dimeric three-electron-bonded (S∴S) radical cation (6) (Scheme 2, reaction 2.5). This reaction is dependent on the primary concentration of R-MTA. At low concentrations of R-MTA (2 × 10-4 M), the population of the dimeric three-electron-bonded (S∴S) radical cations (6) is very low (G(6) ) 0.025 µmol J-1); see inset of Figure 1. Deprotonation of the monomeric radical cation (5) leading to the R-(alkylthio)alkyl radicals (3 and 4) is in competition with the creation of dimeric three-electron-bonded (S∴S) radical cations (6). The radical cation (6) was clearly visible at higher concentrations of R-MTA and its yield increased with concentration (G(6) ) 0.10 µmol J-1 and 0.172 µmol J-1 for 2 × 10-3 M and 10-2 M, respectively). Simultaneously the yield of R-(alkylthio)alkyl radicals (3 and 4) dropped from 0.17 µmol J-1 to 0.031 µmol J-1 for 2 × 10-3 M and 10-2 M R-MTA, respectively. In considering the relative stability of the two R-(alkylthio)alkyl radicals 3 and 4, the former has the possibility for extra stabilization. In the R-(alkylthio)alkyl radical (3), the unpaired electron on the carbon moiety is located between the electrondonating thioether group and the electron-accepting amide group. The observed long half-life (see below) of the radical should be the effect of stabilization from the capto-dative effect.91 Thus the candidate for the long-lived 340 nm absorbing

110

J. Phys. Chem. A, Vol. 114, No. 1, 2010

Wisniowski et al.

CHART 1: Capto-Dative Effect Observed in the r-(Alkylthio)alkyl Radical in the •OH-Induced Oxidation of r-(Methylthio)acetamide in Acid Solution

species is precisely this species (see Chart 1). This extra stabilization of radical 3 relative to radical 4 was confirmed by the respective yields at 2 × 10-3 M R-MTA where G(3) was 0.14 µmol J-1 and G(4) was 0.03 µmol J-1. For R-MTA at pH 1 and 2 × 10-4 M, the measured G(6) ) 0.025 µmol J-1 is significantly smaller than the G value of the intermolecularly three electron bonded dimeric sulfur radical cation of N-acetylmethionine amide G ) 0.13 µmol J-1 that is formed under identical conditions.56 Since these two intermolecularly three electron bonded dimeric sulfur radical cations are formed in competition with the respective R-(alkylthio)alkyl radicals in R-MTA and N-acetylmethionine amide, the relative low yield of 6 in R-MTA speaks to the special stabilization of 3 relative to typical R-(alkylthio)alkyl radicals such as those formed in N-acetylmethionine amide where the carbon centered radicals (AcNH-CH(CH2•CHSCH3)-C(dO)-NH2 and AcNHs CH(CH2CH2S•CH2)sC(dO)sNH2) are stabilized by the presence of sulfur alone. The quantum chemical calculations (Table 1) are in agreement with the pulse radiolysis experiments. The stabilization energy of the R-(alkylthio)alkyl radical (3) is larger than those of the H3Cs•CHsCH3, H3CsSs•CH2, and H3CsSs•CHsCH3 radicals. The small shift in the calculated absorption band maximum λmax ) 318 nm compared to the experimentally observed value of λmax ) 340 nm can be reasonably explained by the fact that the calculations were performed for vertical transitions in the gas phase whereas, in a polar solvent like water, the radical (3) can undergo further stabilization. If this stabilization is larger in its excited state, as opposed to its ground state, then the final adiabatic transition energy will be smaller. Such preferential stabilizations in the excited states of species are characteristic of their, usually, extended excited-state charge distributions. Pulse Radiolysis with UV-vis Detection: pH Effect. At neutral pH and at a low concentration of R-MTA (2 × 10-4 M), the transient absorption spectra observed after irradiation shows the complete absence of the absorption band characteristic of the intermolecularly three-electron-bonded (S∴S) dimeric radical cation at 480 nm, see Figure 4b,c, data (/). The decay of the 340 nm band shows biphasic kinetics, Figure 4a. Rapid decay occurs with a lifetime τ ∼ 2.4 µs followed by a much slower decay in the millisecond time domain. Contrary to the behavior at pH 3.4, the fast decay of 340 nm is not paralleled by a build-up of the 480 nm absorption, but this fast decay is accompanied by a slight increase in absorption in the 260-280 nm region. The calculated τ value (2.4 µs) is in excellent agreement with the τ value calculated as the inverse of the first-order rate constant kd (4.1 × 105 s-1, vide infra). The experimental optical spectrum in Figure 4b, recorded 1.3 µs after the pulse, was best resolved into contributions from the following components: hydroxysulfuranyl radical with G(2) ) 0.46 ( 0.01 µmol J-1 and the R-(alkylthio)alkyl radical with G(3) ) 0.039 ( 0.004 µmol J-1. The spectrum recorded 22.5 µs after the pulse was corrected before the spectral resolution (Figure 4c, corrected data) for the contribution of acetamide radicals (Ac•) with G(Ac•) ) G(H•)

Figure 4. (a) Time profiles of the transient at absorption maxima located at (O) λ ) 270 nm and (b) λmax ) 340 nm after pulse irradiation of a 2 × 10-4 M solution of R-MTA saturated with N2O at pH ) 6.8. Spectral resolution of the transient absorption spectra observed (b) 1.3 µs and (c) 22.5 µs after pulse irradiation of a 2 × 10-4 M solution of R-MTA saturated with N2O at pH ) 6.8. The experimental spectrum in Figure 4c was corrected for the contribution of G(Ac•) ) 0.056 µmol J-1.

) 0.056 µmol J-1. The corrected spectrum can be resolved into contributions from the following radicals: G(3) ) 0.13 ( 0.01 µmol J-1, G(4 and 8) ) 0.26 ( 0.01 µmol J-1, and G(6) ) 0.007 ( 0.001 µmol J-1. It can be seen in Figure 5a that the features of the spectral evolution in neutral solution observed after irradiation at high concentration (10-2 M R-MTA) are similar to those observed at low concentration of R-MTA (vide supra). Surprisingly, even at high concentration of R-MTA (10-2 M, Figure 5a), the presence of an intermolecularly three-electron-bonded (S∴S) dimeric radical cation in the transient spectra is very small at neutral pH. The spectrum recorded 22.5 µs after the pulse (Figure 5a) was corrected before the spectral resolution for the contribution of acetamide radicals (Ac•) with G(Ac•) ) G(H•) ) 0.056 µmol J-1. The corrected spectrum (5b) can be resolved into contributions from the following radicals: G(3) ) 0.062 ( 0.003 µmol J-1, G(4 and 8) ) 0.34 ( 0.01 µmol J-1, and G(6) ) 0.004 ( 0.001 µmol J-1. Pulse Radiolysis with UV-vis Detection: Reaction Rate Constants of the Hydroxysulfuranyl Radical. Since it is relatively easy to separate out the hydroxysulfuranyl radical at 340 nm in the •OH-induced oxidation of R-MTA, it is important to use this opportunity to measure its rate constants with protons and unoxidized R-MTA molecules. The rate constant for the protoncatalyzed oxidation of the hydroxysulfuranyl radical is useful

Radical Formation in Oxidation of Thioether

J. Phys. Chem. A, Vol. 114, No. 1, 2010 111

Figure 6. (a) Pseudo-first-order decay rates of 340 nm traces as a function of proton concentration, keeping the concentration of R-MTA constant at 2 mM. (b) Pseudo-first-order decay rates of 340 nm kinetic traces as a function of R-MTA, keeping the pH ) 7. S ) R-MTA.

Figure 5. (a) Transient absorption spectra and kinetic traces (inset) observed following electron pulse irradiation of a 10-2 M R-MTA aqueous solution saturated with N2O at pH ) 6.45. (b) Spectral resolution of the transient absorption spectra observed 22.5 µs after pulse irradiation of a 10-2 M solution of R-MTA saturated with N2O at pH ) 6.45. The experimental spectrum in Figure 5b was corrected for the contribution of G(Ac•) ) 0.056 µmol J-1.

for quantitative analysis of the reaction scheme (see below) and as an aid in the assignment of the radicals (see above). The reactions to consider are

The three rate constants can be determined by following the initial 340 nm decays as a function of the concentration of protons and as a function of the concentration of R-MTA. These resulting decays were single exponentials. The computed pseudofirst order rates, kobs (reciprocals of the measured lifetimes),

kobs ) kd + kH[H+] + kS[S]

(3)

are plotted in Figure 6a,b vs proton and S ) R-MTA concentrations, respectively. The resulting rate constants kd, kH, and kS are equal to 4.1 × 105 s-1, 1.9 × 1010 M-1 s-1, and 5.4 × 106 M-1 s-1, respectively. Pulse Radiolysis with TRESR Detection. The time-resolved electron spin resonance (TRESR) spectrum following the pulse radiolysis of an N2O-saturated aqueous solution of R-MTA at pH ) 6 is shown in Figure 7. The most important observation concerns the radical which is visible as a doublet with aH ) 14 G and g ) 2.0048. A similar radical, which was stable at room temperature, was observed with standard ESR following

Figure 7. TRESR spectrum of the transient radical observed 1 µs after the pulse irradiation of 2 × 10-3 M aqueous solution of R-MTA saturated with N2O at pH ) 6. Inset: TRESR kinetic trace of the radical.

γ-irradiation of a polycrystalline powder of R-MTA (not shown). This radical appeared as a doublet with the same hyperfine coupling as the radical observed in TRESR. A doublet, having such a hyperfine coupling, is consistent with the structure of a radical containing one unpaired electron centered on a carbon atom bonded to a single proton (see also Computational Results and Table 2). In the case of R-MTA, only the CH3sSs•CHsCONH2 radical is consistent with such a radical. The assignment of the spectrum to the CH3sSs•CHsCONH2 radical is also supported by the agreement of its g factor with those for other R-(alkylthio)alkyl radicals,92-94 i.e., g ) 2.0049 for •CH2SCH3.93,94 The TRESR kinetic trace (inset to Figure 7) shows that the CH3sSs•CHsCONH2 radical decayed slowly, probably over the millisecond time domain. It can be noted that the ESR signal from acetamide radicals arising from SH2 reactions of R-MTA with hydrogen atoms were at a minimum in this pH range.44 Steady-State γ-Radiolysis Combined with Gas Chromatography. The γ-radiolysis experiments were performed in the pH range 0-7. Depending on the pH of N2O-saturated solutions containing 10-2 M R-MTA, irradiation of them leads to the yield of carbon dioxide (CO2) shown in Figure 8. Reaction Mechanism in Neutral Solution. There are further questions that arise and concern kinetic and spectral features of transients forming in neutral solutions. Why is the yield of

112

J. Phys. Chem. A, Vol. 114, No. 1, 2010

Wisniowski et al. SCHEME 3: •OH-Induced Oxidation of r-MTA in Neutral Solution

Figure 8. Yields of CO2 vs pH profile (primary yield of •OH radical normalized) after the irradiation of the N2O-saturated solutions of R-MTA at concentration 10-2 M.

the intermolecularly S∴S-bonded dimeric radical cation at neutral pH very low, even at high concentration of R-MTA (10 mM), and what is the reaction pathway leading to the very unexpected formation of CO2? The first observation might be rationalized by the efficient direct conversion of the hydroxysulfuranyl radicals of R-MTA into its respective C-centered radicals, previously observed for S-ethylthioacetone.40,41 However, formation of CO2 from a molecule that does not contain a carboxylic group seems to involve an interesting mechanistic pathway. The •OH-induced oxidation of R-MTA is strongly dependent on pH and the concentration of the compound. In neutral solutions, the main initial step in the reaction of hydroxyl radicals with thioethers is still the creation of the hydroxysulfuranyl radical ((2) Scheme 3, reaction 3.1). However, at low concentrations of protons and low concentrations of R-MTA, the lifetime of the hydroxysulfuranyl radical is dependent only on its unimolecular decay reaction (Scheme 3, reaction 3.4) with kd ) 4.1 × 105 s-1 (vide supra). At a time delay of 1.3 µs following an electron pulse into an N2O-saturated aqueous solution of 0.2 mM R-MTA at pH ) 6.8, the transient spectrum exhibited only a single band with λmax ) 340 nm (Figure 4b, data (*)). The spectral resolution showed this was mainly due to the hydroxysulfuranyl radical 2 with a small contribution of radical 3, Figure 4b. Under such conditions (low concentrations of protons and low concentrations of R-MTA), the hydroxysulfuranyl radical can be stabilized by an intramolecular hydrogen bond. This bond can be between the hydrogen atom, from the >SOH radical site, and either the oxygen atom ((2a) Scheme 3) or the nitrogen atom ((2b) Scheme 3) of the amide group. Creation of such six-membered-ring structures is well-known in the case of H3CsSsCH2sC(dO)sOCH3.3 Furthermore the rate constant for decay of the 340 nm absorbing transient was 4 times less for R-MTA (kd ) 4 × 105 s-1, vide supra) than for an analogous transient from H3CsSsCH3.3 By analogy to the mechanism proposed for (alkylthio)ethanol derivatives,22 it is suggested that intramolecular hydrogen transfer is the key step in the dissociation of the R-MTA hydroxysulfuranyl radical which decomposes via the conformational structure 2c (Scheme 3). Elimination of water from 2c (Scheme 3, reaction 3.4) generates an imine radical ((7) Scheme 3) that decays via three reaction pathways. (i) β-scission leads to an R-(alkylthio)alkyl radical and isocyanine ((8) and (9), respectively, Scheme 3,

SCHEME 4: •OH-Induced Oxidation of r-MTA in Neutral Solution

reaction 3.5). This is followed by hydrolysis of 9 (Scheme 3, reaction 3.6) and the well-known Hofmann rearrangement, which gives carbon dioxide CO2 and ammonia NH3 as final products (Scheme 3, reactions 3.7 and 3.8). (ii) Intramolecular hydrogen transfer, from the δC carbon atom to the radical imine group via a six-membered ring ((7) Scheme 3, reaction 3.9) (Barton type reaction),95 leads to the R-(alkylthio)alkyl radical ((4) Scheme 3). Finally (iii) a 1,3-hydrogen shift (Scheme 4, reaction 4.2) leads to an R-(alkylthio)alkyl radical ((3) Scheme 4). However, for the processes leading to the formation of both

Radical Formation in Oxidation of Thioether radicals (3) and (4), it is difficult to definitely exclude participation of water molecules in the proton transfer or even the protonation/deprotonation mechanism with subsequent backelectron transfer. Independent support for point (iii) is found in the TRESR spectrum following the pulse radiolysis of an N2O-saturated aqueous solution of R-MTA at pH ) 6. This spectrum was assigned to the radical 3 (Figure 7), vide supra. At neutral pH, the key reaction seems to be intramolecular hydrogen transfer in the conformational structure of the hydroxysulfuranyl radical 2c (Scheme 3, reaction 3.4 or Scheme 4, reaction 4.1). The high efficiency of these reactions competes favorably with formation of the dimeric three-electron-bonded (S∴S) radical cations (6) (Scheme 4, reaction 4.3). The essential feature of the mechanism in Scheme 3 is the possibility of creating the six-membered ring structure. This possibility is not present in the, otherwise, closely related molecule, N-acetylmethionine amide. Under the same conditions of pH and concentration of substrate, the •OH-induced oxidation of Nacetylmethionine amide led to an efficient formation of the dimeric three-electron-bonded (S∴S) radical cations.56 The possibility of suggesting the creation of the imine radical (7) with all its subsequent reaction pathways is supported by the presence of carbon dioxide as a final product. The amount of carbon dioxide depends on the pH (Figure 8). The inflection point in this titration curve is at pH 1.9. From Schemes 2 and 3, this means that, at pH 1.9, the rate constant, kH, of the protoncatalyzed reaction for the dissociation of the hydroxysulfuranyl radical ((2 f 5) Scheme 2, reaction 2.4) and the rate constant, kir, of intramolecular rearrangement (Scheme 3, reaction 3.10) of the hydroxysulfuranyl radical (2) to structures 2a-c (Scheme 3) are equal. The rate constant, kir ) 2.4 × 108 s-1, was calculated from the relation kH[H+] ) kir, taking [H+] ) 1.25 × 10-2 M and kH ) 1.9 × 1010 M-1 s-1 (vide supra). The order of this value is similar to the previously calculated k′ir ) 5 × 107 s-1 in 2-(methylthio)ethanol and 2,2′′-dihydroxydiethyl sulfide.22 Source of R-(Alkylthio)alkyl Radicals. The absorption bands characterized by the absorption maxima located at λmax ) 280-285 nm are evidence for the presence of R-(alkylthio)alkyl radicals (Scheme 3, 4 and/or 8) following the •OH-induced oxidation of R-MTA in neutral solutions. (By the generic term R-(alkylthio)alkyl radicals, we are referring to such radicals, exclusive of radical 3). From Figure 1, radical (3) also absorbs at 280 nm. Using the spectral resolution shown in Figure 4c, the Gε280-285 (22.5 µs) ) 1.19 cm2 J-1 can be resolved into a G(3, 22.5 µs) ) 0.13 µmol J-1 and a G(R-(alkylthio)alkyl radicals) ) 0.26 µmol J-1. The molar absorption coefficient for R-(alkylthio)alkyl radicals was taken as ε280-285 ) 3000 M-1 cm-1 for analogous R-(alkylthio)alkyl radicals.22 If β-scission of the imine radical (7) were the only mechanism for its decay (Scheme 3, reaction 3.5), an equality would be expected between the yield of the R-(alkylthio)alkyl radical G(8) and the yield of carbon dioxide G(CO2). A brief comment can be made on the difference between this limiting maximum yield of R-(alkylthio)alkyl radicals G(4 and 8 22.5 µs) ) 0.34 µmol J-1 (taken from the solution containing 10 mM of R-MTA, Figure 5b) and G(CO2) ) 0.19 µmol J-1 (taken from the solution containing 10 mM of R-MTA, Figure 8). However, G(R-(alkylthio)alkyl radicals) being greater than G(CO2) is still consistent with Scheme 3, where another type of R-(alkylthio)alkyl radical (4) can be produced in a parallel decay of (7) via an intramolecular hydrogen transfer, from the δC carbon atom to the radical imine site (Scheme 3, reaction 3.9).

J. Phys. Chem. A, Vol. 114, No. 1, 2010 113

G(7 f 4) ) G(4&8, 22.5 µs) - G(CO2) ) 0.15 µmol J-1 (4) Thus, the difference in radiation chemical yields is accounted for by the formation of radical (4), which accounts for the excess G(R-(alkylthio)alkyl radicals, 4 and 8) over the G(CO2) in neutral solution containing R-MTA. Radical stabilization presumably plays an important role in this conversion. Source of the R-(Alkylthio)alkyl Radical Stabilized by the Captodative Effect. Another question of interest is whether the radical (3) is formed exclusively as a consequence of the •OH radical abstracting a hydrogen from the methylene moiety of R-MTA (Scheme 3, reaction 3.2) or whether radical (3) is also created in significant quantities by the 1,3-hydrogen shift in the imine radical (7) (Scheme 4, reaction 4.2). The answer to these questions can be given by comparing the transient absorption spectra of irradiated R-MTA in neutral aqueous solutions at time delays (td) of 1.3 and 22.5 µs in Figure 4b,c, respectively. If only the radicals 4 and 8 would be formed from 7 via reactions 3.9 and 3.5, respectively (Scheme 3), then

G(4&8, 22.5 µs) = G(7)

(5)

Since all of the hydroxysulfuranyl radicals (2) form imine radicals (7) (reaction 3.4) (see Scheme 3)

G(7) ) G(2, 1.3 µs) ) 0.46 µmol J-1

(6)

The value for G(2, 1.3 µs) was taken from the spectral resolution in Figure 4b. Combining eqs 5 and 6, G(4 and 8) gives

G(4&8, 22.5 µs) ) 0.46 µmol J-1

(7)

However, from the spectral resolution in Figure 4c, it was found that G(4 and 8, 22.5 µs) ) 0.26 µmol J-1. The shortfall in G(7), when only G(4 and 8, 22.5 µs) is taken into consideration, is significantly improved when the increase in G(3) between 1.3 and 22.5 µs (≈0.09 µmol J-1) is added into the yield of radicals formed from the decay of 7. This semiquantitative agreement with the expectations of Schemes 2 and 3 supports the notion that part of the yield of radical 3 comes from a 1,3-hydrogen shift in the imine radical 7 (Scheme 4, reaction 4.2). Radical 7 is thus one of the origins of radical 3 in neutral aqueous solutions of R-MTA. Source of the Dimeric Three-electron-bonded (S∴S) Radical Cation. A remaining question of interest is what is the source, albeit low, of the dimeric three-electron-bonded (S∴S) radical cation (6) formed in neutral aqueous solution at high concentrations of R-MTA. (See the 480 nm kinetic trace in the inset to Figure 5a.) From Scheme 4, the radical cation (6) is proposed to be formed from 2 via two pathways 4.3 and 4.4/4.5 in the competition with the imine radical (7) through reaction 4.1. These three decay pathways of 2 (two of them leading to 6) are independent pseudo-first-order or first-order processes. Because of their assumed independence, yields can be computed from the rate constants as branching ratios. Under the conditions where [S] ) 10-2 M and [H+] ) 3.55 × 10-7 M, the rates are kd ) 4.1 × 105 s-1, kS[S] ) 5.4 × 104 s-1, and kH[H+] ) 6.7 × 103 s-1, for the pathways 4.1, 4.3, and 4.4/4.5, respectively (vide supra), S denotes R-(methylthio)acetamide. The rate for pathway 4.3 is almost an order of magnitude faster than the rate for pathway 4.4. Moreover, reaction 4.4 cannot compete with reaction 4.6. Therefore, pathways 4.4 and 4.5 can be ignored as a source for radical 6 under the conditions stated. Pathway 4.3 is thus the source of 6. The predicted yield of 6 relative to 7 is, therefore, the ratio of the first two rates: kd/(kS[S]) ) 0.13. This can be tested with the

114

J. Phys. Chem. A, Vol. 114, No. 1, 2010

Wisniowski et al.

TABLE 1: Stabilization Enthalpies (∆Hs, kcal mol-1), Free Energies of Hydration (∆Gaq, kcal mol-1), Absorption Maxima (λmax, nm), Oscillator Strengths (f), and Solvatochromic Shift (∆λaq, nm) Calculated in This Work (See Text) R•

∆Hs

∆GaqIEFPCM

λmax

f

∆λaq

CH3 CH3s•CHsCH3 CH3sSs•CH2 CH3sSs•CHsCH3 • CH2sC(dO)sNH2 CH3sSs•CHsC(dO)sNH2

0.00 5.60 10.02 11.13 -0.57 19.00

1.69 -0.16 -1.28 -1.46 -15.27 -11.31

286 (π f π*) 287 (π f π*) 164 (n f π*; n,π f π*) 318 (π f π*)

0.08 0.06 0.16 0.12

-2.0 -1.5 -5.8 -2.4

•

TABLE 2: Proton Isotropic Fermi Contact Couplings (AH, G) Calculated in This Work (See Text) R•

ARH

CH3 CH3s•CHsCH3 CH3sSs•CH2 CH3sSs•CHsCH3 • CH2sC(dO)sNH2 CH3sSs•CHsC(dO)sNH2

-23 -22 -14; -18 -13 -21; -21 -18

•

AβH 8; 52, 18 11; 44; 11

AγH

6; 5, -1 -1; 3; 4 -3; -2 -1; 6; 4

yields from spectral resolutions. The yield of 6 G(1.3 µs) ) 0.07 µmol J-1 was taken from the spectral resolution (not shown) of the spectrum presented in Figure 5a. The yield of 7 is taken as the sum of G(3, 22.5 µs) and G(4 and 8, 22.5 µs), which amounts to G(7) ) 0.40 µmol J-1 from the spectral resolution in Figure 5b. The ratio G(6) to G(7) by this measure is 0.17. This agreement between the relative yields from the measured rate constants and the relative yields from the spectral resolutions is support for the proposed kinetic scheme and mechanism. Computational Results. The thermodynamic stability of CH3sSs•CHsC(dO)sNH2 and model radicals: •CH3, CH3s • CHsCH3, CH3sSs•CH2, CH3sSs•CHsCH3, •CH2sC(dO)s NH2 as well as their parent compounds was characterized by their stabilization enthalpies ∆Hs(R•). The ∆Hs(R•) is a comparative quantity related to the bond dissociation enthalpy for hydrogen abstraction, leading to the formation of the radical, as compared to the respective bond dissociation enthalpy for the methyl radical. Thus, ∆Hs(R•) can be calculated from91

∆Hs(R•) ) [∆Hf0(•CH3) - ∆Hf0(R•)] [∆Hf0(CH3-H) - ∆Hf0(R-H)] (8) where ∆Hf0(R•), ∆Hf0(R-H), ∆Hf0(•CH3), and ∆Hf0(CH3-H) are the enthalpies of formation of radical R•, its “parent compound”, the methyl radical, and methane, respectively. The G3(MP2)//B3LYP calculated stabilization enthalpies (∆Hs), free energies of hydration (∆Gaq), absorption maxima (λmax), oscillator strengths (f), and solvatochromic shift (∆λaq, nm) are presented in Table 1. ESR parameters calculated from B3LYP/EPR-III$// B3LYP(IEFPCM)/6-31G(d) are presented in Table 2. Conclusion The •OH oxidation of R-MTA in acidic solutions displays typical behavior for an organic thioether due to the efficient formation of the dimeric three-electron-bonded (S∴S) radical cation (6) and the R-(alkylthio)alkyl radical (3) that is stabilized by a capto-dative effect. The assignment of this latter radical structure (see Chart 1) was supported directly by a time-resolved ESR spectrum following pulse radiolysis. The observed hyperfine coupling constant (see Figure 7, 14 G) was the same as computed by DFT (see Table 2, 18 G), and the g factor is typical of other R-(alkylthio)alkyl radicals.92-94 Collaborative evidence was obtained through detailed analysis of time-resolved optical

spectra from pulse radiolysis experiments at different pH and concentrations of R-MTA. From these results, a free-radical mechanism is proposed in Scheme 2 for acidic solutions. In neutral solutions, the •OH oxidation of R-MTA displays different behavior where a six-membered hydroxysulfuranyl radical (2a)-(2c) is formed via an intramolecular hydrogen bond. Radical (2c) eliminates a water molecule and forms an imine radical (7) which, in turn, decays into three independent channels. The first decay channel begins with a β-scission of (7). The β-scission is followed by hydrolysis and a subsequent Hofmann rearrangement. One of the end products of this first decay channel of (7) is CO2, which was detected. The second decay channel of (7) involves an intramolecular hydrogen transfer from the δC carbon atom to the radical imine site of (7) producing the R-(alkylthio)alkyl radical (4). In the third decay channel of (7) there is a 1,3-hydrogen shift in the imine radical (7), which forms the radical (3). (See Scheme 3, reactions 3.5-3.9 and Scheme 4, reaction 4.2.) Direct H-abstractions by •OH radicals (Scheme 3, reactions 3.2 and 3.3) complete the reaction mechanism for the •OHinduced oxidation of R-MTA in neutral aqueous solutions. Acknowledgment. This article is dedicated to the memory of Professor David Armstrong, who contributed to sulfur radical chemistry. The work described herein was supported by Office of Basic Energy Sciences of the U.S. Department of Energy and the Polish Ministry of Scientific Research Grant (3 T09A 066 26). G.L.H. did part of the work as a Fulbright Scholar at AMU. K.B. thanks the Notre Dame Radiation Laboratory for financial support during part of this work and COST Action CM0603 on the Free Radicals in Chemical Biology (CHEMBIO RADICAL). The computation was performed by employing the computing resources of the Interdisciplinary Centre for Mathematical and Computation Modelling Warsaw University, Poland (G28-21). This paper is the Document No. NDRL-4735 from the Notre Dame Radiation Laboratory. References and Notes (1) Asmus, K.-D. Acc. Chem. Res. 1979, 12, 436–442. (2) Asmus, K.-D.; Bahnemann, D.; Fischer, C.-H.; Veltwisch, D. J. Am. Chem. Soc. 1979, 101, 5322–5329. (3) Bobrowski, K.; Scho¨neich, C. J. Chem. Soc., Chem. Commun. 1993, 795–797. (4) Scho¨neich, C.; Aced, A.; Asmus, K.-D. J. Am. Chem. Soc. 1993, 115, 11376–11383. (5) Bonifacic, M.; Mo¨ckel, H.; Bahnemann, D.; Asmus, K.-D. J. Chem. Soc., Perkin Trans. 2 1975, 675–685. (6) Bahnemann, D.; Asmus, K.-D. J. Chem. Soc., Chem. Commun. 1975, 238. (7) Asmus, K.-D.; Bahnemann, D.; Bonifacic, M.; Gillis, H. A. Faraday Discuss. Chem. Soc. 1978, 63, 213–225. (8) Asmus, K.-D. Sulfur-centered three-electron bonded radical species. In Sulfur-Centered ReactiVe Intermediates in Chemistry and Biology; Chatgilialoglu, C., Asmus, K.-D., Eds.; Plenum Press: New York, 1990; Vol. 197; pp 155-172. (9) Asmus, K.-D.; Bonifacic, M. Sulfur-centered reactive intermediates as studied by radiation chemical and complementary techniques. In

Radical Formation in Oxidation of Thioether S-Centered Radicals; Alfassi, Z. B., Ed.; John Wiley & Sons Ltd.: Chichester, U.K., 1999; pp 141-191. (10) Bobrowski, K.; Hug, G. L.; Pogocki, D.; Marciniak, B.; Scho¨neich, C. J. Phys. Chem. B 2007, 111, 9608–9620. (11) Bobrowski, K.; Hug, G. L.; Pogocki, D.; Marciniak, B.; Scho¨neich, C. J. Am. Chem. Soc. 2007, 129, 9236–9245. (12) Mishra, B.; Priyadarsini, K. I.; Mohan, H. Res. Chem. Intermed. 2005, 31, 625–632. (13) Mohan, H.; Mittal, J. P. Radiat. Phys. Chem. 2005, 74, 220–226. (14) Scho¨neich, C.; Pogocki, D.; Hug, G. L.; Bobrowski, K. J. Am. Chem. Soc. 2003, 125, 13700–13713. (15) Varmenot, N.; Remita, S.; Abedinzadeh, Z.; Wisniowski, P.; Strzelczak, G.; Bobrowski, K. J. Phys. Chem. A 2001, 105, 6867–6875. (16) Asmus, K.-D. Heteroatom-centered free radicals. Some selected contributions by radiation chemistry. In Radiation Chemistry:Present Status and Future Prospects; Jonah, C., Rao, B. S. M., Eds.; Elsevier: Amsterdam, 2001; pp 341-393. (17) Scho¨neich, C.; Pogocki, D.; Wisniowski, P.; Hug, G. L.; Bobrowski, K. J. Am. Chem. Soc. 2000, 122, 10224–10225. (18) Bobrowski, K.; Hug, G. L.; Marciniak, B.; Scho¨neich, C.; Wisniowski, P. Res. Chem. Intermed. 1999, 25, 285–297. (19) Gawandi, V. B.; Mohan, H.; Mittal, J. P. J. Chem. Soc., Perkin Trans. 2 1999, 1425–1432. (20) Bobrowski, K.; Pogocki, D.; Scho¨neich, C. J. Phys. Chem. A 1998, 102, 10512–10521. (21) Bobrowski, K.; Hug, G. L.; Marciniak, B.; Miller, B. L.; Scho¨neich, C. J. Am. Chem. Soc. 1997, 119, 8000–8011. (22) Scho¨neich, C.; Bobrowski, K. J. Am. Chem. Soc. 1993, 115, 6538– 6547. (23) Bobrowski, K.; Pogocki, D.; Scho¨neich, C. J. Phys. Chem. 1993, 97, 13677–13684. (24) Steffen, L. K.; Glass, R. S.; Sabahi, M.; Wilson, G. S.; Scho¨neich, C.; Mahling, S.; Asmus, K.-D. J. Am. Chem. Soc. 1991, 113, 2141–2145. (25) Asmus, K.-D.; Go¨bl, M.; Hiller, K.-O.; Mahling, S.; Mo¨nig, J. J. Chem. Soc., Perkin Trans. 2 1985, 641–646. (26) Glass, R. S.; Hojjatie, M.; Wilson, G. S.; Mahling, S.; Go¨bl, M.; Asmus, K.-D. J. Am. Chem. Soc. 1984, 106, 5382–5383. (27) Hiller, K.-O.; Masloch, B.; Go¨bl, M.; Asmus, K.-D. J. Am. Chem. Soc. 1981, 103, 2734–2743. (28) Glass, R. S.; Hug, G. L.; Schoneich, C.; Wilson, G. S.; Kuznetsowa, L.; Lee, T.-M.; Ammam, M.; Lorance, E.; Nauser, T.; Nichol, G. S.; Yamamoto, T. J. Am. Chem. Soc. 2009, 131, 13791-13805. (29) Gawandi, V. B.; Mohan, H.; Mittal, J. P. J. Phys. Chem. A 2000, 104, 11877–11884. (30) Hug, G. L.; Bobrowski, K.; Pogocki, D.; Ho¨rner, G.; Marciniak, B. ChemPhysChem 2007, 8, 2202–2210. (31) Asmus, K.-D. Nukleonika 2000, 45, 3–10. (32) Glass, R. S. Top. Curr. Chem. 1999, 205, 1–87. (33) Glass, R. S. Neighbouring group participation: General principles and application to sulfur-centered reactive species. In Sulfur-centered reactiVe intermediates in chemistry and biology; Chatgilialoglu, C., Asmus, K.-D., Eds.; Plenum Press: New York, 1990; Vol. 97, pp 212-226. (34) Free Radicals in Aging; Yu, B. P., Ed.; CRC Press: Boca Raton, FL, 1993. (35) Dean, R. T.; Fu, S.; Stocker, R.; Davies, M. J. Biochem. J. 1997, 324, 1–18. (36) Oxygen Radicals and the Disease Process; Craig, E. T., Kalyanaraman, B., Eds.; Harwood Academic Publishers: Amsterdam, 1997. (37) Scho¨neich, C. Exp. Gerontology 1999, 34, 19–34. (38) Wardman, P. Thiyl Radicals in Biology: Their Role as a ‘Molecular Switch’ central to Cellular Oxidative Stress. In S-centered Radicals; Alfassi, Z. B., Ed.; Wiley: Chichester, U.K., 1999; pp 289-309. (39) Rauk, A.; Armstrong, D. A.; Fairlie, D. P. J. Am. Chem. Soc. 2000, 122, 9761–9767. (40) Varmenot, N.; Berges, J.; Abedinzadeh, Z.; Scemama, A.; Strzelczak, G.; Bobrowski, K. J. Phys. Chem. A 2004, 108, 6331–6346. (41) Berges, J.; Varmenot, N.; Scemama, A.; Abedinzadeh, Z.; Bobrowski, K. J. Phys. Chem. A 2008, 112, 7015–7026. (42) Bordwell, F. G.; Zhang, X.-M.; Alnajjarr, M. S. J. Am. Chem. Soc. 1992, 114, 7623–7629. (43) Sustmann, R.; Korth, H.-G. AdV. Phys. Org. Chem. 1990, 26, 131– 178. (44) Wisniowski, P.; Bobrowski, K.; Carmichael, I.; Hug, G. L. J. Am. Chem. Soc. 2004, 126, 14468–14474. (45) Wisniowski, P.; Bobrowski, K.; Filipiak, P.; Carmichael, I.; Hug, G. L. Res. Chem. Intermed. 2005, 31, 633–641. (46) von Sonntag, C. The Chemical Basis of Radiation Biology; Taylor and Francis: New York, 1987. (47) Janata, E.; Schuler, R. H. J. Phys. Chem. 1982, 86, 2078–2084. (48) Buxton, G. V.; Greenstock, C. L.; Helman, W. P.; Ross, A. B. J. Phys. Chem. Ref. Data 1988, 17, 513–886.

J. Phys. Chem. A, Vol. 114, No. 1, 2010 115 (49) Schuler, R. H.; Hartzell, A. L.; Behar, B. J. Phys. Chem. 1981, 85, 192–199. (50) LaVerne, J. A.; Pimblott, S. M. J. Phys. Chem. 1991, 95, 3196– 3206. (51) LaVerne, J. A. Radiat. Res. 2000, 153, 196–200. (52) Karolczak, S.; Hodyr, K.; Lubis, R.; Kroh, J. J. Radioanal. Nucl. Chem. 1986, 101, 177–188. (53) Karolczak, S.; Hodyr, K.; Polowinski, M. Radiat. Phys. Chem. 1992, 39, 1–5. (54) Wisniowski, P. B.; Bobrowski, K.; Mirkowski, J. Pulse Radiolysis Laboratory Home Page; Institute of Nuclear Chemistry and Technology: Warsaw, Poland, 2006. (55) Wisniowski, P.; Hug, G. L. Decom2006 2006. (56) Wisniowski, P. B. Neighboring group participation in radiation induced radical reactions of thioethers. Ph.D. Thesis, Warsaw, 2001. (57) Hiller, K.-O.; Asmus, K.-D. Int. J. Radiat. Biol. Relat. Stud. Phys., Chem. Med 1981, 40, 597–604. (58) Verma, N.; Fessenden, R. W. J. Chem. Phys. 1976, 65, 2139–2155. (59) Fessenden, R. W.; Hornak, J. P.; Venkataraman, B. J. Chem. Phys. 1981, 74, 3694–3704. (60) Madden, K. P.; McManus, H. J. D.; Fessenden, R. W. ReV. Sci. Instrum. 1994, 65, 49–57. (61) Duncanson, I. B. Fusion 1993, 40, 26. (62) Hug, G. L.; Fessenden, R. W. J. Phys. Chem. A 2000, 104, 7021– 7029. (63) Wisniowski, P. B.; Carmichael, I.; Fessenden, R. W.; Hug, G. L. J. Phys. Chem. A 2002, 106, 4573–4580. (64) Exploring Chemistry with Electronic Structure Methods, 2nd ed.; Foresman, J., Frisch, A., Eds.; Gaussian Inc.: Pittsburgh, PA, 1995. (65) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A., Jr.; Stratmann, R. E.; Burant, J. C.; Dapprich, S.; Millam, J. M.; Daniels, A. D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.; Barone, V.; Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo, C.; Clifford, S.; Ochterski, J.; Petersson, G. A.; Ayala, P. Y.; Cui, Q.; Morokuma, K.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Cioslowski, J.; Ortiz, J. V.; Baboul, A. G.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komoromi, I.; Gomperts, R.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Gonzalez, C.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Andres, J. L.; Head-Gordon, M.; Replogle, E. S.; Pople, J. A. Gaussian 98, A.7; Gaussian Inc: Pittsburgh, PA, 1998. (66) ISIS Draw, 2.2.4; MDL Information Systems Inc.: Hayward, CA, 2001. (67) Leach, A. R. Molecular Modelling: Principles and Applications; Longman: Harlow, England, 1996. (68) Laaksonen, L. gOpenMol, 2.10; 2001; http://www.csc.fi/∼laaksone/ gopenmol. (69) Walters, P.; Stahl, M. Babel, 1.6; Dolata Research Group, University of Arizona, Tucson, AZ 85721, 1999. (70) Curtiss, L. A.; Raghavachari, K.; Redfern, P. C.; Rassolov, V.; Pople, J. A. J. Chem. Phys. 1998, 109, 7764–7776. (71) Baboul, A. G.; Curtiss, L. A.; Redfern, P. C.; Raghavachari, K. J. Chem. Phys. 1999, 110, 7650–7657. (72) Curtiss, L. A.; Redfern, P. C.; Raghavachari, K.; Rassolov, V.; Pople, J. A. J. Chem. Phys. 1999, 110, 4703–4709. (73) Hohenberg, P. C.; Kohn, W.; Sham, L. J. The beginning and some thoughts on the future. In Density Functional Theory of Many-Fermion Systems; Trickey, S. B., Ed.; Academic Press: San Diego, CA, 1990; pp 7-26. (74) Becke, A. D. J. Chem. Phys. 1993, 98, 5648–5652. (75) Ochterski, J. W. Gaussian: Wallingford, CT, 1999; pp 1-10, www.gaussian.com. (76) Szabo, A.; Ostlund, N. S. Modern Quantum Chemistry. Introduction to AdVanced Electronic Structure Theory; McGraw-Hill: New York, 1989. (77) Atkins, P. W.; Friedman, R. S. Molecular Quantum Mechanics; Oxford University Press: Oxford, New York, Tokyo, 1997. (78) Jensen, F. Introduction to Computational Chemistry; Wiley: Chichester, U.K., 1999. (79) Curtiss, L. A.; Raghavachari, K.; Redfern, P. C.; Pople, J. A. J. Chem. Phys. 1998, 109, 42–56. (80) Barone, V.; Adamo, C.; Russo, N. Int. J. Quantum Chem. 1994, 52, 963–971. (81) Labanowski, J. K. Simplified Introduction to ab initio Basis Sets, Terms and Notation; Ohio Supercomputer Center: Columbus, OH, 2001; pp 1-13. (82) Casida, M. E.; Jamorski, C.; Casida, K. C.; Salahub, D. R. J. Chem. Phys. 1998, 108, 4439–4449. (83) Davidson, E. R.; Feller, D. Chem. ReV. 1986, 86, 681–696. (84) Marcinek, A.; Adamus, J.; Huben, K.; Gebicki, J.; Bartczak, T. J.; Bednarek, P.; Bally, T. J. Am. Chem. Soc. 2000, 122, 437–443. (85) Klamt, A.; Schu¨u¨rman, G. J. Chem. Soc., Perkin Trans. 2 1993, 799–805.

116

J. Phys. Chem. A, Vol. 114, No. 1, 2010

(86) Tomasi, J.; Perisco, M. Chem. ReV. 1994, 94, 2027–2094. (87) Cances, E.; Mennucci, B.; Tomasi, J. J. Chem. Phys. 1997, 107, 3032–3041. (88) Cramer, C. J.; Truhlar, D. G.Chem. ReV. 1999, 99, 21612200. (89) Foresman, J. B.; Head-Gordon, M.; Pople, J. A. J. Phys. Chem. 1991, 96, 135–149. (90) Hayon, E.; Ibata, T.; Lichtin, N. N.; Simic, M. J. Am. Chem. Soc. 1970, 92, 3898–3903. (91) Fossey, J.; Lefort, D.; Sorba, J. Free Radicals in Organic Chemistry; Wiley: Chichester, U.K., 1995.

Wisniowski et al. (92) Biddles, I.; Wiffen, J. T.; Hudson, A. Tetrahedron 1972, 28, 867– 874. (93) Gilbert, B. C.; Hodgeman, D. K. C.; Norman, R. O. C. J. Chem. Soc., Perkin Trans. 2 1973, 1748–1752. (94) Gilbert, B. C. Structure and reaction mechanisms in sulphur-radical chemistry revealed by ESR spectroscopy. In Sulfur-Centered ReactiVe Intermediates in Chemistry and Biology; Chatgilialoglu, C., Asmus, K.-D., Eds.; Plenum Press: New York, 1990; Vol. 197, pp 135. (95) Carruthers, W. Some Modern Methods of Organic Synthesis; Cambridge University Press: New York, 1986.

JP9071026